Dynamic evaluation and parameter adjustment method for feed mixing uniformity

Abstract

This invention provides a method for dynamic evaluation and parameter adjustment of feed mixing uniformity, relating to the field of data processing technology. The method includes: establishing a material characteristic matrix based on raw material characteristic data; storing the material characteristic matrix in correspondence with the initial operating parameters of the mixing equipment to generate first mixing control data; during the mixing process, collecting local vibration amplitude and temperature rise rate at different spatial locations and combining them with the first mixing control data; calculating the difference in vibration amplitude distribution within adjacent sampling periods during the feed mixing process to obtain a spatial fluctuation index; when the spatial fluctuation index continuously decreases and the temperature rise rate experiences a sudden increase, generating local accumulation indication data and triggering short-term acceleration and reversal operations; executing the feed mixing process; and generating a mixing termination signal when the rate of change of the spatial fluctuation index for three consecutive sampling periods is lower than a preset fluctuation threshold. This invention improves the autonomy and accuracy of feed mixing uniformity adjustment.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method for dynamic evaluation and parameter adjustment of uniformity in feed mixing processes. Background Technology

[0002] Existing feed mixing processes typically achieve uniform material distribution through fixed mechanical mixing parameters (such as stirring speed, mixing time, and material feeding sequence). Specifically, horizontal ribbon mixers or twin-shaft paddle mixers are generally used in production. By setting the mixing time and filling coefficient, the materials are continuously tumbled and sheared within the mixing chamber, thereby achieving macroscopic mixing of powders or granules. The mixing effect is usually evaluated after mixing by sampling and testing the content of key components to calculate the coefficient of variation (CV value) to determine whether the mixing uniformity meets the standard. When the test results deviate from the standard, operators adjust the mixing time or equipment parameters based on experience. This method is widely used in the production of conventional compound feeds or premixes.

[0003] However, when trace additives (such as vitamins, minerals, or pharmaceutical additives at proportions below 0.5%) are present in certain formulations, existing technologies are prone to uneven mixing. Taking the mixing of powdered trace elements with high-density pelleted feed as an example, if the mixing parameters are fixed, the materials may stratify during mixing due to differences in particle size and density. Some light powders may suspend and remain in the upper layer under air turbulence, while heavy pellets will settle to the bottom prematurely. Since existing detection methods can only sample and measure after mixing, they cannot identify segregation trends in real time. When material stratification or powder adsorption onto the inner wall of the equipment occurs, even if the final uniformity meets the standard, subsequent re-separation during unloading or conveying may still lead to localized component deviations, resulting in uneven distribution of effective components in the final product. Summary of the Invention

[0004] The purpose of this invention is to provide a method for dynamic evaluation and parameter adjustment of uniformity in the feed mixing process, aiming to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0006] A method for dynamic evaluation and parameter adjustment of feed mixing uniformity, the method comprising:

[0007] Acquire raw material characteristic data, including particle size distribution, bulk density and moisture content of each raw material, and establish a material characteristic matrix based on the raw material characteristic data;

[0008] The material property matrix is ​​stored in correspondence with the initial operating parameters of the mixing equipment to generate the first mixing control data, which is used to guide the initial operation of the mixing process.

[0009] During the mixing process, vibration sensors and temperature sensors arranged in the mixing chamber collect local vibration amplitude and temperature rise rate at different spatial locations, and combine them with the first mixing control data to form mixing state data;

[0010] Based on the mixed state data, the difference in vibration amplitude distribution between adjacent sampling periods during the feed mixing process is calculated to obtain the spatial fluctuation index, which is used to characterize the uniformity trend of material distribution.

[0011] When the spatial fluctuation index continues to decline and the temperature rise rate shows a sudden increase in stages, local accumulation indication data is generated, and a second mixed control data is formed based on the triggering of short-term acceleration and reversal operations.

[0012] Feed mixing is performed based on the second mixing control data. When the rate of change of the spatial fluctuation index is lower than the preset fluctuation threshold for three consecutive sampling cycles, a mixing termination signal is generated.

[0013] Preferably, a material characteristic matrix is ​​established based on raw material characteristic data, including:

[0014] Based on the raw material characteristic data, the particle size distribution, bulk density and moisture content characteristic values ​​of each raw material are extracted, and the particle size distribution is divided into intervals to obtain the particle size interval ratio data.

[0015] Based on the particle size range ratio data and the bulk density of each raw material, the density difference between raw materials is calculated, and the density difference is used as a weighting factor to correct the ratio value in the particle size range ratio data to generate density-weighted data.

[0016] Based on density-weighted data and the moisture content of each raw material, a flowability correction factor is calculated to characterize the adhesion tendency of the raw materials and form flowability correction data.

[0017] The fluidity correction data is weighted according to the proportion of each raw material in the formula to generate a set of raw material feature vectors.

[0018] Based on the set of raw material feature vectors, calculate the correlation coefficients of any two raw materials in three dimensions: particle size difference, density difference, and moisture content difference, and generate raw material flow related data.

[0019] The raw material flow-related data are arranged in a matrix according to the raw material combination to obtain the flow influence coefficient matrix between raw materials, and then output as the material characteristic matrix.

[0020] Preferably, local vibration amplitude and temperature rise rate at different spatial locations are collected and combined with the first mixed control data to form mixed state data, including:

[0021] Based on the preset sampling point locations, multiple vibration signals are acquired to form raw vibration data;

[0022] The original vibration data is processed by time-segmented sampling, and the peak frequency and average amplitude of each sampling segment are extracted to generate local vibration characteristic data.

[0023] Temperature signals at each sampling point are collected synchronously, and the rate of temperature rise is calculated to generate local temperature change data.

[0024] Based on local vibration characteristic data and local temperature change data, time-domain matching and energy response calculation are performed to generate regional coupled response data;

[0025] Based on the energy response values ​​of each monitoring area in the regional coupling response data, a one-to-one correspondence is made with the equipment operating parameters in the first hybrid control data to form parameter matching data;

[0026] Calculate the normalized difference of each region in the parameter matching data, and use the normalized difference as a dynamic weight to correct the corresponding energy response value to generate weighted energy response data;

[0027] The weighted energy response data is cumulatively averaged over time to obtain mixed-state data.

[0028] Preferably, based on the mixed state data, the difference in vibration amplitude distribution between adjacent sampling periods during the feed mixing process is calculated to obtain the spatial fluctuation index, including:

[0029] Based on the mixed-state data, the vibration amplitude distribution information of multiple consecutive sampling periods is extracted to form a periodic dataset;

[0030] The difference in vibration amplitude between adjacent periods in the periodic dataset is calculated, and a weighted average is performed according to the monitoring area to generate difference data.

[0031] The difference data is smoothed by performing a moving average to eliminate instantaneous disturbances and obtain smoothed difference data;

[0032] The spatial volatility index is generated by calculating the change ratio of the current cycle relative to the previous two cycles based on the smoothed difference data.

[0033] Preferably, when the spatial fluctuation index continues to decrease and the temperature rise rate experiences a sudden increase, local accumulation indication data is generated, and based on the triggering of short-term acceleration and reversal operations, second mixed control data is formed, including:

[0034] Based on the spatial fluctuation index and the temperature rise rate information in the mixed state data, the rate of change of the two parameters within a continuous sampling period is calculated and trend fitting is performed to generate periodic trend data.

[0035] Regional difference analysis is performed on the periodic trend data. When the increase in the temperature rise rate of any region exceeds the preset temperature rise increase ratio threshold of the previous period's average and the corresponding vibration amplitude decreases, local accumulation indication data is generated.

[0036] Based on the local accumulation indication data, the stirring shaft section corresponding to the accumulation area is determined, and a short-time acceleration control signal is output to the section to generate acceleration execution data.

[0037] Based on the accelerated execution data, after the control equipment completes the high-speed stirring for a preset time, it performs a reverse rotation operation to generate disturbance recovery data;

[0038] The coordinates of the stacking area in the stacking indication data are matched with the speed increase in the acceleration execution data to generate stacking response matching data;

[0039] Based on the inversion duration in the disturbance recovery data, the accumulated response matching data is dynamically corrected to form control parameter update data;

[0040] The operating parameter values ​​of the corresponding stacking section in the control parameter update data are replaced with the corresponding parameter items in the first mixed control data to generate the second mixed control data.

[0041] Preferably, based on the raw material feature vector set, the correlation coefficients of any two raw materials in three dimensions—particle size difference, density difference, and moisture content difference—are calculated to generate raw material flow-related data, including:

[0042] Based on the raw material feature vector set, extract the particle size range ratio data of any two raw materials, calculate the difference in each particle size range, and generate particle size difference data.

[0043] Based on the particle size difference data and the bulk density values ​​of the two raw materials, the density ratio is calculated, and the particle size difference data is corrected by using the density ratio as a weighting factor to generate density-weighted data.

[0044] Based on the density-weighted data and the difference in moisture content between the two raw materials, and after normalization, the corrected moisture content data is obtained.

[0045] The moisture content corrected data and density weighted data are normalized separately to generate normalized density data and normalized moisture content data.

[0046] The normalized density data and normalized moisture content data are weighted and summed according to a preset ratio, and the weighted average difference value is calculated to generate a comprehensive difference index.

[0047] Based on the numerical range of the comprehensive difference index, the difference range between each raw material is determined, forming raw material difference characteristic data;

[0048] Based on the raw material difference characteristics data, the correlation coefficients of the two raw materials in three dimensions—particle size, density, and moisture content—are calculated to generate raw material flow correlation data.

[0049] The raw material flow-related data for each pair of raw materials are summarized sequentially to generate a raw material flow-related data set.

[0050] Preferably, the energy response values ​​of each monitoring area in the regional coupling response data are matched one-to-one with the equipment operating parameters in the first hybrid control data to form parameter matching data, including:

[0051] Based on the regional coupling response data, the spatial coordinates and corresponding energy response values ​​of each monitoring area are obtained, and energy response distribution data are generated.

[0052] Based on the energy response distribution data and the equipment operating parameters in the first hybrid control data, a corresponding match is made in the time dimension to form time-aligned data;

[0053] Spatial mapping processing is performed on the time-aligned data. A mapping relationship is established between the spatial coordinates of the monitoring area and the blade number of the stirring device to generate regional mapping data.

