EUR
en
Flotation is the most widely used technology for mineral separation and puri fi cation. The fl otation produc-tion process has complex mechanism characteristics and is in fl uenced by multiple variables that are coupled with each other, which has always been a di ffi culty in controlling the bene fi ciation process. The fl otation system of mineral pro-cessing plants mostly relies on manual control, which is in fl uenced by subjective factors such as worker experience, technical level, and sense of responsibility, making it di ffi -cult to optimize control parameters and maximize produc-tion effi ciency. This paper systematically summarized the automation systems of fl otation equipment such as automatic dosing device, automatic liquid level detection device, automatic feed concentration adjustment device, and automatic feed fl ow adjustment device. The accurate extraction methods of physical and dynamic characteristics such as color, texture, size, and moving speed of fl otation froth were reviewed. The traditional data-driven model and image feature-based prediction methods for prediction of the grade, recovery rate, ash content in the concentrate, and tailings were combed. On this basis, a technical route for achieving intelligent fl otation process was proposed with the aim of providing theoretical and practical references through the collaborative operation of fl otation devices, detection sensors, and machine learning algorithms.
fl otation process; fl otation equipment automa-tion; intelligent froth feature extraction; key indicator prediction
Mineral resources are the fundamental material materials on which human society relies for survival and develop-ment. According to the di ff erent physical, chemical, or physicochemical properties of minerals, di ff erent methods are used to separate useful minerals from gangue, and various coexisting useful minerals are separated from each other as much as possible to remove or reduce harmful impurities, in order to obtain raw materials required for smelting or other industries. The separation process is called bene fi ciation (Zhou and Geng 2014). Mineral processing technology mainly includes fl ota-tion, magnetic separation, gravity separation, and electric separation. Among them, fl otation technology is the most widely used technology in the fi eld of mineral processing with nearly 90 % of nonferrous metal minerals using fl ota-tion separation (Jin and Zhang 2021; Zhao and Li 2004). Flotation is a scienti fi c technology that separates minerals based on their di ff erent surface physicochemical properties and their fl oatability. Flotation can be divided into positive fl otation and reverse fl otation according to di ff erent valu-able components. The method of discharging useless minerals (i.e., gangue minerals) as tailings in the slurry is called positive fl otation. On the contrary, it is called reverse fl otation. The commonly used fl otation reagents in fl otation include collectors, frothers, inhibitors, activators, pH adjusters, dispersants, fl occulants, etc. It is necessary to follow the order of adding adjusters, inhibitors or activators, collectors, and frothers. Common fl otation machines include mechanical stirring, in fl atable, and in fl atable mechanical stirring. Flotation is a vast and complex process composed of multiple coupled sub processes (Chen 2020). The fl otation equipment ( fl otation machine and fl otation column) is the core of achieving fl otation production, and its control parameters (mainly including stirring speed, slurry level, reagent addition amount, in fl ation amount, etc.) a ff ect the fl otation production indicators and e ffi ciency (Yu 2022). There are three main types of fl otation production control systems: manual control, automatic control, and intelligent control. In the traditional fl otation process, the adjustment of control parameters mainly depends on the arti fi cial perception of the color, size, fl ow speed, bubble hanging sound, and other apparent characteristics of fl otation froth to determine whether the fl otation process operates nor-mally, and to determine how to adjust the pulp level, reagent amount, maximum amount, and other control parameters according to experience. This manual control method is di ffi cult to optimize control parameters and maximize pro-duction e ffi ciency due to subjective factors such as worker experience, technical level, and sense of responsibility (Xue et al. 2015; Zhang and Yu 2002). Automatic control mainly involves accurately adding dosage through stable feeding properties, and when the feeding properties or system state changes, it relies on manual modi fi cation of corresponding parameters, which only has feedforward control. Intelligent control is the addition of feedback sensors on the basis of automatic control, which can quickly identify changes in feed properties and system status, provide feedback, and optimize control parameters (Jin and Zhang 2021). Numerous studies have shown that achieving intelligent fl otation processes requires starting from both hardware and software aspects: improving the automation and accu-racy of fl otation equipment (devices and sensors) and opti-mizing machine learning adjustment algorithms. Therefore, this article summarizes the current research status of fl otation equipment automation technology and variable detection intelligent technology and proposes a technical route to achieve intelligent fl otation production process, in order to provide theoretical and application references for intelligent fl otation process.
