TECHNICAL FIELD

[001] The present disclosure relates to systems and methods used for a milling system of a mining operation.

BACKGROUND

[002] Mining iron or magnetite ore has a long, traditional, history. Iron ore plays a key part in our society but there are various challenges presented in mining iron ore. There is a need to improve existing systems and methods used for a milling system of a mining operation, such as improving efficiency, stability of the system, and/or reducing manual operation.

[003] Any reference to or discussion of any document, act or item of knowledge in this specification is included solely for the purpose of providing a context for the present invention. It is not suggested or represented that any of these matters or any combination thereof formed at the priority date part of the common general knowledge, or was known to be relevant to an attempt to solve any problem with which this specification is concerned.

SUMMARY OF THE INVENTION

[004] In a first aspect, the present disclosure provides a milling system comprising: a feed assembly configured to transport materials; a milling station arranged to receive the materials from the feed assembly and process the materials to smaller sizes; a separating assembly arranged to receive a product of the milling station and arranged to produce a mineral concentrate therefrom, and a control system arranged to control operation of at least one of the feed assembly, the milling station and the separating assembly, based on a target flow rate of the separating assembly.

[005] In one embodiment, the materials are raw materials, such as crushed ore.

[006] In one embodiment, the mineral concentrate is a magnetite concentrate. [007] In one embodiment, the feed assembly comprises: one or more apron feeders, feed conveyors, and transfer conveyors arranged to transport the materials from a stockpile to the milling system, and between different components of the milling system.

[008] In one embodiment, the feed assembly further comprises one or more pumps.

[009] In one embodiment, the milling station includes: a primary milling device; one or more crushing devices, and one or more associated screening devices arranged to separate coarser materials from a product of the primary milling device for further grinding.

[010] In one embodiment, the primary milling device is an autogenous mill (AG mill), and at least one of the crushing devices includes a pebble crusher.

[011] In one embodiment, the AG mill includes: an inlet arranged to receive the materials from the feed assembly, a rotating body, inside which the materials are processed into smaller sizes, and an outlet, where a product of the AG mill exits from the AG mill.

[012] In one embodiment, the separating assembly includes a cyclone cluster arranged to receive the product of the AG mill, and to separate the product into two streams: an overflow including lighter and finer materials, and an underflow including heavier and coarser materials which require further grinding by the milling station.

[013] In one embodiment, the separating assembly includes a magnetic separator arranged to receive the overflow from the cyclone cluster and to produce the magnetite concentrate, whereas the underflow of the cyclone cluster is circulated back to the AG mill for regrinding.

[014] In one embodiment, the target flow rate of the separating assembly is determined based on a target mass flow rate of the overflow.

[015] In one embodiment, the control system is arranged to control operation of one or more of the feed assembly, the milling station and the separating assembly based on one or more process values, including: weight loading of the AG mill;

AG mill speed; AG mill energy consumption;

AG mill stator current;

AG mill lubrication bearing pressures; weight loading of the feed assembly; calculated mass flow rate of the overflow, and size measurements of the materials.

[016] In one embodiment, the control system includes a processing module arranged to receive the process values, and to perform pre-processing of the process values.

[017] In one embodiment, the pre-processing of the process values include one or more of: filtering, smoothing, calculating a rate of change (ROC), calculating a moving average, predicting a trend, and similar thereof.

[018] In one embodiment, the control system is arranged to control one or more of the following process parameters, to achieve control of the milling system: feed rate of the raw materials to the milling system; weight loading of the AG mill;

AG mill speed;

AG mill energy consumption, feed rate of process water to the AG mill; speed of the pebble crusher; feed rate of the product to the cyclone cluster.

[019] In one embodiment, the control system is configured to implement a supervisory control loop and one or more subsidiary control loops, wherein the supervisory control loop obtains or receives data indicative of the process values, and determines a setpoint or a process value of one or more of the subsidiary control loops in real time, wherein the subsidiary control loops are arranged to control the process parameters.

