METHOD AND SYSTEM FOR ONLINE MONITORING AND OPTIMIZATION OF MINING AND MINERAL PROCESSING OPERATIONS

• US 20200132882A1
• Filed: 10/24/2019
• Published: 04/30/2020
• Est. Priority Date: 10/26/2018
• Status: Active Grant

• Current Assignee: TATA Consultancy Services Limited (Tata Sons Pvt Ltd.)
• Original Assignee: TATA Consultancy Services Limited (Tata Sons Pvt Ltd.)
• Inventors: RUNKANA, Venkataramana, NADIMPALLI, Nagaravi Kumar Varma, PAREEK, Aditya, MASAMPALLY, Vishnu Swaroopji
• Granted Patent: US 11,307,327 B2
• Time in Patent Office: Days
• CPC Class Codes: G01V 20/00 Geomodelling in general
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/01 Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G06Q 10/06393 Score-carding, benchmarking or key performance indicator [KPI] analysis
• CPC Class Codes: G06Q 50/02 Agriculture; Fishing; Forestry; Mining

Abstract

Monitoring and analysis of plurality of operations in mining and mineral processing is critical to achieve optimized performance. Existing tools are specific to one or other individual operations and this individuality introduces limitations for end to end monitoring of entire mining to mineral processing operations. Method and system for online monitoring and optimization of mining and mineral processing operations providing an integrated approach utilizing short-term mining plan data, information generated using established drill and blast design software, simulation models of fragmentation, crushing, screening, grinding and flotation to arrive at an optimized charge plan and set points for controllers is disclosed. The proposed method and system improves key performance indicators such as cost of mining operations, specific energy consumption in comminution circuit, maximizes yield of desired particle size, and maximizes grade and recovery of mineral of interest while considering operational constraints.

Claims

1. A processor implemented method for online monitoring and optimization of mining and mineral processing operations, the processor implemented method comprising:

fetching from a plurality of data sources, by one or more hardware processors, data for a set of variables corresponding to a set of sensors associated with the mining and mineral processing operations;

pre-processing, by the one or more hardware processors, the fetched data corresponding to the set of variables by discarding outliers, performing imputation for adding artificial values at missing positions, organizing data collected at different frequencies to one common frequency, identifying and selecting data based on steady state operation of a process or a sub-process associated with the mining and mineral processing operations;

determining, by the one or more hardware processors, standard operating condition for the mining and mineral processing operations using the pre-processed data corresponding to the set of variables;

segregating each variable among the set of variables into one of drilling-blasting operations, hauling operations, comminution operations, and flotation and concentration operations, wherein segregation is performed in accordance with a master tag list;

generating, by the one or more hardware processors, a set of models based on the segregated set of variables and preprocessed data, wherein the set of models is generated for a set of Key Performance Indicators (KPIs) of interest and a plurality of process parameters of interest associated with the mining and mineral processing operations, wherein the generated set of models comprise a set of Machine learning (ML) models or a set of individualized physics based models or hybrid models, and wherein the set of individualized physics based or hybrid models are generated for the plurality of process parameters based on known physics based models by determining model parameters using non-linear curve fitting;

simulating, by the one or more hardware processors, current operating condition of the mining and mineral processing operations corresponding to the set of variables based on at least one of;

the generated ML models, the set of individualized physics based or hybrid models of the KPIs and the plurality of process parameters, wherein the simulated current operating condition provides current values for KPIs and the plurality of process parameters of interest; and

hierarchically optimizing, by the one or more hardware processors, the KPIs of interest of the mining and mineral processing operations to update the current operating conditions of a subset of variables among the set of variables of the mining and mineral processing operations, wherein the KPIs of interest are from at least one of mining operations, a comminution circuit, and a flotation and concentration circuit.

2. The processor implemented method of claim 1, further comprising monitoring current operating condition of the set of variables in accordance with the pre-processed data of the set of variables corresponding to the mining and mineral processing operations to detect failure of one or more sensors from the set of sensors or failure of a process equipment, wherein the failure and a reason for the failure is indicated to a system administrator.

3. The method of claim 1, wherein the optimization of the KPIs of interest for the at least one of the mining operations, the comminution circuit, and the flotation and concentration circuit comprises:

formulating and solving optimization problem which comprises customizable combination of the KPIs of interest, modeled using the ML models the set of individualized physics based or the set of hybrid models, wherein solution of the optimization problem provides a set of values of the subset of variables, which are further communicated to a plurality of controllers and operators to improve the mining and mineral processing operations.

4. The method of claim 3, wherein the solution of the optimization problem provides:

recommendation of the charge plan including type and amount of explosives to be used; and

set points for the plurality of controllers located at different positions along the comminution circuit, and flotation and concentration circuit.

