Intelligent management system and method for industrial solid wastewater quenching slag fibration production process

Abstract

The invention relates to the technical field of data processing, and discloses an intelligent management system and method for an industrial solid waste water quenched slag fibration production process, and the method comprises the steps: constructing an industrial solid waste water quenched slag crystal morphology mathematical model; constructing a crystal morphology multi-objective optimization model based on the mathematical model to obtain an optimal production process parameter set and an expected optimization objective; key process parameters are screened from the process parameters and divided into different importance levels, corresponding control threshold ranges and strengths are set, and a hierarchical control matrix is established; actually-measured crystal morphology parameters are obtained, and water-quenched slag production process parameters are dynamically optimized and controlled based on the actually-measured parameters and the grading control matrix; according to the method, the crystal morphology of the water-quenched slag can be accurately regulated and controlled, the production cost is reduced while the product quality is guaranteed, and high-value utilization of industrial solid waste resources is promoted.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and more specifically, to an intelligent management system and method for an industrial solid wastewater quenching slag fiberization production process. Background Art

[0002] The Chinese patent with the authorization announcement number CN118011981B proposes a production quality optimization method and system for coal mine support, which mainly controls and optimizes the production process of anchor steel in real time based on the grain density threshold, the grain size uniformity coefficient threshold, and the grain distribution uniformity coefficient threshold. This method optimizes the production quality of anchor steel by dynamically adjusting the production control state, thereby improving the quality reliability and optimization rate of the production process. However, although this method can optimize the production quality, it mainly focuses on the quality control of materials in the production process, and optimizes specific problems in different links of the production process, but lacks in-depth analysis of the correlation between various process parameters in the production process, and fails to effectively integrate the crystal morphology model with real-time data feedback in the production process. Therefore, this technology cannot comprehensively solve the multi-objective optimization problem in complex production processes, especially when faced with the interaction of multiple process parameters, it is difficult to achieve optimal production efficiency.

[0003] The Chinese patent application with publication number CN103433270A provides a comprehensive recycling and utilization method for solid waste water-quenched slag, which aims to reduce environmental pollution and improve the recovery rate of resources by efficiently recycling and utilizing the water-quenched slag. Through the combination of an electromembrane separation concentrator and a crystallization pool, the method successfully converts waste into usable products, meeting the requirements of clean production. However, although the invention has significant environmental advantages in the waste recycling process, its focus is on the physical and chemical treatment process of solid waste, mainly classifying and treating waste through mechanical and chemical methods, and fails to deeply explore how to intelligently manage and dynamically optimize multiple process parameters involved in the production process. Therefore, although this method has improved recycling efficiency, how to more finely adjust the process parameters during the processing process to improve the intelligence and accuracy of the production process is still the shortcoming of this technology.

[0004] In summary, the existing technology mainly adjusts a single optimization target, lacks comprehensive optimization of the complex relationship between multiple process parameters, and it is difficult to achieve the global optimal production plan; in terms of coordination and control between multiple key process parameters, there is a lack of a systematic solution for dynamic optimization. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides an intelligent management system and method for the industrial solid wastewater quenching slag fiberization production process.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] Intelligent management method for industrial solid wastewater quenching slag fiberization production process, including:

[0008] Construct a mathematical model of the crystal morphology of industrial solid waste quenching slag. Based on this mathematical model, construct a multi-objective optimization model of the crystal morphology to obtain the optimal production process parameter set and expected optimization target.

[0009] Based on the mathematical model of the crystal morphology of industrial solid waste quenching slag, key process parameters are screened out from the process parameters; the screened key process parameters are divided into different importance levels; for key process parameters of different importance levels, the corresponding control threshold range and control intensity are set according to the optimal production process parameter set, and a hierarchical control matrix is ​​established;

[0010] The measured crystal morphology parameters of the water-quenched slag production process are obtained, and based on the measured crystal morphology parameters and the hierarchical control matrix, the water-quenched slag production process parameters are dynamically optimized and controlled.

[0011] Furthermore, the construction of a mathematical model of the crystal morphology of industrial solid waste quenching slag includes:

[0012] Collecting historical production data of industrial solid waste quenching slag, extracting a first record set, wherein the first record set includes a grain size time series, a distribution uniformity time series, a temperature time series, a cooling rate time series, and an additive ratio time series;

[0013] According to the first record set, a nonlinear correlation model between grain size and distribution uniformity and process parameters is established to obtain a mathematical model of the crystal morphology of industrial solid waste water quenched slag; the process parameters include temperature, cooling rate and additive ratio.

[0014] Furthermore, the establishment of a nonlinear correlation model between grain size and distribution uniformity and process parameters to obtain a mathematical model of the crystal morphology of industrial solid waste quenching slag includes:

[0015] According to the first record set, a nonlinear correlation model between grain size and process parameters is established, which is marked as the initial grain size model; the initial grain size model includes n1 grain size influencing parameters ,in, is the constant term of the initial model of grain size, for Moment The influence parameters of the process parameters on the grain size at time t, 0≤i≤n1-1, 1≤ ≤n1-1;

[0016] According to the first record set, a nonlinear correlation model between distribution uniformity and process parameters is established, which is marked as the distribution uniformity initial model; the distribution uniformity initial model includes n2 distribution uniformity influencing parameters ,in, is the constant term of the initial model for distribution uniformity, for Moment The influence parameters of the process parameters on the distribution uniformity at time t, 0≤j≤n2-1, 1≤ ≤n2-1;

[0017] The parameters of the initial model of grain size and the initial model of distribution uniformity are solved to obtain and The optimal value of

[0018] Will Substitute the optimal value of into the initial grain size model to obtain the final grain size model;

[0019] Will Substitute the optimal value of into the initial model of distribution uniformity to obtain the final model of distribution uniformity;

[0020] The final model of grain size and the final model of distribution uniformity constitute the mathematical model of the crystal morphology of industrial solid waste quenching slag.

[0021] Furthermore, obtaining the optimal production process parameter set and the expected optimization target includes:

[0022] Based on the mathematical model of crystal morphology of industrial solid waste quenching slag, a multi-objective optimization model of crystal morphology was established with external parameters as decision variables and minimization of grain size and maximization of distribution uniformity as optimization goals.

[0023] Solve the multi-objective optimization model of crystal morphology to obtain the optimal production process parameter set and the expected optimization target, wherein the expected optimization target includes the target value of grain size and the target value of distribution uniformity; the optimal production process parameter set includes the historical optimal values ​​of the production process parameters.

[0024] Furthermore, the step of screening out key process parameters from the process parameters includes:

[0025] Based on the final model of grain size and the final model of distribution uniformity, the partial derivatives of the process parameters on grain size and distribution uniformity are calculated respectively, and the influence coefficient matrix of each process parameter is obtained;

[0026] The influence coefficient matrix is ​​normalized to obtain the influence weights of process parameters on grain size and distribution uniformity;

[0027] The process parameters are sorted in descending order according to their influence weights, and the process parameters whose cumulative weights reach the preset cumulative weight threshold θ1 are selected as key process parameters.

[0028] Furthermore, the screening out key process parameters is divided into different importance levels including:

[0029] Sort the key process parameters in descending order according to their impact weights, and define the key process parameters with the impact weights at the top W1 as the first-level key parameters;

[0030] The key process parameters with influence weights between W1 and W2 are defined as secondary key parameters;

[0031] The key process parameters with influence weights between W2 and W3 are defined as three-level key parameters, where W3>W2>W1, and W1, W2, and W3 are percentages.

[0032] Furthermore, the establishment of a hierarchical control matrix includes:

[0033] Obtain the historical optimal value of each key process parameter from the optimal production process parameter set, and use the historical optimal value of each key process parameter as the standard value of the corresponding key process parameter; set the control threshold range of the first-level key parameter to the first-level key parameter standard value ±M1; the control threshold range of the second-level key parameter to the second-level key parameter standard value ±M2; the control threshold range of the third-level key parameter is the third-level key parameter standard value ±M3; where M3>M2>M1, M1, M2, and M3 are percentages;

[0034] Set the control intensity of the first-level key parameters to level one, the control intensity of the second-level key parameters to level two, and the control intensity of the third-level key parameters to level three;

[0035] A hierarchical control matrix is ​​generated with the importance levels of key process parameters as rows and the control threshold ranges and control intensities as columns.

[0036] Further, the measured crystal morphology parameters include grain size parameters and crystal distribution uniformity parameters;

[0037] The dynamic optimization control of water-quenched slag production process parameters includes:

[0038] Input the grain size parameters and crystal distribution uniformity parameters into the multi-objective optimization model of crystal morphology to obtain the optimal production process parameter values ​​under the current production state;

[0039] Compare the deviations between the measured crystal morphology parameters and the expected optimization targets, dynamically adjust the hierarchical control matrix, and obtain the adjusted hierarchical control matrix;

[0040] Combined with the adjusted hierarchical control matrix, the optimal production process parameter values ​​under the current production status are parsed into actionable process parameter adjustment instructions.

[0041] Furthermore, the adjusted hierarchical control matrix includes:

[0042] Calculate the deviation ΔG between the grain size parameter and the target grain size value, and calculate the deviation ΔU between the crystal distribution uniformity parameter and the target distribution uniformity value;

[0043] If |ΔG|≤ΔG θ And |ΔU|≤ΔU θ , the existing hierarchical control matrix settings remain unchanged; ΔG θ is the preset grain size deviation threshold, ΔU θ is a preset distribution uniformity deviation threshold;

[0044] If |ΔG|>ΔG θ Or |ΔU|>ΔU θ , then dynamically adjust the hierarchical control matrix to obtain the adjusted hierarchical control matrix.

[0045] An intelligent management system for the production process of industrial solid wastewater quenching slag fiberization, which is used to implement the above-mentioned intelligent management method for the production process of industrial solid wastewater quenching slag fiberization, the system includes:

[0046] Model building module: used to build a mathematical model of the crystal morphology of industrial solid waste quenching slag. Based on the mathematical model of the crystal morphology of industrial solid waste quenching slag, a multi-objective optimization model of crystal morphology is built to obtain the optimal production process parameter set and the expected optimization target;

[0047] Hierarchical control module: Based on the mathematical model of the crystal morphology of industrial solid waste quenching slag, key process parameters are screened out from the process parameters; the screened key process parameters are divided into different importance levels; for key process parameters of different importance levels, the corresponding control threshold range and control intensity are set according to the optimal production process parameter set, and a hierarchical control matrix is ​​established;

[0048] Dynamic optimization module: obtains the measured crystal morphology parameters of the water-quenched slag production process, and dynamically optimizes and controls the water-quenched slag production process parameters based on the measured crystal morphology parameters and the hierarchical control matrix.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] The present invention achieves a quantitative description of the crystal growth process by constructing a mathematical model of the crystal morphology of industrial solid waste water quenching slag, reveals the key process parameters affecting the crystal morphology and their action mechanism, and provides a theoretical basis for process optimization.