[020] In one embodiment, at least some of the subsidiary control loops are cascaded, that is, an output of a subsidiary control loop determines a setpoint or a process value of another subsidiary control loop.

[021] In one embodiment, data indicative of setpoint constraints are also obtained or received by the supervisory control loop. [022] In one embodiment, the setpoint constraints are entered in a SCADA system, which are then sent to the supervisory control loop.

[023] In one embodiment, the control system includes a plurality of programmable logic controllers (PLC), and the SCADA system.

[024] In one embodiment, the control system includes a number of PID controllers which are implemented in the PLC controllers.

[025] In one embodiment, the milling system includes a second milling station placed downstream of the separating assembly, and an associated second separating assembly arranged to produce magnetite concentrate.

[026] In a second aspect, the present disclosure provides a control method for a milling system, comprising a feed assembly configured to transport materials, a milling station for processing the raw materials into smaller sizes, and a separating assembly arranged to produce a mineral concentrate, the control method comprises: implementing a supervisory control loop and one or more subsidiary control loops, wherein the supervisory control loop is configured based on a target flow rate of the separating assembly, which determines a setpoint and/or a process value of the one or more of the subsidiary control loops; wherein the subsidiary control loops are arranged to control one or more process parameters of the milling station, the feed assembly, and the separating assembly.

[027] In one embodiment, the mineral concentrate is a magnetite concentrate.

[028] In one embodiment, the feed assembly comprises: one or more apron feeders, feed conveyors, and transfer conveyors arranged to transport the materials to the milling system and between different components of the milling system.

[029] In one embodiment, the milling station includes a primary milling device, one or more crushing devices, and one or more associated screening devices arranged to separate coarser materials from a product of the primary milling device for further grinding by the one or more crushing devices.

[030] In one embodiment, the primary milling device is an autogenous mill (AG mill), and at least one of the crushing devices include a pebble crusher. [031] In one embodiment, the AG mill includes: an inlet arranged to receive the raw materials from the feed assembly, a rotating body, inside which the raw materials are processed into smaller sizes, and an outlet, where a product of the AG mill exits from the AG mill.

[032] In one embodiment, the separating assembly includes a cyclone cluster arranged to receive the product of the AG mill, wherein the cyclone cluster is configured to separate the product into two streams: an overflow and an underflow, wherein the overflow includes lighter materials, and the underflow includes heavier materials which require further milling by the milling station.

[033] In one embodiment, the separating assembly includes a first magnetic separator arranged to receive the overflow from the cyclone cluster and to produce the magnetite concentrate.

[034] In one embodiment, the process values include one or more of the following: weight loading of the AG mill;

AG mill speed;

AG mill energy consumption;

AG mill stator current;

AG mill lubrication bearing pressures; weight loading of the feed assembly; actual overflow mass flow rate of cyclone cluster; size measurements of the raw materials.

[035] In one embodiment, the one or more process parameters include one or more of the following: feed rate of the raw materials to the milling system; weight loading of the AG mill;

AG mill speed;

AG mill energy consumption, feed rate of process water to the AG mill; speed of the pebble crusher; feed rate of the process raw materials to the cyclone cluster. [036] In one embodiment, at least some of the subsidiary control loops are cascaded, that is, a setpoint and/or a process value of a subsidiary control loop is determined by an output of another subsidiary control loop.

[037] Further features and advantages of the present disclosure will become apparent from the following detailed description.

