A computer control method and system for flotation production process

Abstract

The application discloses a kind of computer control methods and systems of floatation production process, it is related to computer floatation control technical field, including, located in floatation pool acquisition multimodal data, including graphic frame data and physical quantity vector, calculate foam reflectivity determine light source brightness color temperature adjustment coefficient, carry out enhancement correction, and construct image super-resolution model, output high physical consistent reconstruction graph;The unit quality of calculation reconstruction graph is calculated, and the collision force of subarea is calculated, all segmented regions are time-advanced and dynamics evolution, and consistency error verification is carried out.The method disclosed in the application carries out enhancement correction to neighborhood pixels by combining the defect correction parameter of calibration, not only improves image detail level and signal-to-noise ratio, but also corrects systematic error caused by equipment itself, so that the quality of super-resolution input image is higher, and the spatial detail information of image is richer and more consistent in physics.

Description

Technical Field

[0001] This invention relates to the field of computer flotation control technology, and in particular to a computer control method and system for a flotation production process. Background Technology

[0002] Flotation technology is an important means in the field of mineral processing, used to separate mineral particles from gangue particles. Its basic principle is to separate mineral particles by having bubbles adhere to them and float them to the liquid surface, while suppressing gangue particles. As the resource industry places higher demands on the efficiency and product quality of flotation processes, automation, intelligence and precise control of the flotation process have become key development directions. At present, computer control technology for the flotation process is gradually being introduced to monitor the state of the froth layer, changes in physical parameters and dynamic fluid behavior in real time, and to adjust the process based on this data.

[0003] However, existing flotation control systems typically rely on the acquisition and analysis of single types of data to establish a comprehensive information interaction model. This makes it difficult to fully understand the flotation state from multiple levels, including enhanced images, thermodynamic parameters, and dynamic fluid behavior. Secondly, there is limited research on foam microdynamics in existing technologies. Although some flotation control technologies have introduced computational models based on foam distribution or size, these models often fail to effectively capture complex dynamic evolution characteristics, thus limiting their ability to describe the partitioning collision force and time evolution response mechanism. Furthermore, existing flotation automation equipment is not yet well-developed in terms of data anomaly monitoring and lacks the ability to analyze anomalies based on the prediction of the master state vector, physical quantity parameters, and rate of change, making it difficult to achieve long-term stability prediction and optimization of the flotation process. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a computer control method for the flotation production process. This addresses the shortcomings of existing flotation control systems, which typically rely on the acquisition and analysis of single-type data to establish a comprehensive information interaction model. These systems cannot fully understand the flotation state from multiple levels, including enhanced images, thermodynamic parameters, and dynamic fluid behavior. Furthermore, existing technologies have limited research on foam microdynamics. While some flotation control technologies introduce computational models based on foam distribution or size, these models often fail to effectively capture complex dynamic evolution characteristics, thus limiting their ability to describe partitioned collision forces and time-evolution response mechanisms. In addition, existing automated flotation equipment lacks robust data anomaly monitoring capabilities, lacking the ability to analyze anomalies based on master state vectors, physical quantity parameters, and rate of change predictions. This makes it difficult to predict and optimize the long-term stability of the flotation process.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a computer control method for a flotation production process, comprising:

[0008] Multimodal data, including graphic frame data and physical quantity vectors, are collected in the flotation cell. The foam reflectivity is calculated to determine the light source brightness and color temperature adjustment coefficient, and enhancement correction is performed. An image super-resolution model is constructed, and a high-resolution physical consistency reconstruction map is output.

[0009] Calculate the unit mass of the reconstructed map and the collision force between partitions. Perform time progression and dynamic evolution on all segmented regions and verify the consistency error.

[0010] The combination of Reynolds number, foam volume fraction, foam coverage and surface tension coefficient is determined as the main state vector. Based on the main state vector, physical quantity vector and disturbance term, the derivative equation of the rate of change is constructed to predict the rate of change and generate a contribution heat map. Combined with the consistency error verification results, the term with the largest contribution is adjusted by proportional integral PI control. After adjustment, the consistency error verification is carried out again.

[0011] Record operational data to build a historical anomaly case library.

[0012] As a preferred embodiment of the computer control method for the flotation production process described in this invention, the output high-resolution physical consistency reconstruction map includes,

[0013] The light intensity on the foam surface is read using a photometer. The gray level of the foam is measured at the center of the foam area in conjunction with the collected graphic frame data. The average gray level of the actual gray area of ​​the foam is determined, and the reflectivity of the foam is determined.

[0014] Multispectral illumination compensation coefficients, including adjustment coefficients for brightness and color temperature, are calculated based on the standard light intensity reference value and standard color temperature reference value of the foam area determined by historical experience.

[0015] The adjustment coefficients based on brightness and color temperature are input and adjusted through the LED lamp power control to complete dimming and color adjustment, and the light correction graphic frame data is re-acquired and enhanced correction is performed according to the neighborhood pixels.

[0016] The watershed algorithm is used to segment the enhanced graphics frame data to obtain different independent bubble partitions at the same time.

[0017] The image super-resolution model is input by aligning the enhanced image frame data with the normalized physical quantity vectors using timestamp data, and training it together with historical enhanced images and high-resolution labeled images, where each sample is labeled with physical parameters.

