Simulation optimization is a scientific tool that is widely used for design and optimization of c... more Simulation optimization is a scientific tool that is widely used for design and optimization of comminution circuits in mineral processing plants. In this research, first of all, in order to determine the suitable d80 for cicuit hydrocyclone underflow, the parameters required for simulator (residence time distribution, breakage function, selection function and Plitt's model calibration) were determined. Then, simulation of comminution circuit by presentation of BMCS ® software under MATLAB environment was performed. Finally, the GA toolbox of MATLAB software was used for optimization of the comminution circuit by determining the simulator as fitness function. The results showed that by determining the optimum value of input parameters (solid content, fresh feed and water rate to the input streams), the genetic algorithm is able to achieve a suitable d80 for circuit cuf (cyclon underflow).
Abstract In the previous study, a mathematical model was proposed to predict retained mass inside... more Abstract In the previous study, a mathematical model was proposed to predict retained mass inside a laboratory Knelson Concentrator bowl. In this part, a model is proposed to describe the particle separation performance of the laboratory Knelson Concentrator. The separation model considers the main forces (fluid drag force, F d , centrifugal force, F c , and buoyancy force, F b ) acting on particles inside the concentrating bowl, which are functions of material properties as well as the operating condition parameters. Several materials with different densities including magnetite, zinc, ferromolybdenum, lead, and tungsten were used as valuable components to simulate minerals with the same densities in two-component synthetic feeds with various size fractions. The modelling of component separation in the Knelson Concentrator was performed based on the ratio of F d / ( F c - F b ) . Based on an extensive experimental database, practical and simple models were proposed as predictors of recovered mass of valuable components from tailings and the mass of quartz (as gangue) in the Knelson Concentrator versus the ratio of F d / ( F c - F b ) for quartz.
The run of mine ore from Aghdarreh gold mine must be comminuted to achieve the desired degree of ... more The run of mine ore from Aghdarreh gold mine must be comminuted to achieve the desired degree of liberation of gold particles. Currently, comminution circuits include a single-stage crushing using a jaw crusher and a single-stage grinding using a Semi-Autogenous Grinding (SAG) mill in closed circuit with a hydrocyclone package. The gold extraction is done by leaching process using cyanidation method through a series of stirred tanks. In this research, an optimization study of Aghdarreh plant grinding circuit performance was done to lower the product particle size (PComputer simulations were performed to assess performance of two proposed new circuit flowsheets. The first proposed flowsheet consists of existing SAG mill circuit and a new proposed ball mill in closed circuit with a new second hydrocyclone package. The second proposed flowsheet consists of existing SAG mill circuit followed by a new proposed ball mill in closed circuit with the existing hydrocyclone package. In all sim...
ABSTRACT Deeper understanding of screening performance of banana screens helps mineral processing... more ABSTRACT Deeper understanding of screening performance of banana screens helps mineral processing engineers to control and optimize them. In this article, screening performance of banana screens using a DEM (Discrete Element Method) solver ‘LIGGGHTS’ is investigated. An industrial double-deck banana screen with five panels and two laboratory single-deck banana screens with three and five panels are simulated. Effects of design parameters including incline angle of each panel, vibration amplitude and vibration frequency on screening performance of both the industrial screen and laboratory screens are studied. Also, the effects of operational parameters including simulation time and feed particle size distribution on screening performance of the industrial screen are examined. Simulation results are expressed in terms of recognized and comprehensible quantities for engineers, i.e., screen efficiency and screening recovery. To validate simulation results, a parameter named partition number is defined. Comparison of the partition numbers of simulations related to the laboratory screens with those available in the literature demonstrates a good agreement which validates the DEM simulations and the used software. Also, the results indicate that the industrial and laboratory screens show a totally different behavior with respect to changes in design parameters.