The automation system of fl otation equipment mainly includes automatic dosing device, liquid level automatic detection device, feed concentration automatic adjustment device, and feed fl ow automatic adjustment device.
Flotation, as an important operational link in the production process of mineral processing plants, is closely related to the rationality of reagent addition. Reasonable automatic dosing is an important means to ensure accurate dosing of drugs. The mining automation technology in Western coun-tries is relatively developed. In the 1950s, countries such as Sweden, the former Soviet Union, and Finland began to develop a series of automatic dosing control devices, mainly including two types: micro metering pumps and di ff erential pressure control valves (Ji 2016). The metering pump dosing machine is equipped with a metering pump at each dosing point, characterized by precise dosing and good corrosion resistance, but the disadvantage is that the operation is relatively cumbersome. The pressure di ff erence control valve dosing machine uses air pressure di ff erence to trans-port medicine liquid, but it also has complex operation and few industrial applications (Wu 2019). In the 1960s, China successively developed cup type automatic dosing devices and siphon type dosing devices. However, due to di ffi culties in achieving precise dosing, pipeline blockage, and siphon disconnection, these two types of automatic dosing devices have gradually been phased out. After the 1970s, many bene fi ciation plants began to use modern equipment controlled by digital computers, such as a diaphragm pump reagent adding machine at a bene fi ciation plant in Queensland, Australia. The fl otation operation was carried out using a segmented dosing method, and the system was added reagents by real-time monitoring of product ash content changes. Currently, the widely used automatic dosing devices in the mineral processing fi eld are electro-magnetic valve type, volumetric pump type, and weighing type technologies, which have advantages such as closed-loop control, precise dosing, and convenient operation.
The electromagnetic valve type dosing device is composed of a storage device, a dosing device, a liquid level alarm, an electromagnetic valve, and a fl oat valve. Its working prin-ciple is based on the basic principle of ori fi ce fl ow and the intermittent dosing method. Currently, there is massive research and application on electromagnetic valve type dosing devices in China. For instance, the KMUST-FDCS series dosing control system developed by Kunming University of Technology has been applied to the Damajianshan bene fi ciation plant of Yunnan Lvchun Mining Co., Ltd., improving the e ffi ciency of reagent utilization (Wang 2009). The BRFS type control system developed by Beijing Research Institute of Mining and Met-allurgy is applied to the sorting system of the Duobaoshan Copper Molybdenum Mine in Heilongjiang Province. Eight reagents, including kerosene, yellow medicine, black medi-cine, Z-200, sodium sul fi de, water glass, terpineol oil, and mercaptoacetic acid, are added to the process. The experi-mental results show that the error between the actual dosage and the designed dosage of all reagents is within ±2 %, indicating a good automatic dosing e ff ect (Yan et al. 2016). It should be pointed out that when the dosage per unit cycle is too large, it is di ffi cult for the fl oat valve to maintain a constant liquid level in a timely and accurate manner, which can cause fl uctuations in the reagent fl ow through the so-lenoid valve and a ff ect the accuracy of dosing. If the dosage is too small and the selection of the solenoid valve is also small, it is easy to cause blockage of the solenoid valve.