BRIEF DESCRIPTION OF THE DRAWINGS

[038] Various preferred embodiments of the present disclosure will now be described, by way of examples only, with reference to the accompanying figures, in which:

Figure 1 illustrates a milling system according to one embodiment of the present disclosure;

Figure 2 shows a schematic of a control system according to one embodiment of the present disclosure;

Figure 3 shows how a control system may be implemented to control operation of a feed assembly according to one embodiment;

Figure 4 shows an example of a control loop used to control operation of a feed assembly;

Figure 5 shows a comparison of measurements obtained by the control system before and after data pre-processing;

Figure 6 shows how the control system may be used to control AG mill speed;

Figure 7 shows a control system used for controlling feed rate of process water to an autogenous mill (AG mill) according to one embodiment;

Figure 8 shows a control system used for controlling pebble crushers of the milling system according to one embodiment;

Figure 9 shows a control system used for controlling a cyclone cluster of the milling system according to one embodiment; Figure 10 shows a schematic of a control loop implemented to control AG mill process water feed according to one embodiment;

Figures 11 a and 11b show improved results of AG mill process water feed control, for energy consumption purpose;

Figure 12 shows a schematic of a control loop implemented to achieve pebble crusher control, and

Figure 13 shows a schematic of a control loop implemented to achieve cyclone cluster control.

DETAILED DESCRIPTION OF EMBODIMENTS

1. A Milling System

[039] Figure 1 illustrates an exemplary milling system 10 according to an embodiment of the invention.

[040] The milling system 10 is primarily used for grinding raw materials into smaller sizes, such that they are suitable for subsequent extraction of a mineral concentrate from the raw materials. The embodiments of the present disclosure have been described with respect to a milling system that processes magnetite ore in order to extract magnetite concentrate therefrom, however, it should be appreciated that the systems and methods disclosed herein can be readily applied to any milling systems generally used by the mining industry. It should be also appreciated that a mining site may include a plurality of such milling systems which operate in the same location and/or in distributed locations in order to process a large quantity of raw mining materials.

[041 ] With reference to Figure 1 , the milling system 10 includes a feed assembly 100 for transporting magnetite ore from a stockpile to the milling system 10, a milling station 200 arranged to process the magnetite ore by way of crushing or grinding the received magnetite ore into smaller sizes, and a separating assembly 300 configured to extract magnetite concentrate from the processed magnetite ore which it receives from the milling station 200.

[042] In this embodiment, the feed assembly 100 includes one or more apron feeders 101 , feeding conveyors 102, and/or transfer conveyors 103 which assist movement of the magnetite ore from the stockpile to the milling system 10, as well as movement between different components of the milling system 10.

[043] As mentioned, the milling station 200 is arranged to process the magnetite ore into smaller sizes such that they are suitable for extraction of magnetite concentrate by the separating assembly 300. The milling station 200 includes a primary milling device, which is an autogenous mill (AG mill) 201 in this example. The AG mill 201 includes an inlet 202 for receiving magnetite ore from a feed conveyor 102, a rotating body portion 203, and an outlet 204 for discharging a magnetite slurry. The received magnetite ore makes its way into the rotating body portion 203 of the AG mill 201 , and as the body portion 203 rotates, larger rocks of ore are moved in a cascading motion which causes breakage and compressive grinding of finer particles. Process water is also injected into the AG mill 201 , which turns crushed magnetite ore and finer particles into magnetite slurry. At this point, the magnetite slurry discharged from the outlet 204 of the AG mill 201 usually still includes larger rocks which require further regrinding and are not suitable to be sent to the separating system 300 yet. Accordingly, at the outlet 204 of the AG mill 201 there is provided a screening device 210 which receives the magnetite slurry from the AG mill 201 , and separates it into two product streams: a first product streams includes large rocks which require further grinding, and a second product stream which is ready to be processed by the separating assembly 300. The first product stream is transported to crushingdevices of the milling station 200, for example pebble crushers 230a, 230b, which are arranged to crush the larger rocks into smaller pieces, and then recirculate the crushed rocks back to the inlet 202 of the AG mill 201 via feed conveyor 102. The second product stream is discharged to a feeding hopper 207 first, and then sent to the separating assembly 300 via a pump 208. The speed of the pump 208 directly determines how much feed is sent to the separation assembly 300.