ABSTRACT Particle size distributions of obtained samples from several sampling campaigns were det... more ABSTRACT Particle size distributions of obtained samples from several sampling campaigns were determined and raw data were mass balanced before being used in simulation studies. After determination of breakage function, selection function, Bond work index, residence time distribution parameters, and Whiten’s model parameters for air separators and diaphragms between the two compartments of tube ball mills, performance of the circuits was simulated for given throughputs and feed particle size distribution. Whiten’s model parameters were determined by GA (genetic algorithm) toolbox of MATLAB software. Based on implemented models for modeling and simulation, optimization of circuits was carried out. It increased nearly 10.5% and 15.8% in fresh feed capacity input to each tube ball mill. In addition, circulating load ratios of circuits are modified to 118% and 127% from low level of 57% and 22%, respectively, and also cut points of air separators are adjusted at 30 and 40 μm from high range of 53 and 97 μm, respectively. All applications helped in well-operation and energy consumption reduction of equipments.
Grinding, particularly at finer sizes to achieve the required degree of liberation or specific su... more Grinding, particularly at finer sizes to achieve the required degree of liberation or specific surface, is a critical unit operation in terms of energy consumption and process optimization. Nowadays, considering the significant developments made in computer hardware and software, using simulation programs to optimize design and operation of various ore treatment plants could be very beneficial to the mineral process engineers. Currently, there are a number of steady-state simulators that can be run under DOS TM or WINDOWS TM environments such as BMCS, MODSIM TM and JKSimMet TM . In this paper, COMSIM, a new simulator which runs under Excel spreadsheet will be introduced which uses a Population Balance Model (PBM) to simulate ball mills. Plitt and Nageswararao models have been used to describe the performance of hydrocyclones, which by linking to the ball mill model allow for closed-circuit simulations. To describe material flow through the ball mills, two tanks-in-series models called Weller's model and one plug flow unit plus n perfect mixers units (1PF + nPM) model have been used. Utilizing Microsoft TM Excel TM capabilities such as graphics and VBA programming to implement new macro-based user functions, the authors developed an easy-to-use comminution simulation environment. By comparing the outputs of COMSIM with measured grinding data and previously existing simulators, particularly BMCS, its performance was firmly tested both in terms of accuracy and precision of obtained results.
Mathematical modeling and simulation techniques are widely used to design and optimize comminutio... more Mathematical modeling and simulation techniques are widely used to design and optimize comminution circuits in mineral processing plants. However, circuit performance predictions are prone to errors due to inaccurate calibration of models used in simulations. To address this problem, the authors applied a method based on genetic algorithms (GA) for estimation of HPGR (high pressure grinding rolls) model parameters. In this research, a simulation algorithm was developed and implemented in MATLAB™ based on published HPGR models to test and demonstrate GA application for model calibration. The GA toolbox of MATLAB was used to obtain the optimal values of HPGR model parameters. The authors successfully validated simulator predictions against HPGR data sets at laboratory and industrial scales. The results indicate that GA is a robust and powerful search method to find the best values of HPGR model parameters that lead to more reliable simulation predictions.
Comminution simulators are extensively used by mineral processing engineers for plant design and ... more Comminution simulators are extensively used by mineral processing engineers for plant design and optimization purposes. Recently, there had been a great progress in developing new and more powerful optimization methods such as Genetic Algorithms (GA), Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO) and Tabu Search Algorithm (TSA). Simulation optimization is required if one wants to find the best steady-state values of important process variables. In this paper, the authors investigated the integration of GA optimization algorithm with a pre-existing grinding circuit simulator called Ball Milling Circuits Simulator (BMCS) in MATLAB TM environment. The BMCS code has been written in ANSI C language and has been validated against real industrial grinding circuit data sets. Various C modules of the BMCS grinding software were restructured under a new single source code file so that it can be imported into MATLAB. Then, a number of input simulation data were identified and selected as possible process variables (e.g., solids flow rate, water addition rate, and number of operating cyclones) which must be optimized in order to achieve a pre-defined process objective (e.g., a specific d 80 of circuit output). The obtained results show that BMCS simulation trials can be successfully optimized by applying evolutionary algorithms via MATLAB toolboxes. This allows the mineral processor to perform automatic repetitive simulations to find the possible solutions of the problem at hand quickly.