5. The method of claim 4, wherein optimizing the KPIs of interest of the mining and mineral processing operations is in accordance with a phase-wise and hierarchical procedure comprising:

creating drilling pattern and a preliminary charge pattern based on short term mining plan;

performing production drilling operations to collect Monitoring While Drilling (MWD) data, wherein the MWD data comprise information of drill rod revolution rate, rotary torque, pull down force, drilling rate, angular velocity and penetration rate;

using the MWD data and drilling pattern to estimate the geotechnical parameters comprising specific fracture energy and blasting parameters, further comprising, air deck length, stemming height for each hole to be blasted;

designing a preliminary charge plan, comprising information of charge mix for each of the hole, based on the estimated geotechnical parameters and drilling pattern;

estimating fragment size distribution (FSD) using the generated models, wherein the generated models utilize the estimated blasting parameters and the designed charge plan, wherein the estimated FSD along with relevant process parameters are used to simulate the expected KPIs of mining operations comprising hauling costs, stockpile throughput, flyrock size, velocity and range and cost of explosives;

utilizing the estimated FSD and the process parameters relevant for the mining and mineral processing operations to estimate KPIs of mineral processing operations comprising specific energy consumption, screening efficiency, particle size distribution at important points during comminution circuit along with recovery and grade of the mineral of interest in flotation and concentration circuit, using ML models, physics based and hybrid models; and

recursively performing either one or all of the above steps to obtain optimized charge plan and set points of process parameters of mining and mineral processing operations that globally minimize a customized objective function expressed in terms of KPIs of interest.

6. The method of claim 1, wherein the set of sensors are associated with:

equipment and instrumentation installed for drilling, blasting or hauling operation required for mining operations or equipment and instrumentation installed in comminution circuit and, flotation and concentration circuit of mineral processing operations;

quality parameters of a plurality of streams of the mining and mineral processing operations, comprising raw material, product and intermediate streams and data corresponding to daily raw material processed;

source and quantity and data related to characterization of the raw materials of different types being processed; and

process records containing data corresponding to equipment breakdown and maintenance history, environmental and weather conditions, and rock and terrain properties.

7. The method of claim 1, wherein the plurality of KPIs of interest for which the set of models are generated comprise cost of blasting operations, fragment size distribution, flyrock formation including size, range and maximum velocity, hauling cost, throughput to stockpile, screening efficiency, specific energy consumption, particle size distribution, comminution circuit throughput, grade and recovery of mineral of interest in output streams of the flotation and concentration circuit.

8. The method of claim 1, wherein the plurality of process parameters of interest for which the set of models is generated are used as soft sensors, and wherein the soft sensors provide real time or near real time prediction of infrequently measured or unmeasured critical process parameter or one or more parameters specific to each mining and mineral processing operation.

9. A system for online monitoring and optimization of mining and mineral processing operations, the system comprising:

a memory storing instructions;

one or more communication interfaces; and

one or more hardware processors coupled to the memory via the one or more communication interfaces, wherein the one or more hardware processors are configured by the instructions to;

fetch from a plurality of data sources data for a set of variables corresponding to a set of sensors associated with the mining and mineral processing operations;

pre-process the fetched data corresponding to the set of variables by discarding outliers, performing imputation for adding artificial values at missing positions, organizing data collected at different frequencies to one common frequency, identifying and selecting data based on steady state operation of a process or a sub-process associated with the mining and mineral processing operations;

determine standard operating condition for the mining and mineral processing operations using the pre-processed data corresponding to the set of variables;

segregate each variable among the set of variables into one of a drilling-blasting operations, hauling operations, comminution operations, and flotation- and concentration operations, wherein segregation is performed in accordance with a master tag list;

generate a set of models based on the segregated set of variables and preprocessed data, wherein the set of models is generated for a set of Key Performance Indicators (KPIs) of interest and a plurality of process parameters of interest associated with the mining and mineral processing operations, wherein the generated set of models comprise a set of Machine learning (ML) models, or a set of individualized physics based models or hybrid models, and wherein the set of individualized physics based models or hybrid models are generated for the plurality of process parameters based on known physics based models by determining model parameters using non-linear curve fitting;

simulate current operating condition of the mining and mineral processing operations corresponding to the set of variables based on at least one of;

the generated ML models, the individualized physics based, the set of hybrid models of the KPIs and the plurality of process parameters, wherein the simulated current operating condition provides current values for KPIs and the plurality of process parameters of interest; and

hierarchically optimize the KPIs of interest of the mining and mineral processing operations to update the current operating conditions of a subset of variables among the set of variables of the mining and mineral processing operations, wherein the KPIs of interest are from at least one of mining operations, a comminution circuit, and a flotation and concentration circuit.

10. The system of claim 9, wherein the one or more hardware processors are further configured to monitor current operating condition of the set of variables in accordance with the pre-processed data of the set of variables corresponding to the mining and mineral processing operations to detect failure of one or more sensor from the set of sensors or failure of a process equipment, wherein the failure and a reason for the failure is indicated to a system administrator.

11. The system of claim 9, wherein the one or more hardware processors are further configured to perform the optimization of the KPIs of interest for the at least one of the mining operations, the comminution circuit, and the flotation and concentration circuit by:

formulating and solving optimization problem which comprises customizable combination of the KPIs of interest, modeled using ML models, the set of individualized physics based models or hybrid models, wherein solution of the optimization problem provides a set of values of the subset of variables, which are further communicated to a plurality of controllers and operators to improve the mining and mineral processing operations.

