CN-121998368-A - Weather disturbance coping strip mine mining and selecting collaborative plan dynamic optimization method and system

CN121998368ACN 121998368 ACN121998368 ACN 121998368ACN-121998368-A

Abstract

The invention provides a method and a system for dynamically optimizing a mining and selecting collaborative plan of a strip mine for coping with weather disturbance, and relates to the technical field of optimization of mining and selecting collaborative production plans of the strip mine. The method specifically comprises the steps of obtaining weather forecast data of all planning periods in a production period to be planned, respectively extracting time characteristics and weather characteristics of each planning period, constructing a double-layer planning model based on data driving prediction and optimization decision coupling, and generating an optimal production plan of the production period by utilizing the double-layer planning model based on data driving prediction and optimization decision coupling based on the time characteristics and weather characteristics of all planning periods in the production period. The method can dynamically respond to weather changes, so that capacity fluctuation caused by weather uncertainty on strip mine production is dealt with, and ore supply stability and production economic benefit are ensured.

Inventors

HU QIAN
WANG QING
HUANG MIN
SONG YANG
WANG XINGWEI
LI XINFENG

Assignees

东北大学
长春黄金设计院有限公司

Dates

Publication Date: 20260508
Application Date: 20260130

Claims (10)

1. The method for dynamically optimizing the mining and selecting collaborative plan of the strip mine for coping with weather disturbance is characterized by comprising the following steps: for any production period to be planned, weather forecast data of all planning periods in the production period are obtained, and time characteristics and weather characteristics of each planning period are extracted respectively; constructing a double-layer planning model based on data-driven prediction and optimization decision coupling; based on the time characteristics and the meteorological characteristics of all planning periods in the production period, generating an optimal production plan of the production period by using a double-layer planning model based on the coupling of data-driven prediction and optimization decision.
2. The method for dynamically optimizing a mining cooperation plan of a strip mine for coping with weather disturbance according to claim 1, wherein for any production period to be planned, weather forecast data of all planning periods in the production period are obtained, and the specific contents of time features and weather features of each planning period are extracted respectively as follows: for any planning period in the production period, weather forecast data of the planning period is obtained; Extracting the time characteristics and the original meteorological characteristics of the planning period from the weather forecast data of the planning period; The time characteristics comprise month and season corresponding to the date; the original meteorological features comprise the highest temperature, the lowest temperature, weather category and wind power level; preprocessing the extracted original meteorological features to obtain preprocessed meteorological features; The extracted time characteristics and the preprocessed weather characteristics are used as the time characteristics and the weather characteristics of the planning period.
3. The method for dynamically optimizing a mining cooperation plan of a strip mine for coping with weather disturbances according to claim 2, wherein the preprocessing includes numerical encoding of weather categories and standardized processing of maximum temperature, minimum temperature and wind power level.
4. The method for dynamically optimizing a mining cooperation plan of a strip mine for coping with weather disturbance according to claim 3, wherein the double-layer planning model based on coupling of data-driven prediction and optimization decision comprises an upper-layer productivity prediction model and a lower-layer mining and stripping transportation planning model; the upper layer productivity prediction model is used for predicting production data in a production period to be planned according to the time characteristics and the meteorological characteristics of the production period; The lower layer mining and stripping transportation plan model is used for generating an optimal production plan of the production cycle by taking production data in the production cycle as constraint and taking a weighted sum of minimized mining and stripping transportation cost deviation and ore feeding quantity deviation as a target.
5. The method for dynamically optimizing a mining cooperation plan of a strip mine for coping with weather disturbance according to claim 4, wherein the method for constructing the upper-layer productivity prediction model is as follows: Collecting historical time features, historical meteorological features and historical production data corresponding to a plurality of historical planning periods, wherein the historical production data is actual daily rock production; Preprocessing the historical meteorological features corresponding to each historical planning period to obtain preprocessed historical meteorological features; taking the historical time characteristics of the same historical planning period and the preprocessed historical meteorological characteristics as a training sample, and taking the historical production data of the historical planning period as a prediction label to construct a training sample set; constructing a quantile regression forest model integrated by multiple decision trees based on a random forest frame, and training the quantile regression forest model by using a training sample set with the aim of minimizing quantile loss to obtain a trained quantile regression forest model; and taking the trained quantile regression forest model as an upper productivity prediction model.
6. The method for dynamically optimizing a mining cooperation plan of a strip mine for coping with weather disturbance according to claim 5, wherein the method is characterized in that based on a random forest frame, a multi-decision tree integrated quantile regression forest model is constructed, and training the quantile regression forest model by using a training sample set with the aim of minimizing quantile loss, and the specific content of the trained quantile regression forest model is as follows: initializing an empty quantile regression forest model; sampling the training sample set by adopting a Bootstrap sampling method to generate a plurality of training subsamples; independently training a decision tree in the quantile regression forest model by using each training subsamples; For any decision tree to be trained, determining the target score of the current decision tree, taking a training subsamples corresponding to the current decision tree as root node data, and recursively executing a node splitting process from a root node, wherein the specific content of each node splitting process is as follows: and judging whether the current node to be split meets a preset splitting stopping condition for any node to be split in the current decision tree, if so, taking the current node to be split as a leaf node to obtain a trained current decision tree, and if not, taking the minimum quantile loss as a node splitting criterion to obtain an optimal splitting point of the current node to be split, and generating a next stage splitting node according to the optimal splitting point.