[0051] Based on the mathematical model, a multi-objective optimization model for crystal morphology is constructed, which can determine the optimal combination of process parameters while ensuring the smallest grain size and the most uniform distribution. Compared with traditional empirical production, the model-driven production method improves the scientificity and controllability of the production process.

[0052] Select key process parameters from the process parameters and divide them into important levels, set the control threshold range and intensity in a targeted manner, and establish a hierarchical control matrix. This hierarchical control strategy can achieve refined management of key process parameters, concentrate resources for key control, and reduce management costs while ensuring product quality.

[0053] By acquiring crystal morphology parameters in real time and inputting them into the optimization model, the water-quenched slag production process parameters can be dynamically optimized and controlled. This closed-loop control method can timely detect and correct deviations in the production process, ensure the stability of crystal morphology, and improve the consistency of product quality.

[0054] The intelligent management system of the present invention realizes the automation and intelligence of the production process, reduces human intervention, and improves production efficiency. Through precise control of process parameters, while ensuring the performance of water-quenched slag, it reduces energy consumption and waste of raw materials, and has significant economic and environmental benefits.

[0055] The present invention provides a new idea and method for the resource utilization of industrial solid waste. By regulating the microstructure of water-quenched slag, its application field is broadened, the high-value utilization of industrial solid waste is achieved, and the coordinated development of industrial production and environmental protection is promoted, which has broad application prospects.

[0056] In summary, the intelligent management system and method for the industrial solid waste water-quenched slag fiberization production process of the present invention can achieve precise control of the crystal morphology of the water-quenched slag, improve the automation and intelligence level of the production process, reduce production costs while ensuring product quality, and provide strong technical support for the high-value utilization of industrial solid waste resources, thereby promoting the greening and sustainable development of industrial production. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0058] Figure 1 It is a principle flow chart of the intelligent management method of the industrial solid wastewater quenching slag fiberization production process in the present invention;

[0059] Figure 2A flow chart of a method for screening key process parameters in the intelligent management method for the industrial solid wastewater quenching slag fiberization production process of the present invention;

[0060] Figure 3 A flow chart of a method for classifying the screened key process parameters into different importance levels in the intelligent management method for the industrial solid wastewater quenching slag fiberization production process of the present invention;

[0061] Figure 4 It is a flow chart of a method for obtaining measured crystal morphology parameters of a water-quenched slag production process in the intelligent management method for a fiberization production process of industrial solid waste water-quenched slag of the present invention;

[0062] Figure 5 It is a flow chart of a method for dynamically optimizing and controlling the production process parameters of water-quenched slag in the intelligent management method for the fiberization production process of industrial solid waste water-quenched slag of the present invention;

[0063] Figure 6 It is a method flow chart of obtaining the adjusted hierarchical control matrix in the intelligent management method for the industrial solid wastewater quenching slag fiberization production process of the present invention;

[0064] Figure 7 It is a flow chart of a method for parsing the optimal production process parameter value under the current production state into an operable process parameter adjustment instruction in the intelligent management method of the industrial solid waste water quenching slag fiberization production process of the present invention;

[0065] Figure 8 It is a functional module diagram of the intelligent management system for the industrial solid wastewater quenching slag fiberization production process in the present invention. DETAILED DESCRIPTION

[0066] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] Example 1

[0068] See also Figure 1 As shown, this embodiment provides an intelligent management method for the industrial solid wastewater quenching slag fiberization production process, including:

[0069] Step S1000, constructing a mathematical model of the crystal morphology of industrial solid waste quenching slag, and based on the mathematical model of the crystal morphology of industrial solid waste quenching slag, constructing a multi-objective optimization model of crystal morphology to obtain an optimal production process parameter set and an expected optimization target;

[0070] Furthermore, step S1000 includes:

[0071] Step S1100, constructing a mathematical model of the crystal morphology of industrial solid waste quenching slag;

[0072] Furthermore, step S1100 includes:

[0073] Step S1110, collecting historical production data of industrial solid waste slag quenching, extracting a first record set, wherein the first record set at least includes a grain size time series, a distribution uniformity time series, a temperature time series, a cooling rate time series, and an additive ratio time series;

[0074] Specifically, industrial solid waste water-quenched slag is a solid granular material formed by a rapid cooling process (usually cooling with water) of high-temperature industrial solid waste (such as blast furnace slag or steel slag). Its crystal structure characteristics (such as grain size and distribution uniformity) are affected by factors such as cooling rate, temperature change and the use of additives. Therefore, analyzing the historical production data of water-quenched slag is helpful to reveal the dynamic evolution of crystal morphology.

[0075] Temperature time series refers to the change of temperature of solid waste materials over time during the water-quenched slag production process. The temperature here mainly refers to the surface temperature of the solid waste water-quenched slag, because it directly reflects the heat conduction and heat exchange rate during the cooling process of the solid waste water-quenched slag. For example, in the water-quenched slag production process, the initial temperature of the solid waste water-quenched slag may be as high as 1500°C, and it will quickly drop to below 100°C after water cooling. The temperature sensor can record the temperature change on the surface of the solid waste water-quenched slag at a fixed time interval (such as every second) to form temperature time series data.

[0076] The cooling rate refers to the temperature change rate of the solid waste quenching slag per unit time, which is usually obtained by dividing the temperature change between two moments by the time interval. The cooling rate time series is a set of dynamic data formed by arranging the calculation results of the cooling rate in chronological order. For example, assuming that at a certain measurement point, the temperature of the solid waste drops from 1200℃ to 1000℃ in 1 second, the cooling rate is (1200-1000) / 1=200℃ / s. The cooling rate time series data is obtained by continuously recording the temperature change per second.

[0077] Additive ratio time series refers to the change of the additive ratio of additives (such as stabilizers or binders) over time in the production process. By recording the amount of additives added and the corresponding time, the additive ratio time series can be formed.

[0078] Grain size refers to the size of the crystal particles in the water-quenched slag, which is usually characterized by the average particle diameter in micrometers (μm). Grain size reflects the microstructural characteristics of the water-quenched slag and directly affects its physical properties (such as strength and durability). Distribution uniformity refers to whether the distribution of grain size in space is uniform, which is usually characterized by standard deviation or coefficient of variation. The higher the distribution uniformity, the smaller the difference in grain size and the more stable the performance of the material. For example, the image of the water-quenched slag is collected through a microscope, the diameter of the particles is measured, and the grain size data is obtained. If the grain size of a batch is 10μm and the standard deviation is 2μm, it means that the distribution of the particles is relatively uniform.

[0079] Temperature and cooling rate: The temperature data of solid waste water quenching slag is recorded in real time at key points of the production line by temperature sensors or thermocouples, and the cooling rate time series is calculated. The grain size and distribution uniformity are quantitatively measured by online microscopic imaging and particle size analyzer to measure the size and distribution characteristics of crystal particles. The additive ratio is recorded by online metering equipment to record the amount and time of additive addition.

[0080] Data with time series characteristics can reveal the cause and effect relationship of crystal morphology evolution (such as the effect of temperature change on grain size). Cooling rate and temperature are key driving factors for rapid crystal growth or solidification. Collecting these data can provide a basis for optimizing production process parameters. Distribution uniformity and grain size directly affect the final performance of the material. Data recording helps to build an accurate prediction model. By collecting these data and generating time series characteristics, the dynamic evolution of crystal morphology during the water-quenched slag production process can be fully reflected, providing high-quality data support for the construction of subsequent mathematical models.

[0081] Step S1120, performing data preprocessing on the time series data in the first record set;

[0082] Specifically, data preprocessing is to eliminate noise and outliers in the collected data and unify data of different dimensions to the same scale for subsequent analysis and modeling. Data preprocessing includes data cleaning, data alignment, and data normalization.

[0083] Data cleaning refers to deleting or correcting erroneous, incomplete or inconsistent data. For example, if a temperature sensor fails at a certain point in time and records an abnormal value (such as -100°C), the data needs to be corrected by interpolation or deleting the abnormal value. Data alignment refers to synchronizing time series data from multiple data sources according to the same time base. For example, cooling rate data and additive ratio data may be collected at different time intervals. Data alignment requires interpolation or resampling so that they have consistent records at the same time point. Data normalization is to eliminate the differences between variables of different dimensions and scale the data to a uniform range (such as 0 to 1). For example, if the temperature range is 100°C to 1500°C, through normalization, the data will be adjusted to the range of 0 to 1, so that all variables are on the same scale for easy analysis.

[0084] Data cleaning can eliminate outliers and noise, and avoid prediction bias caused by data problems during model training. Data alignment ensures the synchronization between multiple variables, allowing time series analysis to accurately reveal the dynamic relationship between variables. Data normalization eliminates the order of magnitude difference between variables of different dimensions, improving the convergence and computational efficiency of the model. Through data preprocessing, the data quality can be significantly improved, providing reliable input data for the construction of subsequent mathematical models.

[0085] Step S1130, based on the first record set, a nonlinear correlation model between grain size and distribution uniformity and process parameters is established to obtain a mathematical model of the crystal morphology of industrial solid waste water quenched slag; the process parameters include temperature, cooling rate and additive ratio.

[0086] Further, step S1130 includes:

[0087] Step S1131: Based on the first record set, a nonlinear correlation model between grain size and process parameters is established, which is marked as the initial grain size model; the initial grain size model includes n1 grain size influencing parameters ,in, is the constant term of the initial model of grain size, for Moment The influence parameters of the process parameters on the grain size at time t, 0≤i≤n1-1, 1≤ ≤n1-1;

[0088] The initial grain size model includes:

[0089]

[0090] in:

[0091] :express Grain size at the moment;

[0092] :express Grain size at the moment;

[0093] :express Uniformity of distribution of moments;

[0094] :express The temperature of the moment;

[0095] :express Cooling rate at the time;

[0096] :express The additive ratio at the time;

[0097] : is the constant term of the initial model of grain size; It is a reference value that provides a starting point or baseline for the initial model of grain size;

[0098] :for The parameters affecting the grain size at time a on the grain size at time t;

[0099] :for The parameters affecting the uniformity of the moment distribution on the grain size at time t;

[0100] :for The parameters affecting the grain size at time t by the temperature at that moment;

[0101] :for The parameters affecting the cooling rate at time t on the grain size at time t;

[0102] :for The parameters affecting the additive ratio at time t on the grain size at time t;

[0103] : is the error term of the initial model of grain size, which indicates the deviation of grain size caused by other factors not considered or the incompleteness of the model. These deviations may come from experimental errors, external interference factors, unknown process parameters, etc.