Positive displacement pump type dosing device is a type of pump that utilizes changes in the volume of the pump cyl-inder to transport liquid. There are two types of pumps: rotor pump and reciprocating pump. When selecting the automatic dosing device, the rotor pump uses a single screw pump for dosing, and the reciprocating pump uses ametering pump for dosing (Wei et al. 2022). The working principle of the single screw pump is that when the motor drives the pump shaft to rotate, the liquid medicine in the sealing chamber will advance by one screw pitch every time the screw rotates. With the continuous rotation of the screw, the liquid medicine will extrude the pump body from the sealing chamber. In the sulfur lead zinc fl otation separation process of Maoping Lead Zinc Ore dressing plant, a single screw pump dosing device was used, which met the requirements of the dressing plant for precise dosing, solved the problem of easy blockage of the dosing system, improved the dosing operating environment, and reduced the oper-ating cost (Ao et al. 2018). The metering pump type dosing device is similar to the electromagnetic valve type dosing device and is composed of a storage tank, dosing device, liquid level alarm, metering pump, fl oat valve, and reagent bu ff er hopper. A metering pump is a special volumetric pump for transporting liquids (with cor-rosive liquids), and a more common type is a diaphragm metering pump. The diaphragm causes the movement of the ball valve, forming vacuum adsorption and squeezing phe-nomena, thereby achieving the purpose of transporting liquid medicine. Dongqu Coal Preparation Plant adopts a diaphragm metering pump dosing device. Compared with manually adjusting the dosage, the daily usage of kerosene was reduced by 54 kg, and the daily usage of secondary octanol was reduced by 178 kg. This saves over 840,000 yuan in reagent costs per year (Liang 2021). It should be noted that the metering pump type dosing technology was suitable for mineral processing plants with large dosage and few dosing points. However, for mineral processing plants with small dosage and many dosing points, it cannot meet production accuracy, and this type of device has the characteristics of high price and production cost.
The weighing type dosing device mainly consists of a dosing box, an upper weighing body, a lower weighing body, a weighing sensor, a dosing valve, and a reagent bu ff er bucket. Its working principle is to record the overall weight of the dosing box and the agent based on the weighing sensor. After opening the dosing valve, when the total weight reduction recorded by the weighing sensor is equal to the production design demand, the dosing valve is closed to complete the dosing. According to the characteristics of the equipment, it can be seen that the sensitivity and accuracy of the weighing sensor directly a ff ect the accuracy of dosing, and it is suitable for mineral processing plants with large dosage and few dosing points. The weighing type dosing technology was used in the iron ore fl otation process of Yunnan Gejiu Chongjing Concentrator. The results showed that the dosing control ac-curacy reached ±0.7 % and was not a ff ected by changes in reagent viscosity, pressure, or equipment wear. The operation eff ect was good and the expected goal was achieved. Production practice has shown that the fl otation auto-matic dosing device and technology have overcome the shortcomings of relying on manual dosing, reduced reagent consumption, reduced labor intensity of workers, and improved the economic bene fi ts of the enterprise. Then, it should be pointed out that there are still problems with the automatic dosing process of fl otation, such as poor dosing environment, equipment malfunction, and operational errors. In the future, the intelligent level of the dosing pro-cess should be further improved.
As an important component of fl otation control, fl otation liquid level detection has a signi fi cant impact on product quality and economic bene fi ts. The automatic liquid level detection device is a device for measuring liquid level and is also a major part of achieving liquid level control. In recent years, automatic liquid level detection methods have grad-ually undergone signi fi cant development, ranging from the simplest manual measurement methods such as glass tube method, dual color water level method, and manual gauge method to fl oat ball method, fl oat tube method, and sink barrel method developed based on the principle of buoy-ancy. With the development of modern electromagnetic technology and wave optics technology, capacitance method, resistance method, ultrasonic method, and resistance method have been successively introduced (Lin 2017). Currently, the common liquid level detection devices at home and abroad can be divided into two types: contact type and noncontact type. The noncontact detection devices are mainly ultrasonic level gauge or radar level gauge. Ultra-sonic level gauge has the advantages of high measurement accuracy, convenient installation, and basically no mainte-nance. However, the ultrasonic speed is a ff ected by the transmission medium, external temperature, pressure, density, and the characteristics of the froth layer on the pulp surface, resulting in too large ultrasonic re fl ection angle and often no measurement or measurement errors (Guo and Wang 2020; Hou 2007). The radar level gauge adopts the working mode of transmitting re fl ecting receiving, which is characterized by high measurement accuracy, safety and reliability, and long service life. However, the radar level gauge is not penetrating and has poor detection e ff ect for the level in the environment containing froth (Xie and Guo 2019). The contact type liquid level detection device is mainly composed of input type or pressure type liquid level gauges, and its detection principle is to convert pressure changes into changes in liquid level. The advantage is that it has penetrability and can detect real liquid levels. The disad-vantage is that it is easy to block and is greatly a ff ected by the concentration of the slurry. Nowadays, due to the in fl uence of slurry adhesion, corrosion, slurry mixing, and froth fl oating, the automatic liquid level detection device cannot accurately and e ff ectively detect the liquid level. A liquid level control system includes a liquid level controller, a liquid level sensor, and a variable speed drive. The main di ffi culty in controlling the fl otation liquid level lies in the highly interconnected system where each fl otation cell is coupled and a ff ects each other, requiring the use of inter-connected controllers to achieve stable control of the liquid level. According to the above research, current detection devices cannot accurately measure the liquid level value, which has always been a challenge in industrial control. Therefore, developing an intelligent detection and control system for fl otation liquid level has important practical signi fi cance.