[0018] Output a high-resolution physically consistent reconstruction map based on the trained image super-resolution model;

[0019] The watershed algorithm is used again to segment the high-resolution physical consistency reconstructed image. Based on the i-th segmentation subdomain after resegmentation, the two point pairs with the largest Euclidean distance are found, and the pixel Euclidean distance of the point pair coordinates is determined as the maximum principal diameter of the reconstructed image partition.

[0020] The overall distribution mean is calculated based on the principal axis length of the reconstructed map partitions and the total number of partitioned regions.

[0021] In a preferred embodiment of the computer control method for the flotation production process described in this invention, the following steps are included: calculating the unit mass of the reconstructed image, calculating the collision force between partitions, performing time-progression and dynamic evolution analysis on all partitioned regions, and verifying consistency errors.

[0022] The simulation environment is built based on the normalized physical quantity vectors, the simulation space size is determined based on the high-resolution physical consistency reconstruction map, and the unit mass is initially calculated based on the standard value of foam density.

[0023] Simultaneously combine the solution dynamic viscosity and the fluid principal direction velocity to calculate the fluid drag force, and calculate the collision force between the zones;

[0024] The surface tension is determined based on the interface length of the segmented region, the surface tension coefficient, and the outward normal direction of the segmented region. The time progression and dynamic evolution of all segmented regions are then performed.

[0025] Based on the location information of each reconstructed map partition after advancement and the maximum principal diameter, mesh mapping is performed to generate a simulated image, and consistency error is verified with the high-resolution physical consistent reconstruction map.

[0026] The difference threshold is based on the sum of the historical mean and standard deviation. If the average pixel difference is less than the difference threshold, it indicates that the error is small.

[0027] In a preferred embodiment of the computer control method for the flotation production process described in this invention, the determination of the combination of Reynolds number, foam volume fraction, foam coverage, and surface tension coefficient as the principal state vector, and the construction of a derivative equation for the rate of change based on the principal state vector, physical quantity vector, and disturbance term to predict the rate of change, includes:

[0028] The foam volume fraction is calculated as the ratio of the total area of ​​the segmented region to the total area of ​​the image, and the Reynolds number is calculated based on the solution flow rate, solution density, and solution dynamic viscosity.

[0029] The foam coverage rate is calculated by the ratio of the number of pixels in the foam region to the total number of pixels in the super-resolution image, and the foam volume fraction, Reynolds number, foam coverage rate and surface tension coefficient are combined into the main state vector;

[0030] The influence coefficients of the main process parameters on the state are determined based on the main state vector, physical quantity vector and temperature data at the corresponding time. At the same time, the image mechanical field coupling differential term is determined based on the gradient field of the reconstructed image partition and the main state vector. Combined with the disturbance term, the derivative equation of the main state vector with respect to time is constructed to represent the rate of change of the system state.

[0031] The rate of change of the master state is calculated using historical data and the discrete difference approximation method.

[0032] Data tables are constructed based on the principal state vector, physical quantity vector, temperature, image gradient, and rate of change, with each row representing data at a given moment.

[0033] The fourth-order Runge-Kutta method (RK4) is used for time advancement to update the master state vector, physical quantity vector, temperature, and image gradient, and the rate of change is predicted based on the derivative equation.

[0034] In a preferred embodiment of the computer control method for the flotation production process described in this invention, the following steps are included: generating a contribution heatmap, combining it with the consistency error verification results, adjusting the largest contribution item using proportional-integral (PI) control, and then re-verifying the consistency error.

[0035] Based on the rate of change of the derivative equation, the influence of the main process parameters on the state, the influence of the image mechanical field coupling differential term and the perturbation term on the rate of change of the main state are calculated by normalization and used as the contribution.

[0036] Based on the master state vector and contribution values, a heatmap data table is compiled. The heatmap is then drawn using a drawing tool. The sum of the historical mean and standard deviation is used as the contribution threshold for each contribution. If the contribution of a data item is greater than or equal to the contribution threshold, the data item needs to be adjusted. The adjustment is controlled by proportional-integral (PI) control. After adjustment, the consistency error is re-verified.

[0037] As a preferred embodiment of the computer control method for the flotation production process described in this invention, the step of collecting multimodal data in the flotation cell includes:

[0038] Multi-angle, high-frame-rate industrial CCD+spectral-enhanced cameras are simultaneously deployed at key locations in the flotation cell to acquire multimodal data on the 3D morphology and texture of the foam surface, and simultaneously collect process big data and graphic frame data such as flow rate, pH, conductivity, bubble particle size distribution, and reagent concentration.

[0039] The collected data, including real-time feed flow rate of the flotation cell, pH value of the flotation cell slurry, slurry conductivity, average diameter of foam in the foam zone, and online concentration of reagents, were preprocessed, timestamped, and combined to form a vector of physical quantities.

[0040] As a preferred embodiment of the computer control method for the flotation production process described in this invention, the method for constructing a historical anomaly case database by recording operation data includes:

[0041] If the average pixel difference value is greater than or equal to the difference threshold, it indicates that the error is large, and the main state vector, physical quantity vector, and temperature data are recorded as abnormal data to build a historical abnormal case library.