[044] The separating assembly 300 comprises a cyclone cluster 301 . The cyclone cluster 301 creates a centrifugal movement of the magnetite slurry that it receives, allowing centrifugal force to push heavier materials in an outward direction, and then downwardly along a wall of the cyclone cluster 301 . Finer materials are moved upward of the cyclone cluster 301 in spiral movement until they exit the cyclone cluster 301 . The coarser materials which are discharged from a bottom of the cyclone cluster 301 are usually referred to as underflow, which is sent back to the AG mill 201 for regrinding. The lighter materials which are moved upwards and arranged to exit the cyclone cluster 301 from a top end thereof are usually referred to as overflow. The overflow is stored in a hopper bin 302 first, and then sent to a magnetic separator 310 at a controlled rate by an associated pump 303. The magnetic separator 310 extracts magnetite concentrate from the overflow, which has already turned into magnetite slurry at this point. [045] In the embodiment shown, the milling system 10 also includes a second milling station, which includes a ball mill 320, and additional separating assemblies which include cyclone cluster 301a, and magnetite separators 310a, 310b.

[046] In existing mining operations, the monitoring and control of the operation of a milling system of a similar form to the embodiment shown in Figure 1 is usually achieved by a SCADA system and a number of PLC controllers. In use, process parameters of the various components of the milling system 10 are manually configured or adjusted by an operator via the SCADA system. The present disclosure provides an improved or an alternative control system, which automatically sets process parameters of one or more of the feed assembly 100, the milling station 200, and the separating assembly 300 without, or with very minimal, human intervention.

2. Control system

[047] According to the present disclosure, and with reference to Figure 2, the control system 400 controls various process parameters of one or more of the feed assembly 100, the milling station 200, and the separating assembly 300, based on a target flow rate of the separating assembly 300. More preferably, the target flow rate is target mass flow rate of the overflow of the cyclone cluster 301 , that is, the target mass flow rate of the magnetite slurry discharged by the cyclone cluster 301 . The present inventors have discovered that the target mass flow rate of the overflow is a critical parameter for the milling system 10, and configuring the control system 400 based on this parameter provides noticeable advantages such as increase of average throughput rates, stability of concentrate product grade, reduced operator intervention, and reduced system variances.

[048] In a more preferred embodiment, the control system 400 receives or obtains data indicative of one or more of the following process values:

• Weight loading of the AG mill Wl (measured in tons);

• AG mill speed SI (rpm);

• AG mill energy consumption JI (measured in MW);

• AG mill stator current II (measured in Amps);

• AG mill lubrication bearing pressures PI (measured in bars);

• Calculated mass flow rate of the overflow from the cyclone cluster 301 FMI (measured in tons/hour), based on density of slurry and measurements provided by flow rate sensors. [049] The control system is implemented such that it includes a supervisory control loop, and one or more subsidiary control loops each arranged to control one or more components of the milling system 10. In a preferred embodiment, the subsidiary control loops may include one or more of the following:

• AG mill 201 weight and fresh feed control

• AG mill 201 speed control

• AG mill 201 energy consumption per ton control

• AG mill 201 process water feed control

• Pebble crusher 230a, 230b power loop control

• Pebble crusher 230a, 230b bypass flow control

• Cyclone cluster 301 feeding hopper control

[050] Figure 2 illustrates an example of the control system 400 architecture. The control system 400 includes a SCADA client 401 , a SCADA server 402, a standby SCADA server 403, and a plurality of PLC controllers 404a, 404b, 404c, 404d, 404e where the supervisory control loop and the subsidiary control loops are programmed and implemented. Data indicative of the process values mentioned above are collected by the plurality of PLC controllers 404a-c, and the output of the PLC controllers 404a-c are then used to track and control operation of the milling system 10, for example, by controlling various process parameters of the system, such as:

• movement speed of the apron feeders 101 , to thereby increase or decrease the AG mill 201 weight loading;

• power settings of the AG mill 201 , to thereby cause the AG mill 201 to rotate faster or slower;

• power settings of the pebble crushers 230a, 230b, to thereby cause the pebble crushers to process the received larger rocks faster or slower;

• feed rate of process water into the AG mill 201 , and

• feed rate of overflow to the cyclone cluster 301 from the feeding hopper 207.