Flotation feed concentration is also an important control parameter in fl otation operation. Reasonable feed concen-tration can make froth carry mineral particles to the maximum extent, ensure fl otation recovery, and reduce resource waste. At present, the main methods for detecting the concentration of fl otation feed are manual interval sampling and instrument testing. The manual interval sampling detection method has high sampling accuracy, simple detection process, and low cost, but it is ine ffi cient, unable to detect data in real-time, and cannot adapt to automated and intelligent production (Lv 2021). With the development of current detection technology, real-time detection technology and equipment for slurry concentra-tion have also made signi fi cant progress, such as γ radiation detector, radioactive nuclide concentration detector, fl oat slurry concentration detector, electromagnetic induction concentration detector, and ultrasonic concentration detector device. The γ-ray detector determines the concen-tration of mineral pulp by detecting the attenuation of γ radiation, and the absorption intensity of γ radiation by mineral pulp is related to the concentration of mineral pulp. The principle of a radioactive nuclide concentration detector is to capture the radiation of radioactive isotopes to a certain extent through the pulp. When the radiation generated by radioactive elements passes through the pulp, the pulp will absorb a portion of the radiation, and the absorbed pulp radiation is related to the pulp concentration. The advantage of this detection method is the noncontact detection method, which can detect data in real-time and the device and fl otation process do not interfere with each other. The disadvantage is that this method is radioactive, and isotope radiation can generate a large amount of radiation, which has a health risk impact on sta ff . The fl oat pulp concentra-tion detector is based on Archimedes ’ principle. When the fl oat fl oats in the pulp and reaches balance, the mass of the fl oat to discharge the liquid is the same as the mass of the fl oat itself. The pulp concentration is obtained according to the two transformations of the inclination angle-current or voltage signal-pulp concentration changed by the fl oat bal-ance. The detection device has a simple structure, conve-nient operation, and can provide real-time feedback on slurry concentration. The disadvantage is that the slurry has a corrosive e ff ect on the direct contact device. In addition, the viscosity of the slurry can also have an adverse impact on the accuracy of the device, requiring manual periodic cleaning of the device. The principle of an electromagnetic induction concentration detector is to convert magnetic signals into electrical signals based on the principle of electromagnetic induction and calculate the size of the slurry concentration through the integrated linkage of slurry concentration, magnetic mineral concentration, and induced current. The advantages of this detection method are simple structure and convenient detection. The disad-vantage is that the detection accuracy needs to be improved. The detection principle of an ultrasonic concentration detector is based on the attenuation of sound waves, atten-uation of sound velocity, and changes in sound impedance when ultrasonic waves pass through the slurry. Under the excitation of continuous sine wave pulses, the ultrasonic emission transducer emits ultrasonic plane longitudinal waves into the slurry in a thickness vibration manner. When the ultrasonic interacts with the slurry, it attenuates and reaches the ultrasonic receiving transducer, causing energy and amplitude to decay. This degree of attenuation is closely related to the slurry concentration, and thus the slurry concentration can be calculated (Han 2014). This method is also noncontact detection for radioac-tivity, which can provide real-time feedback