[0054] Response difference calculations are performed on the regional mapping data to determine the response correlation between energy response values ​​and operating parameters, and preliminary parameter matching data is generated.

[0055] Based on the response relevance of the initial parameter matching data, the matching accuracy is weighted and corrected to form the final parameter matching data.

[0056] Preferably, the normalized difference degree of each region in the parameter matching data is calculated, and the normalized difference degree is used as a dynamic weight to correct the corresponding energy response value, generating weighted energy response data, including:

[0057] Based on the parameter matching data, the energy response values ​​and corresponding operating parameter values ​​of each monitoring area are normalized to generate energy normalized data and operating parameter normalized data.

[0058] The relative response deviation data for each monitoring area is obtained by calculating the deviation ratio between the energy normalized data and the operating parameter normalized data.

[0059] Based on the maximum and minimum values ​​of the relative response deviation data, interval normalization is performed to generate normalized difference data for each region;

[0060] Based on the normalized difference data, dynamic weight coefficients are assigned to each monitoring area to generate dynamic weight data.

[0061] Based on dynamic weight data, the energy response value is weighted and corrected to generate weighted energy response value data;

[0062] The weighted energy response values ​​of each region are cumulatively averaged over time to generate weighted energy response data.

[0063] Preferably, based on local accumulation indication data, the stirring shaft section corresponding to the accumulation area is determined, and a short-term acceleration control signal is output to that section to generate acceleration execution data, including:

[0064] Based on the local accumulation indication data, extract the spatial coordinate information of the accumulation area to generate accumulation positioning data;

[0065] Based on the stacking location data and mixing chamber structural parameters, the relative positional relationship between the stacking area and each blade on the stirring shaft is analyzed to generate segment mapping data;

[0066] The corresponding stirring shaft segment number is determined based on the segment mapping data, and this segment is set as the control target segment to generate target segment data;

[0067] Based on the number of blades and current rotation speed information in the target section data, calculate the theoretical material shearing amount per unit time for that section and generate shearing amount prediction data.

[0068] Based on the spatial fluctuation index change rate and temperature rise rate increase information contained in the accumulation indication data, the accumulation strength parameters are calculated; and combined with the correspondence with the shear quantity prediction data, the speed increase and acceleration duration of short-term acceleration are determined, and acceleration control parameters are generated.

[0069] Based on the acceleration control parameters, the stirring shaft section is driven to perform a short-term acceleration operation, generating acceleration execution data.

[0070] Preferably, the accumulated response matching data is dynamically corrected based on the inversion duration in the disturbance recovery data to form control parameter update data, including:

[0071] Based on the disturbance recovery data, the reversal duration and the corresponding temperature rise rate are extracted to generate reversal feature data;

[0072] Based on the location of the stacking region and the increase in rotational speed in the inversion feature data and the stacking response matching data, a multi-dimensional correspondence between time, velocity and location is established to generate disturbance correlation data.

[0073] The disturbance-related data is dynamically evaluated. When the reversal duration exceeds the preset equilibrium time threshold, the reversal energy decay coefficient is calculated, and a disturbance correction factor is generated.

[0074] Based on the disturbance correction factor, the speed increase in the stacked response matching data is attenuated or compensated to generate corrected response data.

[0075] The difference analysis is performed between the corrected response data and the accumulated response matching data to extract the updated operating parameter values ​​and generate control parameter update data.

[0076] The above-described solution of the present invention has at least the following beneficial effects:

[0077] First, by introducing quantitative modeling of raw material characteristic data, this invention establishes a material characteristic matrix before mixing begins. This matrix incorporates key physical characteristics such as particle size distribution, bulk density, and moisture content of each raw material into the calculations, enabling the mixing equipment to generate suitable initial control parameters based on the flow characteristics and interaction patterns of different materials. This method overcomes the problem in existing processes where fixed mixing time and speed cannot adapt to differences in the physical properties of different formulations. It allows for targeted adjustment of the mixing process from the start-up stage, reducing the initial formation of uneven phenomena such as light powder retention and heavy particle sedimentation.

[0078] Secondly, by deploying vibration and temperature sensors within the mixing chamber, real-time dynamic monitoring of the mixing process is achieved. The mixing state data constructed in this invention synchronously integrates vibration amplitude and temperature rise rate, reflecting the local motion state and frictional energy changes of materials in different regions. When the material distribution tends to be uniform, the vibration signal tends to stabilize between sampling points; however, in cases of accumulation, stratification, or energy concentration, the local differences in temperature rise rate and vibration amplitude will significantly increase. By capturing and calculating these changing characteristics, the trend of uniformity changes can be perceived in real time during mixing, thus overcoming the limitation of existing technologies that can only evaluate the effect through sampling after mixing.

[0079] Furthermore, by calculating the difference in vibration amplitude distribution between adjacent sampling periods during the mixing process, a spatial fluctuation index is formed, enabling dynamic quantification of the changes in mixing uniformity. This index continuously reflects the spatiotemporal distribution trend of materials within the cavity, providing a basis for determining whether the mixing process is stabilizing. When the spatial fluctuation index continuously decreases and the temperature rise rate experiences a sudden, phased increase, local accumulation indicator data is automatically generated, triggering a short-term acceleration or reverse stirring operation to form secondary mixing control data. This dynamic feedback mechanism can adjust the stirring state in time before accumulation or stratification causes severe segregation, achieving local redispersion and ensuring a balanced spatial distribution of mixing energy.

[0080] Finally, this invention introduces a change rate judgment logic for the spatial fluctuation index into the mixing termination determination. When the fluctuation rate of three consecutive sampling periods is lower than a preset threshold, a mixing termination signal is automatically output, avoiding deviations caused by subjective experience in determining the mixing time. Taking the mixing of powdered vitamins and high-density grain particles as an example, traditional processes often require extending the mixing time based on experience to ensure uniformity. However, this invention can automatically terminate the operation after detecting that the mixing state is stabilizing, which not only prevents material breakage and excessive temperature rise caused by over-stirring, but also reduces energy consumption and equipment wear. Through the above steps, this invention achieves closed-loop control from material characteristic modeling and process state identification to adaptive parameter adjustment, enabling the feed mixing process to maintain stable uniformity while improving production efficiency and product quality consistency. Attached Figure Description

[0081] Figure 1 This is a flowchart of a method for dynamic evaluation and parameter adjustment of feed mixing uniformity provided in an embodiment of the present invention. Detailed Implementation

[0082] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0083] like Figure 1 As shown, embodiments of the present invention propose a method for dynamic evaluation and parameter adjustment of feed mixing uniformity, the method comprising:

[0084] Acquire raw material characteristic data, including particle size distribution, bulk density and moisture content of each raw material, and establish a material characteristic matrix based on the raw material characteristic data;

[0085] The material property matrix is ​​stored in correspondence with the initial operating parameters of the mixing equipment to generate the first mixing control data, which is used to guide the initial operation of the mixing process.

[0086] During the mixing process, vibration sensors and temperature sensors arranged in the mixing chamber collect local vibration amplitude and temperature rise rate at different spatial locations, and combine them with the first mixing control data to form mixing state data;

[0087] Based on the mixed state data, the difference in vibration amplitude distribution between adjacent sampling periods during the feed mixing process is calculated to obtain the spatial fluctuation index, which is used to characterize the uniformity trend of material distribution.

[0088] When the spatial fluctuation index continues to decline and the temperature rise rate shows a sudden increase in stages, local accumulation indication data is generated, and a second mixed control data is formed based on the triggering of short-term acceleration and reversal operations.

[0089] Feed mixing is performed based on the second mixing control data. When the rate of change of the spatial fluctuation index is lower than the preset fluctuation threshold for three consecutive sampling cycles, a mixing termination signal is generated.

[0090] In this embodiment of the invention, by dynamically sensing and adjusting parameters in real time the material characteristics, operating status, and environmental changes during the feed mixing process, dynamic evaluation and control of uniformity can be achieved during mixing, making the mixing results more stable and reliable. This method first establishes a material characteristic matrix by acquiring characteristic data such as particle size distribution, bulk density, and moisture content of the raw materials, providing basic information on the flowability and interaction between raw materials during the mixing process. By storing the material characteristic matrix in correspondence with the initial operating parameters of the mixing equipment, first mixing control data for initial control is formed, enabling differentiated start-up control of the mixing process based on raw material differences, avoiding uneven energy utilization caused by a single parameter setting.

[0091] During the mixing process, vibration and temperature sensors arranged within the mixing chamber collect real-time data on local vibration amplitude and temperature rise rate, obtaining material motion state data at different spatial locations. By combining this monitoring data with the initial mixing control data, the resulting mixing state data reflects the dynamic distribution of materials and energy transfer characteristics within the mixing chamber, providing a quantifiable basis for subsequent uniformity calculations. Based on the mixing state data, the difference in vibration amplitude distribution between adjacent sampling periods is calculated to obtain a spatial fluctuation index, which characterizes the mixing uniformity trend. This index reflects the spatiotemporal variation characteristics of material distribution; when its variation slows down, it indicates that the material tends to be stably distributed in each region, thus serving as a basis for judging mixing uniformity.

[0092] During the mixing process, if the spatial fluctuation index continues to decrease while the temperature rise rate experiences a sudden, intermittent increase, it can be determined that local accumulation may occur. At this point, local accumulation indication data is generated, and the operating state of the stirring shaft is adjusted by triggering short-term acceleration and reversal operations, causing the material in the accumulation area to redisperse. This process achieves dynamic feedback adjustment by generating second mixing control data, enabling the mixing equipment to autonomously correct parameters based on actual state changes, ensuring a more balanced energy distribution within the mixing chamber.

[0093] When the rate of change of the spatial fluctuation index is lower than the preset fluctuation threshold for three consecutive sampling periods, it indicates that the distribution of materials in the mixing chamber has reached a stable state, and the mixing process is nearing completion. At this point, a mixing termination signal is automatically generated to stop the mixing equipment, thereby avoiding energy waste and excessive material temperature rise caused by over-stirring.