[051] To configure and commission the control system 400, first, a step test is performed by moving setpoints up and down in a step pattern in order to obtain response data for the modelling of the control system 400. Next, detailed design of the SCADA interface, and programming of the PLC controllers 440a-c will be carried out. Following this, the control system 400 will be put into an online simulation phase, which allows it to suggest recommended setpoints which an operator can manually enter in the SCADA system. Once the simulation phase has completed, the control system 400 enters into active commission phase and takes over real time control of the milling system 10. After commissioning, the control system 400 may enter an evaluation period, to assess efficiency and accuracy of the control system 400, which may cause additional tuning of the system 400.

2.1 AG mill

AG mill weight control

[052] As mentioned above, magnetite ore is received and grinded within the AG mill 201 before magnetite contrate can be produced by the milling system 10. The weight of the magnetite ore within the body portion 203 of the AG mill 201 , that is, the weight loading of the AG mill 201 (Wl) must be closely monitored and controlled, to avoid overloading or underloading of the equipment. The control system 400 achieves control of the AG mill weight loading, via manipulating movement speed of the feed assembly 100, and more preferably, the movement speed of the apron feeders 101. In this embodiment, the apron feeders 101 are driven by one or more variable speed drives (VSD).

[053] Figure 3 and 4 illustrate an example of how the control system 400 may be utilised to control the feed assembly 100, to achieve subsequent control of the AG mill weight. In this embodiment, there are three apron feeders 101 a, 101b, 101c, and each apron feeder is driven by an associated VSD. Raw materials from a stockpile are transported to the AG mill 201 via the three apron feeders 101 , 101 b, 101c, a transfer conveyor 103, then to a feeding conveyor 102, before they are supplied to an inlet of the AG mill 201 .

[054] With reference to Figure 4, the control system 400 includes:

• a first PID controller 420 which is a controller that tracks and controls AG mill weight loading Wl,

• a second PID controller 430 which is a controller that tracks and controls a weight odometer rate of the feed conveyor 102, and

• a third PID controller 440 which is the controller for the mass flow rate of the cyclone cluster 301.

[055] The first, second, and third PID controllers operate as follows:

[056] First, the control system 400 obtains or receives data indicative of the following parameters: AG Mill weight loading Wl, size of the raw materials F80, AG mill stator current II, AG mill power consumption JI, AG mill lubrication bearing pressures PI, and mass flow rate of the cyclone cluster overflow FMI (calculated based on field slurry density and measurements provided by flow rate sensors), and performs pre-processing of some or all of the data received.

[057] The third PID controller 440 uses F80 as its setpoint, and FMI as its process value, the control output of the third PID controller is used as the setpoint of the first PID controller 420.

[058] The first PID controller 420 is the AG mill weight controller, and receives Wl as its process value, the control output of the first PID controller 420 is sent to the second PID controller 430 as its setpoint.

[059] AG mill weight loading Wl is also provided as a feedforward parameter to the second PID controller 430, and measured AG mill stator current II, AG mill power consumption JI, AG mill lubrication bearing pressures PI are used as control limits of the second PID controller 430. Take the AG Mill lubrication bearing pad pressures as example, any pressure value over a high constraint setting of 102 bar will be one of the suppression condition of increasing AG Mill load. The AG mill weight Wl is also provided to the second PID controller 430 as process value.

[060] Finally, the control output of the second PID controller 430 is used to control speed of the VSDs used to drive the apron feeders 101 -101c.

[061] As mentioned above, the control system 400 obtains or retrieves data indicative of various process values, and then performs pre-processing of these data before using them in various control loops. The pre-processing tasks are performed in a number of pre-processing modules 410 as described below.

[062] The pre-processing tasks performed by the module 410 may include noise filtering, data smoothing, calculating a moving average, trend prediction, rate of change calculation, and similar thereof. The processing module 410 is important as it converts the obtained data from an analogue form to a digital form such that they can be used by the control system 400, and it also allows the control system 400 to implement its functions more accurately with errors and fluctuations removed. In a preferred embodiment, the process module 410 may include a group of derived function blocks (DFB).