[0042] Secondly, the present invention provides a computer control system for a flotation production process, comprising,

[0043] The data processing module acquires multimodal data from the flotation cell through sensors and acquisition devices;

[0044] The image reconstruction module constructs an image super-resolution model, generates a high-resolution physically consistent reconstruction map, and calculates the maximum principal diameter of the segmented region and the overall distribution mean.

[0045] The evolution simulation module advances the dynamic evolution of all segmented regions in the time dimension, generates simulated images of foam behavior, and uses the simulated images to verify consistency errors.

[0046] The rate prediction module calculates fluid dynamics-related parameters, generates the master state vector, constructs the derivative equation of the rate of change, and predicts the rate of change of foam and material flow.

[0047] The contribution monitoring module calculates the impact of the main process parameters on the system and generates a heat map of the impact contribution. It also identifies anomalies based on the results of consistency error verification.

[0048] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the computer control method for the flotation production process as described in the first aspect of the present invention.

[0049] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein, when the computer program is executed by a processor, it implements any step of the computer control method for the flotation production process as described in the first aspect of the present invention.

[0050] The beneficial effects of this invention are as follows: By combining calibrated defect correction parameters to enhance and correct neighboring pixels, not only are the image detail levels and signal-to-noise ratio improved, but also the systematic errors caused by the device itself are corrected, resulting in higher quality super-resolution input images with richer spatial detail information and stronger physical consistency. The watershed algorithm is used to segment the enhanced image into foam regions, which can effectively identify independent foam structures that are not connected in the image. The alignment of timestamps and physical quantity vectors enables physical parameters to be accurately mapped to specific foam regions, ensuring the authenticity of training samples and the supervision effect. By calculating the collision force between segmented regions, the elastic interaction within the foam group is simulated, effectively capturing the collision and compression phenomena between foams. This makes the foam motion and deformation in the simulation conform to the complex collective dynamic characteristics of the foam medium. By constructing the derivative equation of the rate of change, the dynamic influence of the master state on itself is reflected, and multi-scale coupling and external disturbances are introduced, enabling the model to respond to real-time changes in environmental and process conditions and fully reflect the complex nonlinear dynamic characteristics of the system. Attached Figure Description

[0051] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a schematic diagram of the computer control method for the flotation production process in Example 1.

[0053] Figure 2 This is a schematic diagram of the computer control system for the flotation production process in Example 1. Detailed Implementation

[0054] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0055] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0056] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0057] Example 1, referring to Figures 1 to 2 This is the first embodiment of the present invention, which provides a computer control method for a flotation production process, including the following steps:

[0058] S1, located in the flotation cell, collects multimodal data, including graphic frame data and physical quantity vectors, calculates foam reflectivity to determine the light source brightness and color temperature adjustment coefficient, performs enhancement correction, constructs an image super-resolution model, and outputs a high-resolution physical consistency reconstruction image.

[0059] Preferably, multimodal data is collected in the flotation cell, including:

[0060] Multi-angle, high-frame-rate industrial CCD + spectral enhancement cameras are simultaneously deployed at key locations in the flotation cell to acquire multimodal data on the 3D morphology and texture of the foam surface. Simultaneously, process big data and graphic frame data such as flow rate, pH, conductivity, bubble particle size distribution (laser particle size analyzer), and reagent concentration are collected. (All devices are synchronized with a unified high-precision clock (using PTP / IEEE1588 clock protocol)).

[0061] The collected data, including real-time feed flow rate of the flotation cell, pH value of the flotation cell slurry, slurry conductivity, average diameter of foam in the foam zone, and online concentration of reagents, were preprocessed, timestamped, and combined into a vector of physical quantities.

[0062] By combining a multi-angle, high-frame-rate industrial CCD with a spectral-enhanced camera, the three-dimensional morphology and texture details of the foam surface can be captured, enabling multi-dimensional characterization of the spatial distribution and interface features of the foam structure. This provides rich and high-quality visual information on the foam's quality and dynamic changes. The application of real-time synchronous acquisition and a unified clock protocol ensures accurate temporal alignment between key process parameters such as flow rate, pH, conductivity, bubble size, and reagent concentration and image frame data. This eliminates time deviations between multi-source data and provides a solid spatiotemporal foundation for multimodal data fusion. After preprocessing and data fusion, the resulting physical quantity vectors systematically reflect the fluid chemical environment, foam physical properties, and process reagent dosing status within the flotation cell.

[0063] Furthermore, output a high-resolution physical consistency reconstruction map, including,

[0064] The light intensity on the foam surface is read using a photometer. The gray level of the foam is measured at the center of the foam area in conjunction with the collected graphic frame data. The average gray level of the actual gray area of ​​the foam is determined. The ratio of the average gray level to the gray level of the foam is calculated to determine the reflectivity of the foam.