12. The system of claim 11, wherein the solution of the optimization problem provides:

recommendation for charge plan including amount and type of explosives to be used; and

set points for the plurality of controllers located at different positions along the comminution circuit, and flotation and concentration circuit.

13. The system of claim 12, wherein the one or more hardware processors are further configured to improve the KPIs of interest for the optimization of the mining and mineral processing operations in a phase-wise and hierarchical procedure comprising:

creating drilling pattern and a preliminary charge pattern based on short term mining plan;

performing production drilling operations to collect Monitoring While Drilling (MWD) data, wherein the MWD data comprise information of drill rod revolution rate, rotary torque, pull down force, drilling rate, angular velocity and penetration rate;

using the MWD data and drilling pattern to estimate the geotechnical parameters comprising specific fracture energy and blasting parameters, further comprising, air deck length, stemming height for each hole to be blasted;

designing a preliminary charge plan, comprising information of charge mix for each of the hole, based on the estimated geotechnical parameters and drilling pattern;

estimating fragment size distribution (FSD) using the generated models, wherein the generated models utilize the estimated blasting parameters and the designed charge plan, wherein the estimated FSD along with relevant process parameters are used to simulate the expected KPIs of mining operations comprising hauling costs, stockpile throughput, flyrock size, velocity and range and cost of explosives;

utilizing the estimated FSD and the process parameters relevant for the mining and mineral processing operations to estimate KPIs of mineral processing operations comprising specific energy consumption, screening efficiency, particle size distribution at important points during comminution circuit along with recovery and grade of the mineral of interest in flotation and concentration circuit, using ML models, physics based and hybrid models; and

recursively performing either one or all of the above steps to obtain optimized charge plan and set points of process parameters of mining and mineral processing operations that globally minimize a customized objective function expressed in terms of KPIs of interest.

14. The system of claim 9, wherein the set of sensors are associated with:

equipment and instrumentation installed for drilling, blasting or hauling operation required for mining operations or equipment and instrumentation installed in the comminution circuit and, the flotation and concentration circuit of mineral processing operations;

quality parameters of a plurality of streams of the mining and mineral processing operations, comprising raw material, final and intermediate products;

source and quantity of raw materials processed and data related to characterization of the raw materials; and

process records containing data corresponding to equipment breakdown and maintenance history, environmental and weather conditions, and rock and terrain properties.

15. The system of claim 9, wherein the plurality of KPIs of interest for which the set of models are generated comprise cost of blasting operations, fragment size distribution, flyrock formation including size, range and maximum velocity, hauling cost, throughput to stockpile, screening efficiency, specific energy consumption, particle size distribution, comminution circuit throughput, grade and recovery of mineral of interest in output streams of the flotation and concentration circuit.

16. The system of claim 9, wherein the plurality of process parameters of interest for which the set of models is generated are used as soft sensors, and wherein the soft sensors provide real time or near real time prediction of infrequently measured or unmeasured critical process parameter or one or more parameters specific to each mining and mineral processing operation.

17. One or more non-transitory machine readable information storage media storing instructions which, when executed by one or more hardware processors, causes the one or more hardware processors to execute a method comprising:

fetching from a plurality of data sources data for a set of variables corresponding to a set of sensors associated with the mining and mineral processing operations;

pre-processing the fetched data corresponding to the set of variables by discarding outliers, performing imputation for adding artificial values at missing positions, organizing data collected at different frequencies to one common frequency, identifying and selecting data based on steady state operation of a process or a sub-process associated with the mining and mineral processing operations;

determining standard operating condition for the mining and mineral processing operations using the pre-processed data corresponding to the set of variables;

segregating each variable among the set of variables into one of drilling-blasting operations, hauling operations, comminution operations, and flotation and concentration operations, wherein segregation is performed in accordance with a master tag list;

generating a set of models based on the segregated set of variables and preprocessed data, wherein the set of models is generated for a set of Key Performance Indicators (KPIs) of interest and a plurality of process parameters of interest associated with the mining and mineral processing operations, wherein the generated set of models comprise a set of Machine learning (ML) models or a set of individualized physics based models or hybrid models, wherein the set of individualized physics based or hybrid models are generated for the plurality of process parameters based on known physics based models by determining model parameters using non-linear curve fitting;

simulating current operating condition of the mining and mineral processing operations corresponding to the set of variables based on at least one of;

the generated ML models, the set of individualized physics based or hybrid models of the KPIs and the plurality of process parameters, wherein the simulated current operating condition provides current values for KPIs and the plurality of process parameters of interest; and

hierarchically optimizing the KPIs of interest of the mining and mineral processing operations to update the current operating conditions of a subset of variables among the set of variables of the mining and mineral processing operations, wherein the KPIs of interest are from at least one of mining operations, a comminution circuit, and a flotation and concentration circuit.