Inorganic mineral casting processing parameter optimization method based on real-time data analysis

Abstract

This invention relates to the field of digital casting technology, specifically to a method for optimizing the processing parameters of inorganic mineral castings based on real-time data analysis. The method includes the following steps: a spectrometer detects the melt composition, which is input into a neural network for analysis to generate component characteristic analysis results; pouring temperature and holding time are extracted; viscosity and filling time are monitored; a pouring parameter configuration information table is generated; temperature and cooling rate are monitored in real time; particle swarm optimization is used to screen cooling parameters; the intensity of the cooling stage is adjusted to determine the demolding timing; parameters are compared to adjust the demolding temperature; and an optimized processing parameter scheme for inorganic mineral castings is output. In this invention, the melt composition and components are acquired in real time; an intelligent algorithm automatically analyzes the pouring parameters; temperature and rheological characteristics are dynamically analyzed; multi-stage temperature and cooling intensity are autonomously adjusted; feedback is used to optimize surface and internal quality; parameter consistency is improved; subjective errors are reduced; performance indicators are balanced; and production stability and resource efficiency are enhanced.

Description

Technical Field

[0001] This invention relates to the field of digital casting technology, and in particular to a method for optimizing the processing parameters of inorganic mineral castings based on real-time data analysis. Background Technology

[0002] The field of digital casting technology involves the use of information technology and automation to collect, analyze and control data throughout the entire casting process. Its core aspects include the digital design of parameters, intelligent monitoring of the production process, real-time analysis of casting data, tracking management of casting quality, and automatic optimization of processes, so as to achieve efficient and intelligent management of the entire process from raw material feeding, melting, forming, cooling to finished product inspection.

[0003] Among them, the traditional method for optimizing the processing parameters of inorganic mineral castings refers to setting key parameters, such as temperature setting, pouring speed, vibration frequency, cooling rate, and stirring time, manually by experts during the processing of castings made of inorganic mineral materials. Combined with limited sample test data, parameters are selected and adjusted to control the casting formation process. The optimization of related parameters is completed by manually testing and adjusting based on existing work specifications and past production experience, or by comparing the physical test results of earlier production batches and manually recording the results.

[0004] Existing technologies rely on manual parameter setting and experience-based adjustments, which are easily affected by personnel's operating habits, knowledge limitations, and the number of samples. Parameter adjustments during the casting process are delayed or incomplete. When faced with fluctuations in raw material composition or subtle changes, it is difficult to achieve synchronous matching between parameters and actual conditions, resulting in slow feedback, large quality fluctuations, and difficulty in tracing the multivariate relationships in the process. This leads to reduced batch-to-batch stability and is not conducive to achieving continuous, efficient, and refined management and rapid response to production needs. Summary of the Invention

[0005] To address the shortcomings of existing technologies that rely on manual parameter setting and experience-based adjustments, which are susceptible to limitations imposed by operator habits, knowledge gaps, and sample quantity, resulting in delayed or incomplete parameter adjustments during the casting process, and difficulties in synchronizing parameters with actual conditions when faced with fluctuations or subtle changes in raw material composition, this invention provides a method for optimizing inorganic mineral casting processing parameters based on real-time data analysis. This method includes the following steps:

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a method for optimizing the processing parameters of inorganic mineral castings based on real-time data analysis, comprising the following steps:

[0007] S1: Real-time detection of silicate content and oxide component ratio in inorganic mineral melt using a spectrometer, monitoring mineral phase transformation temperature and crystallization rate, inputting monitoring data into a neural network model for component characteristic analysis, and generating component characteristic analysis results;

[0008] S2: Based on the analysis results of the component characteristics, obtain the pouring temperature range and holding time, detect the mold preheating temperature data and melt viscosity changes, call the crystallization precipitation rate to calculate the mold filling time, and generate a pouring parameter configuration information table by analyzing the matching relationship between melt viscosity changes and pouring speed.

[0009] S3: Call the pouring temperature and holding time in the pouring parameter configuration information table, monitor the internal temperature distribution data and cooling rate changes of the casting in real time, input the monitoring data into the particle swarm optimization algorithm for optimization calculation, screen the cooling curve and verify it, and generate a cooling parameter optimization scheme.

[0010] S4: Obtain the target temperature and duration of each stage through the cooling parameter optimization scheme, detect the surface hardness of the casting, adjust the cooling intensity of the multi-stage cooling according to the change of cooling rate, calculate the degree of deviation between the adjusted cooling intensity and the critical demolding temperature, and generate demolding timing determination parameters.

[0011] As a further aspect of the present invention, the critical demolding temperature is determined through thermodynamic simulation calculations based on the casting material and a predetermined demolding safety standard;

[0012] The component characteristic analysis results include material compatibility grade, crystallization precipitation rate, thermodynamic stability coefficient and component characteristic label. The casting parameter configuration information table includes casting temperature range, holding time and casting rate. The cooling parameter optimization scheme includes cooling temperature curve, cooling duration and cooling medium flow rate. The demolding timing determination parameters include demolding temperature range, critical temperature difference and surface hardness threshold.

[0013] As a further aspect of the present invention, the specific steps of S1 are as follows:

[0014] S101: The reflectance, transmittance and absorptivity values ​​of inorganic mineral melts are detected in real time by a spectrometer. Based on the spectral characteristics, silicate distribution data are extracted, and the oxide ratio is calculated by combining the elemental response spectral line intensity to generate oxide component content data.

[0015] S102: Based on the oxide component content data, monitor the melt spectrum changes, calculate the crystal formation rate based on the intensity peak shift, absorption threshold change and characteristic bandwidth shrinkage rate, and obtain the set of crystallization response change parameters;

[0016] S103: The set of crystallization response change parameters and oxide component content data are called to construct a feature vector sequence. In the neural network model, the features are mapped through weight coefficients. The corresponding intensity range of components and parameter conditions is calculated according to the matching degree scoring rules to generate component characteristic analysis results.

[0017] As a further aspect of the present invention, the specific steps of S2 are as follows:

[0018] S201: Based on the analysis results of the component characteristics, obtain the upper and lower limits of the pouring temperature and the heat preservation time, detect the mold preheating temperature and screen the data samples that meet the boundary difference conditions, map them with the corresponding heat preservation time parameters and perform a combination operation to generate temperature and heat preservation matching interval values.

[0019] S202: Call the temperature holding matching interval value and the viscosity change sequence of the melt collected under multiple holding time conditions, extract key nodes and calculate the node slope difference, compare the change amplitude with the mold filling boundary value, and obtain the viscosity evolution rate value.

[0020] S203: Construct a rate matching interval table based on the viscosity evolution rate value and the pouring speed curve, filter the speed data within the viscosity boundary range, analyze the speed offset and establish the filling time segment, integrate the associated heat preservation time, pouring temperature and mold preheating temperature parameters, and generate a pouring parameter configuration information table.