on changes in slurry concentration. It has high detection accuracy, low maintenance costs, and does not generate ionizing radiation. However, the disadvantage is that when the slurry concen-tration is high, the attenuation of ultrasound is not linearly related to the concentration. In addition, factors such as the absorption of froth to ultrasonic waves and the in fl uence of slurry temperature change on ultrasonic attenuation will also interfere with the test results. Overall, when measuring the concentration of mineral slurry, it is a three-phase mixture of solid, liquid, and gas, which is a strong acid or alkali environment and also has high viscosity. These factors will have an impact on the corrosion resistance, wear resistance, and measurement accuracy of the testing device, and the above methods cannot detect concen-tration data in real-time for a long time and with high accu-racy. Therefore, developing noncontact high-precision, safe and e ffi cient detection devices will be a key development di-rection in the future. The classi fi cation of fl otation automation equipment and corresponding advantages and disadvantages in the application process is shown in Table 1.
It is necessary to accurately and truly measure the fl ow value in a bene fi ciation plant such as tailings leakage line detection, heavy medium cyclone and hydrocyclone detec-tion, and fl otation process detection. Traditional methods such as electromagnetic fl ow meters, ultrasonic fl ow meters, vortex fl ow meters, wedge fl ow meters, etc. pose a huge challenge in accurately measuring fl ow due to scaling on pipe walls, changes in material properties, calibration deviations, and entrainment of bubbles. Currently, the widely used technology internationally is a new method for fl ow testing based on array sensors. The measurement principle of this method is based on an array algorithm of passive sonar sensors to detect, track, and measure any disturbance velocity of pipeline axial movement. These disturbances can be divided into fl uid transmission, acoustic transmission, and vibration transmission on the inner wall of the pipeline. Each type of disturbance has a di ff erent speed, and the type of disturbance and measurement of disturbance rate can be accurately distinguished through the di ff erence in disturbance speed. The measurement principle based on arrays has proven to have signi fi cant advantages in volume fl ow measurement and gas porosity measurement for various mineral processing applications, especially in situations such as bubble entrainment, pipeline fouling, high wear and corrosion of pipelines, and the presence of magnetic minerals. Currently, this technology is being applied in monitoring over 700 mineral processing processes in 22 countries (VO ’Keefe et al. 2010).
The intelligent technology of variable detection is based on the comprehensive implementation of soft measurement technology and computer hardware technology. The prin-ciple of soft measurement technology is based on intelligent algorithms to construct the correlation between auxiliary variables (input variables) and dominant variables (output variables), thereby achieving real-time estimation of measured variables. Intelligent algorithms mainly include arti fi cial neural network (ANN), support vector machine (SVM), genetic algorithm (GA), particle swarm optimization (PSO), decision tree, and its integrated model. Through these intelligent algorithms, the feature extraction of fl otation froth image, the prediction of concentrate and tailings, and the prediction of ash content can be realized. Therefore, this work focuses on the extraction methods and model training algorithms of fl otation froth characteristics (color, texture, size, dynamic characteristics, etc.), while the prediction methods of fl otation indicators (the grade, recovery rate, ash content in the concentrate and tailings) are also fully introduced.