[0094] Taking a feed premix production scenario as an example, the particle size distribution, bulk density, and moisture content of corn flour, soybean meal flour, and vitamin carrier powder are first collected to establish a material characteristic matrix. Based on the first mixing control data generated by the matrix, the initial speed and stirring direction of the mixing equipment are set. During the mixing process, sensors arranged in the mixing chamber continuously monitor the vibration amplitude and temperature rise rate at various locations, calculate the spatial fluctuation index, and adjust the speed in real time. When a region is detected to have excessively rapid temperature rise and weakened vibration, a short-term acceleration and reversal operation is performed to eliminate accumulation. Finally, when the spatial fluctuation index stabilizes within the threshold range for three consecutive cycles, a mixing termination signal is output, completing the mixing process and achieving a control effect of uniform mixing and reasonable energy consumption.

[0095] In this embodiment of the invention, a preset fluctuation threshold is an important parameter for judging the stability of the mixing process. This threshold represents the lower limit of the allowable range of the spatial fluctuation index change rate, used to determine whether the mixing uniformity has reached the termination condition. This threshold can be set through empirical experiments or historical production data, specifically by statistically analyzing the average change rate range of the spatial fluctuation index when it tends to stabilize under different formulations, particle size combinations, and equipment operating conditions. The system inputs this value into the control module during initial operation as a fixed reference for judging mixing termination. When the fluctuation index change rate is less than this threshold for three consecutive sampling periods, it indicates that the material distribution fluctuation has entered a stable range, triggering a mixing termination signal.

[0096] In a preferred embodiment of the present invention, establishing a material characteristic matrix based on raw material characteristic data includes:

[0097] Based on the raw material characteristic data, the particle size distribution, bulk density and moisture content characteristic values ​​of each raw material are extracted, and the particle size distribution is divided into intervals to obtain the particle size interval ratio data.

[0098] Based on the particle size range ratio data and the bulk density of each raw material, the density difference between raw materials is calculated, and the density difference is used as a weighting factor to correct the ratio value in the particle size range ratio data to generate density-weighted data.

[0099] Based on density-weighted data and the moisture content of each raw material, a flowability correction factor is calculated to characterize the adhesion tendency of the raw materials and form flowability correction data.

[0100] The fluidity correction data is weighted according to the proportion of each raw material in the formula to generate a set of raw material feature vectors.

[0101] Based on the set of raw material feature vectors, calculate the correlation coefficients of any two raw materials in three dimensions: particle size difference, density difference, and moisture content difference, and generate raw material flow related data.

[0102] The raw material flow-related data are arranged in a matrix according to the raw material combination to obtain the flow influence coefficient matrix between raw materials, and then output as the material characteristic matrix.

[0103] In this embodiment of the invention, multidimensional feature analysis and matrix modeling of raw material characteristic data are used to quantitatively characterize the differences in raw material properties during feed mixing, thereby providing accurate basic data support for subsequent mixing control. The process first extracts key physical parameters such as particle size distribution, bulk density, and moisture content of the raw materials, and divides the particle size distribution into multiple intervals, thus constructing a spatial statistical expression of the raw material particle characteristics. By introducing a weighted correction of bulk density based on the particle size interval ratio, the difference in compactness between particles can be reflected in the ratio value, making the obtained density-weighted data more representative of the actual material mixing response characteristics. Further, combined with moisture content correction, the flowability and cohesiveness of the material are taken into consideration, thereby establishing representative flowability correction data. This data, after weighted aggregation, forms a raw material feature vector set, which can comprehensively reflect the behavioral characteristics of different raw materials in the mixing environment. By calculating the correlation coefficient between any two raw materials in the three-dimensional feature space of particle size, density, and moisture content, the flow correlation data between the raw materials is obtained and output as a material characteristic matrix in matrix form. This matrix can not only describe the flow coupling relationship between raw materials, but also serve as direct input data for setting and dynamically adjusting mixing parameters. It solves the problems of the inability to quantify the differences between raw materials and the lack of physical property basis for control parameters in the traditional feed mixing process, thereby making the mixing control process more accurate and predictable.

[0104] In a preferred embodiment of the present invention, based on the raw material characteristic data, the particle size distribution, bulk density, and moisture content characteristic values ​​of each raw material are extracted, and the particle size distribution is divided into intervals to obtain particle size interval ratio data, specifically including:

[0105] In the batching section, particle size distribution data of each raw material is acquired using online or offline detection equipment, such as laser particle size analyzers or sieving methods to determine the proportion of different particle size ranges. The bulk density of the raw materials is obtained by weighing-volume measurement, using the average density of the material in its naturally loose state as the standard. Moisture content can be measured using an infrared moisture analyzer or thermogravimetric analysis. After obtaining the above three types of basic data, the particle size distribution is divided into several intervals in ascending order, with each interval representing a particle size class. Based on the percentage of particles in each interval relative to the total mass, particle size interval proportion data is formed. This data reflects the structural proportion of different particle size components in the material, providing a basis for subsequent density-weighted correction.

[0106] In a preferred embodiment of the present invention, the density difference between raw materials is calculated based on the particle size range ratio data and the bulk density of each raw material, and the density difference is used as a weighting factor to correct the ratio value in the particle size range ratio data to generate density-weighted data, specifically including:

[0107] The bulk density of different raw materials was correlated with their particle size distribution ratios to analyze the impact of density differences on particle flowability. When the density difference between two raw materials is large, stratification may occur during mixing, thus requiring correction of the proportion data. Specifically, the ratio of the bulk density difference of the raw materials to the average density was used as a weighting coefficient to adjust the proportion values ​​of each interval, so that the high-density raw materials were appropriately amplified in the overall proportion, while the low-density raw materials were correspondingly reduced, thereby generating density-weighted data that reflects the true mixing characteristics.

[0108] In a preferred embodiment of the present invention, a flowability correction factor is calculated based on density-weighted data and the moisture content of each raw material to characterize the adhesion tendency of the raw materials, thereby forming flowability correction data, specifically including:

[0109] By comparing the differences in moisture content of various raw materials, the degree of their impact on material flow characteristics is determined. Higher moisture content leads to stronger surface adhesion and greater interparticle friction, making agglomeration more likely during mixing. Therefore, a flowability correction factor is defined, based on a comprehensive evaluation of the relative differences in moisture content and the average value of density-weighted data, to represent the ease of flow of the raw materials during mixing. This correction factor is then applied to the density-weighted data, and its proportions are readjusted to obtain flowability correction data that reflects the true adhesion characteristics of the materials, providing a basis for predicting and controlling mixing uniformity.

[0110] In a preferred embodiment of the present invention, the flowability correction data is weighted according to the proportion of each raw material in the formulation to generate a set of raw material feature vectors, specifically including:

[0111] Based on the feed ratio of each raw material in the formulation design, its corresponding flowability correction data is used as input parameters for weighted calculation. The flowability correction data of each raw material is multiplied by its mass proportion in the formulation to obtain a weighted feature vector. By summing the feature vectors of all raw materials, a raw material feature vector set is formed. This set not only includes material properties such as particle size, density, and moisture content, but also reflects the behavioral differences of each raw material in the mixing system. This set of data will serve as the core input for subsequent calculations of raw material correlations and the establishment of a material characteristic matrix, enabling the mixing equipment to perform more precise control and optimization based on data-driven material characteristics.

[0112] In this embodiment of the invention, the proportions of each raw material in the formula are determined during the feed formulation stage and imported into the system as input conditions for calculating the raw material feature vector set. These preset proportions are derived from formula design standards or settings provided by the automatic feeding system. The system uses these preset weights to correlate the physical properties of different raw materials with their proportions, ensuring that the material characteristic matrix reflects the actual mixing ratio of the formula, thereby improving the targeting and accuracy of mixing control.

[0113] In a preferred embodiment of the present invention, local vibration amplitudes and temperature rise rates at different spatial locations are collected and combined with first mixed control data to form mixed state data, including:

[0114] Based on the preset sampling point locations, multiple vibration signals are acquired to form raw vibration data;

[0115] The original vibration data is processed by time-segmented sampling, and the peak frequency and average amplitude of each sampling segment are extracted to generate local vibration characteristic data.

[0116] Temperature signals at each sampling point are collected synchronously, and the rate of temperature rise is calculated to generate local temperature change data.

[0117] Based on local vibration characteristic data and local temperature change data, time-domain matching and energy response calculation are performed to generate regional coupled response data;

[0118] Based on the energy response values ​​of each monitoring area in the regional coupling response data, a one-to-one correspondence is made with the equipment operating parameters in the first hybrid control data to form parameter matching data;

[0119] Calculate the normalized difference of each region in the parameter matching data, and use the normalized difference as a dynamic weight to correct the corresponding energy response value to generate weighted energy response data;

[0120] The weighted energy response data is cumulatively averaged over time to obtain mixed-state data.

[0121] In this embodiment of the invention, by deploying vibration and temperature sensors within the mixing chamber, real-time monitoring and dynamic data fusion of the mixing state are achieved, constructing a mixing state data system reflecting the energy distribution characteristics of the materials. This process involves acquiring vibration signals from multiple preset sampling points to obtain raw vibration data, which is then segmented over time to extract the peak frequency and average amplitude of each sampling segment, characterizing the local stirring intensity. The synchronously acquired temperature signal, after calculating the temperature rise rate, can be used to reflect changes in thermal energy generated by local friction or accumulation. By performing time-domain matching and energy response calculation on the local vibration and temperature data, the generated regional coupled response data reflects the mixing activity level within each monitoring area. Furthermore, matching the regional coupled response data with equipment operating parameters generates parameter matching data that includes spatial location, energy response, and operating state, providing a complete description of the mixing state in both time and space dimensions. Subsequently, by calculating the normalized difference degree of the parameter matching data and using this difference degree as a dynamic weight to correct the energy response value, weighted energy response data is obtained, thereby achieving adaptive correction of local energy anomalies. This method effectively avoids misjudgments caused by monitoring a single parameter, enabling mixed state data to more realistically reflect material distribution and energy transfer characteristics, providing a quantitative basis for uniformity assessment and control parameter optimization, and improving the accuracy and dynamic feedback capability of mixing process monitoring.

