FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENT RULES, 2003
COMPLETE SPECIFICATION (See Section 10 and Rule 13)
Title of invention:
METHOD AND SYSTEM FOR GRINDING CIRCUIT OPTIMIZATION
Applicant
Tata Consultancy Services Limited A company Incorporated in India under the Companies Act, 1956
Having address:
Nirmal Building, 9th floor,
Nariman point, Mumbai 400021,
Maharashtra, India
Preamble to the description
The following specification particularly describes the invention and the manner in which it is to be performed.

TECHNICAL FIELD [001] The disclosure herein generally relates to optimization of a grinding circuit, and, more particularly, to a method and system for real time optimization of a grinding circuit using an integrated data model and an integrated data generated using the integrated data model.
BACKGROUND [002] Mineral processing is an energy intensive industry where optimization of key unit operations and processes is crucial for achieving profitability. Grinding unit is the major contributor towards energy demands in the ore beneficiation process and minimization of power consumption while maximizing the throughput at desired particle size is a challenging task. Purely first principle-based models become too computationally complex and slow which presents hurdles in real-time optimization. Purely data-based models may not capture the complex physics behind the process. For example, a hold-up model associated with the grinding circuits is traditionally modelled using experimentally measured variables such as residence time distribution (RTD), ball mill weight (using load cells), visible measurements based on transparent ball mills, critical flow rate for onset of overflow, etc. Due to practical implementation challenges, availability of the required data maybe a concern at times, affecting the grinding circuit optimization. When data is collected in a static manner as an alternate approach, it fails to take into consideration dynamic nature of the data, and in turn effectiveness of the grinding circuit optimization being performed.
SUMMARY [003] Embodiments of the present disclosure present technological improvements as solutions to one or more of the above-mentioned technical problems recognized by the inventors in conventional systems. For example, in one embodiment, a processor implemented method for real time optimization of a grinding circuit is provided. The method involves receiving input data comprising a plurality of real-time data and non real-time data from the grinding circuit.

Further, the input data is pre-processed to generate a pre-processed data, wherein the pre-processed data satisfies a plurality of conditions. Further, a simulated data is generated in real time using an integrated model comprising a holdup-model, a particle-size model, and a power draw model. Generating the simulated data involves the following steps. Initially, a hold-up data is generated by processing information on a) flow rate (Q), b) ball filling (J), and c) solid concentration in slurry (c) used in the grinding circuit, using the hold-up model, wherein the hold¬up data comprising a predicted hold-up, a slurry filling (U), and mean residence time. Further, a particle size distribution in a ball mill is determined using the particle-size distribution model, comprising a plurality of selection and breakage functions selected based on material properties and slurry filling from the holdup data. Further, power drawn by the ball mill and an auxiliary equipment in the grinding circuit is predicted using the power draw model, wherein the power drawn by the ball mill and an auxiliary equipment is predicted based on a slurry filling parameter from the hold-up data and a slurry flow rate, wherein the hold¬up data, the predicted particle size distribution data, and the predicted power draw data, form the simulated data. After generating the simulated data, an integrated data is generated by combining the pre-processed data with the simulated data, wherein the integrated data and the integrated model are used for real time grinding circuit optimization.
[004] In another embodiment, the simulated data and the integrated model are used for the real time optimization of the grinding circuit. The optimization involves the following steps. Initially, an input data comprising the plurality of real-time data and the non real-time data is received from the grinding circuit. Further, the grinding circuit is optimized by processing the received input data with the integrated data and integrated model, which further involves the steps of determining one or more decision variables to be optimized in the grinding circuit, by solving an optimization problem using the received input data, wherein the optimization problem is modeled to minimize the power drawn by the grinding circuit and maximize throughput of the grinding circuit, and generating