[063] Figure 5 provides an example of data smoothing function performed by the preprocessing module 410. In this example the analogue smoothing formula is created as: Y=Y_Last + (dt/(dt+Lag)) x (((X + X_Last)/2) - Y_Last); wherein X represents raw data input, and Y represents pre-processed data output; dt is the PLC scan time measured in milliseconds,

Lag is a constant time value coded separately for individual analogue data.

[064] The pre-processing module 410 provides a rate of change (ROC) calculation, for example based on a subtraction result of short term moving average and long term moving average based on PLC scan time, the sampling number and interval varies as required performances for each analogue data. Figure 5 demonstrates that fluctuations and noise signals in the received measurements obtained from field sensors have been successfully removed or reduced, thereby generating a more smooth curve.

[065] Similarly, the size of the raw materials F80 are collected and sent to the pre-processing module 410 to perform data filtering and rate of change (ROC) prediction on the obtained results. The addition of filtered size values and ROC are then weighted, and sent to the third PID controller 440 as its setpoint.

[066] Measurements of the AG mill weight loading Wl are also subject to data pre-processing before they are provided to the second PID controller 430 as a feedforward parameter. For example, a rate of change prediction is performed on the received AG mill weight measurements, the results are then integrated into the second PID controller 430 as adjustable feedforward value to improve stability of the control loops.

[067] Measurements of AG mill stator current II, AG mill lubrication bearing pressures PI, and AG mill power consumption JI are obtained by the control system 400 for safety purpose and for monitoring health condition of the AG mill 201 . These measurements are processed to determine control limits of the third PID controller 430. In addition to the control limits, a set of constraints (max, min) may also be entered by an operator through the associated SCADA, which can be changed online at any time without stopping the milling system 10. When the milling system 10 operates within the set of constraints and control limits, the control system 40 manipulates the control output of the first PID controller 420 as the target setpoint of the second PID controller 430.

[068] Advantageously, the PID controllers 420, 430, 440 are cascaded, meaning a control output of a PID controller is used as a setpoint of a downstream PID controller. This allows the control system 400 to operate based on adjustable setpoints which is a more accurate and robust method of controlling the AG mill weight loading. For example, if the process value of the PID controller 440 (i.e. mass flow rate of the cyclone cluster overflow FMI) is higher than the setpoint, the control output of the PID controller 440 will be automatically adjusted by the PID controller 440, which then cascades down to the PID controller 420, and then to PID controller 430. In other words, the setpoints of the downstream PID controllers 420 and 430 are automatically adjusted, based on the mass flow rate of the cyclone cluster overflow FMI received by the PID controller 440.

AG mill speed control

[069] Figure 6 shows an example of how the control system may be utilised to control the speed of the AG mill 201 . AG Mill speed control focuses on the grinding efficiency and the stability of the milling system shown in Figure 1 . In the past, manual adjustment of the AG mill speed is a source of process disturbance and variability and is generally undesirable. With the present disclosure, two cascaded PID controllers 610, 620 are included in the control loop which automatically controls the AG mill speed as explained further below.

[070] The setpoint of the first PID controller 610 is determined based on particle size of raw materials F80. The pre-processing module 410 again performs data filtering, and rate of change (ROC) prediction as described above, the addition of the filtered value and ROC with individual weight factor constants is the CAL1 result sent to the first PID controller 610 as its setpoint. Weight factor constants are essentially data constants obtained during the commissioning phase of the control system, which indicate the importance of a corresponding data value. The process value of the first PID controller 610 is the cyclone cluster overflow mass FMI, calculated from field slurry density and measurements provided by flowrate sensors. The control output of the first PID controller 610 is then used as the adjustable setpoint of the second PID controller 620, which directly tracks and controls AG mill movement speed.