Simulation optimization is a scientific tool that is widely used for design and optimization of c... more Simulation optimization is a scientific tool that is widely used for design and optimization of comminution circuits in mineral processing plants. In this research, first of all, in order to determine the suitable d80 for cicuit hydrocyclone underflow, the parameters required for simulator (residence time distribution, breakage function, selection function and Plitt's model calibration) were determined. Then, simulation of comminution circuit by presentation of BMCS ® software under MATLAB environment was performed. Finally, the GA toolbox of MATLAB software was used for optimization of the comminution circuit by determining the simulator as fitness function. The results showed that by determining the optimum value of input parameters (solid content, fresh feed and water rate to the input streams), the genetic algorithm is able to achieve a suitable d80 for circuit cuf (cyclon underflow).
Abstract In the previous study, a mathematical model was proposed to predict retained mass inside... more Abstract In the previous study, a mathematical model was proposed to predict retained mass inside a laboratory Knelson Concentrator bowl. In this part, a model is proposed to describe the particle separation performance of the laboratory Knelson Concentrator. The separation model considers the main forces (fluid drag force, F d , centrifugal force, F c , and buoyancy force, F b ) acting on particles inside the concentrating bowl, which are functions of material properties as well as the operating condition parameters. Several materials with different densities including magnetite, zinc, ferromolybdenum, lead, and tungsten were used as valuable components to simulate minerals with the same densities in two-component synthetic feeds with various size fractions. The modelling of component separation in the Knelson Concentrator was performed based on the ratio of F d / ( F c - F b ) . Based on an extensive experimental database, practical and simple models were proposed as predictors of recovered mass of valuable components from tailings and the mass of quartz (as gangue) in the Knelson Concentrator versus the ratio of F d / ( F c - F b ) for quartz.
The run of mine ore from Aghdarreh gold mine must be comminuted to achieve the desired degree of ... more The run of mine ore from Aghdarreh gold mine must be comminuted to achieve the desired degree of liberation of gold particles. Currently, comminution circuits include a single-stage crushing using a jaw crusher and a single-stage grinding using a Semi-Autogenous Grinding (SAG) mill in closed circuit with a hydrocyclone package. The gold extraction is done by leaching process using cyanidation method through a series of stirred tanks. In this research, an optimization study of Aghdarreh plant grinding circuit performance was done to lower the product particle size (PComputer simulations were performed to assess performance of two proposed new circuit flowsheets. The first proposed flowsheet consists of existing SAG mill circuit and a new proposed ball mill in closed circuit with a new second hydrocyclone package. The second proposed flowsheet consists of existing SAG mill circuit followed by a new proposed ball mill in closed circuit with the existing hydrocyclone package. In all sim...
ABSTRACT Deeper understanding of screening performance of banana screens helps mineral processing... more ABSTRACT Deeper understanding of screening performance of banana screens helps mineral processing engineers to control and optimize them. In this article, screening performance of banana screens using a DEM (Discrete Element Method) solver ‘LIGGGHTS’ is investigated. An industrial double-deck banana screen with five panels and two laboratory single-deck banana screens with three and five panels are simulated. Effects of design parameters including incline angle of each panel, vibration amplitude and vibration frequency on screening performance of both the industrial screen and laboratory screens are studied. Also, the effects of operational parameters including simulation time and feed particle size distribution on screening performance of the industrial screen are examined. Simulation results are expressed in terms of recognized and comprehensible quantities for engineers, i.e., screen efficiency and screening recovery. To validate simulation results, a parameter named partition number is defined. Comparison of the partition numbers of simulations related to the laboratory screens with those available in the literature demonstrates a good agreement which validates the DEM simulations and the used software. Also, the results indicate that the industrial and laboratory screens show a totally different behavior with respect to changes in design parameters.