one or more recommendations with respect to one or more decision variables determined using an optimization algorithm.
[005] In yet another embodiment, a system for generating data for real time optimization of grinding circuit is provided. The system includes one or more hardware processors, a communication interface, and a memory storing a plurality of instructions. The plurality of instructions when executed, cause the one or more hardware processors to receive input data comprising a plurality of real-time data and non real-time data from the grinding circuit. The system then pre-processes the input data to generate a pre-processed data, wherein the pre-processed data satisfies a plurality of conditions. Further, a simulated data is generated in real time using an integrated model comprising a holdup-model, a particle-size model, and a power draw model, wherein generating the simulated data involves the following steps. Initially, a hold-up data is generated by processing information on a) flow rate (Q), b) ball filling (J), and c) solid concentration in slurry (c) used in the grinding circuit, using the hold-up model, wherein the hold-up data comprising a predicted hold-up, a slurry filling (U), and mean residence time. Further, a particle size distribution in a ball mill is determined using the particle-size distribution model, comprising a plurality of selection and breakage functions selected based on material properties and slurry filling from the holdup data. Further, power drawn by the ball mill and an auxiliary equipment in the grinding circuit is predicted using the power draw model, wherein the power drawn by the ball mill and the auxiliary equipment is predicted based on a slurry filling parameter from the hold-up data and a slurry flow rate, wherein the hold-up data, the predicted particle size distribution data, and the predicted power draw data, form the simulated data. After generating the simulated data, an integrated data is generated by combining the pre-processed data with the simulated data, wherein the integrated data and the integrated model are used for real time grinding circuit optimization.
[006] In yet another embodiment, the simulated data and the integrated model are used by the system for the real-time optimization of the grinding circuit. The real-time optimization involves the following steps. Initially, an input data

comprising the plurality of real-time data and the non real-time data is received from the grinding circuit. Further, the grinding circuit is optimized by processing the received input data with the integrated data and integrated model, which further involves the steps of determining one or more decision variables to be optimized in the grinding circuit, solving an optimization problem using the received input data, wherein the optimization problem is modeled to minimize the power drawn by the grinding circuit and maximize throughput of the grinding circuit, and generating one or more recommendations with respect to one or more decision variables determined using an optimization algorithm.
[007] In yet another aspect, a non-transitory computer readable medium for real time optimization of a grinding circuit is provided. The non-transitory computer readable medium includes a plurality of instructions, which when executed, cause one or more hardware processors to perform the real time optimization of the grinding circuit by executing the following steps. Initially, an input data comprising a plurality of real-time data and non real-time data is received from the grinding circuit. Further, the input data is pre-processed to generate a pre-processed data, wherein the pre-processed data satisfies a plurality of conditions. Further, a simulated data is generated in real time using an integrated model comprising a holdup-model, a particle-size model, and a power draw model. Generating the simulated data involves the following steps. Initially, a hold-up data is generated by processing information on a) flow rate (Q), b) ball filling (J), and c) solid concentration in slurry (c) used in the grinding circuit, using the hold-up model, wherein the hold-up data comprising a predicted hold-up, a slurry filling (U), and mean residence time. Further, a particle size distribution in a ball mill is determined using the particle-size distribution model, comprising a plurality of selection and breakage functions selected based on material properties and slurry filling from the holdup data. Further, power drawn by the ball mill and an auxiliary equipment in the grinding circuit is predicted using the power draw model, wherein the power drawn by the ball mill and the auxiliary equipment is predicted based on a slurry filling parameter from the hold-up data and a slurry flow rate, wherein the hold-up data, the predicted particle size distribution data,

and the predicted power draw data, form the simulated data. After generating the simulated data, an integrated data is generated by combining the pre-processed data with the simulated data, wherein the integrated data and the integrated model are used for real time grinding circuit optimization.
[008] In yet another aspect, the plurality of instructions in the non-transitory computer readable medium when executed causes the real-time optimization of the grinding circuit using the integrated data and the integrated model using the following steps. The real-time optimization involves the following steps. Initially, an input data comprising the plurality of real-time data and the non real-time data is received from the grinding circuit. Further, the grinding circuit is optimized by processing the received input data with the integrated data and integrated model, which further involves the steps of determining one or more decision variables to be optimized in the grinding circuit, solving an optimization problem using the received input data, wherein the optimization problem is modeled to minimize the power drawn by the grinding circuit and maximize throughput of the grinding circuit, and generating one or more recommendations with respect to one or more decision variables determined using an optimization algorithm.
[009] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
BRIEF DESCRIPTION OF THE DRAWINGS [010] The accompanying drawings, which are incorporated in and
constitute a part of this disclosure, illustrate exemplary embodiments and, together
with the description, serve to explain the disclosed principles:
[011] FIG. 1 illustrates an exemplary architecture of a grinding circuit,
according to some embodiments of the present disclosure.
[012] FIG. 2 is an exemplary block diagram of a system for real time
optimization of the grinding circuit of FIG. 1, according to some embodiments of
the present disclosure.