[0065] Based on historically determined standard luminous intensity and standard color temperature reference values ​​for the foam area, multispectral illumination compensation coefficients, including adjustment coefficients for luminance and color temperature, are calculated and expressed as follows:

[0066]

[0067] Where S LES S represents the adjustment coefficient for the brightness of the LED light source. CT L represents the adjustment coefficient for the color temperature of the LED light source. ref L represents the standard light intensity reference value for the foam area. t Indicates the gloss intensity of the foam surface. R represents the standard color temperature reference value. t Indicates the reflectivity of the foam;

[0068] The adjustment coefficients based on brightness and color temperature are input and adjusted via LED power control to complete dimming and color adjustment. Light correction graphic frame data is then re-acquired and enhanced based on neighborhood pixels, as shown below:

[0069]

[0070] in This represents the pixel value at position (x,y) and band λ in the enhanced image, where N(x,y) represents the number of pixels in the neighborhood centered at (x,y). This represents the set of neighborhood indices surrounding pixel (x, y). α and β represent the defect correction parameters from the device's factory calibration, which can be determined by looking up tables in the LED and camera manuals. t (i,j,λ) represents the pixel value of the original image at position (x,y) and band λ;

[0071] The watershed algorithm is used to segment the enhanced graphics frame data to obtain different independent bubble partitions at the same time.

[0072] The image super-resolution model is input from enhanced image frame data and normalized physical quantity vectors, aligned with timestamp data. It is trained using historical enhanced images and high-resolution labeled maps, where each sample carries a physical parameter label. The training loss is calculated as follows:

[0073]

[0074] D py =∑ i (Si,t -D b,t ) 2 ;

[0075] Where L sup I represents the total loss of the super-resolution reconstruction network. HR,t This represents the high-resolution foam image calibrated at time t. Let λ represent the physically constrained high-resolution bubble image output by the image super-resolution model at time t. py This represents the physical loss weighting factor. Represents the normalized physical quantity vector as well as The physical consistency loss function, S i,t D represents the i-th independent bubble partition at time t. b,t The value represents the physical average particle size of the foam at time t, which can be directly measured by an online particle size analyzer.

[0076] The high-resolution physical consistency reconstruction map is output based on the trained image super-resolution model. The obtained high-resolution physical consistency reconstruction map is a super-resolution structure map with significantly enhanced spatial details and characterization of physical consistency. It can be used for subsequent processes such as arbitrary morphological dynamics back-inference, local bubble analysis, and physical field parameter tracing. Physical constraints ensure the reliability of the structure.

[0077] The watershed algorithm is used again to segment the high-resolution physically consistent reconstructed image. Within the i-th segmented subdomain, the pair of points with the largest Euclidean distance is found, and the pixel Euclidean distance of the point pair coordinates is determined as the maximum principal diameter of the reconstructed image partition, expressed as:

[0078]

[0079] Where (x) a ,y a ),(x b ,y b )∈sub i This represents a pair of pixels belonging to the i-th re-segmented foam subdomain, where (x a ,y a ),(x b ,y b These are the coordinates of two pixels. This represents the maximum principal diameter of the i-th reconstructed graph partition at time t;

[0080] The overall distribution mean is calculated based on the principal axis length of the reconstructed map partitions and the total number of partitioned regions.

[0081] By using a photometer to obtain the actual light intensity on the foam surface and combining it with the mean gray values ​​of the center and local areas of the image grayscale, the foam reflectivity can be accurately calculated. This can effectively correct the grayscale deviation in the foam image caused by the difference in reflectivity, and enhance the accuracy of the image in reflecting the true physical state of the foam.

[0082] Based on the reference values ​​of standard light intensity and color temperature in historical foam regions, brightness and color temperature adjustment coefficients are calculated to achieve dynamic power control and adjustment of the LED light source. This ensures optimal matching of ambient light during acquisition, making the image brightness and color restoration closer to reality. It effectively reduces the impact of ambient light changes on image quality and subsequent analysis accuracy. Combined with calibrated defect correction parameters, neighboring pixels are enhanced and corrected, which not only improves the image detail and signal-to-noise ratio but also corrects systematic errors caused by the device itself. This results in higher quality super-resolution input images with richer spatial detail information and stronger physical consistency. The watershed algorithm is used to segment the enhanced image into foam regions, which can effectively identify independent foam structures that are not connected in the image. The alignment of timestamps and physical quantity vectors ensures that physical parameters can be accurately mapped to specific foam regions, guaranteeing the authenticity of training samples and the supervision effect. Thus, the training of a super-resolution model with physical constraints improves the dual accuracy of reconstructed images in terms of spatial resolution and physical properties.

[0083] During training, a physical consistency loss term is introduced, combined with foam particle size labels measured by a real particle size analyzer. This not only improves the visual quality of the super-resolution model but also ensures a high degree of consistency between the structure and physical properties. This enhances the model's generalization ability and realistic perception of complex foam scenarios. The maximum principal diameter is extracted by re-segmenting the high-resolution reconstructed image. By calculating the maximum Euclidean distance pixel pair, the principal scale of the foam segmentation area is reflected, accurately depicting the spatial characteristics of each foam structure. The overall distribution mean calculated based on the principal scale and quantity of all segmented areas provides a reliable statistical feature index for flotation process parameters and foam dynamic changes. This effectively assists in process optimization decisions and enables real-time status monitoring.

