Ash content real-time linkage variable-speed bubble scraping control method and device for flotation machine, terminal and medium

Abstract

The invention relates to the field of coal preparation of flotation machines, and particularly provides an ash content real-time linkage variable-speed foam scraping control method and device for a flotation machine, a terminal and a medium. The method comprises the steps that firstly, foam product ash content values, foam layer thicknesses and foam form images of all groove chambers of the flotation machine are obtained in real time; secondly, extracting a foam rupture rate feature through image processing, and constructing a comprehensive feature vector by combining a foam amount feature and an ash value accumulated in a time sequence; then, the vector is input into a preset fuzzy inference system, and the system outputs the adjustment amount of the rotating speed of the scraper of each groove chamber based on a fuzzy rule built in a flotation process knowledge base; and finally, controlling the variable frequency motor to drive the scraper to operate according to the adjustment amount. According to the method, the ash content, the foam amount and the foam stability are fused, intelligent decision making of the rotating speed of the driving system is conducted based on fuzzy reasoning, accurate sensing and self-adaptive regulation and control of the flotation state are achieved, and therefore the concentrate quality and yield are optimized while the separation process is stabilized and concentrate loss is prevented.

Description

Technical Field

[0001] This application relates to the field of coal preparation using flotation machines, specifically to a method, device, terminal, and medium for real-time linkage variable speed skimming control of ash content in flotation machines. Background Technology

[0002] Flotation is the core separation process in coal washing. A flotation machine typically consists of multiple cells connected in series, where mineral particles undergo separation. The froth scraper is responsible for timely and stable scraping out the froth product enriched with the target mineral.

[0003] Traditional multi-chamber flotation machines used in industry typically employ a mechanical linkage design, where a single motor drives a long shaft running through all chambers, causing all scrapers to operate at the same fixed speed. Improvements have been proposed by equipping each chamber with an independent drive motor and introducing speed regulation based on parameters such as foam layer thickness, thus enhancing operational flexibility to some extent. However, these solutions primarily rely on foam layer thickness, liquid level, and ash content for adjustment, lacking the ability to perceive and respond to foam stability. The scraper speed regulation is disconnected from foam stability, failing to reduce disturbances when foam is fragile or increase recovery when foam is stable. Furthermore, current speed regulation systems often employ simple PID or threshold control, with static and isolated control logic that fails to integrate and intelligently decide based on multi-dimensional information such as ash content, foam quantity, and foam stability, impacting the accuracy of drive system speed regulation. Summary of the Invention

[0004] To address the aforementioned issues, this invention provides a real-time linkage variable speed frothing control method, device, terminal, and medium for ash content in flotation machines. By integrating multi-source information such as ash content, froth quantity, and froth stability, and making intelligent decisions on the drive system speed based on fuzzy reasoning, it achieves accurate perception and adaptive control of the flotation state, thereby stabilizing the separation process, preventing concentrate loss, and optimizing concentrate quality and yield.

[0005] In a first aspect, the technical solution of the present invention provides a method for real-time linkage variable speed skimming control of ash content in a flotation machine, comprising the following steps: Real-time acquisition of flotation machine data Ash content of foam products in each tank Foam layer thickness value and images of foam morphology ; Images of foam morphology Image processing was performed to extract the feature values ​​of the foam bursting rate. ; Set the foam layer thickness value By performing time-series accumulation, the characteristic value of foam volume per unit time is calculated. ;Merge the ash content values fracture rate characteristic value and foam quantity characteristic value Construct a comprehensive feature vector characterizing the current flotation state of the cell. ; The comprehensive feature vector Input to a pre-defined fuzzy inference system, output the first... Adjustment amount of scraper speed in each tank chamber The fuzzy inference system has built-in fuzzy rules based on the flotation process knowledge base. The preconditions of the fuzzy rules are fuzzy linguistic variables of ash content, foam bursting rate and foam amount, and the conclusion is the fuzzy linguistic variable of scraper speed adjustment. According to the adjustment amount Generate control commands and send them to the first Each slot chamber has a corresponding variable frequency motor to drive the scraper to run at the adjusted speed.

[0006] In one optional implementation, the foam morphology image... Image processing was performed to extract the feature values ​​of the foam bursting rate. Specifically, it includes: The foam morphology images of consecutive frames are sequentially grayscaled and Gaussian filtered to suppress image noise introduced by slurry splashing, light fluctuations and equipment vibration. Calculate the absolute difference map of the preprocessed images of adjacent frames, and binarize the difference map using an adaptive thresholding method to generate a region mask that highlights the dynamic changes of the bubble. Morphological opening operations are performed on the mask of the dynamically changing region to remove noise points, and connected component analysis algorithms are used to filter out connected components with an area greater than a preset threshold, which are identified as valid foam rupture regions; the total pixel area of ​​all identified rupture regions is counted and recorded as the instantaneous foam rupture area. Within a set time window, a linear regression is performed on the time series data of the instantaneous foam rupture area to fit its changing trend. The absolute value of the slope of the trend line is the instantaneous rate of change of the foam area. Dividing the instantaneous rate of change by the total area of ​​the foam regions identified in the same frame image yields the normalized foam bursting rate characteristic value. . In an optional implementation, the synthesized feature vector Input to a pre-defined fuzzy inference system, output the first... Adjustment amount of scraper speed in each tank chamber Specifically, it includes: Based on ash content fracture rate characteristic value and foam quantity characteristic value The pre-defined membership functions will integrate the feature vectors. Each input quantity is converted into the membership degree of the corresponding fuzzy linguistic variable; among which, the gray value The fuzzy linguistic variables include "low ash content", "medium ash content", and "high ash content", used to characterize the quality grade of clean coal; the fracture rate characteristic value The fuzzy linguistic variables include "stable," "moderate," and "violent," used to characterize the stability state of the foam layer; foam quantity characteristic values. The fuzzy linguistic variables include “poor”, “moderate”, and “rich”, which are used to characterize the flotation capacity status. Based on a predefined fuzzy rule base of flotation process expert knowledge, a Mamdani-type inference model is used for decision-making. For each rule in the rule base, the confidence of the preconditions of the rule is calculated by the minimum operator or the product operator, and the activation strength of the rule for the final speed regulation decision is obtained. Using the centroid method, the corresponding scraper rotation speed adjustment is calculated based on the activation intensity of all rules and their corresponding output fuzzy set center values. .