[013] FIG. 3 illustrates an example implementation of the system of FIG. 2, in accordance with some embodiments of the present disclosure.
[014] FIG. 4 depicts an integrated data model used by the system of FIG. 2, for the real time optimization of the grinding circuit, according to some embodiments of the present disclosure.
[015] FIG. 5 depicts components of a self-learning module of the system of FIG. 3, for the real-time optimization of the grinding circuit, according to some embodiments of the present disclosure.
[016] FIGS. 6A and 6B (collectively referred to as FIG. 6) is a flow diagram depicting steps involved in the process of optimizing the grinding circuit, using the system of FIG. 2, in accordance with some embodiments of the present disclosure.
[017] FIG. 7 is a flow diagram depicting steps involved in the process of real-time optimization of the grinding circuit using an integrated data and an integrated data model, using the system of FIG. 2, in accordance with some embodiments of the present disclosure.
DETAILED DESCRIPTION OF EMBODIMENTS [018] Exemplary embodiments are described with reference to the accompanying drawings. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. Wherever convenient, the same reference numbers are used throughout the drawings to refer to the same or like parts. While examples and features of disclosed principles are described herein, modifications, adaptations, and other implementations are possible without departing from the scope of the disclosed embodiments.
[019] Mineral processing is an energy intensive industry where optimization of key unit operations and processes is crucial for achieving profitability. Grinding unit is the major contributor towards energy demands in the ore beneficiation process and minimization of power consumption while maximizing the throughput at desired particle size is a challenging task. Purely

first principle-based models become too computationally complex and slow which presents hurdles in real-time optimization. Purely data-based models may not capture the complex physics behind the process.
[020] The method and system disclosed herein provide an integrated model and an integrated data based approach for real time optimization of the grinding circuit. In this approach, the system receives input data comprising a plurality of real-time data and non real-time data from the grinding circuit. The system then pre-processes the input data to generate a pre-processed data, wherein the pre-processed data satisfies a plurality of conditions. Further, a simulated data is generated in real time using an integrated model comprising a holdup-model, a particle-size model, and a power draw model, wherein generating the simulated data involves the following steps. Initially, a hold-up data is generated by processing information on a) flow rate (Q), b) ball filling (J), and c) solid concentration in slurry (c) used in the grinding circuit, using the hold-up model, wherein the hold-up data comprising a predicted hold-up, a slurry filling (U), and mean residence time. Further, a particle size distribution in a ball mill is determined using the particle-size distribution model, comprising a plurality of selection and breakage functions selected based on material properties and slurry filling from the holdup data. Further, power drawn by the ball mill and an auxiliary equipment in the grinding circuit is predicted using the power draw model, wherein the power drawn by the ball mill and the auxiliary equipment is predicted based on a slurry filling parameter from the hold-up data and a slurry flow rate, wherein the hold-up data, the predicted particle size distribution data, and the predicted power draw data, form the simulated data. After generating the simulated data, an integrated data is generated by combining the pre-processed data with the simulated data, wherein the integrated data and the integrated model are used for real time grinding circuit optimization.
[021] In yet another embodiment, the simulated data and the integrated model are used by the system for the real-time optimization of the grinding circuit. The optimization involves the following steps. Initially, an input data comprising the plurality of real-time data and the non real-time data is received from the

grinding circuit. Further, the grinding circuit is optimized by processing the received input data with the integrated data and integrated model, which further involves the steps of determining one or more decision variables to be optimized in the grinding circuit, solving an optimization problem using the received input data, wherein the optimization problem is modeled to minimize the power drawn by the grinding circuit and maximize throughput of the grinding circuit, and generating one or more recommendations with respect to one or more decision variables determined using an optimization algorithm.
[022] Referring now to the drawings, and more particularly to FIG. 1 through FIG. 7, where similar reference characters denote corresponding features consistently throughout the figures, there are shown preferred embodiments and these embodiments are described in the context of the following exemplary system and/or method.
[023] FIG. 1 illustrates an exemplary architecture of a grinding circuit, according to some embodiments of the present disclosure. In this exemplary architecture, a closed loop circuit is assumed, and that a Fresh Feed (FF) input is available from a coarser grinding unit in the upstream, which is commonly a SAG mill circuit in industrial operation. In typical operation of the grinding circuit, slurry from the sump is subjected to classification at hydrocyclone and the underflow (oversize) is fed to the ball mill. After grinding, the slurry from the ball mill is sent back to the sump where it mixes with fresh feed. Secondary water is added to the sump to facilitate removal of fines at the hydrocyclone as the overflow stream. A centrifugal pump is used to transport the slurry from the sump to the hydrocyclone and provide the necessary pressure head required for the classification. In general, slurry is kept suspended inside the sump using an impeller and hence the sump acts as a single perfect mixer. It is assumed that size changes do not occur in the sump. The pump is fitted with a variable speed motor to manipulate the cyclone feed flow rate. The throughput of the grinding circuit and the total power consumed in the process are inversely proportional to each other.