[0104] Step S1132: Based on the first record set, a nonlinear correlation model between distribution uniformity and process parameters is established, which is marked as the distribution uniformity initial model; the distribution uniformity initial model includes n2 distribution uniformity influencing parameters ,in, is the constant term of the initial model for distribution uniformity, for Moment The influence parameters of the process parameters on the distribution uniformity at time t, 0≤j≤n2-1, 1≤ ≤n2-1;

[0105] The distribution uniformity initial model includes:

[0106]

[0107] in:

[0108] :express Uniformity of distribution of moments;

[0109] : It is the constant term of the initial model of distribution uniformity, providing a benchmark level of distribution uniformity. When all other influencing factors (such as grain size, temperature, etc.) remain unchanged, will be equal to ;

[0110] :for The parameters affecting the grain size at time t on the distribution uniformity at time t;

[0111] :for The parameters affecting the uniformity of distribution at time t;

[0112] :for The influence parameter of the temperature at the moment on the distribution uniformity at the moment t;

[0113] :for The influence parameters of cooling rate at time t on the distribution uniformity at time t;

[0114] :for The parameters affecting the additive ratio at time t on the distribution uniformity at time t;

[0115] : The error term of the initial model for distribution uniformity, which represents the deviation of distribution uniformity due to unconsidered factors or incompleteness of the model.

[0116] Step S1133, solving the parameters of the initial model of grain size and the initial model of distribution uniformity to obtain and The optimal value of

[0117] Step S1134: Substitute the optimal value of into the initial grain size model to obtain the final grain size model;

[0118] Step S1135: Substitute the optimal value of into the initial model of distribution uniformity to obtain the final model of distribution uniformity;

[0119] Step S1136, the final model of grain size and the final model of distribution uniformity constitute a mathematical model of the crystal morphology of industrial solid waste quenching slag.

[0120] The mathematical model of the crystal morphology of industrial solid waste quenching slag is:

[0121] in:

[0122] : grain size;

[0123] : Distribution uniformity;

[0124] :temperature;

[0125] : cooling rate;

[0126] : Additive ratio;

[0127] : Nonlinear function between grain size and temperature, cooling rate and additive ratio;

[0128] : A nonlinear function of distribution uniformity and temperature, cooling rate, and additive ratio.

[0129] Specifically, the grain size is affected by the combined effects of process parameters such as temperature, cooling rate and additive ratio. Due to the complexity of the crystal growth process, there is a nonlinear relationship between the grain size and the various process parameters. Therefore, it is necessary to establish a nonlinear correlation model to describe the quantitative relationship between them. This model takes into account that the grain size at the current moment is not only related to its own state at the previous moment, but also affected by the various process parameters at the previous moment, reflecting the dynamic characteristics of grain evolution. By fitting historical production data, the coefficients of each item in the model can be determined to obtain the initial model of grain size. The establishment of this model lays the foundation for the subsequent multi-objective optimization of crystal morphology.

[0130] The higher the distribution uniformity, the smaller the difference in grain size and the better the isotropy of the material. Similar to grain size, distribution uniformity is also affected nonlinearly by temperature, cooling rate and additive ratio. Through data-driven modeling methods, the quantitative relationship between distribution uniformity and various process parameters can be characterized to obtain an initial model of distribution uniformity. This model also takes into account the dependence of the current distribution uniformity on its own state and various process parameters at the previous moment, reflecting the dynamic evolution of crystal growth uniformity. The initial model of distribution uniformity and the initial model of grain size together constitute the core components of the mathematical model of crystal morphology.

[0131] The initial model of grain size and distribution uniformity reflects that the grain size and distribution uniformity at time t are not only related to their own state at time t-1, but also affected by the temperature, cooling rate and additive ratio at time t-1, reflecting the dynamic characteristics of crystal morphology evolution and multi-parameter coupling effect. The initial model of grain size includes n1 grain size influencing parameters, is the constant term of the initial model of grain size, which represents the grain size when there is no influence of process parameters. The initial model of distribution uniformity includes n2 distribution uniformity influencing parameters , is the constant term of the initial model of distribution uniformity, representing the distribution uniformity when there is no influence of process parameters; preferably, n1=6, n2=6. By establishing these two initial models, the nonlinear correlation between the morphological characteristic parameters and the process parameters during the crystal growth process can be quantitatively described, revealing the key factors affecting the crystal quality. The autoregressive terms in the model (such as G(t-1) and U(t-1)) reflect the continuity and historical dependence of crystal growth, that is, the current state is the continuation and development of the previous state. Exogenous variable terms (such as T(t-1), C(t-1) and A(t-1)) characterize the regulatory effect of process parameters on crystal morphology, reflecting the controllability of the production process. Comprehensively considering the autoregressive terms and exogenous variable terms, the dynamic evolution law of crystal morphology can be fully characterized, providing a theoretical basis and decision-making basis for the optimization control of the crystal growth process.

[0132] Parameter solution refers to the use of mathematical optimization methods to determine the values ​​of each coefficient in the model so that the error between the model's predicted value and the measured value is minimized. Common parameter estimation methods include least squares method, maximum likelihood estimation, etc. For the initial grain size model, it is necessary to solve The optimal values ​​of n1 influencing parameters are as follows; for the initial model of distribution uniformity, it is necessary to solve β j=The optimal values ​​of n2 influencing parameters. Since there are cross terms in the two initial models (such as the grain size is included in the initial model of distribution uniformity), it is necessary to jointly solve the parameters of the two models to ensure the consistency and optimality of the results. This can be achieved by constructing a multi-objective optimization problem, minimizing the sum of square errors of the two models at the same time, and using numerical optimization algorithms (such as gradient descent method, genetic algorithm, etc.) to obtain and The optimal solution of the crystal morphology evolution law is obtained by quantitatively describing the evolution law of the crystal morphology. The parameters of the grain size and distribution uniformity models are jointly solved to fully consider the interaction between the two morphology indicators and obtain the globally optimal parameter estimation results. This method avoids the problems of local optimality and inconsistent results that may occur when solving each model parameter separately, and improves the fitting accuracy and prediction ability of the model. Through joint solution, the effects of process parameters such as temperature, cooling rate, and additive ratio on grain size and distribution uniformity can be quantitatively evaluated.

[0133] After obtaining the optimal estimated values ​​of each influencing parameter, substitute them into the initial model of grain size and the initial model of distribution uniformity to obtain the final model of grain size and the final model of distribution uniformity. These two models can quantitatively reflect the nonlinear relationship between grain size and distribution uniformity and temperature, cooling rate, and additive ratio during crystal growth. Using the final model, the evolution of crystal morphology can be predicted and controlled, production process parameters can be optimized, and product quality and production efficiency can be improved.

[0134] The final model of grain size and the final model of distribution uniformity are combined to form a complete mathematical model of the crystal morphology of industrial solid waste quenching slag, which can comprehensively describe the dynamic evolution of the crystal morphology of industrial solid waste quenching slag. Based on this model, the optimization and control of the crystal growth process can be carried out, and process parameters such as temperature, cooling rate and additive ratio can be adjusted to minimize the grain size and make the distribution most uniform, so as to achieve ideal product performance. At the same time, the model also provides a theoretical basis for the setting of process parameters, avoiding empirical blind regulation and improving production efficiency and product quality consistency.

[0135] The establishment of a mathematical model for the crystal morphology of industrial solid waste slag quenching provides a powerful tool for the intelligent optimization of the crystal growth process. Through the mining and analysis of historical production data, the key process parameters and their action rules that affect the crystal morphology are revealed, and the dynamic evolution characteristics of grain size and distribution uniformity are quantitatively characterized. Based on the mathematical model, the crystal morphology under different process conditions can be predicted, and the optimal combination of process parameters can be determined through a multi-objective optimization method to achieve intelligent regulation of the crystal growth process. Compared with traditional empirical production, the model-driven production method can give full play to the value of data, improve the scientificity and controllability of the production process, and effectively improve the quality and performance of industrial solid waste quenching slag materials.

[0136] Step S1200, based on the mathematical model of the crystal morphology of industrial solid waste quenching slag, a multi-objective optimization model of crystal morphology is constructed to obtain the optimal production process parameter set and the expected optimization target.

[0137] Further, step S1200 includes:

[0138] Step S1210, based on the mathematical model of crystal morphology of industrial solid waste quenching slag, with external parameters as decision variables, the external parameters at least including temperature T, cooling rate C and additive ratio A, with minimizing grain size and maximizing distribution uniformity as optimization goals, a multi-objective optimization model of crystal morphology is established;

[0139] The multi-objective optimization model of crystal morphology includes:

[0140]

[0141] The constraints are:

[0142]

[0143] in:

[0144] : minimum temperature;

[0145] : Maximum value of temperature;

[0146] : Minimum cooling rate;

[0147] : Maximum value of cooling rate;

[0148] : Minimum value of additive ratio;

[0149] : Maximum value of additive ratio;

[0150] : The first objective of the multi-objective optimization model for crystal morphology;

[0151] : The second objective of the multi-objective optimization model of crystal morphology;

[0152] Step S1220, solving the multi-objective optimization model of crystal morphology to obtain an optimal production process parameter set and an expected optimization target, wherein the expected optimization target includes a target value for grain size and a target value for distribution uniformity; the optimal production process parameter set includes historical optimal values ​​of the production process parameters.

[0153] Specifically, the grain size G directly affects the strength and durability of the material. The smaller the grain size, the better the material performance. One of the optimization goals is to minimize the grain size. The distribution uniformity U reflects the stability of the grain size distribution. The higher the distribution uniformity, the more consistent the physical properties of the material. The second optimization goal is to maximize the distribution uniformity. Temperature T refers to the surface temperature of the solid waste water quenching slag, in °C, which is a key parameter affecting the grain size and distribution uniformity during crystal growth. The cooling rate C refers to the temperature change rate of the solid waste water quenching slag per unit time, in °C / s, which affects the crystal solidification rate and microstructure. The additive ratio A refers to the addition ratio of the additive in the production process, in percentage (%), which is used to adjust the crystal morphology. These decision variables are the core of the optimization, and by adjusting them, the optimization of grain size and distribution uniformity can be achieved.

[0154] The constraints are set based on the physical limitations and operating range of the actual process. For example, the temperature cannot be too low to affect crystal growth, the cooling rate cannot be too high to avoid crystal damage, and the additive ratio must be controlled within a reasonable range to ensure production costs and material properties.

[0155] The multi-objective optimization problem is solved by NSGA-II (Non-dominated Sorting Genetic Algorithm II), which is a classic multi-objective optimization algorithm with efficient search capabilities and good global convergence. NSGA-II can quickly search for the Pareto optimal solution set; ensure the diversity of solutions and avoid falling into local optimality; and is suitable for high-dimensional nonlinear optimization problems. Output the optimal production process parameter set, including the optimal temperature, optimal cooling rate, and optimal additive ratio. The optimal temperature value ensures the minimization of grain size; the optimal cooling rate takes into account both grain size and distribution uniformity; the optimal additive ratio improves distribution uniformity.

[0156] The expected optimization targets include the target value of grain size and the target value of distribution uniformity, that is, the crystal morphology performance under the optimal combination of process parameters.