[0007] In one optional implementation, the fuzzy rules include at least: Rule 1, if gray value The characteristic value of "low ash content" and foam bursting rate For "stable" and foam quantity characteristic value For "enrichment", the speed adjustment amount for , This is the preset maximum positive adjustment amount; Rule 2, if the gray value Characteristic value of "high ash content" and foam bursting rate "Intense" and foam volume characteristic value If it is "poor", then the speed adjustment amount for , This is the preset maximum negative adjustment amount; Rule 3, if the gray value It is "medium ash" and the characteristic value of foam bursting rate The foam quantity characteristic value is "medium". If the value is "moderate", then the speed adjustment amount is... =0; Rule 4, if the gray value The characteristic value of "low ash content" and foam bursting rate For "stable" and foam quantity characteristic value If the value is "moderate", then the speed adjustment amount is... for ; Rule 5, if the gray value Characteristic value of "high ash content" and foam bursting rate "Intense" and foam volume characteristic value If the value is "moderate", then the speed adjustment amount is... for ;in, and The proportionality coefficient is less than 1.

[0008] In one optional implementation, the centroid method is used to calculate the corresponding scraper rotation speed adjustment based on the activation intensity of all rules and their corresponding output fuzzy set center values. Specifically, it includes: Based on comprehensive feature vectors Flotation mode identification is performed. Flotation modes include normal separation mode, high ash and difficult separation mode, and high foam flushing mode. Based on the identified flotation mode, and according to the pre-set flotation mode-rule-weight coefficient mapping relationship, the weight coefficient of each rule is obtained; Using the centroid method, the corresponding scraper rotation speed adjustment is calculated based on the weight coefficients, activation intensities, and corresponding output fuzzy set center values ​​of all rules. .

[0009] In an optional implementation, the method further includes the following steps: Record the flotation performance indicators after this adjustment, and then use the comprehensive feature vector. Adjustment amount The flotation performance indicators are used as a set of training samples and stored in the historical database; among them, the flotation performance indicators include at least one of the following: clean coal ash content stability rate, clean coal yield, or unit product power consumption. The membership function parameters or fuzzy rule weights of the fuzzy inference system are periodically optimized and iterated based on sample data in the historical database.

[0010] In one alternative implementation, the fuzzy inference system is optimized iteratively using one of a genetic algorithm, a particle swarm optimization algorithm, or a neural network algorithm.

[0011] Secondly, the technical solution of the present invention provides a real-time linkage variable speed skimming control device for ash content in a flotation machine, comprising: The flotation parameter acquisition module is used to obtain the flotation parameters of the flotation machine in real time. Ash content of foam products in each tank Foam layer thickness value and images of foam morphology ; A comprehensive feature vector construction module is used to analyze foam morphology images. Image processing was performed to extract the feature values ​​of the foam bursting rate. ; Set the foam layer thickness value By performing time-series accumulation, the characteristic value of foam volume per unit time is calculated. ;Merge the ash content values fracture rate characteristic value and foam quantity characteristic value Construct a comprehensive feature vector characterizing the current flotation state of the cell. ; The speed adjustment determination module is used to determine the comprehensive feature vector. Input to a pre-defined fuzzy inference system, output the first... Adjustment amount of scraper speed in each tank chamber The fuzzy inference system has built-in fuzzy rules based on the flotation process knowledge base. The preconditions of the fuzzy rules are fuzzy linguistic variables of ash content, foam bursting rate and foam amount, and the conclusion is the fuzzy linguistic variable of scraper speed adjustment. The speed adjustment module is used to adjust the speed according to the adjustment amount. Generate control commands and send them to the first Each slot chamber has a corresponding variable frequency motor to drive the scraper to run at the adjusted speed.

[0012] Thirdly, the technical solution of the present invention provides a terminal, comprising: The memory is used to store the real-time linkage variable speed skimming control program for ash content in the flotation machine; A processor is configured to implement the steps of the real-time linkage variable speed skimming control method for ash content in a flotation machine as described in any of the preceding claims when executing the ash content real-time linkage variable speed skimming control program for a flotation machine.

[0013] Fourthly, the present invention provides a computer-readable storage medium storing a real-time linkage variable speed skimming control program for ash content in a flotation machine. When the real-time linkage variable speed skimming control program for ash content in a flotation machine is executed by a processor, it implements the steps of the real-time linkage variable speed skimming control method for ash content in a flotation machine as described in any of the above claims.