[0021] As a further aspect of the present invention, the temperature insulation matching range value is a parameter range obtained by combining the preset upper and lower limits of pouring temperature and insulation time parameters, and in conjunction with the boundary conditions of mold preheating temperature.

[0022] The viscosity evolution rate value is a quantitative parameter obtained by calculating the slope difference between key nodes based on the viscosity change sequence of the melt collected under multiple sets of heat preservation time conditions.

[0023] As a further aspect of the present invention, the specific steps of S3 are as follows:

[0024] S301: Call the pouring temperature and holding time set in the pouring parameter configuration information table, bind the current pouring task number, collect the real-time temperature data of the temperature control node, record the node temperature change and calculate the local temperature gradient and cooling rate to obtain the node cooling rate distribution value.

[0025] S302: Divide the node region according to the node cooling rate distribution value, extract the cooling rate curve, construct the input vector input to the particle swarm optimization algorithm, calculate the convergence rate according to the objective function and compare the stability screening curve to obtain the cooling curve screening result set.

[0026] The convergence condition of the particle swarm optimization algorithm is that the algorithm is considered to have converged when the maximum number of iterations is reached and the particle update amplitude is lower than a specified threshold.

[0027] S303: Based on the time points and cooling rates of the curves in the cooling curve screening result set, and combined with the corresponding temperatures and holding times in the task parameter list, construct a cooling control parameter vector, call the particle swarm optimization algorithm to fit and adjust, and generate a cooling parameter optimization scheme.

[0028] As a further aspect of the present invention, the specific steps of S4 are as follows:

[0029] S401: Based on the number of cooling stages, stage cooling time and target temperature set in the cooling parameter optimization scheme, set the combination parameters of the injection temperature, flow rate and injection angle of the cooling medium, record the surface temperature change and duration of the casting during each stage of cooling, and generate a stage temperature time series data set.

[0030] S402: Based on the stage temperature time series data group, extract the stage surface temperature and duration, detect the surface hardness value, calculate the response amplitude of the hardness value and the temperature drop rate, filter the interval where the hardness change amplitude is greater than the hardness response threshold, and obtain the rate change sensitive interval value;

[0031] The hardness response threshold is determined by collecting surface temperature and hardness data at different rates through multiple cooling experiments, optimizing the discrimination error using the gradient descent method, and determining the hardness change range limit that can effectively distinguish the rate-sensitive range.

[0032] S403: Based on the rate change sensitive interval value, extract the cooling intensity adjustment parameter, compare the degree of deviation between the adjusted cooling intensity and the demolding critical temperature, and perform weighted calculation by combining the degree of deviation and the stage time factor to generate demolding timing determination parameter.

[0033] As a further aspect of the present invention, the method includes step S5:

[0034] S5: Determine the demolding time point based on the demolding timing judgment parameters, calculate the demolding temperature setpoint, judge the demolding conditions by comparing the demolding timing judgment parameters with the standard demolding index threshold, adjust the demolding temperature setpoint according to the judgment result, and output the inorganic mineral casting processing parameter optimization scheme.

[0035] The optimization scheme for inorganic mineral casting processing parameters includes demolding temperature, recommended demolding time, and parameter adjustment scheme.

[0036] As a further aspect of the present invention, the specific steps of S5 are as follows:

[0037] S501: Based on the demolding timing determination parameters, collect the temperature change trend during the molding stage, monitor the curing rate and stress distribution range, determine the state conditions based on the critical deformation coefficient, structural stability threshold and stress critical coefficient, and generate a demolding matching judgment value.

[0038] S502: Based on the demolding matching judgment value, calculate the demolding temperature setpoint, call the deviation rate between the demolding temperature setpoint and the demolding matching judgment value, perform segment calculation with the heating rate value, adjust the temperature change range, and generate a staged demolding temperature increment coefficient.

[0039] S503: Call the staged demolding temperature increment coefficient, correct the original set value according to the temperature adjustment coefficient, adjust the temperature range of the region in combination with the boundary stress value, fit the parameter relationship, and obtain the optimization scheme of inorganic mineral casting processing parameters.

[0040] As a further aspect of the present invention, the demolding compatibility judgment value refers to an index value used to determine whether a casting is suitable for demolding, which is calculated comprehensively based on parameters such as temperature change, curing rate and stress distribution during the molding process, combined with the critical deformation coefficient and structural stability threshold.

[0041] The staged demolding temperature increment coefficient refers to the adjustment coefficient for correcting the demolding temperature set value in stages, based on the deviation rate between the demolding matching judgment value and the demolding temperature set value and the heating rate.

[0042] The critical deformation coefficient is obtained by finite element simulation of the deformation response of the casting under different temperature and stress conditions. The corresponding deformation coefficient is extracted as the critical value by statistical analysis of the parameter points where deformation occurs.

[0043] The structural stability threshold is determined by applying the safety factor method based on the material mechanical properties and original parameter data to determine the limit value at which the casting will become structurally unstable.

[0044] The critical stress coefficient is determined by analyzing the inflection point of the stress-strain curve to identify the value at which stress accumulation reaches the material's bearing limit, and this value is used as the critical stress coefficient.

[0045] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0046] In this invention, by acquiring the composition and component ratio of inorganic mineral melt in real time, and combining it with intelligent algorithms to automatically analyze parameter matching, multi-dimensional dynamic analysis of casting temperature, rheological parameters, and cooling curves is achieved. This enables the precise output of casting parameters that highly match the actual state and transformation characteristics of the material. It continuously self-matches and adjusts the temperature, holding temperature, and cooling intensity at multiple stages, achieving synchronous feedback and optimization of surface and internal quality. This greatly enhances the flexibility and consistency of production control, reduces errors caused by subjective human judgment, achieves balanced optimization among casting performance indicators, and significantly improves the stability of finished products and resource utilization efficiency. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a schematic diagram of the steps of the present invention;

[0049] Figure 2 This is a detailed schematic diagram of S1 of the present invention;

[0050] Figure 3 This is a detailed schematic diagram of S2 of the present invention;

[0051] Figure 4 This is a detailed schematic diagram of S3 of the present invention;

[0052] Figure 5 This is a detailed schematic diagram of S4 of the present invention;

[0053] Figure 6 This is a detailed schematic diagram of S5 of the present invention. Detailed Implementation

[0054] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0055] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0056] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0057] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0058] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0059] Please see Figure 1 This invention provides a method for optimizing the processing parameters of inorganic mineral castings based on real-time data analysis, comprising the following steps:

[0060] S1: Real-time detection of silicate content and oxide component ratio in inorganic mineral melt using a spectrometer, monitoring mineral phase transformation temperature and crystallization rate, inputting monitoring data into a neural network model for component characteristic analysis, and generating component characteristic analysis results;

[0061] S2: Based on the component characteristic analysis results, obtain the pouring temperature range and holding time, detect the mold preheating temperature data and melt viscosity changes, call the crystallization precipitation rate to calculate the mold filling time, and generate a pouring parameter configuration information table by analyzing the matching relationship between melt viscosity changes and pouring speed.