In the process of mineral fl otation, the physical and dynamic characteristics (color, texture, size, moving speed) of froth are closely related to the fl otation production indicators, working conditions, and operating variables, which can be used as an important basis for judging the e ff ect of mineral separation operations. Because froth images are formed by the accumulation of mineralized bubbles with di ff erent numbers, sizes, shapes, and colors, the boundaries between bubbles are not clear, and bubbles are piled and squeezed with each other. Bubbles are broken and merged seriously, so it is di ffi cult to obtain e ff ective results with conventional froth image processing methods. The physical and dynamic characteristics of froth can be accurately and quickly extracted through intelligent algorithms, machine vision technology, and image processing technology. For instance, some scholars and teams have harnessed a machine vision system to cull froth ’s visual characteristics, encompassing attributes such as bubble dimensions, uniformity, hue, and texture. These attributes are subsequently subject to scru-tiny via assorted intelligent algorithms, thereby enabling the modeling of the nuanced associations between froth char-acteristics and metallurgical parameters. The eventual consequence is the formulation of predictive models (Jahedsaravani et al. 2014; Massinaei et al. 2020; Mehrabi et al. 2014; Tan et al. 2016).
Due to di ff erent mineralized particles in mineralized froth, froth presents di ff erent colors under di ff erent fl otation con-ditions. The color of froth can re fl ect the type and concen-tration of minerals contained, which is closely related to the fl otation production conditions. Therefore, fl otation produc-tion can be guided by the color characteristics of fl otation froth. In order to accurately extract the color features of fl otation froth, many scholars have carried out plentiful research work on the color feature extraction of froth. Massinaei et al. (2019) used a machine vision system to record the characteristics of froth vision (froth color, bubble size, froth speed), texture (energy, entropy, contrast, uni-formity, and correlation), and metallurgical performance (combustible recovery, re fi ned ash content). The relation-ship between froth characteristics, process, and perfor-mance parameters was analyzed. The results indicate that the developed system can be successfully used to diagnose process conditions and predict process performance under di ff erent operating conditions. Mehrabi et al. (2014) used machine vision technology to monitor froth in iron ore fl otation process, which can successfully extract physical and dynamic characteristic parameters such as froth size, quantity, and froth fl oating speed, and the system has good stability. Kaartinen et al. (2002) obtained the color, bearing rate, speed, stability, and size characteristics of froth in the fl otation process by developing a zinc fl otation control system with multiple cameras and put forward guidance for real-time fl otation control. In addition, froth color extraction methods include color histogram technology (Chuk et al. 2005), comprehensive co-occurrence matrix (Palm 2004), color reference system (Bonifazi et al. 2001), multi-image analysis technology (Bartolacci et al. 2006), and multicolor spatial information fusion technology (Yang et al. 2009a,b). These methods can e ff ectively extract froth color features. Although the current fl otation froth color feature extraction technology has made great progress, industrial production is vulnerable to the in fl uence of on-site envi-ronment, natural lighting, light source attenuation, and other factors, and froth images have high spots, color devi-ation, and other factors that interfere with the accurate extraction of froth features. These aspects still need further optimization research.
Froth texture is the comprehensive performance of froth roughness, contrast, and viscosity. As another key feature characterizing the statistical distribution of froth images, it can be used to describe the change of froth state caused by the change of operating conditions and mineral properties in the fl otation process (Gui et al. 2013). Abundant research work has been done on texture feature extraction of froth images at home and abroad, and the feature extraction methods in the literature of running images are divided into four categories, namely, texture feature extraction methods based on statistical families, texture feature extraction methods based on model families, texture feature extraction methods based on structure families, and texture feature extraction methods based on signal processing families. Among them, texture feature extraction methods based on statistical families mainly focus on statistical data analysis of texture image features, including grayscale co-occurrence matrix method (GLCM), grayscale shape statistics method, grayscale di ff erence sta-tistics method, cross diagonal matrix method, grayscale gradient statistics method, local grayscale statistics method, autocorrelation function method, semi variogram method, and texture spectrum statistics