[0084] S2, calculate the unit mass of the reconstructed map and the collision force between partitions, perform time progression and dynamic evolution for all segmented regions, and verify the consistency error;

[0085] Preferably, the unit mass of the reconstructed map is calculated, and the collision force between partitions is calculated. Time progression and dynamic evolution are performed on all segmented regions, and consistency error verification is conducted, including...

[0086] Based on the normalized physical quantity vectors, a simulation environment is built. The simulation space dimensions are determined based on the high-resolution physical consistency reconstruction image. The element mass is initially calculated based on the standard value of foam density, and expressed as:

[0087]

[0088] Where m i d represents the mass of the i-th simulation unit. i p represents the maximum principal diameter of the i-th reconstructed graph partition. f Indicates foam density;

[0089] Simultaneously combining the solution's dynamic viscosity and the fluid's principal direction velocity, the fluid drag force is calculated, and the collision force between zones is also calculated, expressed as:

[0090] F fl，i =3πηd i (uv i );

[0091]

[0092] Where F fl，i F represents the fluid drag force of the i-th reconstructed graph partition. col，ij Let η represent the collision force of the i-th and j-th reconstructed map partitions, η represent the dynamic viscosity of the solution, u represent the principal velocity of the fluid, and v represent the velocity of the fluid in the principal direction. i Let k represent the fluid velocity in the i-th reconstructed graph partition. n This represents the elastic coefficient of foam interaction, obtained by looking up a table using the standard foam model. ij It means that δ ij d represents the overlap between the i-th and j-th reconstructed map partitions. j Let r represent the maximum principal diameter of the j-th reconstructed graph partition, and r be the maximum principal diameter of the j-th reconstructed graph partition. ij This represents the centroid distance between the i-th and j-th reconstructed map partitions;

[0093] The surface tension is determined based on the interface length of the segmented regions, the surface tension coefficient, and the outward normal direction of the segmented regions. The time progression and dynamic evolution of all segmented regions are then performed, as follows:

[0094] F su,i =r t ·L cu,i ·n i ;

[0095]

[0096] x i (t+Δt)=x i (t)+v i (t+Δt)·Δt;

[0097] Where F su,i r represents the surface tension of the i-th reconstructed graph partition. t The surface tension coefficient is represented by L, which is obtained from a table based on reagent concentration / pH. cu,iIndicates the length of the interface of the i-th reconstructed graph partition, n i Indicates the direction of the outward normal to the segmented region, x i Δt represents the center coordinates of the i-th sub-reconstructed map partition, and Δt represents the time step.

[0098] Based on the post-progression location information and maximum principal diameter of each reconstructed map partition, mesh mapping is performed to generate a simulated image, which is then verified for consistency error with the high-resolution physical consistency reconstruction map, as shown below:

[0099]

[0100] Where E dif This represents the average pixel difference value, where M and N represent the number of pixels in the height and width of the image, respectively. This represents the pixel value at (x, y) in the high-resolution physical consistency reconstructed image. Represents the pixel value at (x, y) in the simulated image;

[0101] The difference threshold is based on the sum of the historical mean and standard deviation. If the average pixel difference is less than the difference threshold, it indicates that the error is small.

[0102] By calculating the unit mass based on the maximum main diameter of the partition and the foam density, an accurate estimate of the mass of each foam structure region in physical space is achieved. When calculating the fluid drag force, the dynamic viscosity of the solution and the velocity in the mainstream direction are combined to reasonably reflect the resistance effect of the fluid on the foam structure, enhancing the realism of the fluid-foam interaction in the simulation. By calculating the collision force between the segmented regions, the elastic interaction within the foam group is simulated, effectively capturing the collision and compression phenomena between foams. This ensures that the foam motion and deformation in the simulation conform to the complex collective dynamic characteristics of the foam medium, improving the model's adaptability to the behavior of multi-foam systems. The surface tension is calculated by combining the interface length and normal direction of the segmented region with the surface tension coefficient, accurately simulating the stretching and contraction forces of the foam film. Time progression based on the above forces is performed simultaneously on all segmented regions. By mapping the simulated partition location information and scale back to the image space, a simulated image is generated. Then, a pixel-level comparison is performed with the high-resolution physical consistency reconstruction image using quantitative methods such as mean square error, achieving direct verification between the simulation results and actual observations, enhancing the credibility and applicability of the simulation model.

[0103] S3. Determine the combination of Reynolds number, foam volume fraction, foam coverage, and surface tension coefficient as the principal state vector. Construct the derivative equation of the rate of change based on the principal state vector, physical quantity vector, and perturbation term.

[0104] Predict the rate of change, generate a contribution heatmap, and combine the consistency error verification results to adjust the largest contribution item using proportional-integral PI control. After adjustment, re-verify the consistency error.

[0105] Preferably, the combination of Reynolds number, foam volume fraction, foam coverage, and surface tension coefficient is determined as the principal state vector. Based on the principal state vector, physical quantity vectors, and disturbance terms, a derivative equation for the rate of change is constructed to predict the rate of change, including...

[0106] The foam volume fraction is calculated as the ratio of the total area of ​​the segmented region to the total area of ​​the image. The Reynolds number is then calculated based on the solution flow rate, solution density, and solution dynamic viscosity, and is expressed as:

[0107]

[0108] Where Re t Represents the Reynolds number, p represents the solution density, and Q represents the solution density. t η represents the solution flow rate, η represents the dynamic viscosity of the solution (obtained from a table based on pH and temperature), and A represents the cross-sectional area of ​​the flow channel.