[0062] S3: Call the pouring temperature and holding time in the pouring parameter configuration information table, monitor the internal temperature distribution data and cooling rate changes of the casting in real time, input the monitoring data into the particle swarm optimization algorithm for optimization calculation, screen the cooling curve and verify it, and generate a cooling parameter optimization scheme.

[0063] S4: Obtain the target temperature and duration of each stage through the cooling parameter optimization scheme, detect the surface hardness of the casting, adjust the cooling intensity of multiple stages according to the change of cooling rate, calculate the degree of deviation between the adjusted cooling intensity and the critical demolding temperature, and generate demolding timing judgment parameters.

[0064] S5: Determine the demolding time point based on the demolding timing judgment parameter, calculate the demolding temperature setpoint, judge the demolding conditions by comparing the demolding timing judgment parameter with the standard demolding index threshold, adjust the demolding temperature setpoint according to the judgment result, and output the inorganic mineral casting processing parameter optimization scheme.

[0065] The component characteristic analysis results include material compatibility grade, crystallization precipitation rate, thermodynamic stability coefficient and component characteristic label. The pouring parameter configuration information table includes pouring temperature range, holding time and pouring rate. The cooling parameter optimization scheme includes cooling temperature curve, cooling duration and cooling medium flow rate. The demolding timing determination parameters include demolding temperature range, critical temperature difference and surface hardness threshold. The inorganic mineral casting processing parameter optimization scheme includes demolding temperature, recommended demolding time and parameter adjustment scheme.

[0066] Please see Figure 2 The specific steps of S1 are as follows:

[0067] S101: The reflectance, transmittance and absorptivity values ​​of inorganic mineral melts are detected in real time by a spectrometer. Based on the spectral characteristics, silicate distribution data are extracted, and the oxide ratio is calculated by combining the elemental response spectral line intensity to generate oxide component content data.

[0068] The reflectance, transmittance, and absorptivity of inorganic mineral melts were detected in real time using a spectrometer. First, the sample was kept in a molten state at a constant temperature. Then, an online high-temperature spectrometer integrating an incident light source and detector was used to acquire its spectral response signal in the visible to near-infrared range. A segmented scan was performed within the wavelength range of 400–2500 nm, with reflectance read in 10 nm increments. transmittance With absorption rate The acquired spectral data was preprocessed to remove high-frequency noise and the effective range was extracted. Based on this, the intensity distribution of different bands was compared according to the typical silicate distribution characteristics. A matching vector was constructed using the standard reflectance spectra of silicates in the database. The presence of silicate components was initially determined by judging whether the sample showed obvious absorption dips in the three characteristic absorption bands of 970~1050nm, 1350~1450nm, and 1900~2200nm. The feature matching degree was further calculated by combining the changes in the absorption band depth and bandwidth of the comparison area. When the absorption band depth exceeded 0.3 and the bandwidth fluctuated in the range of 60~150nm, the feature was considered to exist. The silicate type was determined by the peak band shape fitting and symmetry judgment method. Based on this, the elemental response spectrum corresponding to its absorption peak was extracted. For example, a significant absorption peak was detected at 980nm, corresponding to the element Si. The signal intensity was calculated by combining the band height and full width at half maximum (FWHM) of the peak at different time points. The elemental oxide content was calculated using the following formula:

[0069] ;

[0070] in, For the first The mass percentage of the oxides Its spectral intensity integral value, This represents the total number of all detected oxide species. Taking real-time detection data as an example, if each oxide is extracted from the sample... , , , , The total integral is 820. Substituting this into the formula, we get:

[0071] ;

[0072] ;

[0073] Table 1: Distribution of Oxide Component Content

[0074]

[0075] Table 1 lists the oxide component content data calculated based on spectral integral intensity, reflecting the proportion of the main inorganic components in the melt. Based on this proportion, it can be determined which type of casting formula the melt is suitable for, and oxide component content data can be generated.

[0076] S102: Based on oxide component content data, monitor melt spectrum changes, calculate crystal formation rate based on intensity peak shift, absorption threshold change and characteristic bandwidth shrinkage rate, and obtain crystallization response change parameter set;

[0077] Based on oxide composition data, oxides with representative contents were first selected as the tracking targets. From the obtained distribution, oxides with a mass percentage higher than 10%, namely SiO2 and Al2O3, were selected. Their absorption peak positions were continuously monitored at different temperatures, with the temperature range set from 1100℃ to 1400℃. In 20℃ increments, the absorption peak positions (λ) in the 970–1050 nm and 1350–1450 nm wavelength bands were recorded at each temperature point. m The absorption band depth α and bandwidth Δλ are used to calculate the peak position shift for each set of data. T0 is the initial reference temperature of 1100℃, and the absorption bandwidth compression ratio Δλ / ΔT and the depth increase Δα / ΔT are recorded. Based on the trends of decreasing bandwidth, peak shift direction (blue shift or red shift), and increasing absorption intensity in the spectral response, the interval segmentation method is further used to evaluate the change in crystal formation rate. Specifically, the rate of change of absorption peak under unit temperature change is calculated. and If the crystallization trend is enhanced and the formation rate increases, it is determined that the crystallization trend is enhanced and the formation rate is increased. Taking a certain set of data as an example, if the absorption peak position of SiO2 is λ at a temperature of 1100℃, m =995nm, α=0.28, Δλ=85nm, and λ at 1260℃ m=980nm, α=0.33, Δλ=65nm, then:

[0078] ;

[0079] ;

[0080] The enhancement condition is not met; however, if Al2O3 has an absorption peak at the same temperature range, its absorption peak changes from λ. m =1380nm moved to λ m =1335nm, α increases from 0.25 to 0.31, then:

[0081] ;

[0082] ;

[0083] Since the enhanced judgment conditions are not met, the temperature needs to be further increased and the possibility of entering the crystal formation sensitive region needs to be assessed. Based on the trend of crystal formation rate change, a set of crystallization response change parameters is constructed. This set of parameters includes the absorption peak shift, bandwidth change rate, and absorption depth change rate of the main oxides in different temperature regions. In addition, to improve the discrimination ability, a judgment threshold range needs to be set. The shift threshold is set to below -0.5 nm / ℃, the depth increase threshold is above 0.01 / ℃, and the bandwidth compression rate threshold is below -0.125 nm / ℃. If at least two of the three conditions are met, it is recorded as a positive crystallization response and written as a valid value of the parameter set for subsequent parameter condition matching.