[024] For the purpose of interoperability, throughput of the grinding circuit is the mass flow rate of fines from the hydrocyclone overflow and the total power consumption is calculated as:
Total power consumption = (Ball Mill power consumption + Pump power consumption)
[025] The competing variables, which are objective functions of the optimization, are throughput of the fines and total power consumption. Constraints are placed on slurry concentration, ball filling, fresh feed flowrate, ball mill rotational speed as fraction of critical speed, and ball mill filling. For the formulation of a multi-objective optimization problem on the grinding circuit, some special considerations were made. Since the models for ball mill KPIs are steady-state models, the flowrates and PSD in various streams of the grinding circuit assume only steady state values. This leads to a simple input and output relationship at the sump and hence the sump is treated as a simple mixer. As the exact route of achieving a steady state is not known, the sump height is an indeterminate and it is assumed that the variation in sump height when transitioning from one steady state flowrate to another is within feasible limits. The solid fraction in the hydrocyclone overflow mass flow rate is considered as throughput for this study. Based on the design parameter of hydrocyclone, bypass fraction and hydrocyclone D50 are calculated for each steady state. These two parameters are used to calculate the theoretical efficiency and corrected theoretical efficiency of hydrocyclone for each ith class. In general, the theoretical efficiency depicts the fraction of feed material in underflow stream and efficiency curve can be made showing the variation in efficiency with particle size. This curve is displaced due to feed water present in underflow stream, it is avoided by using corrected efficiency based on the particles arriving in underflow. The overflow and underflow mass flow rate are calculated using sump output and corrected theoretical efficiency of the hydrocyclone.
[026] FIG. 2 is an exemplary block diagram of a system for real time optimization of the grinding circuit of FIG. 1, according to some embodiments of the present disclosure. The system 100 includes or is otherwise in communication

with hardware processors 102, at least one memory such as a memory 104, an I/O interface 112. The hardware processors 102, memory 104, and the Input /Output (I/O) interface 112 may be coupled by a system bus such as a system bus 108 or a similar mechanism. In an embodiment, the hardware processors 102 can be one or more hardware processors.
[027] The I/O interface 112 may include a variety of software and hardware interfaces, for example, a web interface, a graphical user interface, and the like. The I/O interface 112 may include a variety of software and hardware interfaces, for example, interfaces for peripheral device(s), such as a keyboard, a mouse, an external memory, a printer and the like. Further, the I/O interface 112 may enable the system 100 to communicate with other devices, such as web servers, and external databases.
[028] The I/O interface 112 can facilitate multiple communications within a wide variety of networks and protocol types, including wired networks, for example, local area network (LAN), cable, etc., and wireless networks, such as Wireless LAN (WLAN), cellular, or satellite. For the purpose, the I/O interface 112 may include one or more ports for connecting several computing systems with one another or to another server computer. The I/O interface 112 may include one or more ports for connecting several devices to one another or to another server.
[029] The one or more hardware processors 102 may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, node machines, logic circuitries, and/or any devices that manipulate signals based on operational instructions. Among other capabilities, the one or more hardware processors 102 is configured to fetch and execute computer-readable instructions stored in the memory 104.
[030] The memory 104 may include any computer-readable medium known in the art including, for example, volatile memory, such as static random-access memory (SRAM) and dynamic random-access memory (DRAM), and/or non-volatile memory, such as read only memory (ROM), erasable programmable ROM, flash memories, hard disks, optical disks, and magnetic tapes. In an embodiment, the memory 104 includes a plurality of modules 106.