[0109] The foam coverage rate is calculated by the ratio of the number of pixels in the foam region to the total number of pixels in the super-resolution image, and the foam volume fraction, Reynolds number, foam coverage rate and surface tension coefficient are combined into the main state vector;

[0110] The influence coefficients of the main process parameters on the state are determined based on the main state vector, physical quantity vector, and temperature data at corresponding times. Simultaneously, the image mechanical field coupling differential term is determined based on the gradient field of the reconstructed image partition and the main state vector. Combined with the perturbation term, the derivative equation of the main state vector with respect to time is constructed to represent the rate of change of the system state, expressed as:

[0111] F1 = A·S t +B·P t +C·T t ;

[0112]

[0113] Where F1 represents the influence coefficient of the main process parameters on the state, F2 represents the differential term of the image mechanical field coupling, F3 represents the disturbance term, and S t P represents the master state vector. t T represents a vector of physical quantities. t Let A represent the temperature data, B represent the influence coefficient of the principal state vector on the rate of change, C represent the influence coefficient of the physical quantity vector on the principal state change, D represent the influence coefficient of the image gradient on the principal state change, E represent the coupling influence coefficient between principal states, and G represent the influence coefficient of impurity perturbation on the principal state change. Indicates impurity concentration. This represents the image gradient calculated from the high-resolution physical consistent reconstruction graph using the Sobel operator method;

[0114] Using historical data and the discrete difference approximation method, the rate of change of the master state is calculated and expressed as:

[0115]

[0116] Where S t+Δt This represents the main state vector that changes over time Δt+t, where Δt represents the time interval.

[0117] Data tables are constructed based on the principal state vector, physical quantity vector, temperature, image gradient, and rate of change, with each row representing data at a given moment.

[0118] The values ​​of A, B, C, D, E, and G are determined by fitting linear regression using the least squares method;

[0119] The fourth-order Runge-Kutta method (RK4) is used for time advancement to update the master state vector, physical quantity vector, temperature, and image gradient, and the rate of change is predicted based on the derivative equation.

[0120] By calculating the foam volume fraction using the ratio of the total area of ​​the segmented region to the area of ​​the entire image, and combining this with flow rate, density, and dynamic viscosity to calculate the Reynolds number, this method relies heavily on real-time images and process data to ensure that the master state variables accurately reflect the physical characteristics of the process site. It also avoids dependence on complex manual measurements, improving the timeliness and automation of measurements. By combining the foam volume fraction, Reynolds number, coverage, and surface tension coefficient into the master state vector, the model more comprehensively captures multi-dimensional information about the interaction between foam and fluid. Combining master process parameters, temperature, and image gradients, along with other multi-source physical and image information, a rate-of-change derivative equation is constructed, reflecting the dynamic influence of the master state on itself. The model also incorporates multi-scale coupling and external disturbances, enabling it to respond to real-time changes in environmental and process conditions and comprehensively reflect the complex nonlinear dynamic characteristics of the system. By applying historical data and employing the discrete difference approximation method, the rate of change of the master state is accurately calculated, providing a solid numerical foundation for the regression fitting of the coefficient matrix. This ensures that the parameters obtained through the least squares method reflect the actual process history and physical laws, improving the model's fitting quality and generalization ability. The application of the fourth-order Runge-Kutta method to advance the derivative equation over time not only guarantees the high accuracy and stability of numerical calculations but also supports continuous dynamic simulation of the master state vector and its influencing factors, thereby enabling the prediction of future state evolution trends.

[0121] Preferably, a contribution heatmap is generated, and based on the consistency error verification results, the item with the largest contribution is adjusted using proportional-integral (PI) control. After adjustment, the consistency error verification is performed again, including...

[0122] Based on the rate of change of the derivative equation, the influence of the main process parameters on the state, the influence of the image mechanical field coupling differential term and the perturbation term on the rate of change of the main state are calculated by normalization and used as the contribution.

[0123] Based on the master state vector and contribution values, a heatmap data table is compiled. The heatmap is then drawn using a drawing tool. The sum of the historical mean and standard deviation is used as the contribution threshold for each contribution. If the contribution of a data item is greater than or equal to the contribution threshold, the data item needs to be adjusted. The adjustment is controlled by proportional-integral (PI) control. After adjustment, the consistency error is re-verified.

[0124] By normalizing the rate of change of different terms in the derivative equation and quantifying their contribution to the change of the master state, a fine division and dynamic monitoring of the internal mechanism of the process system is effectively achieved. This contribution analysis can clearly distinguish the actual impact of the master process parameters, the coupling of the image mechanical field, and the disturbance terms on the change of the system state. The contribution is combined with the master state vector and organized into a structured data table, which is presented through a heat map. This enhances the visualization ability of complex multidimensional data, enabling operators and managers to intuitively understand the relative importance of each influencing factor and the evolution trend at multiple time points.