[0084] S103: The feature vector sequence is constructed by calling the set of crystallization response change parameters and oxide component content data. In the neural network model, the features are mapped through weight coefficients. The corresponding intensity range of components and parameter conditions is calculated according to the matching degree scoring rules to generate component characteristic analysis results.

[0085] A feature vector sequence is constructed by calling the set of crystallization response change parameters and oxide component content data. First, the content of each feature vector dimension is defined, including oxide type, mass percentage, corresponding absorption peak position, shift rate, bandwidth compression ratio, and absorption depth amplification rate. The parameters corresponding to each oxide are arranged in order to form a feature vector with the following structure:

[0086] ;

[0087] The vector of Al2O3 in the previous section is:

[0088] ;

[0089] The vectors of the main oxides are concatenated to form a complete vector sequence. Then, casting parameter matching requirements are set in a preset condition parameter space. For example, when a certain type of melt is suitable for high-silicate type rapid-cooling castings, the required SiO2 mass percentage is higher than 60%, the absorption peak blue shift rate is greater than -0.5 nm / ℃, the absorption depth growth rate is greater than 0.005 / ℃, and the bandwidth compression rate is less than -0.1 nm / ℃. During the comparison process, each dimension value in the feature vector is first checked to see if it meets the set interval. For example, SiO2 corresponds to... The nm / ℃ value is below the set lower limit, failing to meet the matching requirements, hence a score of 0. In the Al2O3 dimension, the blue shift rate and bandwidth compression rate meet the requirements, but the absorption depth growth rate is below the requirement, resulting in a score of 2 / 3. Each dimension is set to a maximum score of 1 point. The matching score is calculated by accumulating the scores, with weighting coefficients set as follows: mass percentage 0.3, peak shift rate 0.2, absorption depth growth rate 0.3, and bandwidth compression rate 0.2. The total score is:

[0090] ;

[0091] Assuming Al2O3 scores 1, 0.5, 1, and 0.5 in the three dimensions mentioned above, then:

[0092] ;

[0093] By setting a matching score threshold, a total score greater than 0.75 is considered a "strong match", a score between 0.5 and 0.75 is considered a "medium match", and a score below 0.5 is considered a "weak match". The oxide compatibility is scored and ranked to generate a matching evaluation result, which serves as a reference for optimizing casting parameters.

[0094] Please see Figure 3 The specific steps of S2 are as follows:

[0095] S201: Based on the component characteristic analysis results, obtain the upper and lower limits of the pouring temperature and the heat preservation time, detect the mold preheating temperature and screen the data samples that meet the boundary difference conditions, map them with the corresponding heat preservation time parameters and perform a combination operation to generate temperature and heat preservation matching interval values.

[0096] Based on the component characteristic analysis results, the upper and lower limits of the casting temperature for the corresponding materials are first extracted from the evaluation results one by one. For example, if the matching score of a certain oxide component after proportioning is 0.82, the corresponding casting temperature range is 1350℃ to 1420℃, and its lower limit temperature is recorded. Upper limit temperature At the same time, obtain the corresponding recommended heat preservation time period, such as 20 minutes to 30 minutes, and set the heat preservation time range. Next, using this temperature range and recommended holding time as the base data, the mold preheating temperature was measured. Assuming the on-site measurement record was 490℃, this temperature value was compared with the mold preheating temperature in each original sample, and the difference between the mold temperature in each original data set and the current sample temperature was calculated. If the preheating temperature of the mold in a certain group of original samples is 470℃, then Determine whether the difference is within the set boundary tolerance range, for example, the tolerance range is set to ±30℃, if and only if When the original sample is deemed to meet the conditions, the selected samples that meet the conditions are combined with their holding time and pouring temperature into a set of mapping items. For example, if the pouring temperature of the original sample A is 1380℃ and the holding time is 25min, then the mapping combination is (1380, 25). The combinations are arranged in ascending order of pouring temperature to construct a matching temperature-holding interval table. The upper and lower limits of the holding time matched at each temperature point are statistically analyzed to form the temperature-holding matching interval value. For example, the minimum and maximum holding time intervals of the temperature range in the sample set are statistically analyzed. The holding time corresponding to 1350℃ is [20, 22]min, 1380℃ is [25, 27]min, and 1420℃ is [29, 30]min. Then, the holding time boundary interval mapping corresponding to the temperature point is constructed. The mapping result is stored in a key-value pair structure, where the key is the temperature point and the value is the holding time range.

[0097] S202: Call the temperature holding matching interval value and the viscosity change sequence of the melt collected under multiple holding time conditions, extract key nodes and calculate the node slope difference value, compare the change amplitude with the mold filling boundary value, and obtain the viscosity evolution rate value.

[0098] Call the temperature holding interval value, select the corresponding pouring temperature point such as 1380℃ and its holding time range [25, 27] min. Within this time interval, call the viscosity change sequence collected by the sensor. The viscosity data is recorded once every 1 min. An example sequence is as follows: [3.28, 3.35, 3.44] Pa·s, corresponding to holding times of 25, 26, and 27 min respectively. Perform the key node extraction operation in this sequence, and record the viscosity at the beginning and end of the holding time, which are 3.28 Pa·s and 3.44 Pa·s respectively. Calculate the difference in node slope, that is, the viscosity change rate Δμ / Δτ. The calculation formula is:

[0099] ;

[0100] This value represents the trend of melt viscosity change within the current holding time interval. The viscosity change rate is compared with the preset mold filling viscosity boundary. For example, the viscosity change limit required for mold filling is 0.1 Pa·s / min. The current value is less than this boundary, and the rate value is recorded as meeting the filling stability requirements. The viscosity change rates of the holding intervals at multiple temperature points are further summarized in the following table:

[0101] Table 2: Viscosity Evolution Rate Sampling Table

[0102] As shown in Table 2, the viscosity change rates at the three temperature points are 0.085, 0.08, and 0.09 Pa·s / min, respectively. The change rate values ​​are compared with the mold filling critical value of 0.1 Pa·s / min. All of them are less than the critical value and are marked as "acceptable rate". The parameter is then retained for the next step of rate matching.

[0103] S203: Construct a rate matching interval table based on the viscosity evolution rate value and the pouring speed curve, filter the speed data within the viscosity boundary range, analyze the speed offset and establish the filling time segment, integrate the associated heat preservation time, pouring temperature and mold preheating temperature parameters, and generate a pouring parameter configuration information table.