[0122] In a preferred embodiment of the present invention, time-domain matching and energy response calculation are performed based on local vibration characteristic data and local temperature change data to generate regional coupled response data, specifically including:

[0123] During the mixing process, vibration and temperature signals from different spatial locations are collected by vibration and temperature sensors arranged inside the mixing chamber. Since the rotation of the stirring shaft, material tumbling, and local friction do not occur synchronously in time, time-domain matching processing of the two types of signals is necessary. Specifically, the temperature change curve and the vibration amplitude curve are aligned along the time axis, and their response characteristics are analyzed within the same time window. When the local vibration intensity increases, the temperature in the corresponding area usually shows an upward trend; by comparing the correlation between the temperature rise rate and the vibration amplitude change rate, the energy transfer efficiency of that area can be inferred. The energy response value of each area is calculated based on the time-aligned data, i.e., the relative intensity of mechanical motion converted into heat energy per unit time. Combined with the spatial distribution of each monitoring point, regional coupled response data reflecting the degree of local mixing activity is generated. This data can be used to characterize the energy utilization status and material flow intensity of different areas, providing a basis for subsequent mixing uniformity assessment and parameter adjustment.

[0124] In a preferred embodiment of the present invention, the weighted energy response data is cumulatively averaged over time to obtain mixed-state data, specifically including:

[0125] During the mixing process, a set of weighted energy response data is generated for each sampling period, reflecting the instantaneous energy distribution in different monitoring areas. Because the mixing state exhibits periodic fluctuations, single sampling data may be affected by local disturbances or instantaneous impacts; therefore, cumulative averaging over time is necessary. By progressively accumulating the weighted energy response data from multiple consecutive sampling periods and calculating its average value, short-term abnormal signals are smoothed, thus reflecting the overall trend of the material mixing state. As time progresses, if the energy response in each region tends to stabilize and the differences gradually decrease, it indicates that the mixing process is approaching uniformity. The result after time-cumulative averaging is defined as the mixing state data, which comprehensively reflects the energy change patterns in different regions within the mixing chamber.

[0126] In a preferred embodiment of the present invention, based on the mixing state data, the difference in vibration amplitude distribution within adjacent sampling periods during the feed mixing process is calculated to obtain a spatial fluctuation index, including:

[0127] Based on the mixed-state data, the vibration amplitude distribution information of multiple consecutive sampling periods is extracted to form a periodic dataset;

[0128] The difference in vibration amplitude between adjacent periods in the periodic dataset is calculated, and a weighted average is performed according to the monitoring area to generate difference data.

[0129] The difference data is smoothed by performing a moving average to eliminate instantaneous disturbances and obtain smoothed difference data;

[0130] The spatial volatility index is generated by calculating the change ratio of the current cycle relative to the previous two cycles based on the smoothed difference data.

[0131] In this embodiment of the invention, a spatial fluctuation index for describing the mixing uniformity trend is proposed by calculating the difference in vibration amplitude distribution between adjacent sampling periods based on mixed state data, thereby realizing a quantitative assessment of the dynamic uniformity of the mixing process. The method first extracts vibration amplitude distribution information from multiple consecutive sampling periods to construct a periodic dataset to reflect the spatial distribution state of materials at different times. By calculating the difference in vibration amplitude between adjacent periods and weighting the average by monitoring area, the overall difference in material distribution changes can be obtained. Subsequently, a moving average is performed on the difference data to eliminate abnormal fluctuations caused by instantaneous disturbances, making the data change trend smoother and more stable. Finally, by calculating the change ratio of the current period relative to the previous two periods, a spatial fluctuation index is formed, which can comprehensively reflect the changing trend of mixing uniformity over time. When the spatial fluctuation index gradually stabilizes and remains in a low-change range, it indicates that the energy transfer of materials between monitoring areas tends to be balanced, and the mixing state is close to uniform. This method transforms time-series vibration data into a quantifiable trend indicator, shifting the judgment of the uniformity of the mixing process from subjective experience to measurable dynamic assessment. This not only improves the accuracy of mixing termination judgment but also provides a reliable basis for subsequent automatic adjustment and energy efficiency control.

[0132] In a preferred embodiment of the present invention, when the spatial fluctuation index continues to decrease and the temperature rise rate experiences a sudden increase, local accumulation indication data is generated, and based on the triggering of short-term acceleration and reversal operations, second mixed control data is formed, including:

[0133] Based on the spatial fluctuation index and the temperature rise rate information in the mixed state data, the rate of change of the two parameters within a continuous sampling period is calculated and trend fitting is performed to generate periodic trend data.

[0134] Regional difference analysis is performed on the periodic trend data. When the increase in the temperature rise rate of any region exceeds the preset temperature rise increase ratio threshold of the previous period's average and the corresponding vibration amplitude decreases, local accumulation indication data is generated.

[0135] Based on the local accumulation indication data, the stirring shaft section corresponding to the accumulation area is determined, and a short-time acceleration control signal is output to the section to generate acceleration execution data.

[0136] Based on the accelerated execution data, after the control equipment completes the high-speed stirring for a preset time, it performs a reverse rotation operation to generate disturbance recovery data;

[0137] The coordinates of the stacking area in the stacking indication data are matched with the speed increase in the acceleration execution data to generate stacking response matching data;

[0138] Based on the inversion duration in the disturbance recovery data, the accumulated response matching data is dynamically corrected to form control parameter update data;

[0139] The operating parameter values ​​of the corresponding stacking section in the control parameter update data are replaced with the corresponding parameter items in the first mixed control data to generate the second mixed control data.

[0140] In this embodiment of the invention, by jointly analyzing the dynamic trends of the spatial fluctuation index and the temperature rise rate, intelligent identification and adaptive parameter adjustment of local accumulation phenomena in the mixing process are achieved. This method calculates the rate of change of the spatial fluctuation index and the temperature rise rate in a continuous sampling period and performs trend fitting to obtain periodic trend data reflecting the material flow equilibrium and frictional energy changes. When the temperature rise rate in a local area suddenly increases in a short period while the corresponding vibration amplitude decreases, it can identify the presence of material accumulation or local energy stagnation in that area. At this time, local accumulation indication data is generated, and the corresponding control section on the stirring shaft is determined based on its location information. A short-term acceleration signal is output based on the detection data to rapidly increase the rotational speed of the target section, causing the accumulated material to redistribute. Simultaneously, after completing the short-term acceleration operation, reverse rotation is executed, causing the material to recover from disturbance in the local area, thereby accelerating the redistribution of energy between particles. Based on the temperature rise recovery and vibration changes in the accumulation area, accumulation response matching data is further generated and dynamically corrected according to the duration of the reverse rotation, forming control data for updating mixing parameters. This process can automatically identify and correct non-uniform states during the mixing process without relying on human intervention, thereby redistributing mixing energy in space, reducing dead zones caused by accumulation, ensuring the stability and consistency of the mixing effect, and improving the energy utilization efficiency of the mixing process.

[0141] In a preferred embodiment of the present invention, the coordinates of the stacking region in the stacking indication data are matched with the speed increase in the acceleration execution data to generate stacking response matching data, specifically including:

[0142] When accumulation is detected in a certain area, the accumulation detection module outputs the three-dimensional spatial coordinates of the accumulation area. Simultaneously, the acceleration control module calculates the speed increase and acceleration duration based on the degree of accumulation. To achieve targeted local intervention, the coordinates of the accumulation area need to be paired with the corresponding acceleration execution data. Specifically, a mapping relationship between the accumulation location and the blade segment is established using the geometric parameters of the mixing chamber and the distribution information of the stirring shaft blades. Based on this mapping relationship, the accumulation coordinates are associated with the nearest blade number and matched with the corresponding speed increase parameters, thus forming accumulation response matching data. This data reflects the dynamic response strategy corresponding to each accumulation area, providing accurate input for subsequent disturbance recovery and parameter correction.

[0143] In a preferred embodiment of the present invention, the operating parameter values ​​of the corresponding stacking sections in the control parameter update data are replaced with the corresponding parameter items in the first mixed control data to generate the second mixed control data, specifically including:

[0144] After completing a short-term acceleration and reverse stirring in the accumulation zone, the corrected operating parameter values ​​are calculated based on the disturbance recovery data, forming updated control parameter data. This data includes parameters such as the latest rotational speed, stirring direction, acceleration duration, and reverse recovery time of the accumulation zone. By comparing this data with the original first mixing control data, the corresponding operating parameter items for the accumulation zone are replaced and updated. In practice, the control unit corresponding to the spatial location of the accumulation zone in the first mixing control data is first identified, and then the updated data is written to the corresponding location to generate a new parameter set. The updated second mixing control data can be directly called in the next mixing cycle, enabling the mixing equipment to automatically optimize the local stirring strategy based on historical operating results, achieving dynamic self-learning and continuous correction of mixing parameters.

[0145] In a preferred embodiment of the present invention, a preset temperature rise increase ratio threshold is used to identify the thermal characteristic boundary of local accumulation. When the temperature rise rate of a certain area increases significantly compared to the previous sampling period, while the corresponding vibration amplitude decreases, it indicates that compaction or accumulation may exist in that area. This threshold is determined based on the thermal response model of the mixing equipment, by statistically analyzing the average value and standard deviation of the temperature increase under normal mixing conditions, and is set to a multiple of this average value (e.g., 1.5 to 2 times). If the monitored temperature rise increase ratio exceeds this threshold, the accumulation detection logic is triggered. Its function is to avoid misjudging temperature rise fluctuations caused by changes in ambient temperature or instantaneous friction of materials, thereby achieving more accurate accumulation identification.

[0146] In a preferred embodiment of the present invention, based on the raw material feature vector set, the correlation coefficients of any two raw materials in three dimensions—particle size difference, density difference, and moisture content difference—are calculated to generate raw material flow-related data, including:

[0147] Based on the raw material feature vector set, extract the particle size range ratio data of any two raw materials, calculate the difference in each particle size range, and generate particle size difference data.

[0148] Based on the particle size difference data and the bulk density values ​​of the two raw materials, the density ratio is calculated, and the particle size difference data is corrected by using the density ratio as a weighting factor to generate density-weighted data.

[0149] Based on the density-weighted data and the difference in moisture content between the two raw materials, and after normalization, the corrected moisture content data is obtained.

[0150] The moisture content corrected data and density weighted data are normalized separately to generate normalized density data and normalized moisture content data.