method. Most statistical based methods were discovered by Julez (Julesz 1975; Julesz and Caelli 1979). The texture feature extraction method based on model families assumes that the texture is distributed according to a certain model and then analyzes this model. Common model methods include Markov random fi eld (MRF) model method (Woods 1972), Gibbs random fi eld model method (Sivakumar and Goutsias 1999), Wold model (Liu and Picard 1994), Fractai model (Tang et al. 2002), complex network model, mosaic model, etc. The key to feature extraction by model method is to select appropriate models and parameter values. However, because of the complex texture of froth, it is di ffi cult to accurately describe it through a single model, and the calculation is heavy. The texture feature extraction method based on structural family is a texture feature extraction method based on texture primitive theory, which emphasizes the regularity of textures. However, due to the lack of regularity in most textures in nature, this method limits its research depth in texture feature extraction. The texture feature extraction method based on signal processing family is to convert texture images from spatial domain to other transform domains through certain methods, which can also be called fi ltering method. Including Radon transform, ring and wedge fi ltering, discrete cosine transform, local Fourier transform, local Walsh transform and Hadamard transform, Gabor wavelet transform, binary wavelet, multi band wavelet, pyramid wavelet decomposition, ridge wavelet decomposition, curved wave decomposition, Laws texture measurement, feature fi lter, orthogonal mirror fi ltering, optimized FIR fi lter, etc. The Fourier transform method (Stromberg and Farr 1986), Gabor transform method (Bovik et al. 1990), and wavelet transform method (Pun 2003) are commonly used in the above methods. Nowadays, the research on texture feature extraction of fl otation froth has made some progress, and the commonly used method is to characterize the texture feature of froth surface through the energy, entropy, and moment of inertia of the fi eld gray correlation matrix, spatial gray correlation matrix, or gray di ff erence matrix. Gui et al. (2012, 2013) used grayscale co-occurrence matrix (GLCM) to extract full texture features of images. Based on the second order combined conditional probability density function of the estimated image, the gray correlation between the pixels in the image that are related in distance and direction is calculated and counted. This correlation is mainly re fl ected in the comprehensive information of the image in direction, adjacent interval, change amplitude, speed, and so on. The physical meaning of froth ’s visual texture feature parame-ters (energy [thickness], moment of inertia, parameter entropy) is analyzed, and the correlation between froth feature parameters and froth texture is pointed out. Bartolacci et al. summarized the commonly used froth texture feature extraction methods, studied three commonly used froth texture analysis methods through Multivariate Image Analysis (Liu et al. 1990), GLCM (Hu et al. 2006), and wavelet texture analysis (Tang et al. 2011), extracted froth texture features respectively, discussed the classi fi cation of fl otation state based on froth texture fea-tures, and established a concentrate grade prediction model with PLS (Partial Least Squares) (Wold et al. 2001) for fl ota-tion control, but no good conclusive results were obtained (Bartolacci et al. 2006). To sum up, although froth texture feature extraction technology has made great progress, due to the complex working conditions of the fl otation process, the micro heterogeneity and complexity of froth texture, as well as the robustness of conceptual uncertainty and light changes, the precise extraction of froth texture features still need further exploration.
The size of fl otation froth is closely related to pulp pH, pulp concentration, and material fi neness and can re fl ect the production indicators such as fl otation product grade and recovery rate. The size feature extraction technology of froth images often depends on the image segmentation e ff ect. However, due to the mutual extrusion, di ff erent shapes, and uneven distribution of froth, it is di ffi cult for conventional segmentation technology to achieve accurate segmentation eff ect. Scholars have carried out massive research on the size characteristics of fl otation froth, mainly focusing on watershed segmentation method (Bonifazi et al. 2001), reconstruction morphology method (Sadrkazemi and Cil-liers 1997), homogeneous gradient method (Botha et al. 1999), valley bottom edge method (Wang et al. 2003), improved reconstruction transformation and watershed method (Yang et al. 2009a,b; Zhou et al. 2009), morphology technology (processing steps are shown in Figure 8) (Wang and Tang 2002; Zhou et al. 2010), clustering preclassi fi cation and dis-tance high-low precision reconstruction method (Yang et al. 2008), the image segmentation technology based on the seed region boundary growth method (Mou and Zhang 2009) solves the di ffi cult segmentation problems of froth image, such as bubble