[0125] S4, record operation data to build a historical exception case library;

[0126] Furthermore, a historical anomaly case library is built by recording operational data, including:

[0127] If the average pixel difference value is greater than or equal to the difference threshold, it indicates that the error is large, and the main state vector, physical quantity vector, and temperature data are recorded as abnormal data to build a historical abnormal case library.

[0128] By setting a comparison between the average pixel difference value and the difference threshold, it is helpful to effectively identify abnormal data under conditions of large errors, ensuring that abnormal phenomena can be detected and recorded in a timely manner. Recording the main state vector, physical quantity vector, and temperature data not only enriches the data dimensions of the abnormal case library, but also provides multi-angle reference for subsequent analysis, improving the accuracy and reliability of abnormal diagnosis. Building a historical abnormal case library lays a solid foundation for establishing model training and abnormal pattern recognition, which helps to optimize the system's abnormal detection algorithm, thereby improving the overall stability and security of the system.

[0129] This embodiment also provides a computer control system for a flotation production process, including,

[0130] The data processing module acquires multimodal data from the flotation cell through sensors and acquisition devices;

[0131] The image reconstruction module constructs an image super-resolution model, generates a high-resolution physically consistent reconstruction map, and calculates the maximum principal diameter of the segmented region and the overall distribution mean.

[0132] The evolution simulation module advances the dynamic evolution of all segmented regions in the time dimension, generates simulated images of foam behavior, and uses the simulated images to verify consistency errors.

[0133] The rate prediction module calculates fluid dynamics-related parameters, generates the master state vector, constructs the derivative equation of the rate of change, and predicts the rate of change of foam and material flow.

[0134] The contribution monitoring module calculates the impact of the main process parameters on the system and generates a heat map of the impact contribution. It also identifies anomalies based on the results of consistency error verification.

[0135] This embodiment also provides a computer device applicable to the computer control method for the flotation production process, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the computer control method for the flotation production process proposed in the above embodiment.

[0136] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0137] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the computer control method for the flotation production process as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0138] In summary, this invention enhances and corrects neighboring pixels by combining calibrated defect correction parameters, which not only improves image detail and signal-to-noise ratio but also corrects systematic errors caused by the device itself. This results in higher quality super-resolution input images with richer spatial detail and stronger physical consistency. The watershed algorithm is used to segment the enhanced image into foam regions, effectively identifying independent, unconnected foam structures. Alignment of timestamps and physical quantity vectors ensures that physical parameters are accurately mapped to specific foam regions, guaranteeing the authenticity of training samples and the effectiveness of supervision. By calculating the collision force between segmented regions, the elastic interaction within the foam community is simulated, effectively capturing collisions and compression phenomena between foams. This ensures that the foam motion and deformation during simulation conform to the complex collective dynamics of the foam medium. By constructing the derivative equation of the rate of change, the dynamic influence of the master state on itself is reflected, and multi-scale coupling and external disturbances are introduced, enabling the model to respond to real-time changes in environmental and process conditions and comprehensively reflect the complex nonlinear dynamic characteristics of the system.

[0139] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A computer control method for a flotation production process, characterized in that, The method comprises the following steps: Collecting multi-modal data in the flotation tank, including graphical frame data and physical quantity vectors, calculating the adjustment coefficients of foam reflectivity to determine brightness and color temperature, performing enhancement correction, and constructing an image super-resolution model to output high-resolution physically consistent reconstruction images; Calculate the unit quality of the reconstruction image, and calculate the collision force between partitions. Time marching and dynamics evolution are performed on all segmented regions, and consistency error verification is performed. Determine the Reynolds number, foam volume fraction, foam coverage rate, and surface tension coefficient combination as the main state vector. According to the main state vector, the physical quantity vector and the disturbance term, the derivative equation of the change rate is constructed, the change rate of the foam and the material flow is predicted, the contribution thermodynamic diagram is generated, combined with the consistency error verification result, the proportional integral PI control adjustment is performed on the largest contribution item, and after adjustment, the consistency error verification is performed again. Record operation data to build a historical abnormal case library. The output high-resolution physically consistent reconstruction image comprises: Use the light meter to read the foam surface light intensity, cooperate with the collected graphical frame data to measure the actual foam gray scale in the foam region center, determine the actual foam gray scale region gray mean value, and determine the foam reflectivity; Based on the standard light intensity reference value and the standard color temperature reference value of the foam region determined based on historical experience, calculate the multi-spectral illumination compensation coefficient, including the adjustment coefficients of brightness and color temperature; Based on the adjustment coefficients of brightness and color temperature, input and adjust through LED light source control, complete dimming and color adjustment, and re-collect corrected graphical frame data, and perform enhancement correction according to the field pixels; For the enhanced graphical frame data, the watershed algorithm is used for image segmentation to obtain different independent foam partitions at the same time; Based on the enhanced graphical frame data and the normalized physical quantity vector, the timestamp data is aligned, the image super-resolution model is input, and based on the historical enhanced image and the high-resolution labeled image, the model is trained together, wherein each sample is labeled with physical parameters; According to the trained image super-resolution model, a high-resolution physically consistent reconstruction image is output. According to the main shaft length of the reconstruction image partition and the total number of segmented regions, the overall distribution mean value is calculated. The calculation of the unit quality of the reconstruction image, the calculation of the collision force between partitions, the time marching and dynamics evolution of all segmented regions, and the consistency error verification comprise: According to the normalized physical quantity vector, a simulation environment is built, the simulation space size is determined according to the high-resolution physically consistent reconstruction image, and the unit quality is preliminarily calculated according to the foam density standard value; Synchronously calculate the fluid drag force combined with the solution dynamic viscosity and the fluid main direction velocity, and calculate the collision force between partitions. According to the segmented region interface length, the surface tension coefficient, and the segmented region outer normal direction, the surface tension is determined, and time marching and dynamics evolution are performed on all segmented regions. According to the position information of each reconstruction image partition after marching and the maximum main diameter, a grid mapping is performed to generate a simulation image, and the consistency error verification is performed with the high-resolution physically consistent reconstruction image.