[031] The plurality of modules 106 include programs or coded instructions that supplement applications or functions performed by the system 100 for executing different steps involved in the process of real time optimization of grinding circuit being done by the system of FIG. 2. The plurality of modules 106, amongst other things, can include routines, programs, objects, components, and data structures, which performs particular tasks or implement particular abstract data types. The plurality of modules 106 may also be used as, signal processor(s), node machine(s), logic circuitries, and/or any other device or component that manipulates signals based on operational instructions. Further, the plurality of modules 106 can be used by hardware, by computer-readable instructions executed by the one or more hardware processors 102, or by a combination thereof. The plurality of modules 106 can include various sub-modules (not shown). The plurality of modules 106 may include computer-readable instructions that supplement applications or functions performed by the system 100 for the grinding circuit optimization.
[032] The data repository (or repository) 110 may include a plurality of abstracted piece of code for refinement and data that is processed, received, or generated as a result of the execution of the plurality of modules in the module(s) 106.
[033] Although the data repository 110 is shown internal to the system 100, it will be noted that, in alternate embodiments, the data repository 110 can also be implemented external to the system 100, where the data repository 110 may be stored within a database (repository 110) communicatively coupled to the system 100. The data contained within such external database may be periodically updated. For example, new data may be added into the database (not shown in FIG. 1) and/or existing data may be modified and/or non-useful data may be deleted from the database. In one example, the data may be stored in an external system, such as a Lightweight Directory Access Protocol (LDAP) directory and a Relational Database Management System (RDBMS). Functions of the components of the system 100 are now explained with reference to the

components of the block diagrams given in FIG. 3 through FIG. 5 and steps in flow diagrams in FIG. 6 and FIG. 7.
[034] FIG. 3 illustrates an example implementation of the system of FIG. 2, in accordance with some embodiments of the present disclosure. The system in FIG. 2 includes a receiving module, a data pre-processing module, a soft sensor and prediction module, an offline simulation module, a self-learning module, an optimization module, and a self-optimization module, wherein these modules are implemented by the one or more hardware processors 102, and are configured by the plurality of instructions to execute the grinding circuit optimization, which is explained in method 600 depicted in FIG. 6.
[035] At step 602 of the method 600, the system 100 receives, via the receiving module, an input data comprising a plurality of real-time data and non real-time data, with respect to operation of the grinding circuit being monitored for optimization. In an embodiment, the system 100 directly collects data from the grinding circuit by deploying appropriate sensors. In another embodiment, data from the grinding circuit maybe stored in a server external to the grinding circuit and the system 100, and the system 100 may fetch/collect the data from the server. In an embodiment, certain parameters may not be directly measured by the system 100 due to various implementation and/or operational challenges. Such data may be generated using appropriate physics based data models (i.e. non real-time data), and may be fed as input to the system 100.
[036] Further, at step 604, the system 100 pre-processes the input data to generate a pre-processed data, wherein the pre-processed data satisfies a plurality of conditions. The plurality of conditions may be with respect to format, presence of outliers, and so on. For example, the data collected may be in different formats. So to process the data further, the data may be in a particular defined format. This conversion maybe done during the pre-processing. Similarly, the collected data may contain many outliers, which, if not removed, may adversely affect quality of further data processing. So the outlier removal also is done in the pre-processing stage. Applying all the pre-defined pre-processing steps on the input data generates a pre-processed data.

[037] Further, at step 606 of the method 600, the system 100 generates, using the soft sensor and prediction module, a simulated data in real time using an integrated model comprising the hold-up model, the particle size model, and the power draw model. The hold-up model, the particle size model, and the power draw model are interconnected as depicted in FIG. 4. At step 606a, the system 100 generates a hold-up data by processing information on a) flow rate (Q), b) ball filling (J), and slurry concentration (c) used in the grinding circuit, using the hold¬up model, wherein the hold-up data comprising a predicted hold-up, a slurry filling (U), and a mean residence time. At least part of the hold-up data is fed as input to the particle size model, and the power draw model. Further, at step 606b, the system 100 determines a particle size distribution in a ball mill using a particle size distribution model, comprising a plurality of selection and breakage functions selected based on material properties and slurry filling from the hold-up data. Further, at step 606c, the system 100 predicts power drawn by the ball mill and an auxiliary equipment in the grinding circuit, based on U from the hold-up data and a slurry flow rate, using a power draw model. The hold-up data, the predicted particle size distribution data, and the power drawn data form the simulated data.
[038] Further, at step 608 of the method 600, the system 100 generates an integrated data by combining the pre-processed data with the simulated data, wherein the integrated data and the integrated model are used for real time grinding circuit optimization.
[039] FIG. 7 is a flow diagram depicting steps involved in the process of real-time optimization of the grinding circuit using an integrated data and an integrated data model, using the system of FIG. 2, in accordance with some embodiments of the present disclosure.
[040] When the system 100 is deployed to perform real-time monitoring and optimization of the grinding circuit, at step 702 of method 700 in FIG. 7, the system 100 receives input data comprising real-time data and non real-time data from the grinding circuit being monitored. The system 100 may be configured to optimize the grinding circuit such that a desired efficiency/result can be optimized from the grinding circuit. The desired results are defined in terms of values of a