[0104] Based on the viscosity evolution rate results, the corresponding rate interval table is extracted, that is, the rate interval corresponding to each temperature point and holding time. For example, the rate corresponding to 1380℃ is 0.08 Pa·s / min, and a rate matching interval [0.07, 0.09] is constructed. Then, the pouring speed data of the original casting examples that were filled within this rate interval are extracted from the database. Assuming that the pouring speed under the condition of viscosity rate of 0.08±0.01 Pa·s / min is extracted as [2.35, 2.42, 2.45, 2.38] cm / s, its average value is calculated to be 2.4 cm / s, and the standard deviation is 0.04 cm / s. Then, it is determined whether the speed deviation is concentrated, and speeds falling within the range of mean ± 1.5 times the standard deviation are selected. The speed data is labeled as "effective speed points," meaning that data within the range of [2.34, 2.46] cm / s are valid. The maximum and minimum values ​​among the effective data are set as the upper and lower limits of the filling speed under this rate condition. At the same time, the holding time and mold preheating temperature are recorded. For example, under the condition of 1380℃, the corresponding holding time is 25~27 min, and the mold preheating temperature is 480~495℃. The combined information corresponding to the temperature points is integrated into the following structure: {Temperature: 1380℃, Holding time: [25, 27] min, Speed: [2.34, 2.46] cm / s, Mold temperature: [480, 495]℃}. Multiple sets of combined parameters are collected to generate a casting parameter configuration information table, which serves as the available parameter configuration set.

[0105] Please see Figure 4 The specific steps of S3 are as follows:

[0106] S301: Call the pouring temperature and holding time set in the pouring parameter configuration information table, bind the current pouring task number, collect the real-time temperature data of the temperature control node, record the node temperature change and calculate the local temperature gradient and cooling rate to obtain the node cooling rate distribution value.

[0107] When calling the pouring temperature and holding time already set in the pouring parameter configuration information table, it is necessary to read the parameter items associated with the task number one by one, where the pouring temperature is set to... The heat preservation time is set to The two parameters are bound to the current casting task number via a database index. For example, if the task number is set to T202507A001, then the number and parameters are bound together. The data is then bound to the control system task parameter mapping table. The temperature control node data acquisition unit is then activated, with the temperature control node positions numbered P1~P4. The corresponding mold areas are set as the upper surface, middle, bottom, and gate end. Temperature values ​​are acquired in real-time at second intervals, with the acquisition time set from 0 to 1800 seconds after pouring. Each node generates 1800 temperature records. For example, at t=300s, node P1 has a temperature of 1375℃, node P2 has 1358℃, node P3 has 1332℃, and node P4 has 1296℃. Based on this, the local temperature gradient per unit length between nodes is calculated using the spatial temperature gradient calculation formula:

[0108] ;

[0109] in , They are nodes , temperature, Let P1 be the geometric distance between two nodes (in millimeters). For example, if the distance between P1 and P3 is set to 120 mm, then at t = 300 s:

[0110] ;

[0111] After forming a two-dimensional temperature gradient distribution matrix between each pair of nodes, the cooling rate is calculated. The cooling rate is defined as the rate at which the temperature decreases per unit time, using the following formula:

[0112] ;

[0113] Taking node P2 as an example, if at t=310s, If the temperature is 1358℃ at t=300s, then... Cooling rate The cooling rate of the node at the sampling point is repeatedly counted to form a node cooling rate distribution vector, such as:

[0114] ;

[0115] Corresponding to nodes P1 to P4 respectively, and defined by the cooling rate range: low-speed cooling is 0.1~0.3℃ / s, medium-speed is 0.3~0.5℃ / s, and high-speed is >0.5℃ / s, to determine the cooling state of the region where the node is located. For example, as mentioned above, node P4 enters the high-speed cooling region at t=300s. After sampling is completed, Stack the time series data to construct a table recording the distribution values ​​of cooling rates.

[0116] Table 3: Cooling rate sampling data (unit: ℃ / s)

[0117]

[0118] As shown in Table 3, the cooling rate of the temperature control point under different time nodes is recorded, and the distribution value of the node cooling rate can be obtained through the data.

[0119] S302: Divide the node region according to the node cooling rate distribution value, extract the cooling rate curve, construct the input vector to input the particle swarm optimization algorithm, calculate the convergence rate according to the objective function and compare the stability to select the curve, and obtain the cooling curve selection result set.

[0120] Based on the node cooling rate distribution values, when dividing node regions, it is necessary to classify the regions according to both spatial location and cooling rate characteristic values. Initially, the node space is divided into four zones at 30mm intervals. Combining this with the aforementioned cooling rate interval definition, the proportion of cooling rate data points in each zone is statistically analyzed. If more than 60% of the data points in a certain zone are in the medium-speed cooling range, then that zone is defined as the medium-speed zone, and so on. The regional cooling state definitions are then numbered, for example, zone A is low-speed, zone B is medium-speed, zone C is medium-speed, and zone D is high-speed cooling. Next, the cooling rate curves for each zone under the corresponding time series are extracted to form an array.

[0121] Curve A Curve D ;

[0122] Next, the input vector for each cooling curve is constructed, using the time series as the dimension. The corresponding input vector is as follows:

[0123] ;

[0124] Taking curve A as an example, , Next, after the vector input, a curve comparison task needs to be performed. By setting an objective function, such as defining curve stability as the ratio of the absolute value of the maximum first derivative to the average derivative, and setting the judgment threshold to 1.5, if a curve has a maximum slope of 0.03 and an average slope of 0.02, then its stability is 1.5, which just reaches the threshold boundary value and meets the screening criteria. Each curve is evaluated according to standards such as time gradient and cooling uniformity. A sorting operation is used to select the top 50% of curves to form the screening result set. For example, the screening result curve set might be:

[0125] {Curve A, Curve C};

[0126] This step yields a set of cooling curve screening results.

[0127] S303: Based on the time point and cooling rate of the curve in the cooling curve screening result set, combined with the corresponding temperature and holding time in the task parameter list, a cooling control parameter vector is constructed, and the particle swarm optimization algorithm is called to fit and adjust it to generate a cooling parameter optimization scheme.

[0128] Based on the cooling curve filtering result set, read the cooling rate value of the curve at the time node, and align it with the temperature and holding time set in the task parameter list. For example, the time series of curve C is... The cooling rate sequence is If the heat preservation time in the task list is 45 minutes and the pouring temperature is 1530℃, then construct the parameter vector:

[0129] Input vector ;

[0130] This input vector is used in subsequent calculations and requires parameter correction by setting a target control value. Assume the system sets a baseline value for the average cooling rate. When the current average cooling rate is The deviation is 0.0125. This can be corrected by adjusting the cooling fan power parameters. Mold cooling channel flow rate The input parameters are then adjusted. Assuming the current cooling fan power is 60% and the cooling channel flow rate is 1.2 L / min, the system can be set to a linear correction rate of 0.01℃ / s for every 5% increase in fan power. Therefore, to adjust the average cooling rate to 0.38, it needs to be reduced by 0.0125, i.e., a 6.25% reduction in fan power. The optimized output cooling parameters are as follows:

[0131] ;

[0132] This adjustment will be saved in the task control parameter set, forming a modified configuration for a specific cooling curve.

