应用JKSimFloat模拟浮选流程验证超大型浮选机选型

在实验室分批浮选试验的基础上,采用JKSimFloat分批浮选回收率模型,分别计算了某铜矿石中不同矿物的浮选动力学参数。参照现场浮选流程,应用JKSimFloat软件进行了不同浮选机选型方案的浮选流程模拟。对比模拟结果和实际生产指标,表明JKSimFloat模拟结果的精度较好。因此,JKSimFloat软件可以作为超大型浮选机选型结果的有效验证工具。     

Laboratory flotation separation of INCO bulk matte

Laboratory batch flotation tests were performed on a sample of bulk slow cooled converter matte (Cu:Ni = 1) to optimize the time in each of the separation stages. Numerical simulations were carried out to estimate a continuous circuit material balance for the conditions selected. An experimental simulation of a continuous circuit (locked cycle test) readily came to steady state and there was reasonable agreement between the numerical and experimental results. The use of nitrogen as the flotation gas and the addition of collector to the grinding mill minimized the oxidation of chalcocite, which has previously resulted in increasing circulating loads, and led to a stable circuit with very low weights in the recycle streams.     

Effects of hydrodynamic parameters on entrainment and flotation performance

The separation selectivity and efficiency of the flotation process depends not only on differences in the physico-chemical surface properties of various minerals within an ore, but also on the hydrodynamics of flotation. The hydrodynamics relate directly to many sub-processes of the flotation, such as suspension of particles in the pulp, dispersion of the air into bubbles and the probability of particle–bubble collision. Mechanical entrainment plays a major role in the quality of the flotation concentrate and the separation selectivity. Therefore, the effects of some important hydrodynamic parameters on both the entrainment and the flotation performance were investigated by means of a fractional factorial experimental design. Furthermore, many flotation tests were performed in order to generate sufficient data for use in empirical modelling and neural network based modelling, in light of the results of the experimental design. The adaptability and reliability of the models developed, depending on the treatment of available experimental data, have been studied comprehensively. Evaluation of the observed and predicted results demonstrated that the effect of some chemical and hydrodynamic parameters of the flotation process on both the metallurgical performance and entrainment in the training region, can successfully be predicted, with an error of less than 6%, by using the developed neural networks models without particular assumptions and additional experiments. In order to prove the validation of the empirical and neural networks models developed and to compare their performances, a few additional tests were conducted under predetermined flotation conditions. The results of these additional flotation tests indicated that the neural network models were consistently more accurate than the empirical models with a negligible error.     

Calculation of locked cycle flotation test results

Locked cycle test are frequently used in mineral processing laboratories during the flowsheet development exercise in order to confirm batch results and to get an approximation of a continuous circuit material balance. A review of the literature indicates that there are at least four different methods for calculating the results of a locked cycle test. Only two of these methods offer the opportunity to examine whether or not the test reached steady state and only the cycle-by-cycle method intrinsically calculates the circulating streams. Because of the difficulty and expense of executing locked cycle tests and the frequent failure of the tests to reach steady state it should be of paramount importance to determine whether or not the test reached steady state; thus, the preferred method of calculation is the cycle-by-cycle technique.     

A strategy for the identification of optimal flotation circuits

This article presents a strategy for the identification of optimal flotation circuits based on experimental data and the assumption that the flotation circuit structure are not very sensitive to stage recovery. The main objective is to find a set of optimal flotation circuit configurations, including the best metallurgical conditions for the process and cell design. The optimization process maximizes the Net Present Value (NPV). This methodology consists of two steps. The first step of the methodology entails obtaining data from the laboratory. Different metallurgical conditions are tested for different flotation stages. Each metallurgical condition is evaluated to determine its kinetics. The second step of the methodology is the optimization process. The optimization process achieves the best solution by optimizing cell design, flotation circuit structure and metallurgical conditions. The optimization process has three phases. The first phase assumes that every flotation stage has the same residence time and calculates feasible flotation circuits, metallurgical conditions and cell volumes. This process is performed for several residence times and, therefore, a set of feasible solutions is generated. The second phase takes the previous set of feasible solutions and calculates the residence time at each flotation stage. Then, for each phase two solution, the optimal flotation circuit, metallurgical conditions and cell design are calculated. The final product is a set of optimal solutions than can be considered for further study. The procedure is illustrated by the design of a zinc flotation plant, considering seven flotation stages and five species.     