plurality of key performance indicators/parameters (KPI) of the grinding circuit i.e. to achieve the intended result, the values of each of the KPIs are required to meet a threshold. Configuration of this intended result is termed as optimization problem configuration. Values of any/all of the KPIs falling below the corresponding threshold may be interpreted as a deviation by the system 100, and this in turn acts as a trigger for initiating optimization process.
[041] During the optimization of the grinding circuit, the optimization module is configured to optimize a plurality of key performance indicators/parameters (KPI) of the grinding circuit using the plurality of physics-based and data-driven models in the integrated model. The plurality of key performance parameters (KPI) of the grinding circuit comprises throughput, yield, efficiency, power draw, slurry concentration, slurry density, particle size distribution, D80, and cyclone separator efficiency. The optimization module further comprises an optimization configuration module, an optimization execution module, and a recommendation module (not shown, but may be part of the modules 106). The optimization configuration module is configured to enable configuring of optimization models/optimizer specific to the grinding circuit. The optimizer may be configured after a predefined time interval, when the key performance parameters of the grinding circuit cross the predefined thresholds, or by manual intervention.
[042] Configuration of the optimization problem involves choosing the type of optimization problem (single objective vs multi objective), direction of optimization (maximize or minimize), one or more objective functions, one or more constraints and their lower and upper limits, one or more manipulated variables and their lower and upper limits, and one or more groups of manipulated variables. Inputs for configuring the optimization model may be taken from the user via the user interface and various configured optimization models are stored in the model repository. The objective functions and constraint functions can be chosen from the plurality of key performance parameters of the grinding circuit. They can also be derived from or be a combination of the plurality of key

performance parameters of the plant. A sample optimization problem for the
grinding circuit is:
Minimize /(U) =min. (Total power), min. (—1 x throughput)
U=. (c, J,.,Q). ^T
s. t. (c _min, J_min,. _min,. FF. _min)<
(c,J,., FF)< . (c. _max,J_max,._max,. FF._max)
Parameter Symbol Minimum Maximum
Slurry concentration C 0.6 0.694
Ball filling J 0.3 0.345
Fresh feed (kg/min) FF 2.15 43.04
Ball Mill speed 0.6 0.8
Ball Mill filling VB 0.18 1.0
where U is the ball mill filling, and Q is the scaled flow rate through the
ball mill.
[043] In the preferred embodiment, the recommendation module is configured to provide at least one recommendation generated using the configured optimizer to the grinding circuit via the server and the user interface. The recommendations comprise of optimal settings of a plurality of manipulated variables. The generated recommendations are provided to optimize the key performance parameters of the grinding circuit. The plurality of manipulated variables of the grinding circuit comprises flow rates of input raw materials, ball filling, slurry concentration, and ball mill speed.
[044] The self-optimization module of the system 100 receives the optimized numerical values of recommended variables from the optimization module. The numerical values of the recommended variables are compared with a data of recommended variable (stored as expected values) available in a knowledge data base in the memory 104. If the recommended numerical values are deviating from the expected values beyond a predefined threshold value, the self-optimization module is triggered. The self-optimization module makes changes to the optimization models used in the optimization module by re-tuning at least one of 1. changing the objective function, 2. changing the values of the

constraints, 3. changing the parameters such as tolerance or convergence criteria of optimization algorithm, and 4. choosing a different optimization algorithm.
[045] The self-learning module of the system 100 of FIG. 3 is configured to monitor the performance of the plurality of physics-based and data-driven models of the grinding circuit and retune/retrain the models in case of a drift in their performance. For physics-based process models, tunable parameters such as ball mill holdup model coefficients, tuning parameters for breakage rates, tuning parameters for selection function, etc., are re-tuned in case of performance drift. For data-based models, either hyper-parameters of the models are re-tuned or models are re-trained in case of a performance drift. The re-tuned and re-trained models are stored in a model repository and may be activated for prediction of values of different parameters, in the prediction module. According to an embodiment of the disclosure, the self-learning module as in FIG. 5 comprises a model performance monitoring module, a drifted KPI selection module, a plant KPI re-training module, a particle size prediction re-tuning module, a mill holdup prediction model re-tuning module, a self-learning assessment module, and a model recommendation module. The self-learning module interacts with the grinding circuit data sources, knowledge database, plurality of databases, model repository, the server, the prediction module, and the user interface as and when required.
[046] The written description describes the subject matter herein to enable any person skilled in the art to make and use the embodiments. The scope of the subject matter embodiments is defined by the claims and may include other modifications that occur to those skilled in the art. Such other modifications are intended to be within the scope of the claims if they have similar elements that do not differ from the literal language of the claims or if they include equivalent elements with insubstantial differences from the literal language of the claims.
[047] The embodiments of present disclosure herein address unresolved problem of real time optimization of grinding circuit. The embodiment, thus provides a way of generating an integrated model by combining and synchronizing operation of data driven and physics based models to generate an