[0133] Please see Figure 5 The specific steps of S4 are as follows:

[0134] S401: Based on the number of cooling stages, stage cooling time and target temperature set in the cooling parameter optimization scheme, set the combination parameters of the cooling medium's injection temperature, flow rate and injection angle, record the surface temperature change and duration of the casting during each stage of cooling, and generate a stage temperature time series data set.

[0135] Based on the cooling parameter optimization scheme, which sets the number of cooling stages, stage cooling time, and target temperature, the combined parameters of the cooling medium's injection temperature, flow rate, and injection angle are sequentially set for each cooling stage. The number of cooling stages is divided into four segments according to the total cooling time of the casting and the required stage cooling rate variation. Each segment corresponds to a target surface temperature, such as setting the target temperature for the first stage to 650℃, the second stage to 480℃, the third stage to 300℃, and the fourth stage to 200℃, with corresponding stage times of 60s, 90s, 120s, and 180s, respectively. In actual execution, based on the target temperature and stage time for each stage, the injection temperature is set to the initial exit temperature of the cooling medium, specifically 20℃. The injection flow rate is set in segments based on the heat exchange area of ​​the casting surface and the target cooling rate at 15℃, 10℃, and 10℃. For example, with a heat exchange area of ​​1.2m², the corresponding flow rates for each stage are 12L / min, 14L / min, 10L / min, and 8L / min, respectively. The injection angle is adjusted according to the surface structure characteristics and cooling uniformity requirements, and is set to 45°, 60°, 75°, and 60°, respectively. During the cooling process, the surface temperature is recorded in real time using surface thermocouples, once every 1 second. The recording results for each stage generate a set of surface temperature and time-corresponding data points, which are numbered and archived in the data acquisition system to form a temperature time-series data group for each stage. The data example is shown in the table below:

[0136] Table 4: Cooling Stage Parameters and Surface Temperature Record Table

[0137]

[0138] As shown in Table 4, during the entire cooling process, by setting the spray conditions and target temperature according to the stages and continuously recording the surface temperature data of the casting, the temperature drop changes and corresponding durations at different stages can be obtained, forming four sets of stage temperature time sequence data.

[0139] S402: Based on the stage temperature time series data group, extract the stage surface temperature and duration, detect the surface hardness value, calculate the response amplitude of hardness value and temperature drop rate, screen the interval where the hardness change amplitude is greater than the hardness response threshold, and obtain the rate change sensitive interval value.

[0140] Based on the time-series temperature data, the average temperature and duration of each stage are extracted. The time-temperature change curves are then subjected to point-by-point differencing to calculate the rate of temperature decrease per unit time. In the example, the second stage is used, where the surface temperature drops from 652℃ to 480℃ in a total time of 90 seconds. Therefore, its average temperature decrease rate is... The surface hardness of the casting after cooling at this stage was tested. Ten hardness testing points were arranged at different locations on the casting, and the test values ​​were recorded as Brinell hardness (HB). Example test values ​​are [168, 170, 166, 172, 165, 168, 169, 170, 171, 167]. The average value was calculated to be 168.6 HB. Then, the response amplitude of the temperature drop rate to the surface hardness was calculated. The method was to normalize the average hardness value of the stage and the rate point-to-point and then calculate the difference to obtain the response amplitude value sequence. The intervals where the hardness change amplitude is greater than the set hardness response threshold are selected. The threshold is set to 0.15 (normalized difference). Under this condition, the amplitude difference between the first stage (rate 1.13℃ / s, average hardness 150HB) and the second stage (rate 1.91℃ / s, average hardness 168.6HB) is 0.18, which meets the threshold condition. This stage interval is determined to be the rate change sensitive interval. This judgment process requires comparing the rate and hardness responses of multiple stages one by one, and comparing the results with the threshold one by one.

[0141] S403: Based on the sensitive interval value of rate change, extract the cooling intensity adjustment parameter, compare the degree of deviation between the adjusted cooling intensity and the critical demolding temperature, and perform weighted calculation by combining the degree of deviation and the stage time factor to generate the demolding timing judgment parameter.

[0142] Based on the rate change sensitive range, the cooling intensity parameters used in the corresponding stage are extracted, namely the combined set values ​​of spray temperature, flow rate, and angle. Taking the second stage as the sensitive range, the extracted parameters are a spray temperature of 15℃, a flow rate of 14L / min, and a spray angle of 60°. These parameters are then offset from the set value of the demolding critical temperature. If the demolding critical temperature is set to 180℃, and the final temperature after stage cooling is 200℃, then the offset value is... The system determines whether the offset value is within the allowable range. Assuming the upper limit of the allowable offset is 30℃, the current offset value of 20℃ falls within the allowable range. Next, it weights the offset by a stage time factor, which is defined as the proportion of the second stage cooling time to the total cooling time. The offset value is multiplied and weighted by the time factor, and the result is... As a parameter for determining the demolding timing, if the threshold value of the determination parameter is set to 5, then the current result 4.0 does not exceed the threshold value, and it is determined that this stage cannot be used as a direct demolding node. In subsequent stages, the temperature will continue to be monitored, and the above determination process will be repeated.

[0143] Please see Figure 6 The specific steps of S5 are as follows:

[0144] S501: Based on the demolding timing determination parameters, collect the temperature change trend during the molding stage, monitor the curing rate and stress distribution range, determine the state conditions based on the critical deformation coefficient, structural stability threshold and stress critical coefficient, and generate demolding matching judgment value.

[0145] Based on the parameters for determining the demolding timing, temperature monitoring points at the start and end points of the curing process are first determined. Thermocouples or infrared temperature sensors are placed on the inner and outer surfaces of the mold, and continuous temperature data is collected at key locations such as the center, corners, and thick-walled areas. A 5-minute interval is set as a sampling cycle to obtain the temperature change trend during the molding stage. Subsequently, the temperature data at the monitoring points within the molding cycle are subjected to curve fitting and first-order difference processing to obtain the temperature rise rate and stable time period. Then, combined with the exothermic reaction characteristics of thermosetting inorganic mineral materials such as magnesium phosphate cement in the range of 80℃ to 180℃, the local temperature increase and maximum temperature difference are calculated to determine the curing rate. The slope of the temperature-time curve is compared with the critical rate (e.g., 1.2℃ / min). If the rate exceeds the threshold, it is considered a rapid curing area. Further, the temperature difference between monitoring points is matched with the geometric boundary mapping area of ​​the mold. Combined with data collected by stress sensors, such as the strain and stress distribution at the monitoring points, the area is divided to determine the structural stress concentration areas. The non-uniform cooling trend is assessed by comparing the set critical deformation coefficient (e.g., material linear expansion difference exceeding 0.0025), structural stability threshold (e.g., residual stress exceeding 20MPa), and stress critical coefficient (e.g., critical ratio reaching 1.35) item by item. The stability requirements are determined using Boolean judgment for each of the three parameters. Regions meeting at least two of the conditions are defined as matching critical demolding regions. These regions are assigned numerical levels (e.g., 0 for unsuitable, 1 for partially suitable, 2 for suitable). A weighted average is then used to generate the demolding matching judgment value for the current overall casting. In the central region, the temperature rise rate is 1.5℃ / min, the peak stress is 18MPa, and the temperature difference is 12℃. The calculated critical deformation coefficient is 0.0023, below the threshold, while the stress value does not exceed the critical value. The structural stability coefficient is approximately 1.28, thus the region meets two standard conditions and is assigned a value of 2. The average demolding matching value for the entire workpiece is approximately 1.6, indicating a tendency towards a suitable demolding state.