The interaction of flash flotation with closed circuit grinding

Flash flotation is an important unit operation within many grinding circuits. It provides an opportunity for the valuable mineral to be recovered as early as possible in the processing plant. This prevents liberated valuable mineral from building up in the recirculating load of the mill, and reduces the occurrence of overgrinding. Overgrinding can place a significant limitation on overall recovery, because it causes the production of valuable fines that are difficult to recover by flotation.The flash flotation cell is fed by the cyclone underflow in a closed grinding circuit. This feed stream represents the optimal particle size distributions of valuable mineral and gangue for flotation kinetics, selectivity and grade. The flash cell treats the recirculating load of the ball mill, and therefore influences the performance of both the mill and the hydrocyclone classifier. The effect of a flash cell in the grinding circuit is difficult to determine since flash cells in Australia are introduced at the design stage and so no plant data is available before and after the introduction of the flash flotation unit. By establishing a computer simulation of the closed-circuit grinding with flash flotation, the interaction of flash flotation with grinding can be estimated. The models can be used to predict the effect of various changes to the operating conditions on circuit performance as well as the expected grinding performance in the absence of flash flotation. This enables the effect of flash flotation and any possible benefits to be evaluated.The model was developed from unit models of the ball milling, hydrocyclone classification and flash flotation processes. An empirical model was used for the flash cell, and generic models were fitted to the ball mill and hydrocyclone based on the matrix model and the Plitt model respectively. The data required for the development of the models was obtained from plant surveys of the Kanowna Belle gold mine and laboratory batch grinding and flotation tests.The model accurately represents the plant grinding and flash flotation circuit while operating under normal conditions. Simulation of the circuit using the model enabled the effect of variations to flash cell operating conditions on the flash concentrate, recirculating load and cyclone overflow to be determined.     

JKSimFloat模拟闭路循环浮选回路

在实验室分批浮选试验的基础上建立了浮选回收率模型,通过回归分析获得了某磷铁矿石浮选动力学参数。应用JKSimFloat软件建立了一个闭路循环浮选回路,将获得的参数输入到软件中,并对其它设备参数进行了合理的优化设置后,模拟计算了该浮选回路的选别效果。与实验室闭路浮选试验结果相比,模拟的精矿品位基本相当,而模拟的精矿回收率偏低。     

工业浮选系统的动态仿真研究

通过大量的浮选试验，考察了主要变量和连续浮选特征参数之间的关系。采用流程拓朴矩阵协调各单元子模型之间的关系，以浮选回路系统模型编制成一个通用的仿真器及相应的数据处理软件包。利用分批浮选试验得的参数及现场数据进行了计算机动态仿真研究，提出了全浮回路的主要动态特性；这为控制系统的设计提供了有力的依据。     

Investigation of seasonal metallurgical shift at inco's clarabelle mill

The seasonal metallurgical shift at Inco's Clarabelle mill, Copper Cliff, Ontario represents a significant variation in pentlandite and precious metals recovery from summer to winter. The decrease in metals recovery in winter has been a subject of research and debate in the past decades. To understand this seasonal phenomenon, on-site testwork using a portable batch flotation column was conducted in two winter seasons and two summer seasons. On-site batch tests on the main flotation feed were used to obtain flotation rate and selectivity data. Other data such as dissolved ions and thiosalts in process and feed waters that could affect flotation was also collected.One of the most important findings from the present investigation was that both sulfide and gangue minerals were floated significantly faster in winter than in summer. The pentlandite /pyrrhotite and pentlandite / rock selectivity was, however, considerably decreased in winter. The process and feed water analysis indicated that the high concentration of dissolved nickel ions in the winter process water could have activated all minerals, leading to the higher flotation rates. The increase in pyrrhotite and rock flotation rates was, however, larger than the increase in pentlandite flotation rate. This resulted in the reduced pentlandite /pyrrhotite and pentlandite / rock selectivity in winter, which was the main cause for the seasonal metallurgical shift.     