[0157] There is a complex nonlinear relationship between grain size and distribution uniformity. Directly adjusting process parameters through experiments is costly and inefficient. By combining the optimal process parameters, the grain size can be significantly reduced, the crystal distribution uniformity can be improved, and the strength, toughness and stability of the material can be improved. The optimization model can be used to systematically analyze the influence of process parameters, and finding the optimal solution to balance various objectives can accurately adjust process parameters, avoid excessive energy consumption and waste of additives, and thus reduce production costs. The NSGA-II algorithm provides powerful search capabilities, which transforms production process optimization from empirical to data-driven, improves the scientificity and reliability of decision-making, can quickly find the global optimal solution, reduce trial and error time, and realize intelligent and efficient production processes. By constructing and solving a multi-objective optimization model for crystal morphology, intelligent optimization of the production process can be achieved, providing an efficient and accurate solution for the resource utilization of industrial solid waste quenching slag.

[0158] Step S2000, based on the mathematical model of the crystal morphology of industrial solid waste quenching slag, select key process parameters from the process parameters; divide the selected key process parameters into different importance levels; for the key process parameters of different importance levels, according to the optimal production process parameter set, set the corresponding control threshold range and control intensity, and establish a hierarchical control matrix;

[0159] Furthermore, step S2000 includes:

[0160] Step S2100, based on the mathematical model of the crystal morphology of industrial solid waste quenching slag, calculate the influence weight of the process parameters on the grain size and distribution uniformity, and screen out the key process parameters;

[0161] Furthermore, if Figure 2 As shown, step S2100 includes:

[0162] Step S2110, based on the final model of grain size and the final model of distribution uniformity, respectively calculate the partial derivatives of the process parameters on the grain size and the distribution uniformity, and obtain the influence coefficient matrix of each process parameter;

[0163] Step S2120, normalizing the influence coefficient matrix to obtain the influence weights of the process parameters on the grain size and distribution uniformity;

[0164] Step S2130 , sorting the process parameters in descending order according to their influence weights, and selecting the process parameters whose cumulative weights reach a preset cumulative weight threshold θ1 as key process parameters.

[0165] Specifically, the goal of step S2100 is to determine which process parameters have a significant impact on grain size and distribution uniformity through quantitative analysis, so as to provide a decision-making basis for the subsequent hierarchical control strategy. Process parameters may include cooling rate, temperature, additive ratio, etc. These parameters may affect the morphological characteristics of the crystal in actual production, such as grain size, distribution uniformity, etc. By constructing a mathematical model, the impact of these process parameters on grain size and distribution uniformity can be quantitatively analyzed, and the impact weight of each process parameter can be calculated.

[0166] The influence weight of a process parameter indicates the contribution of a process parameter to the crystal morphology (such as grain size and distribution uniformity). The larger the influence weight, the more important the control effect of the parameter on the crystal morphology. Grain size refers to the average size of the crystal in the material, while distribution uniformity describes whether the distribution of grains in the entire material is uniform. An ideal water-quenched slag crystal should have an appropriate grain size and a relatively uniform distribution, which helps to improve the mechanical properties and aging resistance of the material. θ1 (usually selected as 80%) is a cumulative weight threshold, which means that the process parameters with the top 80% weight are selected as key process parameters through sorting. This means that only those process parameters with the most significant impact will be selected as key process parameters to ensure the effectiveness of management and control.

[0167] By screening out the process parameters with significant impact, resources can be concentrated for precise regulation and control to avoid interference from irrelevant parameters. Such steps not only improve production efficiency, but also optimize material properties. For example, by analyzing the effect of cooling rate on grain size, if it is found that cooling rate has a significant effect on grain size (with a large influence), this parameter can be further precisely controlled to improve the quality of water-quenched slag.

[0168] Step S2200, classifying the screened key process parameters into different importance levels;

[0169] Furthermore, if Figure 3 As shown, step S2200 includes:

[0170] Step S2210, sorting the key process parameters in descending order according to the influence weights, and defining the key process parameters with the influence weights at the top W1 as the first-level key parameters;

[0171] Step S2220, defining the key process parameters with influence weights between W1 and W2 as secondary key parameters;

[0172] Step S2230, defining the key process parameters with influence weights between W2 and W3 as three-level key parameters, where W3>W2>W1, and W1, W2, and W3 are percentages.

[0173] Specifically, in step S2200, the key process parameters screened in step S2100 are classified according to their importance, divided into different levels (level one, level two, level three), and the specific management and control strategies for each level are determined. Through this classification, more scientific management and control plans can be formulated for process parameters of different importance.

[0174] Preferably, W1=40%, W2=70%, W3=100%. The first-level key parameters (the first 40% of the influence weight) have the most direct and significant impact on the crystal morphology, and usually need to be controlled first. For example, the cooling rate usually has an important influence on the size and distribution uniformity of the grains, so the cooling rate may be classified as a first-level key parameter. The first-level parameters usually require more precise control and real-time monitoring. The second-level key parameters (the influence weight is between 40% and 70%) have a relatively small impact on the crystal morphology, but still have a certain impact. For example, the proportion of additives may belong to this category. Although its impact is not as direct as the cooling rate, in certain cases, adjusting the proportion of additives may also have an optimization effect on the crystal morphology. The third-level key parameters (the influence weight is between 70% and 100%) have a relatively weak impact on the crystal morphology, but in most cases they still need to be monitored and controlled. For example, temperature may belong to this category. The effect of temperature on grain size is relatively indirect, so the requirements can be appropriately relaxed when controlling.

[0175] The influence weight of the process parameter is calculated by the partial derivative in step S2100. The larger the absolute value of the partial derivative, the more significant the influence of the process parameter on the crystal morphology. The sorted influence weights can help us reasonably classify these process parameters to ensure that key factors receive sufficient attention. In the actual production process, not all process parameters are equally important, and grading helps to set appropriate management strategies for each parameter. The first-level key parameters should be strictly controlled, while the third-level parameters can appropriately relax the control requirements. Reasonable grading standards are the basis for implementing effective control.

[0176] By dividing process parameters into different levels of importance, more precise control can be achieved and waste of resources can be avoided. For example, if it is found that the cooling rate has a much greater impact on grain size than the temperature, more resources can be invested in the control of cooling rate, while the control of temperature can be relatively relaxed. This hierarchical management not only improves process efficiency, but also optimizes production costs and quality control.

[0177] For example, it is assumed that it is found during the production process that the cooling rate has a significant effect on the grain size, so the cooling rate has a greater influence weight than other parameters. According to step S2200, the cooling rate is defined as a primary key parameter, which is required to be kept within a precise range during the production process. At the same time, the additive ratio has a smaller impact, so it is classified as a secondary or tertiary key parameter, and its control requirements are relatively loose.

[0178] Step S2300, for key process parameters of different importance levels, set corresponding control threshold ranges and control strengths according to the optimal production process parameter set, and establish a hierarchical control matrix.

[0179] Furthermore, step S2300 includes:

[0180] Step S2310, obtaining the historical optimal value of each key process parameter from the optimal production process parameter set, and taking the historical optimal value of each key process parameter as the standard value of the corresponding key process parameter; setting the control threshold range of the first-level key parameter to the first-level key parameter standard value ±M1; setting the control threshold range of the second-level key parameter to the second-level key parameter standard value ±M2; setting the control threshold range of the third-level key parameter to the third-level key parameter standard value ±M3; wherein M3>M2>M1, M1, M2, and M3 are percentages;

[0181] Step S2320, setting the control intensity of the first-level key parameters to level one, setting the control intensity of the second-level key parameters to level two, and setting the control intensity of the third-level key parameters to level three;

[0182] Step S2330, generating a hierarchical control matrix with the importance levels of key process parameters as rows and the control threshold ranges and control intensities as columns.

[0183] Specifically, the core goal of step S2300 is to implement differentiated control strategies for key process parameters according to different levels of importance, so as to ensure that each key parameter in the production process fluctuates within a reasonable range, thereby ensuring the stability of crystal morphology and consistency of product quality.

[0184] The optimal production process parameter set includes the historical optimal values ​​of the production process parameters, which are used as the standard values ​​of each key process parameter. Using the historical optimal values ​​as the standard values ​​of each key process parameter provides a clear benchmark to ensure that the subsequent production process can proceed towards the best effect. These historical optimal values ​​are obtained through long-term production data and optimization experience accumulation, and usually have high credibility and representativeness. By setting standard values, the production process will not become chaotic or unstable in the absence of goals. By using historical optimal values ​​as standard values, fluctuations in the production process can be effectively reduced and process stability can be improved. These optimal values ​​reflect the parameters that can achieve the best quality under given conditions, and can provide a stable reference in the production process to ensure that each production is as close to the optimal state as possible. The historical optimal values ​​are usually obtained through experiments and multiple adjustments and optimizations, and these values ​​have proven that they can effectively improve product quality and production efficiency. Using these historical optimal values ​​as new production standards can avoid re-conducting large-scale experiments or trial and error processes every time production is carried out, saving time and resources.

[0185] In step S2310, different control threshold ranges are set for key process parameters of each level according to the importance levels (level 1, level 2, level 3) divided in step S2200. These threshold ranges represent the fluctuation range allowed for process parameters during the production process, ensuring that they operate within the control range, thereby avoiding adverse effects on crystal morphology.

[0186] Since the primary key parameters have the most significant impact on the crystal morphology, the primary key parameters need to be strictly controlled. The control threshold range is set to ±M1 of the standard value, where M1 is usually small (for example, ±1%), ensuring that the parameter changes very little and maintaining the stability of the crystal morphology. For example, if the standard value of the cooling rate is 200℃ / s and M1 is set to 1%, its control range is 198℃ / s to 202℃ / s. The secondary key parameters are less influential than the primary key parameters, and the control threshold range is set to ±M2 of the standard value, where M2 is greater than M1 (for example, ±3%). This means that the parameter has a large fluctuation range within a certain range, but it still needs to be kept within a reasonable range to ensure good performance of the crystal morphology. For example, the standard value of the additive ratio is 5%, and M2 is set to 3%, then the control range is 4.85% to 5.15%. The third-level key parameters have a more indirect or smaller impact on the crystal morphology, and the control threshold range is set to ±M3 of the standard value, where M3 is greater than M2 (for example, ±5%). The fluctuation of such parameters has little impact on product quality, so the control range can be appropriately relaxed. For example, if the standard temperature value is 1000°C and M3 is set to 5%, the control range is 950°C to 1050°C.

[0187] By setting different levels of control threshold ranges, optimal resource allocation can be achieved. The primary key parameters have a significant impact on the crystal morphology and need to be strictly controlled to ensure the stability of product quality; while the secondary and tertiary key parameters can be appropriately relaxed under the premise of ensuring quality, so as to improve production efficiency and reduce control costs. In addition, hierarchical control helps to reduce energy waste and material consumption caused by excessive control, and improve the economic benefits of overall production.

[0188] The purpose of step S2320 is to set the control intensity of the key process parameters according to their importance level. The control intensity usually includes aspects such as monitoring frequency, control accuracy and emergency response measures to ensure that parameters at all levels are properly managed.