[0014] As can be seen from the above technical solutions, this application has the following advantages: Firstly, by acquiring and fusing gray values ​​in real time... Foam volume Foam bursting rate These three feature parameters are used to construct a comprehensive feature vector. This invention can comprehensively and accurately perceive the real-time sorting status of each cell. Through a built-in fuzzy inference system, it transforms complex process expert knowledge into executable control rules, intelligently deciding the scraping speed based on the real-time comprehensive status, thus improving the accuracy of the decision results. Simultaneously, it introduces image processing technology to extract foam bursting rate feature values, automatically reducing the scraping speed when foam bursts violently or the state is unstable, minimizing disturbance to the foam layer, preventing valuable mineral particles from falling off, stabilizing the sorting process, and improving operational reliability. By integrating multi-source information such as ash content, foam quantity, and foam stability, and using fuzzy inference to make intelligent decisions on the driving system speed, this invention achieves accurate perception and adaptive control of the flotation state, thereby stabilizing the sorting process, preventing concentrate loss, and optimizing concentrate quality and yield. Attached Figure Description

[0015] To more clearly illustrate the technical solution of this application, the accompanying drawings used in the description will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the overall structure of the variable speed skimmer for a 3-cell flotation machine.

[0017] Figure 2 This is a schematic diagram of a real-time linkage variable speed skimming control method for ash content in a flotation machine, provided by an embodiment of the present invention.

[0018] Figure 3 This is a schematic block diagram of a real-time linkage variable speed skimming control device for ash content in a flotation machine, provided as an embodiment of the present invention.

[0019] Figure 4 This is a schematic diagram of the structure of a terminal provided in an embodiment of the present invention. Detailed Implementation

[0020] To make the purpose, features, and advantages of this application more apparent and understandable, specific embodiments and accompanying drawings will be used to clearly and completely describe the technical solution protected by this application. Obviously, the embodiments described below are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] Unless otherwise defined, all technical and scientific terms used in this application have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this application and in the specification of this invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

[0022] Figure 1 This is a schematic diagram of the overall structure of the variable speed frothing device for a 3-cell flotation machine, including the flotation cell body 1, a variable frequency motor drive system 2, a flotation machine scraper device 3, an online ash content analyzer 4, a flotation concentrate ash content measuring sensor 5, a foam thickness detection system 6, and an optical level sensor 7. Each cell's scraper has an independent drive system; specifically, a variable frequency motor drives the scraper in each cell independently, allowing for independent speed control of the scraper's rotation. An online ash content analyzer 4 is installed between adjacent cells, monitoring the ash content changes of the foam product in real time via the flotation concentrate ash content measuring sensor 5. A foam thickness detection system 6 is installed in each cell, collecting foam layer thickness data in real time via the optical level sensor 7. The system 6 or a backend system calculates the cumulative foam volume based on the average foam layer thickness.

[0023] Figure 2 This is a schematic flowchart of a real-time linkage variable speed skimming control method for ash content in a flotation machine, provided by an embodiment of the present invention. Figure 1 The executing entity can be a real-time linkage variable speed frothing control device for ash content in a flotation machine. The real-time linkage variable speed frothing control method for ash content in a flotation machine provided in this embodiment of the invention is executed by computer equipment; correspondingly, the real-time linkage variable speed frothing control device for ash content in the flotation machine operates within the computer equipment. Depending on different requirements, the order of the steps in this flowchart can be changed, and some steps can be omitted.

[0024] like Figure 2 As shown, the method includes the following steps.

[0025] S1, Real-time acquisition of flotation machine number Ash content of foam products in each tank Foam layer thickness value and images of foam morphology .

[0026] S2, Images of foam morphology Image processing was performed to extract the feature values ​​of the foam bursting rate. ; Set the foam layer thickness value By performing time-series accumulation, the characteristic value of foam volume per unit time is calculated. ;Merge the ash content values fracture rate characteristic value and foam quantity characteristic value Construct a comprehensive feature vector characterizing the current flotation state of the cell. .

[0027] S3, the comprehensive feature vector Input to a pre-defined fuzzy inference system, output the first... Adjustment amount of scraper speed in each tank chamber The fuzzy inference system has built-in fuzzy rules based on the flotation process knowledge base. The preconditions of the fuzzy rules are fuzzy linguistic variables of ash content, foam bursting rate and foam amount, and the conclusion is the fuzzy linguistic variable of scraper speed adjustment.

[0028] S4, according to the adjustment amount Generate control commands and send them to the first Each slot chamber has a corresponding variable frequency motor to drive the scraper to run at the adjusted speed.

[0029] As a refinement and extension of the specific implementation of the above embodiments, in order to fully explain the specific implementation process of this embodiment, the following will provide possible embodiments to describe the specific implementation of the above steps in a non-limiting manner.

[0030] Step S1: Obtain the ash content of the foam product in real time. This is achieved through an online ash analyzer; real-time acquisition of foam layer thickness values ​​is also possible. This is achieved through an optical liquid level sensor 7; real-time acquisition of foam morphology images. Through industrial cameras ( Figure 1 (Not shown in the image) is implemented.

[0031] Step S2 involves analyzing the foam morphology image. Image processing was performed to extract the feature values ​​of the foam bursting rate. This includes: performing differential operations on consecutive frames of foam morphology images to identify dynamic changes in the foam region; statistically analyzing the rate of decrease in the foam region area per unit time and normalizing it to obtain the characteristic value of the foam bursting rate. Specifically, it includes the following steps.

[0032] S21, the foam morphology images of consecutive frames are sequentially grayscaled and Gaussian filtered to suppress image noise introduced by slurry splashing, light fluctuations and equipment vibration.

[0033] S22, calculate the absolute difference map of the preprocessed images of adjacent frames, and binarize the difference map using an adaptive thresholding method to generate a region mask that highlights the dynamic changes of the bubble.