[0146] S502: Based on the demolding compatibility judgment value, calculate the demolding temperature setpoint, call the deviation rate between the demolding temperature setpoint and the demolding compatibility judgment value, perform segmented calculations with the heating rate value, adjust the temperature change range, and generate a staged demolding temperature increment coefficient.

[0147] Based on the demolding matching judgment value, firstly, extract the target demolding temperature setpoint for each monitoring point at the planned demolding time point, such as 140℃ for zone T and 135℃ for the boundary zone T. Then, retrieve the demolding matching judgment value for zone J, which is 1.6. Compare the deviation rate between the target setpoint and the judgment value, using the formula: Deviation Rate = To perform the calculation, taking the central area as an example, if... , The deviation rate is This indicates that the current set temperature is too low. Next, the heating rate value for this region during the last 5 minutes of molding is retrieved. For example, if the actual heating rate is 1.1℃ / min, the deviation rate and heating rate are matched using a piecewise function. For instance, a deviation rate < -0.1 and a heating rate < 1.2℃ / min are defined as requiring a significant temperature increase, so the set temperature change range is +6℃. After segmented calculation, a staged demolding temperature increment coefficient is generated. Substitute into the formula:

[0148] ;

[0149] in, Represents the coefficient of temperature increment during the demolding process. This represents the heating weight parameter for the i-th segment. This represents the demolding temperature setting value for the i-th section. This represents the demolding matching value of the i-th segment. This represents the global temperature offset correction factor. The average value of the demolding temperature setpoint for the representative section. The average value of the demolding matching judgment value of the representative section. Represents the total number of segments. This represents the overall adjustment coefficient.

[0150] The formula is explained in detail below: This is the heating weight for the i-th segment. For example, if segment T is thicker, it is set to 1.2, and segment J is set to 1.1. Set the temperature (e.g., 140℃, 138℃, 135℃). The judgment value is converted to temperature (e.g., 160℃, 150℃, 145℃). The global correction factor is set to 0.8. , , , The calculation process is as follows:

[0151] ;

[0152] ;

[0153] ;

[0154] Obtain the staged demolding temperature increment coefficient The value of ℃ indicates that the demolding temperature needs to be increased by approximately 16.22℃ from the current setting, requiring further adjustments to the demolding strategy.

[0155] S503: Call the staged demolding temperature increment coefficient, correct the original set value according to the temperature adjustment coefficient, adjust the temperature range of the region in combination with the boundary stress value, fit the parameter relationship, and obtain the optimization scheme of inorganic mineral casting processing parameters.

[0156] The calculated staged demolding temperature increment coefficient is called, and the current demolding temperature setpoint sequence is extracted from the parameter control unit. The setpoint is then corrected, that is, the original setpoint temperature of each zone is added to... If the original temperature was 140℃, the corrected temperature would be 156.22℃. This is a regional temperature adjustment. Subsequently, the boundary stress value database is retrieved, and the stress values ​​collected at the corresponding monitoring points in the boundary area (e.g., the stress at the mold edge is 22 MPa) are compared with the corrected temperature in the central region to form a temperature gradient difference. , Represents the temperature gradient between the boundary and the center. Represents the temperature of the boundary region. The temperature represents the central region. If it exceeds 10℃, such as 142℃ after edge correction, the temperature difference is -14.22℃, indicating a tendency for non-uniform stress accumulation in this region. By adding a heating power coefficient to correct the boundary heating rate, the temperature difference is controlled within ±8℃, ensuring synchronous heating. Then, the functional relationship between temperature and stress is fitted, for example, using a linear fitting model. Calculate the temperature recovery value under the influence of boundary stress and fit it. , If the target stress is 20 MPa, then the target temperature is calculated as follows: The temperature is adjusted again based on the actual deviation. Combining the above regional adjustments and global fitting processing, a set of temperature control strategies covering all monitoring points is formed. This strategy corresponds to the processing optimization parameter table, as shown below:

[0157] Table 5: Optimization Table of Processing Parameters

[0158]

[0159] As shown in Table 5, by introducing phased temperature increments and a regional stress temperature difference correction mechanism, a set of temperature parameter optimization schemes suitable for the whole life cycle processing control of inorganic mineral castings has been formed.

[0160] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for optimizing machining parameters of inorganic mineral castings based on real-time data analysis, characterized in that, Includes the following steps: S1: Real-time detection of silicate content and oxide component ratio in inorganic mineral melt using a spectrometer, monitoring mineral phase transformation temperature and crystallization rate, inputting monitoring data into a neural network model for component characteristic analysis, and generating component characteristic analysis results; S2: Based on the analysis results of the component characteristics, obtain the pouring temperature range and holding time, detect the mold preheating temperature data and melt viscosity changes, call the crystallization precipitation rate to calculate the mold filling time, and generate a pouring parameter configuration information table by analyzing the matching relationship between melt viscosity changes and pouring speed. S3: Call the pouring temperature range and holding time in the pouring parameter configuration information table, monitor the internal temperature distribution data and cooling rate changes of the casting in real time, input the monitoring data into the particle swarm optimization algorithm for optimization calculation, screen the cooling curve and verify it, and generate a cooling parameter optimization scheme. S4: Obtain the target temperature and duration of each stage through the cooling parameter optimization scheme, detect the surface hardness of the casting, adjust the cooling intensity of the multi-stage stage according to the change of cooling rate, calculate the degree of deviation between the adjusted cooling intensity and the critical demolding temperature, and generate demolding timing judgment parameters. S5: Determine the demolding time point based on the demolding timing judgment parameters, calculate the demolding temperature setpoint, judge the demolding conditions by comparing the demolding timing judgment parameters with the standard demolding index threshold, adjust the demolding temperature setpoint according to the judgment result, and output the inorganic mineral casting processing parameter optimization scheme. The optimization scheme for inorganic mineral casting processing parameters includes demolding temperature, recommended demolding time, and parameter adjustment scheme.