[0189] Since the primary key parameters have the most significant impact on the crystal morphology, the control intensity of the primary key parameters is set to "primary", which means that the highest level of monitoring and control is required. For example, as a primary key parameter, the cooling rate needs to be equipped with a high-precision sensor to monitor its changes in real time and fine-tune it through an automatic control system to ensure that it always remains within the set control range. In addition, a rapid response mechanism should be established to promptly handle any abnormal situation that exceeds the threshold to avoid adverse effects on the crystal morphology. The control intensity of the secondary key parameters is set to "secondary", which means that moderate monitoring and control are required. For example, as a secondary key parameter, the additive ratio needs to be checked regularly and controlled by automatic metering equipment to ensure that it is within the set threshold range. Although it does not require high-frequency monitoring like the primary parameters, it is still necessary to maintain a certain control accuracy and emergency response capability to cope with possible fluctuations in the production process. The control intensity of the third-level key parameters is set to "third level", indicating that the monitoring and control requirements for this parameter are relatively loose. For example, as a third-level key parameter, although temperature has an impact on the crystal morphology, its impact is relatively indirect, and regular monitoring and basic control measures can be adopted to ensure that it does not fluctuate significantly. The monitoring frequency of such parameters can be relatively low, and the control accuracy can be appropriately relaxed, thereby reducing production costs and control complexity.

[0190] By setting different levels of control intensity, we can achieve refined management of key process parameters, ensuring that key parameters receive adequate attention and control, while reducing unnecessary resource investment for non-critical parameters while ensuring quality. This hierarchical management strategy not only improves the stability and reliability of the production process, but also optimizes the allocation of management resources, reduces production costs, and improves overall production efficiency.

[0191] The purpose of step S2330 is to integrate the control threshold range and control strength determined in the first two sub-steps into a unified hierarchical control matrix for intuitive application and management in the production process. The specific operations are as follows:

[0192] Matrix structure: The rows of the matrix represent key process parameters of different importance levels (primary, secondary, and tertiary), and the columns include the control threshold range and control intensity.

[0193] Control threshold range: For each level of key process parameters, fill in the corresponding control threshold range. For example, the control range of the first-level key parameter cooling rate is ±1%, the control range of the second-level key parameter additive ratio is ±3%, and the control range of the third-level key parameter temperature is ±5%.

[0194] Control intensity: According to the level setting, the first-level key parameters are set as "first-level control", the second-level key parameters are set as "second-level control", and the third-level key parameters are set as "third-level control".

[0195] An example of a hierarchical control matrix is ​​shown in Table 1:

[0196] Table 1 List of hierarchical control matrix

[0197]

[0198] By establishing a hierarchical control matrix, key process parameters of different levels and their corresponding control strategies can be intuitively displayed in a unified framework, which is convenient for production managers to refer to and apply in actual operations. This matrix management method helps to standardize the production process, ensure that each process parameter operates according to the established control strategy, and reduce human operation errors and management confusion. In addition, the hierarchical control matrix is ​​also flexible and can be dynamically adjusted and optimized according to the feedback from production practice and the needs of process optimization, further improving the intelligence and efficiency of the production process.

[0199] For example, it is assumed that in a certain industrial solid wastewater quenching slag production process, the following key process parameters and their importance levels are determined through steps S2200 and S2100:

[0200] Cooling rate (first-level key parameter, influence weight 40%);

[0201] Additive ratio (secondary key parameter, influence weight 20%);

[0202] Temperature (third-level key parameter, influence weight 20%);

[0203] According to steps S2310 and S2320, the control threshold range and control intensity of each parameter are set:

[0204] Cooling rate:

[0205] Control threshold range: ±1%;

[0206] Control intensity: Level 1 control;

[0207] Additive ratio:

[0208] Control threshold range: ±3%;

[0209] Control intensity: Level 2 control;

[0210] temperature:

[0211] Control threshold range: ±5%;

[0212] Control intensity: three levels of control;

[0213] The final generated hierarchical control matrix is ​​shown in Table 2:

[0214] Table 2 List of the final generated hierarchical control matrix

[0215]

[0216] In the actual production process, if the cooling rate monitoring data shows that it deviates from the standard value of 200℃ / s by more than ±1%, the first-level control will be triggered immediately, and the automatic adjustment device will be started to adjust the cooling rate back to the range of 198℃ / s to 202℃ / s. At the same time, the abnormality is recorded and the cause (such as equipment failure or operating error) is analyzed to take preventive measures to prevent similar situations from happening again. For deviations in additive ratios and temperatures, corresponding adjustment measures are taken according to the second-level and third-level control intensity to ensure that they remain within the set control range.

[0217] Through the application of this hierarchical control matrix, the production process can achieve refined management of key process parameters, ensure the stability of crystal morphology and consistency of product quality, while optimizing resource allocation and improving production efficiency and economic benefits.

[0218] Step S3000, obtaining the measured crystal morphology parameters of the water-quenched slag production process, and dynamically optimizing and controlling the water-quenched slag production process parameters based on the measured crystal morphology parameters and the hierarchical control matrix.

[0219] Furthermore, step S3000 includes:

[0220] Step S3100, obtaining measured crystal morphology parameters of the water-quenched slag production process; the measured crystal morphology parameters include grain size parameters and crystal distribution uniformity parameters;

[0221] Furthermore, if Figure 4 As shown, step S3100 includes:

[0222] Step S3110, obtaining a microscopic image and an X-ray diffraction spectrum of crystals during the water-quenched slag production process;

[0223] Step S3120, performing digital image processing on the microscopic image to extract grain size parameters;

[0224] Step S3130, performing curve fitting on the X-ray diffraction pattern to obtain crystal distribution uniformity parameters.

[0225] The curve fitting of the X-ray diffraction pattern comprises:

[0226]

[0227] in:

[0228] : Diffraction intensity, expressed at different diffraction angles The diffraction intensity under the condition of X-ray diffraction is measured by X-ray diffraction instrument. It reflects the intensity of the diffraction phenomenon and can reveal the structural information inside the crystal, such as the size of the grains, the crystal orientation, etc.

[0229] : Correction coefficient, which affects the height of the diffraction peak and represents the intensity of the diffraction peak. The larger the value, the higher the diffraction intensity in that direction and the more uniform the distribution of the crystal in that direction.

[0230] : Full Width at Half Maximum (FWHM) correction coefficient, which controls the width of the diffraction peak. The larger it is, the wider the orientation distribution of the crystal is and the greater the non-uniformity of the crystal in this direction. Related to the distribution of grain size and orientation.

[0231] B: The amplitude coefficient of the Gaussian function, which controls the amplitude of the diffraction peak shape. The Gaussian function is usually used to describe a relatively uniform crystal structure. The larger B is, the more concentrated the crystal distribution in this direction is.

[0232] : The standard deviation of the Gaussian function determines the width of the diffraction peak. Smaller means better uniformity of grain size; Larger means wider grain size distribution and non-uniform crystal size.

[0233] : The quadratic term coefficient reflects the quadratic effect behind the diffraction peak. This term is added to take into account the non-ideal phenomenon of grain distribution, such as defects inside the crystal or material inhomogeneity.

[0234] : The center position of the diffraction peak, indicating the angle at which the maximum value of the diffraction peak appears. Represents the angle at which the main orientation or peak of the crystal lies.

[0235] Represents an exponential function with the natural constant e as the base.

[0236] (half-maximum width) and (Standard deviation): These two parameters can be obtained by curve fitting of the diffraction pattern, reflecting the grain size distribution and crystal orientation distribution. Controls the orientation distribution of the crystal, It reflects the distribution width of the grain size. Usually, these two parameters can be obtained by fitting the diffraction pattern obtained by the experiment with the least square method.

[0237] , B and : These parameters are usually obtained by fitting the diffraction pattern and reflect the amplitude and shape of the diffraction peak.

[0238] : By analyzing the diffraction pattern, the angular position of the maximum diffraction intensity can be determined, and the .

[0239] Diffraction Angle Determines the diffraction intensity With the change of angle The change in the intensity of the diffraction pattern It will experience rises and falls, depending on the width and position of the diffraction peak. Reflected in different angles The diffraction intensity under the condition of and ) decided The specific form of changes in the dependent variable There will be different trends: When it is larger, the diffraction peak becomes broader; when When the size is larger, the distribution of grain size will be broader, resulting in an increase in the width of the diffraction peak. (the width of the crystal orientation distribution) and (the width of the grain size distribution) increases, the width of the diffraction peak increases, and the range of variation of the diffraction intensity expands, indicating that the distribution of the crystals is becoming more and more uneven. and The diffraction peak becomes narrower, indicating that the uniformity of the crystal is improved and the size distribution of the grains is more concentrated.

[0240] and They are two key parameters that reflect the uniformity of crystal distribution. Reflects the uniformity of the orientation distribution of the crystal in different directions. The uniformity of grain size is reflected by fitting these parameters, which can quantitatively evaluate the uniformity of grains in the sample. The combination of Gaussian function and Lorentz function helps to simultaneously describe the grain size distribution (given by ) and the orientation distribution of the crystal (determined by This joint fitting can accurately capture the complex characteristics of crystal distribution and avoid the limitations of a single model. The crystal distribution uniformity parameters obtained by fitting can be used to feedback and adjust the process parameters (such as cooling rate, temperature, etc.) in the water-quenched slag production process. or The larger the value of , the more uneven the crystals are. The crystal morphology can be optimized by adjusting process parameters such as cooling rate or temperature.

[0241] Specifically, the goal of step S3100 is to achieve dynamic optimization and intelligent control of the production process by acquiring and analyzing the crystal morphology parameters in real time during the production process, thereby ensuring the quality and consistency of the final product. In step S3110, advanced detection equipment is used to acquire the microscopic image and X-ray diffraction spectrum of the crystal in the water-quenched slag production process in real time to comprehensively monitor the changes in the crystal morphology parameters. The online microscopic imager is arranged at the key process point of the production line and can continuously take microscopic images of the crystal. Through microscopic imaging, the size, shape and distribution of the grains can be observed, and the grain size time series data can be obtained. After processing, these image data can be used to quantitatively analyze the grain size and distribution uniformity. The X-ray diffractometer is used to obtain the X-ray diffraction spectrum of the crystal. X-ray diffraction technology can reflect the internal structure and crystal plane orientation of the crystal. By analyzing the diffraction spectrum, the growth orientation and uniformity of the crystal can be evaluated. After curve fitting processing, the data of these diffraction spectra can quantitatively describe the uniformity parameters of the crystal distribution. By acquiring microscopic images and X-ray diffraction spectra in real time, the dynamic changes of the crystal morphology during the production process can be fully and accurately monitored. This multi-dimensional monitoring method not only improves the comprehensiveness and accuracy of the data, but also can promptly detect abnormal situations that may occur during the production process, ensuring that the crystal morphology parameters always remain within the predetermined control range, thereby improving product quality and production stability.