[0034] Calculate the absolute difference map of adjacent frames. , is represented as ,

[0035] in, The frame interval time. , Indicates the first The first slot in the Time, Number Images of moments This represents the grayscale conversion function. For difference images... Binarization is performed to obtain the mask of the dynamically changing region. .

[0036] S23, perform morphological opening operation on the mask of the dynamically changing region to remove noise points, and use the connected component analysis algorithm to filter out connected regions with an area greater than a preset threshold and identify them as valid foam rupture regions; count the total pixel area of ​​all identified rupture regions and record it as the instantaneous foam rupture area.

[0037] S24. Within a set time window, perform linear regression on the time series data of the instantaneous foam rupture area to fit its changing trend. The absolute value of the slope of the trend line is the instantaneous rate of change of the foam area.

[0038] S25, divide the instantaneous rate of change by the total area of ​​the foam regions identified in the same frame image to obtain the normalized foam bursting rate feature value. .

[0039] The characteristic value of the foam burst rate is expressed as follows:

[0040] in, The total area of ​​the foam region in the image is used to normalize the bursting rate and eliminate the influence of the image field of view size. The instantaneous fracture area, It represents the absolute value of the rate of change of the ruptured area over time, and is used to characterize the severity of foam rupture.

[0041] In step S2, the foam layer thickness value is... By performing time-series accumulation, the characteristic value of foam volume per unit time is calculated. Specifically, it includes the following steps.

[0042] S26, continuously acquire foam layer thickness values ​​at a fixed sampling frequency. .

[0043] The fixed frequency can be 1Hz.

[0044] S27, in each control cycle, the system maintains a length of A first-in, first-out (FIFO) data buffer is used to store past data. Historical thickness data over a period of time.

[0045] in, , The sampling interval is denoted as .

[0046] S28, by calculating the arithmetic mean of all data in the buffer, the characteristic value of the amount of foam per unit time can be obtained. , is represented as ,

[0047] This embodiment obtains the result through time-series accumulation. This smooths out short-term measurement noise and instantaneous fluctuations in the foam layer (such as shaking caused by the bursting of a single large bubble), thereby extracting a characteristic quantity that more stably reflects the foam generation trend and the tank load.

[0048] In step S3, the comprehensive feature vector is... Input to a pre-defined fuzzy inference system, output the first... Adjustment amount of scraper speed in each tank chamber Specifically, it includes the following steps.

[0049] S31, based on ash content value fracture rate characteristic value and foam quantity characteristic value The pre-defined membership functions will integrate the feature vectors. Each input quantity is converted into the membership degree of the corresponding fuzzy linguistic variable.

[0050] Among them, ash content The fuzzy linguistic variables include “low ash content”, “medium ash content”, and “high ash content”, which are used to characterize the quality grade of clean coal; their membership functions adopt trapezoidal or trigonometric functions.

[0051] Domain: Defined according to the quality standards and specific process requirements of refined coal products. For example, the domain range is [a1, a4] (unit: %).

[0052] "Low gray content" variable: A trapezoidal membership function is used, with the core interval being [a1, a2]. When the gray content is lower than a1, the membership degree is 1; when it is between a1 and a2, the membership degree linearly decreases from 1 to 0.5; when it is between a2 and a3, the membership degree linearly decreases from 0.5 to 0.

[0053] The "medium gray" variable uses a triangular membership function with its peak point located at a2. The range that completely belongs to the "medium gray" is [a2-δ, a2+δ].

[0054] "High gray content" variable: A trapezoidal membership function is used, with its core interval being [a3, a4]. When the gray content value is higher than a4, the membership degree is 1; when it is between a3 and a4, the membership degree increases linearly from 0 to 1.

[0055] The specific values ​​of parameters a1, a2, a3, a4, and δ are set by experts based on actual flotation process parameters.

[0056] fracture rate characteristic value The fuzzy linguistic variables include “stable”, “moderate”, and “violent”, which are used to characterize the stability state of the foam layer.

[0057] Domain: Set according to the statistical range of the image processing results. For example, the domain of the normalized feature values ​​is [0, 1].

[0058] "Stable" variables: employ a descending trapezoidal membership function with a core interval of [0, b1]. When When the value is lower than b1, the membership degree is 1; when it is between b1 and b2, the membership degree decreases linearly from 1 to 0.

[0059] "Medium" variables: use a symmetrical triangular membership function with a peak point at b2 and a span covering [b1, b3].

[0060] "Serious" variables: An ascending trapezoidal membership function is used, with a core interval of [b3, 1]. When When the value is higher than b3, the membership degree is 1; when it is between b2 and b3, the membership degree increases linearly from 0 to 1.

[0061] The specific values ​​of parameters b1, b2, and b3 were determined by cluster analysis of historical bubble image data.

[0062] Foam quantity characteristic value The fuzzy linguistic variables include “poor”, “moderate”, and “rich”, which are used to characterize the flotation capacity status.

[0063] Domain: Determined based on the volume of the flotation tank, the maximum skimming load, and historical data statistics. For example, the domain is [q1, q4] (the unit can be a normalized value or an actual physical quantity).

[0064] "Poor" variables: A trapezoidal membership function is used, with its core interval being [q1, q2]. When... When the value is below q1, the membership degree is 1; when it is between q1 and q2, the membership degree decreases linearly from 1 to 0.

[0065] "Moderate" variables: adopt a triangular membership function with a peak point at q2 and a support interval of [q1, q3].