[0151] The normalized density data and normalized moisture content data are weighted and summed according to a preset ratio, and the weighted average difference value is calculated to generate a comprehensive difference index.

[0152] Based on the numerical range of the comprehensive difference index, the difference range between each raw material is determined, forming raw material difference characteristic data;

[0153] Based on the raw material difference characteristics data, the correlation coefficients of the two raw materials in three dimensions—particle size, density, and moisture content—are calculated to generate raw material flow correlation data.

[0154] The raw material flow-related data for each pair of raw materials are summarized sequentially to generate a raw material flow-related data set.

[0155] In this embodiment of the invention, by establishing a computational model for the multidimensional characteristic differences between raw materials, the physical interaction relationships of different raw materials during the mixing process are quantitatively expressed, providing a calculable basis for the construction of the material characteristic matrix. This method extracts key parameters such as particle size range ratio, bulk density, and moisture content from the raw material feature vector set. By calculating the difference between any two raw materials in each particle size range, particle size difference data is obtained to characterize the influence of particle size differences on mixing behavior. Furthermore, by introducing a density ratio as a weighting factor, the particle size difference data is corrected to obtain density-weighted data that better reflects physical reality, allowing the data to reflect the compaction characteristics between materials. Combined with the moisture content difference between the two raw materials, moisture content correction data is generated, thus incorporating the material's flowability and binding characteristics into the modeling process. Subsequently, the density-weighted data and moisture content correction data are normalized and weighted to obtain a weighted average difference value and generate a comprehensive difference index, used to characterize the overall interaction degree between raw materials. After determining the difference range between raw materials based on this index, a three-dimensional correlation coefficient can be further calculated to obtain raw material flow-related data. By summarizing the results for all raw material pairs, a raw material flow-related data set is formed, thereby establishing a material property matrix that can be used for modeling the mixing process. This process enables the numerical description of the flow characteristics, particle distribution, and moisture differences of different raw materials, providing precise input for the adaptive control of mixing equipment parameters.

[0156] In a preferred embodiment of the present invention, a density ratio is calculated based on particle size difference data and the bulk density values ​​of the two raw materials, and the particle size difference data is corrected using the density ratio as a weighting factor to generate density-weighted data, specifically including:

[0157] First, preliminary testing yielded particle size difference data and corresponding bulk density values ​​for the two raw materials across various particle size ranges. Since density differences significantly affect particle trajectory and flow behavior during mixing, density ratios are used to correct for these differences. Specifically, the bulk densities of the two raw materials are compared, their relative ratios are calculated, and this ratio is used as a weighting factor to adjust the proportions of each range in the particle size difference data. When the density difference between the two raw materials is large, the material with the higher density has a stronger impact on particle distribution during mixing, thus amplifying its weight in the particle size difference data; conversely, if the densities are similar, the adjustment is relatively smaller. This weighted correction method generates density-weighted data that not only includes information on particle size distribution differences but also comprehensively reflects the distribution trend of materials under gravity and inertia, making the data more closely reflect the actual physical characteristics of the mixing process.

[0158] In a preferred embodiment of the present invention, moisture content correction data is obtained by normalizing the density-weighted data and the moisture content difference between the two raw materials, specifically including:

[0159] First, the average moisture content data of the two raw materials were extracted from the tests, and their difference was calculated to reflect the degree of difference in moisture content. Since moisture content directly affects the friction and adhesion between particles, thus altering their flow state during mixing, a correction term needs to be introduced in the data processing. The moisture content difference was combined with density-weighted data, and the weighted data for each particle size range was corrected proportionally to the moisture content difference. When the moisture content difference is large, the correction magnitude is increased accordingly to reflect the difference in flow resistance caused by moisture. To eliminate the influence of numerical magnitude between different raw materials, the correction results were further normalized, converting all data into a relative numerical range between 0 and 1. The final moisture content correction data reflects both the combined effect of particle size and density and comprehensively considers the changes in the material surface adhesion characteristics, providing a unified standard data basis for subsequent multidimensional difference analysis.

[0160] In a preferred embodiment of the present invention, the degree of difference between each raw material is determined based on the numerical range of the comprehensive difference index, forming raw material difference characteristic data, specifically including:

[0161] After performing particle size, density, and moisture content correction calculations on multiple raw material pairs, a comprehensive difference index is obtained for each pair, characterizing the overall degree of difference between the two raw materials in multidimensional properties. For ease of classification and calculation, the comprehensive difference index is divided into several intervals based on its distribution range, such as high difference, medium difference, and low difference. Each interval corresponds to a different level of physical compatibility: the high difference zone indicates significant mismatch in particle size or density; the medium difference zone indicates good mixability but still requires adjustment of operating parameters; and the low difference zone indicates similar raw material properties and direct mixing. Based on the interval to which each raw material pair belongs, the corresponding difference level is recorded and output as raw material difference characteristic data. This data not only reflects the overall difference between raw materials but also serves as a reference for raw material formulation design and mixing parameter setting, providing structured input for establishing the material property matrix.

[0162] In a preferred embodiment of the present invention, based on the raw material difference characteristic data, the correlation coefficients of the two raw materials in three dimensions—particle size, density, and moisture content—are calculated to generate raw material flow correlation data, specifically including:

[0163] Using raw material difference data as input parameters, the correlation between the two raw materials is calculated across three dimensions: particle size, density, and moisture content. Specifically, in the particle size dimension, the trend of the proportion of different particle size ranges is analyzed; a higher correlation is indicated when the proportions of different particle size ranges change in the same direction, and a lower correlation when they change in opposite directions. In the density dimension, the density distribution gradients of the two raw materials are compared; a consistent distribution trend indicates better mixing and flow coordination. In the moisture content dimension, the similarity of their adhesion trends is judged based on the degree to which the difference in surface moisture affects flowability. Finally, the correlation results from the three dimensions are weighted to obtain a comprehensive correlation coefficient, generating raw material flow correlation data. This data reflects the synergistic flow characteristics of different raw materials during the mixing process, providing a quantitative basis for dynamic control and energy parameter setting of the mixing process.

[0164] In a preferred embodiment of the present invention, the energy response values ​​of each monitoring area in the regional coupling response data are matched one-to-one with the equipment operating parameters in the first hybrid control data to form parameter matching data, including:

[0165] Based on the regional coupling response data, the spatial coordinates and corresponding energy response values ​​of each monitoring area are obtained, and energy response distribution data are generated.

[0166] Based on the energy response distribution data and the equipment operating parameters in the first hybrid control data, a corresponding match is made in the time dimension to form time-aligned data;

[0167] Spatial mapping processing is performed on the time-aligned data. A mapping relationship is established between the spatial coordinates of the monitoring area and the blade number of the stirring device to generate regional mapping data.

[0168] Response difference calculations are performed on the regional mapping data to determine the response correlation between energy response values ​​and operating parameters, and preliminary parameter matching data is generated.

[0169] Based on the response relevance of the initial parameter matching data, the matching accuracy is weighted and corrected to form the final parameter matching data.

[0170] In this embodiment of the invention, by establishing a spatiotemporal correspondence between energy response values ​​and equipment operating parameters, high-precision matching between mixing state data and equipment control information is achieved, providing a foundation for subsequent dynamic weighting and energy correction. This method extracts the spatial coordinates and energy response values ​​of each monitoring area from the regional coupling response data to form energy response distribution data, enabling precise recording of the stirring energy state at different locations within the mixing chamber. This distribution data is then time-aligned with the equipment operating parameters to ensure a one-to-one correspondence between operating parameters and corresponding energy responses within the same sampling period, thereby generating time-aligned data. Subsequently, through spatial mapping processing, the monitoring areas are associated with the blade numbers of the stirring device, linking the response value of each spatial coordinate point to a specific stirring structural unit, generating regional mapping data. By calculating the response differences in the mapping data, the correlation between energy responses and operating parameters in different regions can be assessed, generating preliminary parameter matching data. Further weighted correction of the matching accuracy based on the response correlation yields more accurate final parameter matching data. This process enables the hybrid system to simultaneously capture spatial distribution characteristics and operational state changes, integrating previously separate sensor signals and control parameters into a unified whole. This provides a quantitative basis for subsequent energy response correction and dynamic control, significantly improving the spatiotemporal consistency and control correlation of hybrid state data, and ensuring the accuracy of uniformity assessment and the coordination of hybrid control.

[0171] In a preferred embodiment of the present invention, response difference calculation is performed on the regional mapping data to determine the response correlation between the energy response value and the operating parameters, and preliminary parameter matching data is generated, specifically including:

[0172] During the mixing process, a one-to-one correspondence between monitoring areas and equipment operating components (such as stirring shaft blades or impeller sections) has been established through spatial mapping. For each monitoring area, its energy response value (reflecting the vibration energy intensity or frictional heat change of the material in that area) and corresponding operating parameter values ​​(such as stirring speed, torque, or current) are acquired. By comparing the trends of these two values ​​over the same time period, the influence of operating parameter changes on the energy response is analyzed. When operating parameters increase, if the energy response increases synchronously and with similar magnitudes, the correlation between the two is high; if the energy response changes lagging or inversely, the correlation is low. By comparing the rate of change of energy response with the rate of change of operating parameters over multiple sampling periods, the degree of difference between the two is calculated, and the strength of the correlation is determined based on the magnitude of the difference, thus generating preliminary parameter matching data. This data is used to reflect the adaptive relationship between the energy response of each monitoring area and the equipment operating status, providing a basis for subsequent dynamic correction.

[0173] In a preferred embodiment of the present invention, the matching accuracy is weighted and corrected based on the response relevance of the preliminary parameter matching data to form the final parameter matching data, specifically including:

[0174] After obtaining the initial parameter matching data, a weighted correction is applied to improve the overall matching accuracy. First, based on the response correlation of each monitoring region, it is divided into three levels: high, medium, and low. High correlation regions indicate that changes in operating parameters accurately reflect changes in energy response; medium correlation regions show some delay or deviation; and low correlation regions may be affected by localized accumulation or energy transfer obstacles. Different weighting coefficients are set according to the level, giving high correlation regions a greater weight in subsequent data fusion. Next, the initial matching data is weighted and averaged to make the overall result more representative of the actual operating state. If certain regions exhibit stable high correlation over multiple consecutive cycles, their weight is automatically increased to enhance the temporal continuity of dynamic matching. The final output parameter matching data has higher spatiotemporal consistency, accurately describing the dynamic coupling relationship between energy transfer and equipment operation within the mixing chamber, providing precise input for energy correction and control feedback.