7. The method for dynamically optimizing a mining cooperation plan of a strip mine in response to weather disturbances of claim 4, wherein the objective function of the underlying mining and stripping transportation plan model is to minimize a weighted sum of mining and stripping transportation cost deviation and ore supply deviation.
8. The method for dynamically optimizing a mining cooperation plan of a strip mine in response to weather disturbances according to claim 4, wherein constraints of the underlying mining-stripping transportation plan model include: stope equipment mining capacity constraints, planning period Stope Not exceeding the planned period Stope An upper maximum capacity limit of the plant; Mine capacity constraint and planning period The sum of rock production at all stopes within the planning period Within a viable throughput interval; Stope reserve constraint The sum of rock production in all planning phases in the production cycle to be planned does not exceed the stope An upper rock mass limit of (2); Stripping ratio constraint for any planning period Stope The total amount of waste rock transported to the dumping site is equal to that of the secondary stope Total amount of ore mined in and stope Is the product of the stripping ratio of (2); wherein the secondary stope The total amount of ore mined in the stope is Ore quantity and secondary stope of (2) The sum of the amounts of ore transported to inventory; Capacity constraint of dumping, namely all transportation to dumping sites Sum of the amount of waste stone not exceeding the dumping site An upper capacity limit of (2); Material balance constraint for any planning period Stope Is equal to the mining area Ore quantity, stope The amount of ore transported to stock and all secondary stopes The sum of the total waste rock amount transported to the dumping site; Ore dressing plant grade constraint for any metal At any planning stage In the total amount of ore from all stopes to the concentrating mill Is positioned in the concentration plant for metal The required metal content is in a feasible interval; inventory balance constraint, planning period Stope The stock quantity of (2) is equal to the last planning period Stope To the stock of (1) plus the planning period Stope The amount of ore transported to the inventory is subtracted from the projected period From stopes from ores in stock for transport to concentrating mills Ore amount of (2); maximum inventory constraint, planning period The sum of stock amounts of all stopes does not exceed stock for stopes An upper limit of ore storage; Inventory reservation constraints, planning period Stope Inventory for stope is not less than inventory Lower limit of ore storage amount; ore amount positive deviation constraint, planning period The positive deviation of the internal ore feeding amount and the demand amount of the concentrating mill is equal to the planning period The amount of ore that all stopes are carrying to inventory; negative deviation constraint of ore quantity, planned period The difference between the ore demand of the internal concentrating mill and the total supply of all the stopes is compared with zero and takes a larger value, and the obtained larger value is taken as the planning period Negative deviation of the internal ore supply amount from the demand amount of the concentrating mill; cost bias constraint, planning period The total cost of the plan in the system is equal to the plan period Total cost in plus planning period Plus deviation of total cost from planned cost, minus planned period Negative deviation of the total internal cost from the projected cost; ore direct transport constraint for any planning period All the slaves The amount of mine transported to the mill and the amount of ore transported from the inventory to the mill do not exceed the maximum daily ore throughput of the mill; The utilization rate of the dumping site is balanced and constrained, namely for any dumping site Dumping site Total waste rock amount in all planning period, at dumping site The waste stone quantity receiving interval allowed by the balanced utilization rate is within; Exit mutual exclusion constraint in any planning period Inside stope The amount of ore to stock and the amount of ore from the stock to the mill from the stope The ore quantity of (2) cannot be positive at the same time; Non-negative constraint, planning period Positive and negative deviation and planning period of internal ore feeding quantity and ore dressing plant demand quantity Plus or minus deviation of total cost and planning cost, planning period Stope Is the rock exploitation amount and planning period of (a) Stope Ore quantity, planning period of (2) Stope Ore quantity and planning period to stock From stopes from ores in stock for transport to concentrating mills The ore amounts of (a) are all non-negative values.
9. The method for dynamically optimizing a mining cooperation plan of a strip mine in response to weather disturbance according to claim 1, wherein the optimal production plan includes mining amounts of various stopes, transportation paths, inventory storage conditions and planning costs.
10. A weather disturbance-coping strip mine mining cooperation plan dynamic optimization system for implementing the weather disturbance-coping strip mine mining cooperation plan dynamic optimization method according to any one of claims 1 to 9, characterized in that the system comprises: The data acquisition module is used for acquiring weather forecast data of all planning periods in any production period; The feature extraction module is used for respectively extracting the time feature and the weather feature of each planning period according to the weather forecast data of all the planning periods; the model construction module is used for constructing a double-layer planning model based on data-driven prediction and optimization decision coupling; And the plan generation module is used for generating an optimal production plan of the production period by using a double-layer planning model based on the coupling of data-driven prediction and optimization decision based on the time characteristics and the meteorological characteristics of all the plan periods in the production period.