integrated data from data collected from the grinding circuit. Moreover, the embodiments herein further provides a mechanism of optimizing performance of the grinding circuit, based on a defined optimization problem.
[048] It is to be understood that the scope of the protection is extended to such a program and in addition to a computer-readable means having a message therein; such computer-readable storage means contain program-code means for implementation of one or more steps of the method, when the program runs on a server or mobile device or any suitable programmable device. The hardware device can be any kind of device which can be programmed including e.g., any kind of computer like a server or a personal computer, or the like, or any combination thereof. The device may also include means which could be e.g., hardware means like e.g., an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a combination of hardware and software means, e.g., an ASIC and an FPGA, or at least one microprocessor and at least one memory with software processing components located therein. Thus, the means can include both hardware means and software means. The method embodiments described herein could be implemented in hardware and software. The device may also include software means. Alternatively, the embodiments may be implemented on different hardware devices, e.g., using a plurality of CPUs.
[049] The embodiments herein can comprise hardware and software elements. The embodiments that are implemented in software include but are not limited to, firmware, resident software, microcode, etc. The functions performed by various components described herein may be implemented in other components or combinations of other components. For the purposes of this description, a computer-usable or computer readable medium can be any apparatus that can comprise, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
[050] The illustrated steps are set out to explain the exemplary embodiments shown, and it should be anticipated that ongoing technological development will change the manner in which particular functions are performed.

These examples are presented herein for purposes of illustration, and not limitation. Further, the boundaries of the functional building blocks have been arbitrarily defined herein for the convenience of the description. Alternative boundaries can be defined so long as the specified functions and relationships thereof are appropriately performed. Alternatives (including equivalents, extensions, variations, deviations, etc., of those described herein) will be apparent to persons skilled in the relevant art(s) based on the teachings contained herein. Such alternatives fall within the scope of the disclosed embodiments. Also, the words “comprising,” “having,” “containing,” and “including,” and other similar forms are intended to be equivalent in meaning and be open ended in that an item or items following any one of these words is not meant to be an exhaustive listing of such item or items, or meant to be limited to only the listed item or items. It must also be noted that as used herein and in the appended claims, the singular forms “a,” “an,” and “the” include plural references unless the context clearly dictates otherwise.
[051] Furthermore, one or more computer-readable storage media may be utilized in implementing embodiments consistent with the present disclosure. A computer-readable storage medium refers to any type of physical memory on which information or data readable by a processor may be stored. Thus, a computer-readable storage medium may store instructions for execution by one or more processors, including instructions for causing the processor(s) to perform steps or stages consistent with the embodiments described herein. The term “computer-readable medium” should be understood to include tangible items and exclude carrier waves and transient signals, i.e., be non-transitory. Examples include random access memory (RAM), read-only memory (ROM), volatile memory, nonvolatile memory, hard drives, CD ROMs, DVDs, flash drives, disks, and any other known physical storage media.
[052] It is intended that the disclosure and examples be considered as exemplary only, with a true scope of disclosed embodiments being indicated by the following claims.

WE CLAIM:
1. A processor implemented method (600) for real time optimization of a grinding circuit, comprising:
receiving (602) input data comprising a plurality of real-time data
and non real-time data from the grinding circuit;
pre-processing (604) the input data to generate a pre-processed
data, wherein the pre-processed data satisfies a plurality of
conditions;
generating (606) a simulated data in real time using an integrated
model comprising a holdup-model, a particle-size model, and a
power draw model, wherein generating the simulated data
comprising:
generating (606a) a hold-up data by processing information on a) flow rate (Q), b) ball filling (J), and c) solid concentration in slurry (c) used in the grinding circuit, using the hold-up model, wherein the hold-up data comprising a predicted hold-up, a slurry filling (U), and mean residence time;
determining (606b) a particle size distribution in a ball mill using the particle-size distribution model, comprising a plurality of selection and breakage functions selected based on material properties and slurry filling from the holdup data; and
predicting (606c) power drawn by the ball mill and an auxiliary equipment in the grinding circuit using the power draw model, wherein the power drawn by the ball mill and the auxiliary equipment is predicted based on a slurry filling parameter from the hold-up data and a slurry flow rate,