[0175] In a preferred embodiment of the present invention, the normalized difference degree of each region in the parameter matching data is calculated, and the normalized difference degree is used as a dynamic weight to correct the corresponding energy response value, thereby generating weighted energy response data, including:

[0176] Based on the parameter matching data, the energy response values ​​and corresponding operating parameter values ​​of each monitoring area are normalized to generate energy normalized data and operating parameter normalized data.

[0177] The relative response deviation data for each monitoring area is obtained by calculating the deviation ratio between the energy normalized data and the operating parameter normalized data.

[0178] Based on the maximum and minimum values ​​of the relative response deviation data, interval normalization is performed to generate normalized difference data for each region;

[0179] Based on the normalized difference data, dynamic weight coefficients are assigned to each monitoring area to generate dynamic weight data.

[0180] Based on dynamic weight data, the energy response value is weighted and corrected to generate weighted energy response value data;

[0181] The weighted energy response values ​​of each region are cumulatively averaged over time to generate weighted energy response data.

[0182] In this embodiment of the invention, by calculating the normalized difference degree and dynamically weighting the parameter matching data, adaptive adjustment of the uneven local energy distribution in the mixing process is achieved, making the evaluation results of the mixing state more realistic and stable. The method first normalizes the energy response value and the corresponding operating parameter value, converting them into dimensionless relative indicators to eliminate the influence of differences in physical magnitude and units between different monitoring areas. Then, by calculating the deviation ratio between the two, relative response deviation data reflecting the energy utilization efficiency of a local area is obtained. This deviation data reveals the degree of coordination between energy input and material response in each area. When the deviation ratio is large, it indicates that the energy transfer efficiency in that area is low or there is accumulation, requiring dynamic compensation. Based on the interval distribution of the relative response deviation data, the normalized difference degree is calculated to establish a quantitative expression of the energy imbalance in each area. Then, dynamic weights are assigned to each monitoring area based on the normalized difference degree, so that areas with high differences receive a greater influence factor in energy correction. After correcting the energy response value using this weighted data, weighted energy response data is generated, making the overall energy distribution more balanced. Finally, the weighted energy response values ​​are cumulatively averaged over time to effectively suppress the impact of instantaneous fluctuations and ensure stable evaluation results. This process enables a closed-loop energy feedback system from the sensing layer to the control layer, allowing the mixing process to continuously correct energy deviations during operation, thereby improving the accuracy of mixing uniformity calculations and the sensitivity of dynamic control.

[0183] In a preferred embodiment of the present invention, the relative response deviation data of each monitoring area is obtained by calculating the deviation ratio between the energy normalized data and the operating parameter normalized data, specifically including:

[0184] After normalizing the energy response and operating parameter values, two sets of dimensionless data were obtained, facilitating comparison at the same scale. For each monitoring area, the deviation ratio between the normalized energy response and the normalized operating parameters within the same sampling period was calculated to reflect the degree of matching between the energy input and the actual response in that area. When the deviation ratio is close to zero, it indicates high energy transfer efficiency and good input-output matching; when the deviation ratio is large, it indicates a mismatch between energy input and response, possibly caused by local accumulation, voids, or frictional differences. The deviation ratio was continuously tracked for multiple periods to generate relative response deviation data for each monitoring area. This data can intuitively reflect the energy utilization status within the mixing chamber, providing a reliable basis for judging the uniformity of material mixing, and also providing a reference for subsequent allocation of dynamic weights and adjustment of energy distribution strategies.

[0185] In a preferred embodiment of the present invention, dynamic weighting coefficients are assigned to each monitoring area based on normalized difference data to generate dynamic weighting data, specifically including:

[0186] Based on the normalized difference data calculated in the previous stage, the consistency of energy distribution in each monitoring area is assessed. When the normalized difference of a certain area is large, it indicates that its energy response deviates significantly from the overall mixing state, and this area is identified as requiring key adjustment. Dynamic weight coefficients are assigned to each area according to the relative magnitude of the difference, giving higher weights to areas with high differences and lower weights to stable areas. This allocation process can be implemented using linear proportions or interval decreasing methods to ensure that the weight changes are proportional to the degree of energy deviation. Subsequently, these weight coefficients are used to weight and correct the energy response values, giving areas with high differences a larger proportion in the overall energy adjustment, thereby compensating for local energy distribution imbalances. The generated dynamic weight data can be updated in real time to continuously optimize energy distribution and mixing control parameters in subsequent mixing cycles, thereby improving the stability and overall uniformity of the mixing process.

[0187] In a preferred embodiment of the present invention, based on local accumulation indication data, the stirring shaft segment corresponding to the accumulation area is determined, and a short-term acceleration control signal is output to that segment to generate acceleration execution data, including:

[0188] Based on the local accumulation indication data, extract the spatial coordinate information of the accumulation area to generate accumulation positioning data;

[0189] Based on the stacking location data and mixing chamber structural parameters, the relative positional relationship between the stacking area and each blade on the stirring shaft is analyzed to generate segment mapping data;

[0190] The corresponding stirring shaft segment number is determined based on the segment mapping data, and this segment is set as the control target segment to generate target segment data;

[0191] Based on the number of blades and current rotation speed information in the target section data, calculate the theoretical material shearing amount per unit time for that section and generate shearing amount prediction data.

[0192] Based on the spatial fluctuation index change rate and temperature rise rate increase information contained in the accumulation indication data, the accumulation strength parameters are calculated; and combined with the correspondence with the shear quantity prediction data, the speed increase and acceleration duration of short-term acceleration are determined, and acceleration control parameters are generated.

[0193] Based on the acceleration control parameters, the stirring shaft section is driven to perform a short-term acceleration operation, generating acceleration execution data.

[0194] In this embodiment of the invention, by introducing spatial positioning of the accumulation region, blade structure mapping, and shear prediction mechanisms, targeted identification and proactive intervention control of local accumulation phenomena within the mixing chamber are achieved. After detecting local accumulation indication data, the spatial mapping between the accumulation region coordinates and the geometric parameters of the mixing chamber is first used to accurately determine the correspondence between the accumulation position and the stirring shaft blades, generating segment mapping data. This process avoids the reliance on experience-based judgment or fixed-segment acceleration methods in traditional technologies, transforming the control objective from overall speed regulation to localized directional speed regulation, thereby significantly reducing energy waste and ineffective stirring.

[0195] After obtaining the target section, the theoretical shear rate per unit time is calculated using information on the number, size, and rotational speed of the blades, predicting the mixing capacity of that section. Then, combined with the spatial fluctuation index change rate and temperature rise rate increase reflected in the accumulation indication data, the accumulation intensity parameter is calculated, thereby quantifying the severity of accumulation and the level of energy transfer obstruction. Based on the matching relationship between accumulation intensity and shear capacity, the rotational speed increase and acceleration duration are dynamically determined, achieving adaptive generation of short-term acceleration control parameters. Unlike traditional fixed acceleration strategies, this process can adjust energy output in real time according to the accumulation intensity, allowing the equipment to eliminate accumulation while avoiding re-separation caused by excessive disturbance.

[0196] Through the above steps, regionalized, closed-loop control of the mixing equipment during operation is achieved. Once local acceleration is complete, the stirring shaft section immediately returns to its original speed, ensuring both mixing efficiency and energy balance. This method effectively solves the problem of localized deposition or agglomeration of high-density and lightweight materials within the mixing chamber, prevents material stratification, improves overall mixing uniformity and equipment response stability, and provides a dynamically adjustable and automatically executed precise control mechanism for the feed mixing process.

[0197] In a preferred embodiment of the present invention, based on the stacking positioning data and the mixing chamber structural parameters, the relative positional relationship between the stacking region and each blade on the stirring shaft is analyzed to generate segment mapping data, specifically including:

[0198] Upon detecting localized accumulation, the spatial location information of the accumulation area is first obtained from the accumulation detection module. This location information is typically represented in three-dimensional coordinates, including the radial position, axial height, and angular offset from the center of the mixing chamber. Simultaneously, the structural parameter data of the mixing equipment is retrieved, including the geometric dimensions of the mixing chamber, the installation position of the stirring shaft, and the number, length, tilt angle, and spacing of the blades. Next, through geometric mapping, the coordinates of the accumulation area are matched with the blade distribution model of the stirring shaft to determine which blade's effective range or adjacent blades the accumulation point falls within. If the accumulation area covers multiple blade effective zones, the overlap ratio between each blade and the accumulation area is calculated based on the accumulation volume percentage and the location of the accumulation center point. This overlap ratio is then used as an influence weight to generate a one-to-one correspondence between the accumulation area and the blade number. Through this spatial mapping process, segment mapping data is finally output. This data accurately indicates the stirring shaft segment where the accumulation area is located, providing a clear execution target for subsequent short-term acceleration control. Compared to the traditional method of determining the accumulation position based on manual experience, this method achieves automated alignment from sensor detection to structural modeling, enabling the equipment to adjust the stirring intensity of specific areas.

[0199] In a preferred embodiment of the present invention, based on the number of blades and the current rotational speed information in the target section data, the theoretical material shearing amount of the section per unit time is calculated, and shearing amount prediction data is generated, specifically including:

[0200] After determining the target section, the number, shape parameters (such as length, width, and tilt angle), and installation position of the corresponding blades are extracted from the section mapping data. Combined with real-time rotation speed information provided by the equipment operation module, the number of rotations of the blades in that section per unit time and the volume of material within its effective range are calculated. Based on the surface area of ​​the blades in contact with the material during rotation and their trajectory, the shear force generated by each blade per unit time is estimated. The shear force is related not only to the blade geometry but also to the material's flowability, viscosity, and bulk density. Therefore, the theoretical shear value is corrected based on relevant parameters in the material characteristic matrix to make the prediction result closer to reality. Then, the shear forces of all blades in the section are superimposed, and combined with the real-time rotation speed trend, to obtain the comprehensive theoretical shear force of the entire target section per unit time, i.e., the shear force prediction data. This data can be used to assess whether the mixing capacity of the current section is sufficient to eliminate local accumulation. If the predicted shear force is lower than a set threshold, the acceleration amplitude or acceleration duration of that section will be automatically increased, thereby ensuring effective disturbance of the accumulation area and redispersal of the material, improving the dynamic response efficiency of the mixing process.