2. The method for optimizing the machining parameters of inorganic mineral castings based on real-time data analysis according to claim 1, characterized in that, The critical demolding temperature is determined through thermodynamic simulation calculations based on the casting material and predetermined demolding safety standards. The component characteristic analysis results include material compatibility grade, crystallization precipitation rate, thermodynamic stability coefficient and component characteristic label. The casting parameter configuration information table includes casting temperature range, holding time and casting rate. The cooling parameter optimization scheme includes cooling temperature curve, cooling duration and cooling medium flow rate. The demolding timing determination parameters include demolding temperature range, critical temperature difference and surface hardness threshold.

3. The method for optimizing the machining parameters of inorganic mineral castings based on real-time data analysis according to claim 1, characterized in that, The specific steps of S1 are as follows: S101: The reflectance, transmittance and absorptivity values ​​of inorganic mineral melts are detected in real time by a spectrometer. Based on the spectral characteristics, silicate distribution data are extracted, and the oxide ratio is calculated by combining the elemental response spectral line intensity to generate oxide component content data. S102: Based on the oxide component content data, monitor the melt spectrum changes, calculate the crystal formation rate based on the intensity peak shift, absorption threshold change and characteristic bandwidth shrinkage rate, and obtain the set of crystallization response change parameters; S103: The set of crystallization response change parameters and oxide component content data are called to construct a feature vector sequence. In the neural network model, the features are mapped through weight coefficients. The corresponding intensity range of components and parameter conditions is calculated according to the matching degree scoring rules to generate component characteristic analysis results.

4. The method for optimizing the machining parameters of inorganic mineral castings based on real-time data analysis according to claim 3, characterized in that, The specific steps of S2 are as follows: S201: Based on the analysis results of the component characteristics, obtain the upper and lower limits of the pouring temperature and the heat preservation time, detect the mold preheating temperature and screen the data samples that meet the boundary difference conditions, map them with the corresponding heat preservation time parameters and perform a combination operation to generate temperature and heat preservation matching interval values. S202: Call the temperature holding matching interval value and the viscosity change sequence of the melt collected under multiple holding time conditions, extract key nodes and calculate the node slope difference, compare the change amplitude with the mold filling boundary value, and obtain the viscosity evolution rate value. S203: Construct a rate matching interval table based on the viscosity evolution rate value and the pouring speed curve, filter the speed data within the viscosity boundary range, analyze the speed offset and establish the filling time segment, integrate the associated heat preservation time, pouring temperature and mold preheating temperature parameters, and generate a pouring parameter configuration information table.

5. The method for optimizing the machining parameters of inorganic mineral castings based on real-time data analysis according to claim 4, characterized in that, The temperature insulation matching range value is a parameter range obtained by combining the preset upper and lower limits of pouring temperature and insulation time parameters, and in conjunction with the boundary conditions of mold preheating temperature. The viscosity evolution rate value is a quantitative parameter obtained by calculating the slope difference between key nodes based on the viscosity change sequence of the melt collected under multiple sets of heat preservation time conditions.

6. The method for optimizing the machining parameters of inorganic mineral castings based on real-time data analysis according to claim 4, characterized in that, The specific steps for S3 are as follows: S301: Call the pouring temperature and holding time set in the pouring parameter configuration information table, bind the current pouring task number, collect the real-time temperature data of the temperature control node, record the node temperature change and calculate the local temperature gradient and cooling rate to obtain the node cooling rate distribution value. S302: Divide the node region according to the node cooling rate distribution value, extract the cooling rate curve, construct the input vector input to the particle swarm optimization algorithm, calculate the convergence rate according to the objective function and compare the stability screening curve to obtain the cooling curve screening result set. S303: Based on the time points and cooling rates of the curves in the cooling curve screening result set, and combined with the corresponding temperatures and holding times in the task parameter list, construct a cooling control parameter vector, call the particle swarm optimization algorithm to fit and adjust, and generate a cooling parameter optimization scheme.

7. The method for optimizing the machining parameters of inorganic mineral castings based on real-time data analysis according to claim 6, characterized in that, The specific steps of S4 are as follows: S401: Based on the number of cooling stages, stage cooling time and target temperature set in the cooling parameter optimization scheme, set the combination parameters of the injection temperature, flow rate and injection angle of the cooling medium, record the surface temperature change and duration of the casting during each stage of cooling, and generate a stage temperature time series data set. S402: Based on the stage temperature time series data group, extract the stage surface temperature and duration, detect the surface hardness value, calculate the response amplitude of hardness value and temperature drop rate, filter the interval where the hardness change amplitude is greater than the hardness response threshold, and obtain the rate change sensitive interval value. S403: Based on the rate change sensitive interval value, extract the cooling intensity adjustment parameter, compare the degree of deviation between the adjusted cooling intensity and the demolding critical temperature, and perform weighted calculation by combining the degree of deviation and the stage time factor to generate demolding timing determination parameter.

8. The method for optimizing the machining parameters of inorganic mineral castings based on real-time data analysis according to claim 7, characterized in that, The specific steps of S5 are as follows: S501: Based on the demolding timing determination parameters, collect the temperature change trend during the molding stage, monitor the curing rate and stress distribution range, determine the state conditions based on the critical deformation coefficient, structural stability threshold and stress critical coefficient, and generate a demolding matching judgment value. S502: Based on the demolding matching judgment value, calculate the demolding temperature setpoint, call the deviation rate between the demolding temperature setpoint and the demolding matching judgment value, perform segment calculation with the heating rate value, adjust the temperature change range, and generate a staged demolding temperature increment coefficient. S503: Call the staged demolding temperature increment coefficient, correct the original set value according to the temperature adjustment coefficient, adjust the temperature range of the region in combination with the boundary stress value, fit the parameter relationship, and obtain the optimization scheme of inorganic mineral casting processing parameters.

9. The method for optimizing the machining parameters of inorganic mineral castings based on real-time data analysis according to claim 8, characterized in that, The demolding compatibility judgment value refers to the index value calculated based on the temperature change, curing rate and stress distribution parameters during the molding process, combined with the critical deformation coefficient and structural stability threshold, to determine whether the casting is suitable for demolding. The staged demolding temperature increment coefficient refers to the adjustment coefficient for correcting the demolding temperature set value in stages, based on the deviation rate between the demolding matching judgment value and the demolding temperature set value and the heating rate. The critical deformation coefficient is obtained by finite element simulation of the deformation response of the casting under different temperature and stress conditions. The corresponding deformation coefficient is extracted as the critical value by statistical analysis of the parameter points where deformation occurs. The structural stability threshold is determined by applying the safety factor method based on the material mechanical properties and original parameter data to determine the limit value at which the casting will become structurally unstable. The critical stress coefficient is determined by analyzing the inflection point of the stress-strain curve to identify the value at which stress accumulation reaches the material's bearing limit, and this value is used as the critical stress coefficient.

Images