[0066] "Enriched" variables: A trapezoidal membership function is used, with the core interval being [q3, q4]. When When the value is higher than q4, the membership degree is 1; when it is between q3 and q4, the membership degree increases linearly from 0 to 1.

[0067] The specific values ​​of parameters q1, q2, q3, and q4 are determined by the design load and optimal operating range of the flotation machine.

[0068] Input variables (represent , , One of them belongs to a certain fuzzy linguistic value. membership degree The membership function is used to calculate the membership degree of the precise input value, which is then converted into a fuzzy set.

[0069] S32, based on a predefined fuzzy rule base of flotation process expert knowledge, a Mamdani-type inference model is used for decision-making; for each rule in the rule base, the confidence of the preconditions of the rule is calculated by the minimum operator or the product operator, and the activation strength of the rule for the final speed regulation decision is obtained.

[0070] In some alternative implementations, the fuzzy rules include at least: Rule 1, if gray value The characteristic value of "low ash content" and foam bursting rate For "stable" and foam quantity characteristic value For "enrichment", the speed adjustment amount for , This is the preset maximum positive adjustment amount; Rule 2, if the gray value Characteristic value of "high ash content" and foam bursting rate "Intense" and foam volume characteristic value If it is "poor", then the speed adjustment amount for , This is the preset maximum negative adjustment amount; Rule 3, if the gray value It is "medium ash" and the characteristic value of foam bursting rate The foam quantity characteristic value is "medium". If the value is "moderate", then the speed adjustment amount is... =0; Rule 4, if the gray value The characteristic value of "low ash content" and foam bursting rate For "stable" and foam quantity characteristic value If the value is "moderate", then the speed adjustment amount is... for ; Rule 5, if the gray value Characteristic value of "high ash content" and foam bursting rate "Intense" and foam volume characteristic value If the value is "moderate", then the speed adjustment amount is... for ;in, and The proportionality coefficient is less than 1.

[0071] A Mamdani-type fuzzy inference model is adopted. For each rule... Its activation intensity It is obtained by taking the minimum or multiplicative membership degree of the preconditions, and is expressed as follows:

[0072] or

[0073] in, express Regarding the rules Gray fuzzy linguistic values ​​defined in The membership degree is calculated similarly for other variables.

[0074] S33, using the centroid method, calculates the corresponding scraper speed adjustment based on the activation intensity of all rules and their corresponding output fuzzy set center values. .

[0075] The corresponding scraper speed adjustment amount is calculated. Represented as,

[0076] in, It is the total number of fuzzy rules. It is the first The rule's conclusion outputs the center value of the fuzzy set; for example, it outputs the linguistic variable "speed adjustment amount". for The coordinates of the center point of the corresponding triangle membership function.

[0077] In some alternative implementations, based on the comprehensive feature vector The flotation process involves identifying the flotation mode, dynamically adjusting the weight of each rule in the fuzzy rule base based on the identified mode, and then performing weighted fuzzy inference. Specifically: S331, based on comprehensive feature vectors The flotation operating mode is identified, which includes normal separation mode, high ash and difficult separation mode, and high foam flushing mode.

[0078] Operating condition pattern recognition is a rule-based classifier whose input is a comprehensive feature vector. The output is the identifier of the most likely operating condition.

[0079] Normal sorting mode: Its characteristic is gray value Within the target range, the rate of foam bursting Low and stable, foam volume Moderate. In this mode, the system's core objectives are stable production and ensuring product quality.

[0080] High gray difficulty selection mode: Its characteristic is gray value The level remains high, and the rate of foam bursting is also high. The amount of foam may increase due to the deterioration of mineral surface properties. The amount may be too small. In this mode, the system should focus on enhancing secondary enrichment and reducing concentrate ash content as its core objectives.

[0081] High foam volume flushing mode: characterized by high foam volume Abnormally high, far exceeding normal load, and at the same time, the foam bursting rate... The ash content may increase due to an excessively thick or unstable foam layer. The temperature may rise due to increased mechanical entrainment. In this mode, the system should focus on stabilizing the foam layer, preventing overflow accidents, and avoiding deterioration of concentrate quality.

[0082] The identification process involves judging... This is accomplished by checking whether each component falls within a pre-defined threshold range of characteristic values ​​for different modes set by process experts. For example: IF AND THEN Operating Condition = "High Gray Difficulty Selection Mode"; IF THEN Operating Condition = "High Foam Flushing Mode"; ELSE condition = "Normal sorting mode".

[0083] S332: Based on the identified flotation mode, and according to the pre-set flotation mode-rule-weight coefficient mapping relationship, obtain the weight coefficient of each rule.

[0084] A pre-defined flotation operating condition mode-rule-weight coefficient mapping table is provided. This mapping table defines the weight coefficient corresponding to each rule in the fuzzy rule base under each operating condition mode. (in ≥1). The magnitude of the weighting coefficient directly reflects the importance and priority of the rule under the current working conditions.

[0085] S333 uses the centroid method to calculate the corresponding scraper speed adjustment based on the weight coefficients, activation intensities, and corresponding output fuzzy set center values ​​of all rules. .

[0086] The corresponding scraper speed adjustment amount is calculated. Represented as,

[0087] in, The first one obtained in step S332 The weighting coefficient of each rule.

[0088] This specific implementation can "sense" the main contradiction in current production (whether it's poor quality or unstable foam) and prioritize the expert strategy that best resolves the contradiction, thus improving the purposefulness and effectiveness of control decisions. By differentiating the weights of rules under different modes, the system avoids the problem of decision-making difficulties when multiple rule suggestions conflict, ensuring the clarity and consistency of control instructions. In the face of abnormal operating conditions, it can automatically switch to a different strategy, improving the stability and adaptability of the entire flotation process when raw material properties fluctuate.