ABSTRACT Particle size distributions of obtained samples from several sampling campaigns were det... more ABSTRACT Particle size distributions of obtained samples from several sampling campaigns were determined and raw data were mass balanced before being used in simulation studies. After determination of breakage function, selection function, Bond work index, residence time distribution parameters, and Whiten’s model parameters for air separators and diaphragms between the two compartments of tube ball mills, performance of the circuits was simulated for given throughputs and feed particle size distribution. Whiten’s model parameters were determined by GA (genetic algorithm) toolbox of MATLAB software. Based on implemented models for modeling and simulation, optimization of circuits was carried out. It increased nearly 10.5% and 15.8% in fresh feed capacity input to each tube ball mill. In addition, circulating load ratios of circuits are modified to 118% and 127% from low level of 57% and 22%, respectively, and also cut points of air separators are adjusted at 30 and 40 μm from high range of 53 and 97 μm, respectively. All applications helped in well-operation and energy consumption reduction of equipments.
Grinding, particularly at finer sizes to achieve the required degree of liberation or specific su... more Grinding, particularly at finer sizes to achieve the required degree of liberation or specific surface, is a critical unit operation in terms of energy consumption and process optimization. Nowadays, considering the significant developments made in computer hardware and software, using simulation programs to optimize design and operation of various ore treatment plants could be very beneficial to the mineral process engineers. Currently, there are a number of steady-state simulators that can be run under DOS TM or WINDOWS TM environments such as BMCS, MODSIM TM and JKSimMet TM . In this paper, COMSIM, a new simulator which runs under Excel spreadsheet will be introduced which uses a Population Balance Model (PBM) to simulate ball mills. Plitt and Nageswararao models have been used to describe the performance of hydrocyclones, which by linking to the ball mill model allow for closed-circuit simulations. To describe material flow through the ball mills, two tanks-in-series models called Weller's model and one plug flow unit plus n perfect mixers units (1PF + nPM) model have been used. Utilizing Microsoft TM Excel TM capabilities such as graphics and VBA programming to implement new macro-based user functions, the authors developed an easy-to-use comminution simulation environment. By comparing the outputs of COMSIM with measured grinding data and previously existing simulators, particularly BMCS, its performance was firmly tested both in terms of accuracy and precision of obtained results.
Mathematical modeling and simulation techniques are widely used to design and optimize comminutio... more Mathematical modeling and simulation techniques are widely used to design and optimize comminution circuits in mineral processing plants. However, circuit performance predictions are prone to errors due to inaccurate calibration of models used in simulations. To address this problem, the authors applied a method based on genetic algorithms (GA) for estimation of HPGR (high pressure grinding rolls) model parameters. In this research, a simulation algorithm was developed and implemented in MATLAB™ based on published HPGR models to test and demonstrate GA application for model calibration. The GA toolbox of MATLAB was used to obtain the optimal values of HPGR model parameters. The authors successfully validated simulator predictions against HPGR data sets at laboratory and industrial scales. The results indicate that GA is a robust and powerful search method to find the best values of HPGR model parameters that lead to more reliable simulation predictions.
Comminution simulators are extensively used by mineral processing engineers for plant design and ... more Comminution simulators are extensively used by mineral processing engineers for plant design and optimization purposes. Recently, there had been a great progress in developing new and more powerful optimization methods such as Genetic Algorithms (GA), Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO) and Tabu Search Algorithm (TSA). Simulation optimization is required if one wants to find the best steady-state values of important process variables. In this paper, the authors investigated the integration of GA optimization algorithm with a pre-existing grinding circuit simulator called Ball Milling Circuits Simulator (BMCS) in MATLAB TM environment. The BMCS code has been written in ANSI C language and has been validated against real industrial grinding circuit data sets. Various C modules of the BMCS grinding software were restructured under a new single source code file so that it can be imported into MATLAB. Then, a number of input simulation data were identified and selected as possible process variables (e.g., solids flow rate, water addition rate, and number of operating cyclones) which must be optimized in order to achieve a pre-defined process objective (e.g., a specific d 80 of circuit output). The obtained results show that BMCS simulation trials can be successfully optimized by applying evolutionary algorithms via MATLAB toolboxes. This allows the mineral processor to perform automatic repetitive simulations to find the possible solutions of the problem at hand quickly.
Uploads
Papers by A. Farzanegan