System and method for activating deep raffinate injection based on column test predictive model
The method may comprise receiving historical data (e.g., mineralogy data, irrigation data, raffinate data, heat data, lift height data, geographic data on ore placement and/or blower data); training a predictive model using the historical data to create a trained predictive model; adding future assumption data to the trained predictive model; running the forecast engine for a plurality of parameters to obtain forecast data for a mining production target; comparing the forecast data for the mining production target to the actual data for the mining production target; determining deviations between the forecast data and the actual data, based on the comparing; and changing each of the plurality of parameters from the forecast data to the actual data to determine a contribution to the deviations for each of the plurality of parameters.
1. A method comprising:
obtaining, by a processor, days under leach, chemistry data and mineralogy data of mineralogy from a column test of a column of ore from a section of a stockpile;
simulating, by the processor, leaching in the column of ore in the section of the stockpile from the days under leach, the chemistry data and the mineralogy data;
adjusting, by the processor, process parameters applied to the column of ore,
wherein the adjusting of the process parameters impacts the simulating of the leaching in the column of ore;
obtaining, by the processor, adjusted chemistry data and adjusted mineralogy data based on the adjusting of the process parameters and the simulating of the leaching in the column of ore using the adjusted process parameters;
providing, by the processor, the days under leach, the adjusted chemistry data and the adjusted mineralogy data to a machine learning model to build a column test predictive model;
determining, by the processor using the machine learning model, estimated remaining amount of mineral in the section of the stockpile by using the column test predictive model;
generating, by the processor using the machine learning model, a prediction of an amount of mineral recovery from the estimated remaining amount of mineral in the section of the stockpile;
generating, by the processor using the machine learning model, a mineral recovery curve to estimate an amount of mineral recovery from actual leaching over a period of time, by using the prediction of the amount of mineral recovery; and
activating, by the processor and based on the mineral recovery curve, deep injection of raffinate at a depth to target the estimated remaining amount of mineral in the section of the stockpile to implement the actual leaching over the period of time to achieve the estimated amount of mineral recovery,
wherein the activating causes a drilling machine to drill holes at the depth to target the estimated remaining amount of mineral in the section of the stockpile, and
wherein the activating causes a pumping machine to pump the raffinate under pressure to implement the actual leaching over the period of time to achieve the estimated amount of mineral recovery.
2. The method of claim 1 , wherein the activating the deep injection of the raffinate comprises sending a signal to the drilling machine to drill the holes at the depth to target the estimated remaining amount of mineral in the section of the stockpile.
3. The method of claim 1 , wherein the activating the deep injection of the raffinate comprises adjusting chemistry of the raffinate for the actual leaching to react with the mineralogy of the estimated remaining amount of mineral in the section of the stockpile, wherein the raffinate is used for the deep injection of the raffinate.
4. The method of claim 1 , wherein the activating the deep injection of the raffinate comprises enhancing the raffinate pumped into the estimated remaining amount of mineral in the section of the stockpile with at least one of sulfuric acid, ferric ions, air bubbles, oxygen bubbles, heat, or microbes.
5. The method of claim 1 , wherein the activating the deep injection of the raffinate comprises enhancing the raffinate pumped into the estimated remaining amount of mineral in the section of the stockpile with at least one of an acid, oxidant or leach enhancing additive.
6. The method of claim 1 , wherein the activating the deep injection of the raffinate comprises pumping the raffinate under the pressure into a raffinate well within the estimated remaining amount of mineral in the section of the stockpile.
7. The method of claim 1 , wherein the chemistry data is stored in a laboratory information management system (LIMS).
8. The method of claim 1 , wherein the process parameters include at least one of type of film, a period of time for implementing the leaching, total depth of the section of the stockpile, raffinate application flowrate, raffinate temperature, air application rate, air wet bulb temperature, evaporation percentage, shortwave infrared radiation absorptivity, longwave absorbed infrared radiation emissivity, longwave out radiation emissivity, exotherm heat generation, heat transfer by convection at a surface of the stockpile or average temperature at an end of the period of time for the leaching.
9. The method of claim 1 , wherein the machine learning model is a multi-layer perceptron (MLP).
10. The method of claim 1 , further comprising determining, by the processor using the machine learning model, a location of the estimated remaining amount of mineral in the section of the stockpile.
11. The method of claim 1 , wherein the chemistry data and the mineralogy data include at least one of an amount of Fe in the raffinate, Fe 2 , raffinate acid, raffinate additive, raffinate temperature, an amount of Cu in the raffinate, acid soluble copper, an indicator of a presence of leachable minerals in the ore, application rate or cure acid.
12. The method of claim 1 , wherein the machine learning model incorporates at least one of a value of an agglomeration acid concentration or a value of an agglomeration additive concentration applied in the column test.
13. The method of claim 1 , further comprising adjusting, by the processor, agglomeration acid concentration applied in the column test.
14. The method of claim 1 , further comprising adjusting, by the processor, a plan for mining the section of the stockpile to capture the estimated remaining amount of mineral.
15. The method of claim 1 , further comprising transmitting, by the processor, the estimated remaining amount of mineral in the section of the stockpile to an ore map.
16. The method of claim 1 , further comprising inputting, by the processor, laboratory records of column test performance into the column test predictive model, for training the column test predictive model.
17. The method of claim 1 , further comprising transmitting, by the processor, the estimated remaining amount of mineral in the section of the stockpile to a tool for tracking mine material.
18. The method of claim 1 , further comprising transmitting, by the processor, the estimated remaining amount of mineral in the section of the stockpile to a tool for tracking mine material, wherein the tool for tracking mine material provides data to the stockpile and section mapping tool, and wherein the stockpile and the section mapping tool provides provide data to at least one of an ore map or a predictive model.