[0242] Step S3120 involves performing digital image processing on the acquired microscopic image to extract grain size parameters. This process generally includes the following steps:

[0243] Image segmentation: Use image processing algorithms (such as threshold segmentation, edge detection, region growing, etc.) to separate the grain area from the background in the microscopic image to generate a binary image. The purpose of image segmentation is to distinguish the grain part of interest from the irrelevant background part for subsequent analysis.

[0244] Feature extraction: Extract the geometric features of the grains from the segmented image, such as the area, perimeter, diameter, etc. Commonly used methods include Hough transform and morphological processing. These geometric features are the basis for quantitative analysis of grain size. Morphological processing is the use of mathematical morphological operations (such as dilation and erosion) to process images to enhance or extract specific structures.

[0245] Grain size calculation: Based on the extracted geometric features, the average grain size and the width of the grain size distribution are calculated. Grain size is usually expressed as an average diameter (μm), while the width of the grain size distribution can be quantified by the standard deviation or coefficient of variation.

[0246] Data statistics and analysis: Statistical analysis is performed on the data of multiple microscopic images to obtain the temporal variation trend of grain size, providing a basis for subsequent crystal morphology evaluation and model optimization.

[0247] Through digital image processing technology, grain size parameters can be extracted efficiently and accurately, eliminating the subjectivity and inefficiency of manual measurement. This automated data processing method not only improves the speed and accuracy of data acquisition, but also enables real-time analysis of large-scale production data, providing reliable data support for dynamic optimization of the production process.

[0248] Step S3130 mainly extracts crystal distribution uniformity parameters by curve fitting of the X-ray diffraction pattern; this process includes the following steps:

[0249] Spectrum preprocessing: Perform denoising, baseline correction and other preprocessing on the acquired X-ray diffraction spectrum to ensure the accuracy and usability of the data. The preprocessing step helps to eliminate instrument noise and background signals, and improve the accuracy of subsequent analysis. Denoising refers to removing random noise in the spectrum through filtering and other methods. Baseline correction refers to adjusting the background signal of the diffraction spectrum to align it with the ideal baseline.

[0250] Curve fitting: Use appropriate mathematical models (such as Gaussian function, Lorentz function, etc.) to fit the diffraction peaks to determine the distribution of crystal plane orientation. The purpose of curve fitting is to simplify the complex diffraction spectrum into a series of standard curves for subsequent quantitative analysis.

[0251] Calculation of crystal distribution uniformity: Calculate the uniformity parameters of crystal distribution based on the fitted diffraction peak width and shape. Commonly used indicators include the standard deviation of crystal plane orientation, full width at half maximum (FWHM), etc. These indicators can reflect the orientation consistency of the crystal in different directions. Full width at half maximum (FWHM) refers to the width of the diffraction peak at half the peak height, reflecting the dispersion of the crystal orientation. The standard deviation measures the degree of dispersion of the crystal plane orientation distribution. The smaller the standard deviation, the more uniform the distribution.

[0252] Data analysis and evaluation: Statistical analysis is performed on the data of multiple diffraction patterns to evaluate the temporal variation trend of crystal distribution uniformity and provide a basis for dynamic optimization of crystal morphology.

[0253] By fitting the curve of the X-ray diffraction pattern, the uniformity parameters of the crystal distribution can be accurately quantified, providing more in-depth internal structure information than microscopic images. This method can reveal the changes in the orientation and uniformity of the crystal at different growth stages, helping production managers to adjust process parameters in a timely manner, optimize crystal morphology, and improve the physical properties and consistency of the product.

[0254] By acquiring crystal morphology parameters in real time, we can promptly understand the state changes in the production process. Combined with the hierarchical control matrix, we can dynamically adjust the process parameters to ensure that the production process always runs in the best state.

[0255] Step S3200, dynamically optimizing and controlling the water-quenched slag production process parameters based on the measured crystal morphology parameters and the hierarchical control matrix.

[0256] Furthermore, if Figure 5 As shown, step S3200 includes:

[0257] Step S3210, inputting the grain size parameters and the crystal distribution uniformity parameters into the crystal morphology multi-objective optimization model to obtain the optimal production process parameter values ​​under the current production state;

[0258] Specifically, the multi-objective optimization model of crystal morphology comprehensively considers the grain size parameters and crystal distribution uniformity parameters, and calculates the optimal combination of multiple process parameters (such as temperature, cooling rate, and additive ratio) through optimization algorithms (such as particle swarm optimization algorithm, genetic algorithm, etc.). The goal of the model is to optimize the grain size and distribution uniformity at the same time, so as to obtain the optimal production process parameters that can meet the quality requirements. The optimal production process parameters refer to the parameter values ​​that can make the grain size and distribution uniformity reach the expected optimal state under the current production conditions. Usually, these parameters include temperature, cooling rate, and additive ratio.

[0259] By inputting the measured parameters into the optimization model, a set of optimal process parameter combinations can be obtained to ensure the optimization of crystal morphology quality. These optimal process parameters provide a clear basis for subsequent adjustments and controls, thereby improving the stability of the production process and product quality. For example, if the grain size is found to be uneven in the current production process, the optimization model can obtain a balanced solution by adjusting parameters such as cooling rate and additive ratio to ensure that the final product quality meets expectations.

[0260] Step S3220, comparing the deviation between the measured crystal morphology parameters and the expected optimization target, dynamically adjusting the hierarchical control matrix, and obtaining an adjusted hierarchical control matrix;

[0261] Furthermore, if Figure 6 As shown, step S3220 includes:

[0262] Step S3221, calculating the deviation ΔG between the grain size parameter and the target grain size value, and calculating the deviation ΔU between the crystal distribution uniformity parameter and the target distribution uniformity value;

[0263] Step S3222, determine whether |ΔG| exceeds a preset grain size deviation threshold ΔG θ , determine whether |ΔU| exceeds the preset distribution uniformity deviation threshold ΔU θ ;

[0264] Step S3223, if |ΔG|≤ΔG θ And |ΔU|≤ΔU θ , the existing hierarchical control matrix settings remain unchanged;

[0265] Step S3224, if |ΔG|>ΔG θ Or |ΔU|>ΔU θ , then dynamically adjust the hierarchical control matrix to obtain the adjusted hierarchical control matrix.

[0266] Further, step S3224 includes:

[0267] Step S32241, narrowing the control threshold ranges of key process parameters of different importance levels in the hierarchical control matrix to obtain adjusted control threshold ranges;

[0268] Step S32242, improving the control intensity of key process parameters of different importance levels in the hierarchical control matrix to obtain adjusted control intensity;

[0269] Step S32243, with the importance levels of key process parameters as rows, and the adjusted control threshold ranges and adjusted control intensities as columns, to generate an adjusted hierarchical control matrix.

[0270] Specifically, the purpose of step S3221 is to calculate the deviation ΔG between the grain size parameter and its target value, and the deviation ΔU between the crystal distribution uniformity parameter and the target value. By calculating these deviations, the degree of deviation of the current production status can be quantitatively evaluated, thereby providing a basis for the next step of management and control adjustments. The deviation of the grain size reflects whether the size of the grains in the current production process has reached the expected target value. If the absolute value of ΔG is large, it means that the control effect of the grain size is not good, and parameters such as the cooling rate may need to be adjusted. Crystal distribution uniformity reflects whether the crystals are evenly distributed in the material. The deviation of the ΔU value shows whether the distribution in the current production meets the optimization goal. If the ΔU deviation is large, it means that the crystal distribution is uneven and other process parameters need to be adjusted, such as temperature or additive ratio.

[0271] Step S3222 is used to determine whether the deviations ΔG and ΔU exceed the preset thresholds. This step determines whether the current hierarchical control matrix needs to be adjusted. θ is the preset threshold of grain size deviation, which defines the maximum deviation range allowed for grain size. θ , which means that the grain size is outside the expected range and the production process needs to intervene to correct the deviation. ΔU θ is the preset threshold value of the distribution uniformity deviation, which defines the maximum deviation range allowed for the crystal distribution uniformity. θ , which means that the crystal distribution is uneven, which may affect the quality of the final product and require adjustment of relevant process parameters in the production process.

[0272] If the grain size deviation ΔG and the crystal distribution uniformity deviation ΔU do not exceed the set threshold (i.e. |ΔG|≤ΔG θ And |ΔU|≤ΔU θ ), it means that the crystal morphology quality in the current production process meets the optimization target and no adjustment is needed. Therefore, the hierarchical control matrix on the production line will continue to remain unchanged. This situation shows that the existing process parameters and control measures are effective, and the production process does not require too much intervention, and can stably produce products that meet quality requirements. If it is found that |ΔG|>ΔG θ Or |ΔU|>ΔU θ , indicating that there are situations in the production process that do not meet the quality requirements, and it is necessary to correct the deviation by adjusting the hierarchical control matrix to ensure that the crystal morphology meets the expected standards.

[0273] In order to more accurately control the key process parameters in the production process, it is first necessary to narrow the control threshold range of key process parameters of different importance levels in the hierarchical control matrix. For example, if the control deviation of the cooling rate is too large, its control threshold can be narrowed to ±0.5% of the standard value, so as to more strictly control its fluctuations. For those process parameters with higher importance (such as cooling rate, additive ratio, etc.), increase their control intensity. For example, if the current control intensity is "level 2", it can be increased to "level 1", which means that these process parameters will be checked more frequently and adjusted in real time. Then generate a new hierarchical control matrix to ensure that the control threshold range and control intensity of each key process parameter have been adjusted to the value that best suits the current production status.

[0274] The purpose of dynamically adjusting the hierarchical control matrix is ​​to detect deviations in time during the production process and make corrections, so as to ensure that the quality of the crystal morphology is always stable and meets the optimization goals. By narrowing the threshold range and increasing the control intensity, the key process parameters in the production process can be finely controlled, quality fluctuations can be reduced, and product consistency can be improved. For example, if it is found that the control of the cooling rate is not accurate, timely adjusting the control intensity to "level 1" and reducing the control threshold can effectively improve the uniformity of grain size and distribution, and ensure that the final product quality meets the expected standards.

[0275] Step S3230, in combination with the adjusted hierarchical control matrix, the optimal production process parameter values ​​under the current production status are parsed into operable process parameter adjustment instructions.