低碱度铜硫分离高效抑制剂的研究

对有机抑制剂DP-1、无机抑制剂DP-2和DP-3浮选分离德兴铜矿一段铜硫混合精矿进行了试验研究。结果表明,DP-1、DP-2和DP-3都是铜硫分离时硫的有效抑制剂,但DP-3的综合性能要优于DP-1和DP-2抑制剂。闭路浮选试验结果表明,当DP-3总用量为500g/t时,可获得铜精矿中铜品位28.43%、铜回收率97.71%和钼品位0.212%、钼回收率80.56%的二段分离指标,与石灰工艺相比,铜、钼、金、银的回收率分别提高了0.75%、31.38%、2.76%和8.31%,表明低碱度浮选工艺对于伴生金属的回收具有十分明显的优势。生产综合样验证试验进一步证明捕收剂Mac-12和抑制剂DP-3可望实现德兴铜矿铜硫低碱度高效浮选分离。     

Effect of grinding media on zinc depression in a lead cleaner circuit

Laboratory testwork, investigating the effect of high chrome grinding media in a lead regrind application has on subsequent metallurgical performance, was conducted at a large silver–lead–zinc operation in Australia. The initial data showed that the use of a more inert grinding media could have increased zinc losses to the lead cleaner concentrate if careful attention was not paid to alloy selection.Diagnostic tests showed that iron hydroxide surface coatings generated by grinding media corrosion reactions are an effective depressant for sphalerite in this ore body, even though it is known that an excess of these coatings could depress both galena and sphalerite flotation. These tests demonstrated that a 1% chrome alloy produced the desired pulp chemical conditions to yield an increase in lead concentrate grade through the rejection of sphalerite from the lead circuit.A plant trial was conducted in one of the two parallel grinding/flotation trains, and data collected for statistical analysis. During the plant trial, pulp chemical surveys of the regrind circuit were also taken to compare the effect of grinding media on the cleaner one feed slurry pulp potentials, dissolved oxygen, pH, temperature and EDTA extractable iron.The statistical analysis showed clearly that the change to 1% chrome grinding media had a significant positive impact on improving galena/sphalerite selectivity during lead cleaner flotation and improved the lead concentrate grade. The improved metallurgical performance is explained in terms of modified pulp chemistry.     

Characterization of large size flotation cells

Design and operating conditions of large size mechanical flotation cells were evaluated by comparing it with the actual operating conditions in a plant. The objective was to determine the time scale-up factor, typically based on empirical rules. Experiments were conducted on the rougher flotation circuit at Minera Escondida Ltd. The circuit consisted of self-aerated mechanical cells of 160 m³, arranged in six parallel banks with nine cells each.The rougher circuit flotation kinetics was evaluated from direct sampling and local mass balances around each cell of the bank. Adjusted overall mass balances were also developed. This information was used to fit different kinetic flotation models, and it was found that the rectangular distribution function was the most appropriate to describe the distributed rate constant for industrial operation. Then, a rougher flotation simulator was developed to describe the actual operation in terms of the operating variables (mass flow rate, solid percentage, feed grade) and the actual volumetric flow rate entering to each cell. In this study feed pulp samples were taken in parallel from the rougher circuit and were simultaneously floated in laboratory. The kinetic behavior was then modeled at a laboratory batch scale in order to determine the time scale-up factor between laboratory batch flotation data and industrial size flotation. The time scale-up factor observed for large sized cells, 160 m³, was found reasonably similar to those previously determined for self-aerated mechanical cells, but of lower size, operating at similar recoveries. In addition, the relative effect of mixing, between laboratory batch and an industrial flotation bank was quantified by the ϕ parameter, separating the impact that kinetic and mixing changes have on the time scale-up factor.In general, the rougher flotation operation was found to reach the predicted metallurgical target, and that the optimal separability criterion was also respected.The diagnostic generates information about the internal state of the process and helps to identify potential improvements for design, operation and control of the circuit.     