[071] As illustrated, a number of feed forward parameters are also sent to the second PID controller 620, such as AG mill weight loading Wl, AG mill power JI, AG mill stator current II, and AG mill lubrication bearing pressures PI. These parameters are used for prediction of AG mill overload, which should be carefully avoided at all times. As the second PID controller 620 is used to track and control AG mill speed, its process value is provided by AG mill speed SI (rpm).

AG mill process water feed control

[072] Process water is supplied to the AG mill 201 to assist with grinding of magnetite ore and turning fine particles into magnetite slurry. In existing milling systems, the supply of process water is controlled by a control valve adjusted manually based on experience and understanding of the process. The control system 400 of the present disclosure provides automatic control of the supply of process water to the AG mill 201 , through direct control of an associated valve 205 as illustrated in Figure 7.

[073] Figure 10 shows a schematic of a control loop implemented to control AG mill process water feed, in which two cascaded PID controllers 1010 and 1020 are employed. The first PID controller 1010 is an AG mill unit power consumption controller, and its process value is determined based on AG Mill power JI (MW) and calculated cyclone cluster overflow mass FMI (t/h). The setpoint of the first PID controller 1010 is determined based on the size of the raw materials F80. The control output of the first PID controller 1010 is then used as an adjustable setpoint of the second PID controller 1020.

[074] A number of feed forward parameters are sent to the second PID controller 1020, such as AG mill weight Wl, AG mill power JI, AG mill stator current II, and AG mill lubrication bearing pressures PI. In Figure 10, CAL1 is the ROC trending prediction of AG Mill weight loading Wl, CAL2 is the ROC trending prediction of AG Mill power JI (MW) and AG Mill stator current II (Amp). CAL1 and CAL2 are created as feed forward parameters for AG Mill overload prediction.

[075] Figures 11 a and 11 b show improved results of AG mill process water feed control, for energy consumption purpose. Each table includes measurements obtained from six different milling systems (L1 to L6), and only L4 includes the control system 400 of the present disclosure. Comparing the performance of L4 and L5, which both have similar amount of fresh feed of raw materials, the unit energy consumption of the AG mill has been reduced by 0.3 KWH/T in Figure 11 a, and reduced by 0.8 KWH/T in Figure 11b, which is a significant improvement.

2.2. Pebble crusher control

[076] Many existing pebble crushers rely on manual operation, which is the main source of process disturbance and variability. In the past, pebble crusher bin level control was done manually by operators via changing setpoints of pebble crusher or AG mill controllers, which puts additional complexity to normal operations.

[077] Figures 8 and 12 demonstrate how the control system 400 may be implemented to achieve control of the pebble crushers 230a, 230b, through control of variable speed drives used to drive the pebble crushers 230a, 230b. [078] The control system 400 obtains measurements of the weight loading of the AG mill Wl, bin levels LI as process values. The target mass flow rate of the cyclone cluster 301 and a set of setpoint constraints are also entered into the control system 400 to implement its control functions of the pebble crushers 230a, 230b. Similar to other control loops, process values obtained from various field sensors are subject to pre-processing by the pre-processing module 410. For example, the control system introduces CAL1 and CAL2, multiplied by weight factor constants to set the setpoint of pebble crusher current PID controller 1210. CAL1 is the ROC trending prediction of AG Mill weight loading Wl, CAL2 is the ROC trending prediction of AG Mill power JI (MW) and AG Mill stator current II (Amp). In addition, PID controller 1220 monitors and controls the level of materials stored in bin 1 and 2 associated with the pebble crushers 230a, 230b respectively, and its control output is used to set the setpoint of PID controller 1230 which controls bypass conveyor weight flow rate. In some embodiments, the bypass conveyor 240 is located between pebble crushers and the pebble return conveyor, and is used to transport materials within the pebble crushers back to the AG mill feed conveyor if the level of materials within the pebble crusher bins is too high. In previous systems, the bypass conveyor 240 was manually operated to adjust the speed of the conveyor. With the control system 400, the bypass conveyor 240 will be controlled by pebble bin level PID controller. In Figure 12, CAL3 is the ROC trending prediction of pebble crusher return rate and is sent as a feedforward parameter to the PID controller 1230.