2. A computer control method for a flotation production process according to claim 1, characterized by: The Reynolds number, the foam volume fraction, the foam coverage and the surface tension coefficient are combined into a main state vector, a derivative equation of a change rate is constructed according to the main state vector, a physical quantity vector and a perturbation term, the change rate of the foam and the material flow is predicted, including, The foam volume fraction is obtained by the ratio of the total area of the segmented area to the total area of the image, and the Reynolds number is calculated according to the solution flow, the solution density and the solution dynamic viscosity; The foam coverage is calculated by the ratio of the number of the foam area pixels to the total number of pixels in the super-resolution image, and the foam volume fraction, the Reynolds number, the foam coverage and the surface tension coefficient are combined into a main state vector; The influence coefficient of the main process parameter on the system state is determined according to the main state vector, the physical quantity vector and the temperature data of the corresponding time, and the image mechanics field coupling differential term is determined according to the gradient field of the reconstructed image partition and the main state vector, and the derivative equation of the main state vector with respect to time is constructed by combining the perturbation term, which represents the change rate of the system state; The change rate of the main state is calculated using the historical data and the discrete difference approximation method; A data table is constructed according to the main state vector, the physical quantity vector, the temperature, the image gradient and the change rate, wherein each row is the data of a time point; The fourth-order Runge-Kutta method (RK4) is used for time advancing, the main state vector, the physical quantity vector, the temperature and the image gradient are updated, and the change rate of the foam and the material flow is predicted according to the derivative equation.

3. A method of computer control of a flotation production process according to claim 2, characterized in that: The contribution thermodynamic diagram is generated, the results of the consistency error verification are combined, the largest contribution term is adjusted by proportional integral (PI) control, and after the adjustment, the consistency error verification is re-performed, including, The influence coefficient of the main process parameter on the system state, the image mechanics field coupling differential term and the perturbation term are extracted, and the proportions of the influence coefficient, the image mechanics field coupling differential term and the perturbation term in the change rate of the main state are calculated as the contribution degrees by normalization; According to the main state vector and the contribution degrees, a thermodynamic diagram data table is arranged, the thermodynamic diagram is drawn by a drawing tool, the sum of the historical mean value and the standard deviation is taken as the contribution threshold value of each contribution degree, if the contribution degree of the data item is greater than or equal to the contribution threshold value, it is judged that the data item needs to be adjusted, and proportional integral (PI) control is used for adjustment, and after the adjustment, the consistency error verification is re-performed.

4. A computer control method for a flotation production process according to claim 3, characterized by: The multi-modal data are collected at the flotation tank, including, Multi-angle high-frame-rate industrial CCD cameras and spectrum-enhanced cameras are synchronously deployed at key positions of the flotation tank to obtain 3D topography and texture multi-modal data of the foam surface, real-time flow of the flotation tank, slurry pH value, slurry conductivity, average diameter of the foam in the foam area, online concentration of the reagent and graphic frame data are synchronously collected, data preprocessing is performed, time stamps are unified, and a physical quantity vector is formed.

5. A computer control method for a flotation production process as claimed in claim 4, characterized by: The operation data are recorded to construct a historical abnormal case library, including, If the average pixel difference value is greater than or equal to the difference threshold value, it indicates that the error is large, and the main state vector, the physical quantity vector and the temperature data are recorded as abnormal data to construct a historical abnormal case library.

6. A computer control system for a flotation production process, based on the computer control method for a flotation production process according to any one of claims 1 to 5, characterized in that: including, The data processing module obtains multi-modal data of the flotation tank through sensors and collection equipment; An image reconstruction module constructs an image super-resolution model, generates a high-resolution physically consistent reconstructed image, and calculates the maximum main path of each segmented region and the overall distribution mean value; An evolution simulation module simulates the dynamics evolution of all segmented regions in the time dimension, generates a simulation image of the foam behavior, and verifies the consistency error using the simulation image; A rate prediction module calculates fluid dynamics related parameters, generates a main state vector, constructs a derivative equation of the change rate, and predicts the change rate of the foam and material flow; A contribution monitoring module calculates the influence coefficient of the main process parameters on the system state, generates an influence contribution thermodynamic map, and identifies abnormalities according to the consistency error verification result. 7.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is characterized in that: The processor executes the computer program to realize the steps of the computer control method of the flotation production process according to any one of claims 1-5.

8. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to realize the steps of the computer control method of the flotation production process according to any one of claims 1-5.

Images