[0089] In some alternative implementations, the method further includes the following steps.

[0090] S5, record the flotation performance indicators after this adjustment, and then convert the comprehensive feature vector... Adjustment amount The flotation performance indicators are used as a set of training samples and stored in the historical database; among them, the flotation performance indicators include at least one of the following: clean coal ash content stability rate, clean coal yield, or unit product power consumption.

[0091] S6. Periodically optimize and iterate the membership function parameters or fuzzy rule weights of the fuzzy inference system based on sample data in the historical database.

[0092] The optimization and iteration of the fuzzy inference system can be performed using one of the following algorithms: genetic algorithm, particle swarm optimization algorithm, or neural network algorithm.

[0093] The following uses a genetic algorithm as an example to illustrate how a genetic algorithm can be used to periodically optimize and iterate the parameters of a fuzzy inference system based on sample data from a historical database. The specific steps include:

[0094] S61. Initialization and Encoding.

[0095] The optimization objective is defined as minimizing the overall control error of historical samples. This error can be the deviation between the setpoint and actual value of clean coal ash content, or the negative correlation function between clean coal yield and expected value, or a weighted composite index of the above indicators and unit product power consumption. The parameters to be optimized in the fuzzy inference system (such as the shape parameter of the membership function, rule weight coefficients, etc.) Each parameter is encoded as a chromosome. Each parameter is encoded as a gene on the chromosome using real-number encoding. An initial population of P individuals is generated, where P is the population size.

[0096] S62, Fitness Assessment.

[0097] For each individual in the population (i.e., a specific set of parameter combinations), it is decoded and configured into the fuzzy inference system. This is achieved using N sets of sample data from a historical database. (in Using the actual flotation performance indicators as the test set, forward simulation calculations were performed: the samples... The input is fed into the configured fuzzy system to obtain the inferred speed adjustment amount. And compare it with the optimal adjustment amount actually implemented in the sample. Compare the samples. Calculate the individual's fitness value across all samples. :

[0098] fitness value The smaller the value, the closer the fuzzy system output under this set of parameters is to the historical optimal decision, and the higher the individual fitness.

[0099] S63, Select Operation.

[0100] Selection is achieved using either roulette wheel selection or tournament selection, based on an individual's fitness value. Individuals with higher fitness (i.e., smaller margin of error) have a greater probability of being selected for the next generation. This ensures that superior genes have a higher chance of being preserved and inherited.

[0101] S64, Cross Operation.

[0102] For the selected parent individuals, a preset crossover probability is used. Simulated binary crossover or other real-number crossover algorithms are performed to exchange some genetic information and produce new offspring. The crossover operation aims to explore new regions of the parameter space and maintain population diversity.

[0103] S65, Mutation operation.

[0104] For the offspring produced after crossover, with a smaller probability of mutation Performing polynomial mutations or other mutation operations randomly changes the values ​​of one or more genes on a chromosome. Mutation operations introduce random perturbations, which help to escape local optima and expand the search range.

[0105] S66. Judgment of the formation and termination of a new generation of populations.

[0106] The new individuals generated through selection, crossover, and mutation operations form a new generation population. Steps S62 to S65 are repeated for iterative evolution. The population continues until the preset maximum number of generations, G, is reached. max The algorithm iteration terminates when the fitness value of the best individual in the population no longer improves significantly over several consecutive generations.

[0107] S67, System parameters updated.

[0108] The individual with the highest fitness in the final generation of the population is decoded to obtain the optimized optimal parameter combination, which is then used to update the membership function parameters and / or rule weight coefficients of the online fuzzy inference system. This completes a closed loop of self-learning and performance improvement.

[0109] The foregoing has described in detail an embodiment of a real-time linkage variable speed skimming bubble control method for ash content in a flotation machine. Based on the real-time linkage variable speed skimming bubble control method for ash content in a flotation machine described in the above embodiment, this invention also provides a real-time linkage variable speed skimming bubble control device for ash content in a flotation machine corresponding to the method.

[0110] Figure 3 This is a schematic block diagram of a real-time linkage variable speed frothing control device for ash content in a flotation machine, provided by an embodiment of the present invention. In this embodiment, the real-time linkage variable speed frothing control device 300 for ash content in a flotation machine can be divided into multiple functional modules according to the functions it performs, such as... Figure 3 As shown. The module referred to in this invention is a series of computer program segments that can be executed by at least one processor and perform a fixed function, and which are stored in memory.

[0111] Flotation parameter acquisition module 310 is used to acquire the flotation parameters of the flotation machine in real time. Ash content of foam products in each tank Foam layer thickness value and images of foam morphology .

[0112] The comprehensive feature vector construction module 320 is used for analyzing foam morphology images. Image processing was performed to extract the feature values ​​of the foam bursting rate. ; Set the foam layer thickness value By performing time-series accumulation, the characteristic value of foam volume per unit time is calculated. ;Merge the ash content values fracture rate characteristic value and foam quantity characteristic value Construct a comprehensive feature vector characterizing the current flotation state of the cell. .

[0113] The speed adjustment determination module 330 is used to determine the comprehensive feature vector. Input to a pre-defined fuzzy inference system, output the first... Adjustment amount of scraper speed in each tank chamber The fuzzy inference system has built-in fuzzy rules based on the flotation process knowledge base. The preconditions of the fuzzy rules are fuzzy linguistic variables of ash content, foam bursting rate and foam amount, and the conclusion is the fuzzy linguistic variable of scraper speed adjustment.