2.3. Cyclone cluster control

[079] As mentioned above, the input of the cyclone cluster 301 is provided by feeding hopper 207 and its pump 208. The stability of the feeding hopper level is critical for maintaining normal operation of the cyclone cluster 301 and the milling system 10. Figure 13 illustrates how the control system 400 may be implemented to achieve control of level and density of the feeding hopper 207.

[080] According to the present disclosure, an auto swapping function has been developed for hopper bin dual level sensors, based on environmental conditions and maintenance experience. Each individual level sensor includes a bad channel check, data frozen check and rate of changes (ROC) check, which are arranged to trigger auto swapping to another level sensor when a predefined condition has been detected. An additional hopper pressure sensor may also be included to assist with sensor swapping function if both level sensor data escapes from condition checks, or as a differential value between the dual level sensors exceeds a certain amount. [081] With reference to Figure 13, particles sizes of raw materials F80 are collected to the preprocessing module 410 which performs data filtering, smoothing functions and rate of change (ROC) prediction. The addition of F80 values, and ROC with individual weight factor constant is the pre-processed result send to PID controller 1310, which tracks the cyclone cluster overflow mass flow rate FMI. This PID controller 1310 is introduced on top of cyclone cluster overflow pressure PID controller 1320, to automatically determine its adjustable setpoint. The cyclone cluster overflow pressure PID controller 1320 then cascades down to a cyclone cluster feeding flow rate controller 1330, to automatically adjust its variable setpoint. The PID controller 1330 is arranged to control a variable speed drive associated with the pump 208.

[082] In addition, the control system 400 introduces a cyclone cluster feeding slurry density controller 1340 on the top of a hopper level controller 1350, which directly controls the rate of process water being supplied to the feeding hopper 207 by adjusting operation of a control valve.

[083] The operation of an AG mill relies heavily on the ore itself as a grinding medium. Accordingly, characteristics of the ore, such as hardness of the ore, and particle sizes, have a great impact on grinding efficiency. The capacity and power consumption of the milling system often fluctuate as different types of raw materials are processed. Other factors, such as frequent manual intervention, manual adjustment of various components of the milling system, may also have an impact on the capacity and power efficiency of the milling system. In the past, it is difficult to achieve or maintain an optimal griding efficiency, due to the complex nature of the incoming ore. According to the present invention, logical relationships between various parameters that affect the grinding efficiency and power consumption of the milling system have been established, and implemented in the control system described herein. The control system allows the control parameters to be adjusted automatically, based on characteristics of the incoming ore, and the current state of the milling system, such that the operation of the milling system is more stable and efficient.

[084] In this specification, adjectives such as left and right, top and bottom, hot and cold, first and second, and the like may be used to distinguish one element or action from another element or action without necessarily requiring or implying any actual such relationship or order. Where context permits, reference to a component, an integer or step (or the alike) is not to be construed as being limited to only one of that component, integer, or step, but rather could be one or more of that component, integer or step. [085] In this specification, the terms ‘comprises’, ‘comprising’, ‘includes’, ‘including’, or similar terms are intended to mean a non-exclusive inclusion, such that a method, system or apparatus that comprises a list of elements does not include those elements solely, but may well include other elements not listed.

[086] The above description relating to embodiments of the present disclosure is provided for purposes of description to one of ordinary skill in the related art. It is not intended to be exhaustive or to limit the disclosure to a single disclosed embodiment. As mentioned above, numerous alternatives and variations to the present disclosure will be apparent to those skilled in the art from the above teaching. Accordingly, while some alternative embodiments have been discussed specifically, other embodiments will be apparent or relatively easily developed by those of ordinary skill in the art. The present disclosure is intended to embrace all modifications, alternatives, and variations that have been discussed herein, and other embodiments that fall within the spirit and scope of the above description.