[0276] Furthermore, if Figure 7 As shown, step S3230 includes:

[0277] Step S3231, matching and comparing the optimal production process parameter value under the current production state with the adjusted control threshold range and the adjusted control intensity of the corresponding key process parameter in the adjusted hierarchical control matrix;

[0278] Step S3232: if the optimal production process parameter value under the current production state is within the adjusted control threshold range of the corresponding key process parameter, a fixed value control instruction of the corresponding key process parameter is generated, and the optimal production process parameter value is directly used as the set value of the corresponding key process parameter in the production process;

[0279] Step S3233: if the optimal production process parameter value under the current production state exceeds the adjusted control threshold range of the corresponding key process parameter, a fixed value control instruction or fuzzy control rule of the corresponding key process parameter is generated according to the adjusted control intensity of the corresponding key process parameter;

[0280] Further, step S3233 includes:

[0281] Step S32331, for the key process parameters with the adjusted control intensity of level 1, the corresponding optimal production process parameter value and the average value of the adjusted control threshold range are used as the instruction value to generate a fixed value control instruction for the corresponding key process parameter;

[0282] Step S32332, for the key process parameters with adjusted control intensity of level 2 or 3, generate fuzzy control rules for the corresponding key process parameters according to the degree of deviation of the optimal production process parameter value beyond the corresponding adjusted control threshold range.

[0283] Step S3234, converting the fixed value control instruction or the fuzzy control rule into a process parameter adjustment instruction.

[0284] Specifically, the purpose of step S3230 is to convert the optimal production process parameter values ​​(such as optimal temperature, cooling rate and additive ratio) obtained in step S3210 into operable process parameter adjustment instructions in combination with the adjusted hierarchical control matrix. The key to this step is to connect the theoretically calculated optimal values ​​with the actual control constraints and generate adjustment instructions that can be actually executed in the production process.

[0285] The hierarchical control matrix provides the control threshold range and control intensity of different process parameters at different importance levels. By comparing with the optimal production process parameter values, it is possible to determine whether these parameters are within a reasonable control range and generate corresponding adjustment instructions. If the optimal production process parameter value is within the corresponding control threshold range, a fixed value control instruction is directly generated; if the optimal value exceeds the threshold range, an adjustment control instruction is generated according to the control intensity.

[0286] The goal of step S3231 is to match and compare the optimal production process parameter value with the control threshold range and control intensity in the adjusted hierarchical control matrix. This process is the prerequisite for generating specific adjustment instructions, and it can determine whether the optimal production process parameter value meets the existing control standards. For each key process parameter, the adjusted control threshold range represents the acceptable fluctuation range of the parameter. For example, the control threshold of the cooling rate may be ±1% of the standard value, while the control threshold of the temperature may be ±5%. The control intensity determines the management intensity of different process parameters. For example, the control intensity of the first-level key parameters is higher and requires more precise control; while the control intensity of the third-level key parameters is lower, allowing greater fluctuations. By comparing the optimal parameters with the control threshold and control intensity, it can be determined whether the production parameters need to be further adjusted.

[0287] When the matching result in step S3231 shows that the optimal production process parameter value is within the control threshold range, it means that the process parameters in the current production process meet the optimization requirements and do not need further adjustment. In this case, the system will directly generate a fixed value control instruction, and use the optimal production process parameter value directly as the set value of the corresponding process parameter in the production process. Fixed value control instructions mean that the system adjusts the corresponding process parameters (such as temperature, cooling rate, etc.) to the optimal value to ensure that the production process reaches the optimal state. For example, if the calculated optimal cooling rate is 50℃ / s and the value is within the control threshold range, the system will set the cooling rate to 50℃ / s and maintain it at this value. Directly setting the optimal process parameters as fixed values ​​reduces the complexity of human intervention and adjustment, and ensures stable operation of the production process.

[0288] If it is found in step S3231 that the optimal production process parameter value exceeds the control threshold range, the production process needs to be adjusted to a certain extent. In this case, the corresponding adjustment control instruction is generated according to the adjusted control intensity. For the key process parameters of the first-level control intensity (such as cooling rate), if its optimal value exceeds the control range, it will be replaced by the average value of the control threshold range where the parameter is located, and a fixed value control instruction will be generated. For the key process parameters of the second and third-level control intensity (such as temperature), fuzzy control rules are generated according to the degree of deviation between the optimal parameter value and the control threshold range. Fuzzy control rules include adjustment direction and adjustment amplitude. For example, if the optimal temperature value is 105°C and its control threshold range is 100°C to 104°C, the optimal temperature value exceeds the upper limit. The system generates a fuzzy control rule indicating that the temperature should be "slightly reduced" or "reduced" by a certain amplitude so that the temperature gradually approaches the maximum value of the control threshold range.

[0289] The purpose of step S3234 is to convert the generated fixed value control instructions and fuzzy control rules into specific process parameter adjustment instructions. These instructions will be sent to the field control equipment through the distributed control system to achieve dynamic adjustment of the process parameters.

[0290] Step S3200 dynamically optimizes and controls the water-quenched slag production process parameters by combining the measured crystal morphology parameters and the real-time data input of the hierarchical control matrix, providing an accurate and flexible quality optimization mechanism for the production process. Its beneficial effects include: improving the automation and intelligence level of the production process, achieving continuous dynamic optimization of quality, balancing resource utilization and cost-effectiveness, ensuring that each batch of products meets quality standards, and reducing fluctuations and unstable factors in the production process. This optimization control system not only improves product quality, but also shows significant advantages in production efficiency and cost control, and ultimately promotes the continuous improvement and upgrading of the water-quenched slag production process.

[0291] Example 2

[0292] This embodiment provides an intelligent management system for the production process of industrial solid wastewater quenching slag fiberization based on the embodiment 1, such as Figure 8 As shown, including:

[0293] Model building module: used to build a mathematical model of the crystal morphology of industrial solid waste quenching slag. Based on the mathematical model of the crystal morphology of industrial solid waste quenching slag, a multi-objective optimization model of crystal morphology is built to obtain the optimal production process parameter set and the expected optimization target;

[0294] Hierarchical control module: Based on the mathematical model of the crystal morphology of industrial solid waste quenching slag, key process parameters are screened out from the process parameters; the screened key process parameters are divided into different importance levels; for key process parameters of different importance levels, the corresponding control threshold range and control intensity are set according to the optimal production process parameter set, and a hierarchical control matrix is ​​established;

[0295] Dynamic optimization module: obtains the measured crystal morphology parameters of the water-quenched slag production process, and dynamically optimizes and controls the water-quenched slag production process parameters based on the measured crystal morphology parameters and the hierarchical control matrix.

[0296] In the model building module, the construction of the industrial solid waste quenching slag crystal morphology mathematical model includes:

[0297] Step S1110, collecting historical production data of industrial solid waste quenching slag, extracting a first record set, the first record set including a grain size time series, a distribution uniformity time series, a temperature time series, a cooling rate time series and an additive ratio time series;

[0298] Step S1120, performing data preprocessing on the time series data in the first record set;

[0299] Step S1130, based on the first record set, a nonlinear correlation model between grain size and distribution uniformity and process parameters is established to obtain a mathematical model of the crystal morphology of industrial solid waste water quenched slag; the process parameters include temperature, cooling rate and additive ratio.

[0300] The step S1130 includes:

[0301] Step S1131: Based on the first record set, a nonlinear correlation model between grain size and process parameters is established, which is marked as the initial grain size model; the initial grain size model includes n1 grain size influencing parameters ,in, is the constant term of the initial model of grain size, for Moment The influence parameters of the process parameters on the grain size at time t, 0≤i≤n1-1, 1≤ ≤n1-1;

[0302] Step S1132: Based on the first record set, a nonlinear correlation model between distribution uniformity and process parameters is established, which is marked as the distribution uniformity initial model; the distribution uniformity initial model includes n2 distribution uniformity influencing parameters ,in, is the constant term of the initial model for distribution uniformity, for Moment The influence parameters of the process parameters on the distribution uniformity at time t, 0≤j≤n2-1, 1≤ ≤n2-1;

[0303] Step S1133, solving the parameters of the initial model of grain size and the initial model of distribution uniformity to obtain and The optimal value of

[0304] Step S1134: Substitute the optimal value of into the initial grain size model to obtain the final grain size model;

[0305] Step S1135: Substitute the optimal value of into the initial model of distribution uniformity to obtain the final model of distribution uniformity;

[0306] Step S1136, the final model of grain size and the final model of distribution uniformity constitute a mathematical model of the crystal morphology of industrial solid waste quenching slag.

[0307] In the model building module, obtaining the optimal production process parameter set and the expected optimization target includes:

[0308] Step S1210, based on the mathematical model of crystal morphology of industrial solid waste quenching slag, with temperature T, cooling rate C and additive ratio A as decision variables, and with minimizing grain size and maximizing distribution uniformity as optimization goals, a multi-objective optimization model of crystal morphology is established;

[0309] Step S1220, solving the multi-objective optimization model of crystal morphology to obtain an optimal production process parameter set and an expected optimization target, wherein the expected optimization target includes a target value for grain size and a target value for distribution uniformity; the optimal production process parameter set includes historical optimal values ​​of the production process parameters.

[0310] In the hierarchical control module, the key process parameters selected from the process parameters include:

[0311] Step S2110, based on the final model of grain size and the final model of distribution uniformity, respectively calculate the partial derivatives of the process parameters on the grain size and the distribution uniformity, and obtain the influence coefficient matrix of each process parameter;

[0312] Step S2120, normalizing the influence coefficient matrix to obtain the influence weights of the process parameters on the grain size and distribution uniformity;

[0313] Step S2130 , sorting the process parameters in descending order according to their influence weights, and selecting the process parameters whose cumulative weights reach a preset cumulative weight threshold θ1 as key process parameters.

[0314] In the hierarchical control module, the screened key process parameters are divided into different importance levels including:

[0315] Step S2210, sorting the key process parameters in descending order according to the influence weights, and defining the key process parameters with the influence weights at the top W1 as the first-level key parameters;

[0316] Step S2220, defining the key process parameters with influence weights between W1 and W2 as secondary key parameters;

[0317] Step S2230, defining the key process parameters with influence weights between W2 and W3 as three-level key parameters, where W3>W2>W1, and W1, W2, and W3 are percentages.

[0318] In the hierarchical control module, the establishment of a hierarchical control matrix includes:

[0319] Step S2310, obtaining the historical optimal value of each key process parameter from the optimal production process parameter set, and taking the historical optimal value of each key process parameter as the standard value of the corresponding key process parameter; setting the control threshold range of the first-level key parameter to the first-level key parameter standard value ±M1; setting the control threshold range of the second-level key parameter to the second-level key parameter standard value ±M2; setting the control threshold range of the third-level key parameter to the third-level key parameter standard value ±M3; wherein M3>M2>M1, M1, M2, and M3 are percentages;

[0320] Step S2320, setting the control intensity of the first-level key parameters to level one, setting the control intensity of the second-level key parameters to level two, and setting the control intensity of the third-level key parameters to level three;

[0321] Step S2330, generating a hierarchical control matrix with the importance levels of key process parameters as rows and the control threshold ranges and control intensities as columns.

[0322] In the dynamic optimization module, the dynamic optimization control of water-quenched slag production process parameters includes:

[0323] Step S3100, obtaining measured crystal morphology parameters of the water-quenched slag production process; the measured crystal morphology parameters include grain size parameters and crystal distribution uniformity parameters;

[0324] Step S3200, dynamically optimizing and controlling the water-quenched slag production process parameters based on the measured crystal morphology parameters and the hierarchical control matrix.