[0114] Speed ​​adjustment module 340, used to adjust according to the adjustment amount Generate control commands and send them to the first Each slot chamber has a corresponding variable frequency motor to drive the scraper to run at the adjusted speed.

[0115] The real-time linkage variable speed frothing control device for ash content in the flotation machine of this embodiment is used to implement the aforementioned real-time linkage variable speed frothing control method for ash content in the flotation machine. Therefore, the specific implementation of this device can be found in the embodiment section of the real-time linkage variable speed frothing control method for ash content in the flotation machine above. So, its specific implementation can be referred to the description of the corresponding embodiments, and will not be described in detail here.

[0116] Furthermore, since the ash real-time linkage variable speed skimming control device for flotation machine in this embodiment is used to implement the aforementioned ash real-time linkage variable speed skimming control method for flotation machine, its function corresponds to the function of the above method, and will not be repeated here.

[0117] Figure 4 This is a schematic diagram of a terminal 400 provided in an embodiment of the present invention, including: a processor 410, a memory 420, and a communication unit 430. The processor 410 is used to implement the process steps of the above-described embodiment of the real-time linkage variable speed skimming control method for ash content in a flotation machine when implementing the ash content real-time linkage variable speed skimming control program for a flotation machine stored in the memory 420.

[0118] The terminal 400 includes a processor 410, a memory 420, and a communication unit 430. These components communicate via one or more buses. Those skilled in the art will understand that the server structure shown in the figure does not constitute a limitation of the present invention. It can be a bus topology or a star topology, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0119] The memory 420 can be used to store the execution instructions of the processor 410. The memory 420 can be implemented by any type of volatile or non-volatile storage terminal 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 read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. When the execution instructions in the memory 420 are executed by the processor 410, the terminal 400 is able to perform some or all of the steps in the above method embodiments.

[0120] The processor 410 serves as the control center of the storage terminal, connecting various parts of the electronic terminal via various interfaces and lines. It executes software programs and / or modules stored in the memory 420, and calls data stored in the memory to perform various functions of the electronic terminal and / or process data. The processor can be composed of integrated circuits (ICs), such as a single packaged IC or multiple packaged ICs with the same or different functions connected together. For example, the processor 410 may consist only of a central processing unit (CPU). In this embodiment of the invention, the CPU may have a single processing core or include multiple processing cores.

[0121] The communication unit 430 is used to establish a communication channel, enabling the storage terminal to communicate with other terminals. It can receive user data sent by other terminals or send user data to other terminals.

[0122] This invention also provides a computer storage medium, which may be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc. The computer storage medium stores a real-time linked variable speed bubble scraping control program for a flotation machine. When executed by a processor, this program implements the process steps of the above-described embodiment of the real-time linked variable speed bubble scraping control method for a flotation machine.

[0123] Those skilled in the art will clearly understand that the techniques in the embodiments of the present invention can be implemented using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium such as a USB flash drive, mobile hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, or other media capable of storing program code. It includes several instructions to cause a computer terminal (which may be a personal computer, server, or a second terminal, network terminal, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

[0124] In the embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0125] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0126] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0127] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An ash real-time linkage variable speed bubble scraping control method for a flotation machine, characterized in that, The method comprises the following steps: Real-time acquisition of flotation machine data Ash content of foam products in each tank Foam layer thickness value and images of foam morphology ; Image processing is performed on the foam morphology image to extract a bubble size value Image processing is performed to extract a bubble size value Image processing is performed to extract a bubble size value A time-series accumulation is performed to calculate a bubble size value per unit time The ash content value , the bubble size value , and the bubble size value per unit time are fused to construct a comprehensive feature vector representing the current state of the flotation tank ​ The comprehensive feature vector is input to a preset fuzzy inference system, and an adjustment amount of the rotating speed of the first slot chamber scraper is output . The fuzzy inference system is built-in with fuzzy rules based on a flotation process knowledge base, premise conditions of the fuzzy rules are fuzzy language variables of ash content, froth collapse rate and froth volume, and a conclusion is a fuzzy language variable of scraper speed adjustment amount; According to the adjustment amount , a control instruction is generated and sent to the variable frequency motor corresponding to the first chamber to drive the scraper to run at the adjusted rotational speed.

2. The method of claim 1, wherein, Image processing is performed on the foam morphology images Image processing is performed to extract the foam collapse rate characteristic value , specifically comprising: The froth pattern images of continuous frames are sequentially subjected to grayscale and Gaussian filtering processing to suppress image noise introduced by pulp splashing, light fluctuation and equipment vibration; An absolute difference image of the preprocessed images of adjacent frames is calculated, and the difference image is binarized using an adaptive threshold method to generate a region mask highlighting dynamic changes in froth; The dynamic change region mask is subjected to morphological opening operation to remove noise points, and a connected region analysis algorithm is used to screen out connected regions with an area greater than a preset threshold, which are identified as effective froth collapse regions; the total pixel area of all identified collapse regions is counted and recorded as the instantaneous froth collapse area; Within a set time window, linear regression is performed on the time series data of the instantaneous froth collapse area to fit its trend, and the absolute value of the slope of the trend line is the instantaneous change rate of the froth area; Dividing the said instantaneous change rate by the total area of the identified foam region in the same frame image, finally obtaining the normalized foam breakage rate eigenvalue .