adhesion, and boundary blur. Among the above methods, the valley edge method, watershed seg-mentation method, and morphological technology are the three commonly used segmentation methods at home and abroad. However, each of these three methods has its own limitations. Although the edge detection algorithm has a relatively simple template and is easy to operate, it also has problems such as inaccurate edge positioning, nonsingle pixel wide edges, and poor operator noise resistance (Ren 2008; Wang 2003). The commonly used methods for water-shed segmentation are distance-based, gradient-based, and marker-based watershed image segmentation methods. But the most common problem is severe undersegmentation and oversegmentation (Li 2011; Zhang 2013). In addition, the morphological parameters in morphological algorithms should be selected based on the characteristics of the object, and in practical industrial applications, the fl otation condi-tions are unpredictable, which can easily lead to the loss of robustness of fi xed morphological parameter patterns and cause analysis errors. In recent years, deep learning has become more and more well-known. Horn et al. (2017) found that convolutional neural networks (CNNs) performed better than traditional methods in feature extraction. Montes-Atenas et al. (2016) employed deep neural networks (DNNs) to predict bubble size and bubble rate using data obtained from validated computational fl uid dynamics (CFD) computations. Pure water and slurry (in conditions similar to those employed in mineral froth fl otation) case studies are evaluated. It is found that the DNN can predict the CFD results accurately when using four hidden layers, describing discontinuities in the bubbly fl ow regime. The relative errors computed between the CFD data and the prediction obtained by the DNN is as low as 8.8 % and 1.8 % for bubble size and bubble rate, respectively. Furthermore, Fu and Aldrich (2018) found that better results can be obtained with deeper CNN. Zhang et al. (2022) developed a generative adversarial network (GAN)-based setpoint computation model, and its e ff ective-ness is veri fi ed in experiments on zinc fl otation data. It should be noted that these methods require a large amount of o ffl ine data to train a reliable model, which includes various sources of data such as fl otation froth images, pro-cess variables, and manipulated variables. The image acquisition equipment in the industrial fl otation process is usually composed of hardware equipment and software systems for image acquisition, display, storage, online analysis, etc.
The change of air recovery rate and froth size determined by the dynamic characteristics (moving speed) of fl otation froth directly a ff ects the fl otation production e ffi ciency and fl otation separation indicators. In view of the fl ow distortion such as froth collapse, fragmentation, overlap, and merger in the fl otation process, the traditional image processing algorithm is di ffi cult to identify and track froth and cannot achieve accurate extraction of froth dynamic characteristics. Focusing on the accurate extraction of froth dynamic char-acteristics, many scholars have carried out more research on froth dynamic characteristics. Mou (2012) compared and analyzed the matching e ff ect of froth images based on optical fl ow, macroblock tracking method, and phase correlation method. The results showed that macroblock tracking technology combining phase cor-relation and grayscale template matching was more suitable for estimating the motion characteristics of froth images. Ventura-A-Iedina and Cilliers (2000) found that under certain conditions, froth fl ow speed can be used as a quan-titative feature of fl otation performance. Neethling (2008) discussed the in fl uence of froth movement speed charac-teristics on recovery. By analyzing froth structure parame-ters and froth surface area fl ow rate, froth texture spectrum characteristics are used to estimate froth size changes. Zhang and Liu et al. (2016) proposed to improve the matching conditions of SIFT algorithm according to the size and direction of froth speed, eliminate mismatches through random sampling consistent algorithm, and extract dynamic features according to the matching results. Fu and Aldrich (2018) used convolutional neural network (CNN) to develop fl otation froth image sensor and identi fi ed froth status through a large number of image sample training. Feng (2017) achieved accurate mapping of the velocity fi eld through block matching, pairwise matching, PROSAC screening matching, and other strategies for adjacent frame froth images and also achieved quantitative description of froth stability. Holtham and Nguyen 2002 proposed pixel tracking technology to obtain froth velocity characteristics. Kaartinen et al. (2002) used froth cross-correlation peak value to describe froth speed and applied it to zinc fl otation process. Brown et al. (2001) used the fl ow rate of froth to evaluate the
Bookmark
Daniel Féau processes personal data in order to optimise communication with our sales leads, our future clients and our established clients.
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.