[0201] In a preferred embodiment of the present invention, the packing strength parameters are calculated based on the spatial fluctuation index change rate and temperature rise rate increase information contained in the packing indication data; and the short-term acceleration speed increase and acceleration duration are determined by combining the correspondence with the shear rate prediction data, thereby generating acceleration control parameters, specifically including:

[0202] During the mixing process, the vibration amplitude and temperature changes in various regions within the mixing chamber are continuously monitored, generating accumulation indication data. This data includes two key parameters: the rate of change of the spatial fluctuation index and the increase in the rate of temperature rise. The rate of change of the spatial fluctuation index reflects the change in the uniformity of material distribution within adjacent sampling periods; when its rate of decline accelerates, it indicates that the material flow is stagnating or localized adhesion has occurred. The increase in the rate of temperature rise characterizes the degree of localized friction or compaction; a sudden increase usually means that the material is being squeezed or accumulated in that area. Correlation analysis of these two signals is performed, and the intensity of accumulation is determined by evaluating the ratio of the decrease in the fluctuation index to the increase in the rate of temperature rise. If both show drastic changes, it is defined as a high-intensity accumulation region; if only the temperature rise changes significantly while the fluctuation index decreases slowly, it is moderate accumulation; if both changes are small, it is slight accumulation. The analysis results are output as an accumulation intensity parameter, used to quantitatively characterize the severity of the accumulation state.

[0203] Subsequently, the packing strength parameter is matched with the previously calculated predicted shear rate. The predicted shear rate reflects the theoretical mixing capacity of the stirring shaft section per unit time, while the packing strength parameter represents the degree of local material congestion. By comparing the relative difference between the two, it is determined whether the current mixing capacity is sufficient to remove the packing. When the shear rate is lower than the packing strength, it is determined that the rotational speed needs to be increased; when the shear rate is close to the packing strength, a slight acceleration is maintained to prevent energy overshoot. Based on the correspondence between the two, the target rotational speed and duration required for short-term acceleration are dynamically calculated. This process comprehensively considers the geometric characteristics of the mixing chamber and the response delay of the stirring shaft to ensure that the acceleration time can effectively break up the packing without causing violent material agitation. Finally, the calculation results are output as acceleration control parameters to drive the controller to perform short-term acceleration operations, realizing quantitative intervention and adaptive energy adjustment for local packing, thereby improving the dynamic uniformity of the mixing process and the stability of equipment operation.

[0204] In a preferred embodiment of the present invention, the accumulated response matching data is dynamically corrected based on the inversion duration in the disturbance recovery data to form control parameter update data, including:

[0205] Based on the disturbance recovery data, the reversal duration and the corresponding temperature rise rate are extracted to generate reversal feature data;

[0206] Based on the location of the stacking region and the increase in rotational speed in the inversion feature data and the stacking response matching data, a multi-dimensional correspondence between time, velocity and location is established to generate disturbance correlation data.

[0207] The disturbance-related data is dynamically evaluated. When the reversal duration exceeds the preset equilibrium time threshold, the reversal energy decay coefficient is calculated, and a disturbance correction factor is generated.

[0208] Based on the disturbance correction factor, the speed increase in the stacked response matching data is attenuated or compensated to generate corrected response data.

[0209] The difference analysis is performed between the corrected response data and the accumulated response matching data to extract the updated operating parameter values ​​and generate control parameter update data.

[0210] In this embodiment of the invention, by dynamically analyzing the reversal duration and temperature rise rate in the disturbance recovery data, adaptive updates of parameters after local accumulation correction in the mixing process are achieved, enabling the control strategy to perform self-learning optimization based on the actual disturbance effect. The method first extracts the duration and corresponding temperature rise rate during the reversal operation to generate reversal feature data, reflecting the energy decay and heat release process after reverse mixing. By correlating this data with the accumulation position and speed increase in the accumulation response matching data, a multi-dimensional correspondence of time, speed, and position is established to obtain disturbance correlation data. Based on this data, the reversal process is dynamically evaluated. When the reversal duration exceeds a preset equilibrium time threshold, the reversal energy decay coefficient is calculated to obtain a disturbance correction factor, which measures the effectiveness of the disturbance operation in restoring energy distribution. Then, based on the disturbance correction factor, the speed increase in the accumulation response matching data is compensated or attenuated to generate corrected response data, reflecting the adaptive adjustment of the mixing to the disturbance response. By performing difference analysis between the corrected response data and the original accumulation response data, updated operating parameter values ​​are extracted and control parameter update data is formed, enabling automatic correction of the control strategy based on historical response conditions. This process enables the mixing to have dynamic optimization capabilities, allowing it to adjust subsequent control parameters according to the effects of disturbances, reduce the probability of repeated accumulation, achieve continuous adaptive optimization of the mixing state and improve energy utilization, thereby improving the control accuracy and process robustness of the entire mixing process.

[0211] In a preferred embodiment of the present invention, the disturbance correlation data is dynamically evaluated. When the reversal duration exceeds a preset equilibrium time threshold, the reversal energy attenuation coefficient is calculated, and a disturbance correction factor is generated. Specifically, this includes:

[0212] After completing the short-term acceleration and reversal operations in the accumulation zone, real-time operational data during the reversal phase is collected, including reversal duration, stirring shaft torque change rate, material temperature rise change rate, and vibration amplitude trend. Analyzing this data allows for the assessment of energy transfer attenuation characteristics within the mixing chamber. When the reversal duration is short, energy is mainly concentrated in the local disturbance area, and the material remains in an unstable redistribution state. However, when the reversal duration exceeds a certain threshold, energy gradually dissipates, leading to a decrease in mixing intensity. Therefore, based on the relationship between energy consumption rate and time, the degree of energy attenuation during reversal is calculated to determine the disturbance correction factor. This correction factor can be understood as a compensation coefficient reflecting the equipment's ability to redistribute materials during the reversal phase, and its value is inversely proportional to the reversal duration and energy attenuation rate. When the reversal energy attenuates too quickly, the disturbance correction factor is increased to compensate for subsequent parameter adjustment errors, thereby ensuring that the mixing process remains stable during the energy recovery phase after disturbance.

[0213] In a preferred embodiment of the present invention, the rotational speed increase in the stacking response matching data is attenuated or compensated according to the disturbance correction factor to generate corrected response data, specifically including:

[0214] After obtaining the disturbance correction factor, it is calculated against the parameters in the stacking response matching data. Using the correction factor as the weight, the rotational speed increase is dynamically adjusted. If the correction factor is less than the preset benchmark value, it indicates that the energy attenuation during the reversal phase is small, and the rotational speed increase is moderately reduced to avoid excessive acceleration leading to material re-separation. If the correction factor is greater than the benchmark value, it indicates that the energy loss during reversal is large, and the rotational speed increase is automatically compensated or the acceleration duration is extended to restore the shear strength of the target section. This correction process can be continuously executed over multiple sampling periods, gradually stabilizing the rotational speed control parameters. The final generated correction response data includes the corrected rotational speed increase, reversal duration, and their dynamic correlation parameters, accurately reflecting the energy compensation status of the equipment during the disturbance recovery phase and providing an accurate basis for parameter updates. This process solves the problem of difficulty in quantifying excessive or insufficient reversal in traditional control, achieving adaptive adjustment of disturbance energy.

[0215] In a preferred embodiment of the present invention, difference analysis is performed between the corrected response data and the accumulated response matching data to extract the updated operating parameter values ​​and generate control parameter update data, specifically including:

[0216] After generating the corrected response data, it is compared and analyzed with the previous stacking response matching data. First, the changes in the same parameters (such as rotational speed, acceleration duration, and reversal time) in the two sets of data are compared, and the difference value of each parameter is calculated. Second, based on the direction and magnitude of the difference value, it is determined whether the operating parameters need to be adjusted: if the value of the corrected response data is higher than the original matching data, it indicates that the disturbance energy in the previous stage was insufficient, and the operating intensity of this section needs to be increased; if the corrected value is lower than the original matching data, it indicates that the reversal was excessive, and the operating power should be appropriately reduced or the reversal time shortened. After analyzing all parameters, the updated values ​​are written into the operating parameter table to form the control parameter update data. Finally, this updated data will replace part of the original first mixing control data to guide the operation of the equipment in the next mixing cycle. Through the above steps, dynamic closed-loop optimization of mixing parameters is achieved, enabling the equipment to automatically correct parameter deviations after each mixing, thereby continuously maintaining the stability of material uniformity in subsequent operations.

[0217] In a preferred embodiment of the present invention, a preset equilibrium time threshold is used to determine whether the duration of the reversal operation is sufficient to restore energy balance within the mixing chamber. When the reversal duration exceeds this threshold, the system considers energy transfer to be stable and material flow to have returned to normal. This threshold is determined by analyzing the reversal energy decay curve of the equipment and is typically set as the time required for the energy response to decrease to 20% to 30% of its initial value. Its purpose is to avoid insufficient disturbance due to excessively short reversals or material re-stratification due to excessively long reversals.

[0218] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for dynamic evaluation and parameter adjustment of feed mixing uniformity, characterized in that, The method includes: Acquire raw material characteristic data, including particle size distribution, bulk density and moisture content of each raw material, and establish a material characteristic matrix based on the raw material characteristic data; The material property matrix is ​​stored in correspondence with the initial operating parameters of the mixing equipment to generate the first mixing control data; During the mixing process, vibration sensors and temperature sensors arranged in the mixing chamber collect local vibration amplitude and temperature rise rate at different spatial locations, and combine them with the first mixing control data to form mixing state data; Based on the mixed state data, the difference in vibration amplitude distribution between adjacent sampling periods during the feed mixing process is calculated to obtain the spatial fluctuation index. When the spatial fluctuation index continues to decline and the temperature rise rate shows a sudden increase in stages, local accumulation indication data is generated, and a second mixed control data is formed based on the triggering of short-term acceleration and reversal operations. Feed mixing is performed based on the second mixing control data. When the rate of change of the spatial fluctuation index is lower than the preset fluctuation threshold for three consecutive sampling cycles, a mixing termination signal is generated.