Description

Weather disturbance coping strip mine mining and selecting collaborative plan dynamic optimization method and system Technical Field The invention relates to the technical field of optimization of a mining and selecting collaborative production plan of a strip mine, in particular to a dynamic optimization method and a system of the mining and selecting collaborative plan of the strip mine for coping with weather disturbance. Background In the field of optimization of strip mine production scheduling, the prior art is mainly researched aiming at uncertainty factors such as geology, price, supply and the like. The typical technical scheme aiming at the geological uncertainty is a two-stage random integer programming method (Stochastic Integer Programming, SIP), and the scheme builds a random optimization model by quantifying uncertain parameters such as geological grade, tonnage and the like, and brings the geological uncertainty into a long-term production plan generation process to realize optimal exploitation scheduling decision. The implementation process proposed by Ramazan S, Dimitrakopoulos R. Production scheduling with uncertain supply: a new solution to the open pit mining problem [J]. Optimization and engineering, 2013, 14: 361-380. is that firstly, the probability distribution of uncertain variables such as grade, tonnage and the like is quantified through geological exploration data, then a two-stage random planning model is constructed, the initial exploitation plan is determined in the first stage, the subsequent exploitation strategy is adjusted in the second stage according to actual geological condition feedback, and finally, the production plan with both stability and economy is output. According to the scheme, probability distribution of parameters such as geological grade and tonnage is used as core input, and a mining plan is generated through two-stage planning, so that the influence of geological parameter fluctuation on a production target can be effectively reduced. However, the core defect is that the dynamic interference of weather factors on the actual capacity of the strip mine is not considered at all, so that the actual capacity is easy to be disjointed in the planned execution process. From the technical principle, the constraint condition of the scheme is designed only around static parameters such as geological reserves, equipment theoretical capacity and the like, and the nonlinear influence of weather factors such as extreme temperature, rainfall, strong wind and the like on the mining efficiency and the transportation safety is not included. When the weather is encountered in actual production, the plan generated according to the scheme can generate contradiction that the theoretical capacity reaches the standard but the actual capacity is suddenly reduced, so that the problems of ore supply shortage, equipment idling, disordered scheduling of a dumping site and the like are caused, and the stable production requirement of the strip mine under the condition of weather uncertainty can not be met. Aiming at uncertainty of market price, the existing mature technology is a long-term planning method combining price path simulation and empire competition algorithm (IMPERIALIST COMPETITIVE ALGORITHM, ICA), the scheme simulates a metal price fluctuation path through models such as GARCH, GBM and the like, and utilizes the empire competition algorithm to solve an optimal production plan under uncertainty of price, so that cost control and income maximization are realized. The implementation process proposed by ：Mokhtarian Asl M, Sattarvand J. Integration of commodity price uncertainty in long-term open pit mine production planning by using an imperialist competitive algorithm [J]. Journal of the Southern African Institute of Mining and Metallurgy, 2018, 118 (2): 165-172. comprises the steps of constructing a price prediction model based on historical price data to generate a multi-scenario price fluctuation path, constructing an optimization model containing constraints such as exploitation quantity, transportation path, equipment scheduling and the like with the aim of long-term benefit maximization, and outputting production plans adapting to different price scenarios through iterative solution of an empire competition algorithm. The scheme focuses on the influence of market price fluctuation on the income, realizes cost control through simulating price paths and algorithm optimization, and can effectively cope with market price fluctuation risks. However, the method also does not relate to the direct influence of weather factors on the productivity of each link of strip mining and transportation of the strip mine, namely, two main core defects exist, namely, firstly, the productivity parameter setting is separated from the actual scene of the weather influence, the daily productivity of the strip mine is stabilized at the theoretical maximum value by the scheme default, the prod