Optimisation of a multistage flotation plant using plant survey data

The purpose of the investigation was to develop a procedure for optimisation of complex flotation circuits, using steam assays and pulp densities alone (i.e. no additional measurements or batch tests). Conventional models, based on pulp kinetics, were regressed directly to steady-state data from a nine-stage platinum flotation plant, to provide a basis for optimisation studies. ‘Froth factors’ were obtained by regression to characterise the froth in each stage and a ‘depressant factor’ was also required for gangue minerals, to account for effect of additional depressant in the cleaners. These factors improved the fit, but significant differences between plant data and the model were still present, particularly in the cleaning circuit.The model for the froth was modified, to incorporate two rate constants for the recycle of particles from the froth. These parameters provided a workable approximation of the distribution of probabilities of particles being recycled in plant cells, depending upon where particles appeared at the surface. In order to maintain an acceptable number of model parameters, only two pulp kinetic parameters and two recycle parameters were used for the three mineral classes. The fit to the data was improved very significantly, when compared to the use of conventional pulp-based models with several pulp rate parameters. This relatively simple model did not require additional data from batch tests and it was suitable for plant simulations.Simulations of various strategies for plant operation were tested, allowing the simulator to vary the concentrate flows from all or some of the stages, in order to maximise platinum recovery. The simulator was imbedded in an automatic optimiser and constraints on chrome content or mass flow of final concentrate were used. The simulations of the existing circuit demonstrated that an improvement in recovery of about 0.5 per cent could be achieved by optimising all concentrate flows. The mass pull of concentrate per unit area was used as a guide for maintaining efficient cell operation and in some cases cleaner cells were re-allocated to modify the volume of a stage. The addition of an extra cleaning stage was simulated, resulting in a further increase in recovery. Plant tests are in progress to verify the simulations.     

基于正交试验和BP人工神经网络的浮选药剂制度优化

探索了人工神经网络在浮选药剂制度优化中的应用。研究通过浮选药剂三因素三水平正交试验采集具有一定代表性的实验数据,然后采用BP人工神经网络,建立了输入为pH值调整剂碳酸钠的用量、抑制剂硅酸钠的用量和捕收剂的用量,输出为选矿效率的BP人工神经网络模型,并通过样本检验模型的准确性。试验结果:F检验确定的药剂制度为pH值调整剂用量4 kg/t、抑制剂用量3 kg/t、捕收剂用量1 kg/t,选矿效率为19.75%,BP人工神经网络确定的药剂制度为pH值调整剂用量4 kg/t、抑制剂用量2.3 kg/t、捕收剂用0.9 kg/t,选矿效率为19.47%,两种药剂制度的选矿效率差别很小,但BP网络确定的药剂制度药剂用量更少。结果表明利用BP人工神经网络可用于浮选药剂制度的优化,可减少试验量,可通过建立模型找出更优的药剂制度,同时结合正交试验减少人工神经网络的学习样本数量,增加样本代表性,优化神经网络的学习过程。     

基于长程时空特征与多尺度外观特征的锌精选工况识别北大核心

锌精选作为锌浮选的最后一道流程,其工况直接决定锌浮选最终产品质量。现有基于卷积网络的浮选工况识别方法具备挖掘隐藏特征的能力,取得了良好效果,但仍存在表征能力有限、模型参数大等问题。为此,提出了基于长程时空特征与外观特征的锌精选工况识别模型。首先,提出基于分离三维卷积网络(Separable 3D Convolutional Neural Network,S3D CNN)与注意力机制的泡沫视频相邻帧间短程时空特征提取方法,获得特征聚焦的泡沫视频相邻帧间短程时序信息。然后,在短程时空特征的基础上采用双向卷积长短时记忆网络(Bi-directional Convolutional Long Short-Term Memory,BiConvLSTM)提取泡沫视频帧间的长程时空特征,获取泡沫视频帧间的长程动态时序信息。最后,采用基于残差网络和迁移学习的二维卷积网络提取泡沫图像的多尺度外观特征,并融合长程时空特征,对锌精选工况进行识别。实验结果表明,与现有卷积网络方法相比,所提模型在工况识别精度和模型参数上性能更佳。     

Development of a flotation model incorporating liberation characteristics

A liberation/flotation model has been developed to evaluate flotation performance, in which the grade distribution of composite particles is incorporated explicitly. The grade distribution of particles at a given size fraction was calculated using a predictive liberation model developed by the authors. Flotation behaviour of these particles in narrow size fractions were evaluated using batch flotation tests. The recovery of particles in narrow size/grade classes at various flotation times under a given flotation environment were obtained from SEM images of mounted specimens of feed, concentrate and tailings samples using image analysis techniques.By describing the size, grade and recovery data of particles in narrow size/grade classes by separate matrices, a transformation matrix technique has been developed that can predict the flotation performance of a composite sample of an ore comprising various size classes. It has been shown that the predicted results agree well with the observed results.This liberation/flotation model has wide applications in geometallurgical testing and design, where the downstream flotation performance is to be determined with limited availability of ore samples and their liberation characteristics. It also alleviates the deficiencies involved with flotation models that assumes first order behaviour.     