3. The method of claim 1, wherein, The comprehensive feature vector is input to a preset fuzzy inference system, and an adjustment amount of the rotation speed of the first slot chamber scraper is output , and specifically comprises: According to the ash value , the breakage rate characteristic value and the froth amount characteristic value , the membership functions respectively preset, each input quantity in the comprehensive characteristic vector is converted into the membership of the corresponding fuzzy language variable; wherein, the fuzzy language variable of the ash value includes "low ash", "medium ash", "high ash", used to represent the quality grade of clean coal; the fuzzy language variable of the breakage rate characteristic value includes "stable", "moderate", "severe", used to represent the stability state of the froth layer; the fuzzy language variable of the froth amount characteristic value includes "poor", "moderate", "rich", used to represent the flotation capacity state; Based on the fuzzy rule base of the pre-defined flotation process expert knowledge, a Mamdani type inference model is used for decision-making; for each rule in the rule base, the premise condition confidence of the rule is calculated by using a minimum operator or a product operator to obtain the activation strength of the rule for the final speed regulation decision; The adjusting amount of the scraper rotating speed is calculated according to the activation intensity of all rules and the corresponding output fuzzy set center value by using the barycentric method .

4. The method of claim 3, wherein, The fuzzy rules at least include: Rule one, if the ash value is "low ash" and the foam collapse rate characteristic value is "stable" and the foam amount characteristic value is "rich", then the rotational speed adjustment amount is , , , , , is the preset maximum positive adjustment amount; Rule two, if the ash value is "high ash" and the foam breakage rate characteristic value is "fast" and the foam amount characteristic value is "poor", then the rotation speed adjustment amount is is a preset maximum negative adjustment amount.​​​​​ Rule 3, if the gray value The characteristic value of the foam bursting rate is "medium ash content". The foam quantity characteristic value is "medium". If it is "moderate", then the speed adjustment amount is... =0; Rule four, if the ash value is "low ash" and the foam collapse rate characteristic value is "stable" and the foam amount characteristic value is "moderate", then the rotation speed adjustment amount is ;​​​​ Rule 5, if the gray value Characteristic value of "high ash content" and foam bursting rate "Violent" and foam volume characteristic value If it is "moderate", then the speed adjustment amount is... for ;in, and The proportionality coefficient is less than 1.

5. The method according to claim 3 or 4, characterized in that, The adjusting amount of the scraper rotating speed is calculated according to the activation intensity of all rules and the corresponding output fuzzy set center value by using the barycentric method , and specifically comprises: Based on comprehensive feature vectors Performing flotation condition mode recognition, the flotation condition mode including a normal separation mode, a high-ash difficult separation mode, and a high-foam amount scouring mode. According to the identified flotation condition mode, the weight coefficients of each rule are obtained according to a pre-set mapping relationship between the flotation condition mode, the rule and the weight coefficient; The adjusting amount of the scraper rotating speed is calculated according to the weight coefficient of all rules, the activation intensity and the corresponding output fuzzy set center value by using the barycentric method .

6. The method of claim 1, wherein, The method further comprises the following steps: record the flotation effect index after the present regulation, and store the comprehensive feature vector , the adjustment amount and the flotation effect index as a group of training samples into a historical database; wherein the flotation effect index includes at least one of clean coal ash stability rate, clean coal yield, or unit product power consumption. Periodically, the membership function parameters or the fuzzy rule weights of the fuzzy inference system are optimized and iterated based on sample data in the historical database.

7. The method of claim 6, wherein, The optimization and iteration of the fuzzy inference system use one of a genetic algorithm, a particle swarm optimization algorithm or a neural network algorithm.

8. An ash real-time linkage variable speed bubble scraping control device for a flotation machine, characterized by, Comprise: The flotation parameter acquisition module is configured to acquire, in real time, a froth product ash value, a froth layer thickness value, and a froth morphology image of the first tank chamber of the flotation machine. ​​​​ Comprehensively The feature vector construction module is configured to construct a feature vector of the froth morphology image The image processing is performed to extract a bubble breakage rate feature value The froth layer thickness value The time series accumulation is performed to calculate a froth amount feature value per unit time The ash content value , the bubble breakage rate feature value and the froth amount feature value are fused to construct a comprehensive feature vector representing the current tank flotation state ​ A rotating speed adjustment amount determination module is configured to input the comprehensive feature vector into a preset fuzzy inference system, and output an adjustment amount of the rotating speed of the first slot chamber scraper ;​​ The fuzzy inference system is built-in with fuzzy rules based on a flotation process knowledge base, premise conditions of the fuzzy rules are fuzzy language variables of ash content, froth collapse rate and froth volume, and a conclusion is a fuzzy language variable of scraper speed adjustment amount; The speed adjustment module is used to adjust the speed according to the adjustment amount. Generate control commands and send them to the first Each slot chamber has a corresponding variable frequency motor to drive the scraper to run at the adjusted speed.

9. A terminal, characterized by comprising: Comprise: A memory is configured to store an ash real-time linkage variable-speed froth scraping control program for a flotation machine; A processor is configured to execute the ash real-time linkage variable-speed froth scraping control program for the flotation machine to implement the steps of the ash real-time linkage variable-speed froth scraping control method for the flotation machine according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The readable storage medium stores the ash real-time linkage variable-speed froth scraping control program for the flotation machine, and the ash real-time linkage variable-speed froth scraping control program for the flotation machine is executed by the processor to implement the steps of the ash real-time linkage variable-speed froth scraping control method for the flotation machine according to any one of claims 1 to 7.

Images