wherein the integrated model is utilized to generate the simulated data, comprising the hold-up data, the predicted particle size distribution data, and the predicted power draw data; and generating (608) an integrated data by combining the pre-processed data with the simulated data, wherein the integrated data and the integrated model are used for the real time optimization of the grinding circuit.
2. The method as claimed in claim 1, wherein the real-time data from the grinding circuit comprises temperature, pressure, slurry concentration, the ball filling, slurry density, and ball mill speed.
3. The method as claimed in claim 1, wherein the non real-time data from the grinding circuit comprises lab experimental data comprising material properties comprising hardness, chemical composition, and feed and product particle size distributions .
4. The method as claimed in claim 1, wherein pre-processing the input data comprises a) identification and removal of outliers, b) imputation of missing data, and c) synchronization and integration of a plurality of variables from one or more data sources to a consistent time scale for all the data of the grinding plant in which the grinding circuit is present.
5. The method as claimed in claim 1, wherein the real time optimization of the grinding circuit using the integrated data and the integrated model comprising:
receiving (702) input data comprising the plurality of real-time data and the non real-time data from the grinding circuit; and

optimizing (704) the grinding circuit by processing the received input data with the integrated data and the integrated model, comprising:
determining (704a) one or more decision variables to be optimized in the grinding circuit, solving an optimization problem using the received input data, wherein the optimization problem is modeled to minimize the power drawn by the grinding circuit and maximize throughput of the grinding circuit; and
generating (704b) one or more recommendations with respect to one or more decision variables determined using an optimization algorithm, for the real time optimization of the grinding circuit.
6. A system (100) for real time grinding circuit optimization, comprising:
one or more hardware processors (102);
a communication interface (112); and
a memory (104) storing a plurality of instructions, wherein the
plurality of instructions when executed, cause the one or more
hardware processors to:
receive input data comprising a plurality of real-time data and non real-time data from the grinding circuit; pre-process the input data to generate a pre-processed data, wherein the pre-processed data satisfies a plurality of conditions;
generate a simulated data in real time using an integrated model comprising a holdup-model, a particle-size model, and a power draw model, wherein generating the simulated data comprising:
generating a hold-up data by processing information on a) flow rate (Q), b) ball filling (J),

and c) solid concentration in slurry (c) used in the grinding circuit, using the hold-up model, wherein the hold-up data comprising a predicted hold-up, a slurry filling (U), and mean residence time; determining a particle size distribution in a ball mill using the particle-size distribution model, comprising a plurality of selection and breakage functions selected based on material properties and slurry filling from the holdup data; and predicting power drawn by the ball mill and an auxiliary equipment in the grinding circuit using the power draw model, wherein the power drawn by the ball mill and the auxiliary equipment is predicted based on a slurry filling parameter from the hold-up data and a slurry flow rate,
wherein the integrated model is utilized to generate
the simulated data, comprising the hold-up data, the
predicted particle size distribution data, and the
predicted power draw data; and
generate an integrated data by combining the pre-processed
data with the simulated data, wherein the integrated data
and the integrated model are used for the real time
optimization of the grinding circuit.
7. The system as claimed in claim 6, wherein the real-time data from the grinding circuit comprises temperature, pressure, slurry concentration, the ball filling, slurry density, and ball mill speed.
8. The system as claimed in claim 6, wherein the non real-time data from the grinding circuit comprises lab experimental data comprising material

properties comprising hardness, chemical composition, and feed and product particle size distributions.
9. The system as claimed in claim 6, wherein the one or more hardware processors are configured to pre-process the input data by a) identification and removal of outliers, b) imputation of missing data, and c) synchronization and integration of a plurality of variables from one or more data sources to a consistent time scale for all the data of the grinding plant in which the grinding circuit is present.
10. The system as claimed in claim 6, wherein the one or more hardware processors are configured to perform the real time optimization of the grinding circuit using the integrated data and the integrated model, by:
receiving input data comprising the plurality of real-time data and
the non real-time data from the grinding circuit; and
optimizing the grinding circuit by processing the received input
data with the integrated data and the integrated model, comprising:
determining one or more decision variables to be optimized
in the grinding circuit, solving an optimization problem
using the received input data, wherein the optimization
problem is modeled to minimize the power drawn by the
grinding circuit and maximize throughput of the grinding
circuit; and
generating one or more recommendations with respect to one or more decision variables determined using an optimization algorithm, for the real time optimization of the grinding circuit.