冶金过程人工神经网络建模及其参数优化策略

鉴于人工神经（ANN）网络在处理复杂系统方面卓越的能力,采用人工神经网络来处理冶金过程的系统建模问题,分析并建立了人工神经网络建模的框架及应用方法。在用网络模型仿真实际系统的基础上,详细论述了基于网络模型的3种系统参数优化策略。将已建立的模型应用于冶金实际生产,结果表明,预测结果与实际生产数值相当吻合。     

Flotation scale up: use of separability curves

Separability curves (mineral recovery versus yield) have been used to characterize the copper flotation process both at batch laboratory scale and industrial plant scale (rougher banks). Then, to approach the scale-up problem the rougher bank operation and the batch were compared using the corresponding separability curves. Comparison was made at the maximum separation efficiency point in both operations. Thus, a time factor was established for optimal technical separation. The time factor can then be used for kinetic scale-up models, together with the ratio between minerals recovery in both operating scales. Experience from several tests recorded over a period of 10 months in an industrial concentrator showed a good consistency for scaling-up the rougher flotation recovery from batch tests within a 1% absolute error range.The effect of particle size and air flowrate in laboratory batch tests was evaluated in the space of separability curves, regarding their effect on recovery at the optimum separability point. Also, the effect of pulp level and particle size on the bank flotation kinetics was evaluated in an industrial flotation circuit. Thus, estimation of recovery changes due to variations in mineral characteristics and operating conditions was explored.     

A novel approach to the geometallurgical modelling of the Collahuasi grinding circuit

Compañía Minera Doña Inés de Collahuasi SCM initiated the development of a new geometallurgical model to characterize its Rosario deposit in terms of its comminution circuit capacity and flotation performance. The comminution component of the model is now complete and is described in detail in this article.The model uses a combination of simulation and power-based approaches to relate ore hardness and flotation feed size to grinding circuit throughput. Ore hardness values are provided by Collahuasi’s block model which has been populated with ore characterisation data derived from Bond ball work index tests, JK drop-weight tests and SMC Tests^® on diamond drill cores. The influences of planned and unplanned maintenance downtime are taken into account both in terms of frequency and duration.The accuracy of the model is demonstrated using weekly production data taken from the period January–December 2008, showing an average relative error of 5.2% and an R² value of 0.95.The model is now in successful routine use at the mine for planning purposes and is currently in the process of being further developed to incorporate the response of the flotation circuit.     

Interfering ions in the flotation of a phosphate ore in a batch column

The “Barreiro” carbonatite complex is located in Araxá, MG, Brazil. The production of apatite concentrate from this ore started in 1978 and the current production capacity is 800,000 tonnes.The ore is subjected to the following unit operations in the concentrator: grinding, classification, low field magnetic separation, desliming, apatite and barite flotation, high field magnetic separation, filtration and drying.Barite and apatite flotations were performed in 300 ft³ (8.5 m³) Wemco mechanical cells until 1992. The replacement of the mechanical cells by flotation columns occurred in two stages: fine fraction flotation circuit in 1993 and coarse fraction flotation circuit in 1994. All the 66 mechanical cells were replaced by 6 flotation columns (3.0 m × 4.5 m × 14.5m), resulting in improved selectivity and recovery and significant reductions in the consumption of the collectors for barite and, mainly, for apatite.The flotation process control monitoring in laboratory bench scale mechanical cells was then inadequate due to the discrepancy with the industrial practice. A specific methodology was developed for batch column flotation testing. The technique was checked by means of comparison with the industrial plant practice of apatite and barite flotation, showing its reproducibility and reliability for process development testwork.The batch flotation column was utilised to evaluate the effect of ions (dosed in the conditioning stage) on the flotation of barite and apatite. Flotation tolerance limits were established for the ions calcium, magnesium, phosphate, fluoride and rice bran oil soaps. Due to the fact that these ions were dosed as sodium salts (in the case of anions) or as chlorides (in the case of cations), the effect of sodium and chloride ions was also assessed (tests with the addition of sodium chloride).The results led to the environmental benefit of a full process of water recirculation in the concentrator without impairing the flotation performance.