2. The method for dynamic evaluation and parameter adjustment of feed mixing uniformity according to claim 1, characterized in that, A material property matrix is ​​established based on raw material property data, including: Based on the raw material characteristic data, the particle size distribution, bulk density and moisture content characteristic values ​​of each raw material are extracted, and the particle size distribution is divided into intervals to obtain the particle size interval ratio data. Based on the particle size range ratio data and the bulk density of each raw material, the density difference between raw materials is calculated, and the density difference is used as a weighting factor to correct the ratio value in the particle size range ratio data to generate raw material particle size density weighted data. Based on the particle size and density weighted data of the raw materials and the moisture content of each raw material, a flowability correction factor is calculated to characterize the adhesion tendency of the raw materials and form flowability correction data. The fluidity correction data is weighted according to the proportion of each raw material in the formula to generate a set of raw material feature vectors. Based on the set of raw material feature vectors, calculate the correlation coefficients of any two raw materials in three dimensions: particle size difference, density difference, and moisture content difference, and generate raw material flow related data. The raw material flow-related data are arranged in a matrix according to the raw material combination to obtain the flow influence coefficient matrix between raw materials, and then output as the material characteristic matrix.

3. The method for dynamic evaluation and parameter adjustment of feed mixing uniformity according to claim 1, characterized in that, Local vibration amplitudes and temperature rise rates at different spatial locations are collected and combined with the first mixed control data to form mixed state data, including: Based on the preset sampling point locations, multiple vibration signals are acquired to form raw vibration data; The original vibration data is processed by time-segmented sampling, and the peak frequency and average amplitude of each sampling segment are extracted to generate local vibration characteristic data. Temperature signals at each sampling point are collected synchronously, and the rate of temperature rise is calculated to generate local temperature change data. Based on local vibration characteristic data and local temperature change data, time-domain matching and energy response calculation are performed to generate regional coupled response data; Based on the energy response values ​​of each monitoring area in the regional coupling response data, a one-to-one correspondence is made with the equipment operating parameters in the first hybrid control data to form parameter matching data; Calculate the normalized difference of each region in the parameter matching data, and use the normalized difference as a dynamic weight to correct the corresponding energy response value to generate weighted energy response data; The weighted energy response data is cumulatively averaged over time to obtain mixed-state data.

4. The method for dynamic evaluation and parameter adjustment of feed mixing uniformity according to claim 1, characterized in that, Based on the mixed state data, the difference in vibration amplitude distribution between adjacent sampling periods during the feed mixing process is calculated to obtain the spatial fluctuation index, including: Based on the mixed-state data, the vibration amplitude distribution information of multiple consecutive sampling periods is extracted to form a periodic dataset; The difference in vibration amplitude between adjacent periods in the periodic dataset is calculated, and a weighted average is performed according to the monitoring area to generate difference data. The difference data is smoothed by performing a moving average to eliminate instantaneous disturbances and obtain smoothed difference data; The spatial volatility index is generated by calculating the change ratio of the current cycle relative to the previous two cycles based on the smoothed difference data.

5. The method for dynamic evaluation and parameter adjustment of feed mixing uniformity according to claim 1, characterized in that, When the spatial fluctuation index continues to decline and the temperature rise rate experiences a sudden increase, local accumulation indicator data is generated. Based on the triggering of short-term acceleration and reversal operations, a second set of mixed control data is formed, including: Based on the spatial fluctuation index and the temperature rise rate information in the mixed state data, the rate of change of the two parameters within a continuous sampling period is calculated and trend fitting is performed to generate periodic trend data. Regional difference analysis is performed on the periodic trend data. When the increase in the temperature rise rate of any region exceeds the preset temperature rise increase ratio threshold of the previous period's average and the corresponding vibration amplitude decreases, local accumulation indication data is generated. Based on the local accumulation indication data, the stirring shaft section corresponding to the accumulation area is determined, and a short-time acceleration control signal is output to the section to generate acceleration execution data. Based on the accelerated execution data, after the control equipment completes the high-speed stirring for a preset time, it performs a reverse rotation operation to generate disturbance recovery data; The coordinates of the stacking area in the stacking indication data are matched with the speed increase in the acceleration execution data to generate stacking response matching data; Based on the inversion duration in the disturbance recovery data, the accumulated response matching data is dynamically corrected to form control parameter update data; The operating parameter values ​​of the corresponding stacking section in the control parameter update data are replaced with the corresponding parameter items in the first mixed control data to generate the second mixed control data.

6. The method for dynamic evaluation and parameter adjustment of feed mixing uniformity according to claim 2, characterized in that, Based on the raw material feature vector set, the correlation coefficients between any two raw materials in three dimensions—particle size difference, density difference, and moisture content difference—are calculated to generate raw material flow-related data, including: Based on the raw material feature vector set, extract the particle size range ratio data of any two raw materials, calculate the difference in each particle size range, and generate particle size difference data. Based on the particle size difference data and the bulk density values ​​of the two raw materials, the density ratio is calculated, and the particle size difference data is corrected by using the density ratio as a weighting factor to generate density-weighted data on particle size difference between raw materials. Based on the density-weighted data of particle size difference between raw materials and the difference in moisture content between the two raw materials, and after normalization, the moisture content correction data is obtained. The moisture content correction data and the density weighted data of particle size difference between raw materials were normalized to generate normalized moisture content data and normalized density data. The normalized density data and normalized moisture content data are weighted and summed according to a preset ratio, and the weighted average difference value is calculated to generate a comprehensive difference index. Based on the numerical range of the comprehensive difference index, the difference range between each raw material is determined, forming raw material difference characteristic data; Based on the raw material difference characteristics data, the correlation coefficients of the two raw materials in three dimensions—particle size, density, and moisture content—are calculated to generate raw material flow correlation data. The raw material flow-related data for each pair of raw materials are summarized sequentially to generate a raw material flow-related data set.

7. The method for dynamic evaluation and parameter adjustment of feed mixing uniformity according to claim 3, characterized in that, Based on the energy response values ​​of each monitoring area in the regional coupling response data, a one-to-one correspondence is performed with the equipment operating parameters in the first hybrid control data to form parameter matching data, including: Based on the regional coupling response data, the spatial coordinates and corresponding energy response values ​​of each monitoring area are obtained, and energy response distribution data are generated. Based on the energy response distribution data and the equipment operating parameters in the first hybrid control data, a corresponding match is made in the time dimension to form time-aligned data; Spatial mapping processing is performed on the time-aligned data. A mapping relationship is established between the spatial coordinates of the monitoring area and the blade number of the stirring device to generate regional mapping data. Response difference calculations are performed on the regional mapping data to determine the response correlation between energy response values ​​and operating parameters, and preliminary parameter matching data is generated. Based on the response relevance of the initial parameter matching data, the matching accuracy is weighted and corrected to form the final parameter matching data.

8. The method for dynamic evaluation and parameter adjustment of feed mixing uniformity according to claim 3, characterized in that, Calculate the normalized dissimilarity of each region in the parameter-matched data, and use the normalized dissimilarity as a dynamic weight to correct the corresponding energy response values, generating weighted energy response data, including: Based on the parameter matching data, the energy response values ​​and corresponding operating parameter values ​​of each monitoring area are normalized to generate energy normalized data and operating parameter normalized data. The relative response deviation data for each monitoring area is obtained by calculating the deviation ratio between the energy normalized data and the operating parameter normalized data. Based on the maximum and minimum values ​​of the relative response deviation data, interval normalization is performed to generate normalized difference data for each region; Based on the normalized difference data, dynamic weight coefficients are assigned to each monitoring area to generate dynamic weight data. Based on dynamic weight data, the energy response value is weighted and corrected to generate weighted energy response value data; The weighted energy response values ​​of each region are cumulatively averaged over time to generate weighted energy response data.

9. The method for dynamic evaluation and parameter adjustment of feed mixing uniformity according to claim 5, characterized in that, Based on local accumulation indication data, the stirring shaft section corresponding to the accumulation area is determined, and a short-term acceleration control signal is output to this section to generate acceleration execution data, including: Based on the local accumulation indication data, extract the spatial coordinate information of the accumulation area to generate accumulation positioning data; Based on the stacking location data and mixing chamber structural parameters, the relative positional relationship between the stacking area and each blade on the stirring shaft is analyzed to generate segment mapping data; The corresponding stirring shaft segment number is determined based on the segment mapping data, and this segment is set as the control target segment to generate target segment data; Based on the number of blades and current rotation speed information in the target section data, calculate the theoretical material shearing amount per unit time for that section and generate shearing amount prediction data. Based on the spatial fluctuation index change rate and temperature rise rate increase information contained in the accumulation indication data, the accumulation strength parameters are calculated; and combined with the correspondence with the shear quantity prediction data, the speed increase and acceleration duration of short-term acceleration are determined, and acceleration control parameters are generated. Based on the acceleration control parameters, the stirring shaft section is driven to perform a short-term acceleration operation, generating acceleration execution data.

10. The method for dynamic evaluation and parameter adjustment of feed mixing uniformity according to claim 5, characterized in that, Based on the inversion duration in the disturbance recovery data, the accumulated response matching data is dynamically corrected to generate control parameter update data, including: Based on the disturbance recovery data, the reversal duration and the corresponding temperature rise rate are extracted to generate reversal feature data; Based on the location of the stacking region and the increase in rotational speed in the inversion feature data and the stacking response matching data, a multi-dimensional correspondence between time, velocity and location is established to generate disturbance correlation data. The disturbance-related data is dynamically evaluated. When the reversal duration exceeds the preset equilibrium time threshold, the reversal energy decay coefficient is calculated, and a disturbance correction factor is generated. Based on the disturbance correction factor, the speed increase in the stacked response matching data is attenuated or compensated to generate corrected response data. The difference analysis is performed between the corrected response data and the accumulated response matching data to extract the updated operating parameter values ​​and generate control parameter update data.

Images