[0325] The step S3100 includes:

[0326] Step S3110, obtaining a microscopic image and an X-ray diffraction spectrum of crystals during the water-quenched slag production process;

[0327] Step S3120, performing digital image processing on the microscopic image to extract grain size parameters;

[0328] Step S3130, performing curve fitting on the X-ray diffraction pattern to obtain crystal distribution uniformity parameters.

[0329] The step S3200 includes:

[0330] Step S3210, inputting the grain size parameters and the crystal distribution uniformity parameters into the crystal morphology multi-objective optimization model to obtain the optimal production process parameter values ​​under the current production state;

[0331] Step S3220, comparing the deviation between the measured crystal morphology parameters and the expected optimization target, dynamically adjusting the hierarchical control matrix, and obtaining an adjusted hierarchical control matrix;

[0332] Step S3230, in combination with the adjusted hierarchical control matrix, the optimal production process parameter values ​​under the current production status are parsed into operable process parameter adjustment instructions.

[0333] The step S3220 includes:

[0334] Step S3221, calculating the deviation ΔG between the grain size parameter and the target grain size value, and calculating the deviation ΔU between the crystal distribution uniformity parameter and the target distribution uniformity value;

[0335] Step S3222, determine whether |ΔG| exceeds a preset grain size deviation threshold ΔG θ , determine whether |ΔU| exceeds the preset distribution uniformity deviation threshold ΔU θ ;

[0336] Step S3223, if |ΔG|≤ΔG θ And |ΔU|≤ΔU θ , the existing hierarchical control matrix settings remain unchanged;

[0337] Step S3224, if |ΔG|>ΔGθ Or |ΔU|>ΔU θ , then dynamically adjust the hierarchical control matrix to obtain the adjusted hierarchical control matrix.

[0338] The step S3224 includes:

[0339] Step S32241, narrowing the control threshold ranges of key process parameters of different importance levels in the hierarchical control matrix to obtain adjusted control threshold ranges;

[0340] Step S32242, improving the control intensity of key process parameters of different importance levels in the hierarchical control matrix to obtain adjusted control intensity;

[0341] Step S32243, with the importance levels of key process parameters as rows, and the adjusted control threshold ranges and adjusted control intensities as columns, to generate an adjusted hierarchical control matrix.

[0342] The step S3230 includes:

[0343] Step S3231, matching and comparing the optimal production process parameter value under the current production state with the adjusted control threshold range and the adjusted control intensity of the corresponding key process parameter in the adjusted hierarchical control matrix;

[0344] Step S3232: if the optimal production process parameter value under the current production state is within the adjusted control threshold range of the corresponding key process parameter, a fixed value control instruction of the corresponding key process parameter is generated, and the optimal production process parameter value is directly used as the set value of the corresponding key process parameter in the production process;

[0345] Step S3233: if the optimal production process parameter value under the current production state exceeds the adjusted control threshold range of the corresponding key process parameter, a fixed value control instruction or fuzzy control rule of the corresponding key process parameter is generated according to the adjusted control intensity of the corresponding key process parameter;

[0346] Step S3234, converting the fixed value control instruction or the fuzzy control rule into a process parameter adjustment instruction.

[0347] The step S3233 includes:

[0348] Step S32331, for the key process parameters with the adjusted control intensity of level 1, the corresponding optimal production process parameter value and the average value of the adjusted control threshold range are used as the instruction value to generate a fixed value control instruction for the corresponding key process parameter;

[0349] Step S32332, for the key process parameters with adjusted control intensity of level 2 or 3, generate fuzzy control rules for the corresponding key process parameters according to the degree of deviation of the optimal production process parameter value beyond the corresponding adjusted control threshold range.

[0350] In addition, the parts of the above-mentioned technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0351] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent management method for the production process of industrial solid wastewater quenching slag fiberization, characterized in that: The method comprises: Construct a mathematical model of the crystal morphology of industrial solid waste quenching slag. Based on this mathematical model, construct a multi-objective optimization model of the crystal morphology to obtain the optimal production process parameter set and expected optimization target. Based on the mathematical model of the crystal morphology of industrial solid waste quenching slag, key process parameters are screened out from the process parameters; the screened key process parameters are divided into different importance levels; for key process parameters of different importance levels, the corresponding control threshold range and control intensity are set according to the optimal production process parameter set, and a hierarchical control matrix is ​​established; The measured crystal morphology parameters of the water-quenched slag production process are obtained, and based on the measured crystal morphology parameters and the hierarchical control matrix, the water-quenched slag production process parameters are dynamically optimized and controlled.

2. The intelligent management method for industrial solid wastewater quenching slag fiberization production process according to claim 1 is characterized in that: The construction of the mathematical model of the crystal morphology of industrial solid waste quenching slag includes: Collect historical production data of industrial solid waste quenching slag and extract the first record set; According to the first record set, a nonlinear correlation model between grain size and distribution uniformity and process parameters is established, and a mathematical model of the crystal morphology of industrial solid wastewater quenching slag is obtained.

3. The intelligent management method for industrial solid wastewater quenching slag fiberization production process according to claim 2 is characterized in that: The method of establishing a nonlinear correlation model between grain size and distribution uniformity and process parameters to obtain a mathematical model of the crystal morphology of industrial solid waste quenching slag includes: According to the first record set, a nonlinear correlation model between grain size and process parameters is established, which is marked as an initial grain size model; According to the first record set, a nonlinear correlation model between distribution uniformity and process parameters is established, which is marked as an initial distribution uniformity model; Solving the parameters of the initial model of grain size and the initial model of distribution uniformity to obtain the final model of grain size and the final model of distribution uniformity; The final model of grain size and the final model of distribution uniformity constitute the mathematical model of the crystal morphology of industrial solid waste quenching slag.

4. The intelligent management method for industrial solid wastewater quenching slag fiberization production process according to claim 3 is characterized in that: The optimal production process parameter set and expected optimization target are obtained as follows: Based on the mathematical model of crystal morphology of industrial solid waste quenching slag, a multi-objective optimization model of crystal morphology was established with external parameters as decision variables and minimization of grain size and maximization of distribution uniformity as optimization goals. Solve the multi-objective optimization model of crystal morphology to obtain the optimal production process parameter set and the expected optimization target, wherein the expected optimization target includes the target value of grain size and the target value of distribution uniformity; the optimal production process parameter set includes the historical optimal values ​​of the production process parameters.

5. The intelligent management method for industrial solid wastewater quenching slag fiberization production process according to claim 4 is characterized in that: The key process parameters screened out from the process parameters include: Based on the final model of grain size and the final model of distribution uniformity, the partial derivatives of the process parameters on grain size and distribution uniformity are calculated respectively, and the influence coefficient matrix of each process parameter is obtained; The influence coefficient matrix is ​​normalized to obtain the influence weights of process parameters on grain size and distribution uniformity; The process parameters are sorted in descending order according to their influence weights, and the process parameters whose cumulative weights reach the preset cumulative weight threshold θ1 are selected as key process parameters.

6. The intelligent management method for industrial solid wastewater quenching slag fiberization production process according to claim 5 is characterized in that: The screening of key process parameters is divided into different levels of importance, including: Sort the key process parameters in descending order according to their impact weights, and define the key process parameters with the impact weights at the top W1 as the first-level key parameters; The key process parameters with influence weights between W1 and W2 are defined as secondary key parameters; The key process parameters with influence weights between W2 and W3 are defined as three-level key parameters, where W3>W2>W1, and W1, W2, and W3 are percentages.

7. The intelligent management method for industrial solid wastewater quenching slag fiberization production process according to claim 6 is characterized in that: The establishment of a hierarchical control matrix includes: Obtain the historical optimal value of each key process parameter from the optimal production process parameter set, and use the historical optimal value of each key process parameter as the standard value of the corresponding key process parameter; set the control threshold range of the first-level key parameter to the first-level key parameter standard value ±M1; the control threshold range of the second-level key parameter to the second-level key parameter standard value ±M2; the control threshold range of the third-level key parameter is the third-level key parameter standard value ±M3; where M3>M2>M1, M1, M2, and M3 are percentages; Set the control intensity of the first-level key parameters to level one, the control intensity of the second-level key parameters to level two, and the control intensity of the third-level key parameters to level three; A hierarchical control matrix is ​​generated with the importance levels of key process parameters as rows and the control threshold ranges and control intensities as columns.

8. The intelligent management method for industrial solid wastewater quenching slag fiberization production process according to claim 7 is characterized in that: The measured crystal morphology parameters include grain size parameters and crystal distribution uniformity parameters; The dynamic optimization control of water-quenched slag production process parameters includes: Input the grain size parameters and crystal distribution uniformity parameters into the multi-objective optimization model of crystal morphology to obtain the optimal production process parameter values ​​under the current production state; Compare the deviations between the measured crystal morphology parameters and the expected optimization targets, dynamically adjust the hierarchical control matrix, and obtain the adjusted hierarchical control matrix; Combined with the adjusted hierarchical control matrix, the optimal production process parameter values ​​under the current production status are parsed into actionable process parameter adjustment instructions.

9. The intelligent management method for industrial solid wastewater quenching slag fiberization production process according to claim 8, characterized in that: The adjusted hierarchical control matrix includes: Calculate the deviation ΔG between the grain size parameter and the target grain size value, and calculate the deviation ΔU between the crystal distribution uniformity parameter and the target distribution uniformity value; If |ΔG|≤ΔG θ And |ΔU|≤ΔU θ , the existing hierarchical control matrix settings remain unchanged; ΔG θ is the preset grain size deviation threshold, ΔU θ is a preset distribution uniformity deviation threshold; If |ΔG|>ΔG θ Or |ΔU|>ΔU θ , then dynamically adjust the hierarchical control matrix to obtain the adjusted hierarchical control matrix.

10. An intelligent management system for the production process of industrial solid wastewater quenching slag fiberization, which is used to implement the intelligent management method for the production process of industrial solid wastewater quenching slag fiberization according to any one of claims 1 to 9, characterized in that: The system comprises: Model building module: used to build a mathematical model of the crystal morphology of industrial solid waste quenching slag. Based on the mathematical model of the crystal morphology of industrial solid waste quenching slag, a multi-objective optimization model of crystal morphology is built to obtain the optimal production process parameter set and the expected optimization target; Hierarchical control module: Based on the mathematical model of the crystal morphology of industrial solid waste quenching slag, key process parameters are screened out from the process parameters; the screened key process parameters are divided into different importance levels; for key process parameters of different importance levels, the corresponding control threshold range and control intensity are set according to the optimal production process parameter set, and a hierarchical control matrix is ​​established; Dynamic optimization module: obtains the measured crystal morphology parameters of the water-quenched slag production process, and dynamically optimizes and controls the water-quenched slag production process parameters based on the measured crystal morphology parameters and the hierarchical control matrix.

Images