CN116467958A

CN116467958A - Salt lake brine numerical model construction and water supplementing and mineral dissolving efficiency calculation method

Info

Publication number: CN116467958A
Application number: CN202310250646.4A
Authority: CN
Inventors: 汪子涛; 王建萍; 袁小龙; 余冬梅
Original assignee: Qinghai Institute of Salt Lakes Research of CAS
Current assignee: Qinghai Institute of Salt Lakes Research of CAS
Priority date: 2023-03-15
Filing date: 2023-03-15
Publication date: 2023-07-21

Abstract

The invention discloses a method for constructing a salt lake brine numerical model and calculating water-replenishing and ore-dissolving efficiency. The construction method comprises the following steps: performing space discretization and time discretization; constructing a generalized model comprising a seepage sub-model and a solute transport sub-model; writing and converting the generalized model into a numerical model; solving a numerical model; and iteratively adjusting the parameter values in the generalized model to convergence to obtain the salt lake brine numerical model. The invention fully considers the characteristics of salt lake brine, and avoids the conditions of variable density, variable permeability and the like which are not considered in the traditional modeling; the evaluation and calculation method for the efficiency of water supplementing and ore dissolving based on the numerical simulation of the salt lake brine is designed based on the numerical model, is more visual, avoids the interference of other factors, has stronger degree of freedom, can continuously adjust the scheme of obtaining the maximum efficiency by simulation before water supplementing and ore dissolving, and then is applied to reality, thereby reducing the risks of low efficiency and resource waste caused by improper direct water supplementing without pre-simulation.

Description

Salt lake brine numerical model construction and water supplementing and mineral dissolving efficiency calculation method

Technical Field

The invention relates to the technical field of mineral resource exploration, in particular to a method for constructing a salt lake brine numerical model and calculating water supplementing and mineral dissolving efficiency.

Background

The salt lake brine is rich in abundant salt mineral resources, and is an important raw material source in the industries of modern chemical industry, agriculture, metallurgy, construction and the like. Since the last century salt lake brine is industrially exploited, the underground brine water level in the salt lake region is seriously reduced, the extraction difficulty is increased, and even the ground surface subsidence and stratum collapse occur in serious cases. In addition, because the water level is lowered, a large amount of solid minerals in the stratum space above the submergence surface cannot be extracted, and resource waste is caused. In order to relieve the contradiction between supply and demand of water resources and optimize resource exploitation, salt lake enterprises fill fresh water or old brine into the underground space of the salt lake, and dissolve solid minerals in the area above the original diving surface while raising the water level.

However, due to the complex underground environment, fresh water or old brine poured into the underground often "escapes" or does not achieve the expected effect, so that the efficiency of water-replenishing and ore-dissolving is low. Therefore, a set of integration scheme capable of measuring the water replenishing and ore dissolving efficiency and a numerical model capable of conforming to the characteristics of the salt lake are urgently needed by salt lake enterprises so as to realize fine characterization of the underground environment and serve as reference bases for evaluating the ore dissolving efficiency and optimizing resource exploitation.

The current water-replenishing and mineral-dissolving efficiency calculation is not uniform, most of the calculation is only based on the later estimation of products, the early numerical simulation of salt lake brine is less, even if a small amount of early simulation exists, a conventional groundwater modeling scheme is adopted, for example, direct modeling is performed by using mature commercial software such as Groundwater Modeling System (GMS) or Visual Modflow Flex, and typical references are, for example, numerical simulation research [ D ]. University of Chinese academy of sciences, 2017 of brine dynamic evolution mechanism under a water-replenishing background.

However, the product end calculation water-replenishing and ore-dissolving efficiency of the prior method is based on the estimation of the product end, is not combined with the geological conditions of the research area, and is not considered from the change of the resource storage capacity of the research area. In addition, the method can only give evaluation after ore dissolution, belongs to subsequent evaluation, cannot predict before engineering implementation, and cannot repeatedly test errors in advance to optimize layout.

In addition, the current numerical simulation of a small amount of prior art on salt lake brine does not consider the influence of the concentration of brine on a seepage sub-model, and only considers the fluid medium as the inherent characteristic that the change cannot be found, so that the problems of solid salt dissolution, aquifer permeability coefficient change and the like occurring during water supplementing and ore dissolving cannot be solved, and the potential special problems of salt lake regions such as dominant channels and the like are not considered in modeling. The simulation and calculation results obtained are not ideal for achieving high efficiency of the water-soluble mineral replacement.

Disclosure of Invention

Aiming at the defects of the prior art, the invention aims to provide a method for constructing a salt lake brine numerical model and calculating water-replenishing and ore-dissolving efficiency.

In order to achieve the purpose of the invention, the technical scheme adopted by the invention comprises the following steps:

in a first aspect, the invention provides a method for constructing a salt lake brine numerical model, comprising the following steps:

1) Performing space discretization and time discretization on a research area to generate a plurality of discrete cells, and dividing the research area into a plurality of time steps on a time scale;

2) Constructing a generalization model comprising a seepage sub-model and a solute transport sub-model, wherein information exchange exists between the seepage sub-model and the solute transport sub-model, and the simulation conditions in the generalization model are generalized, and the simulation conditions comprise inherent characteristics of the research area and boundary conditions; assigning the simulation conditions to the discrete cells, and setting the density and permeability coefficient of the salt lake brine in the discrete cells as variables based on different time steps;

3) Writing and converting the generalized model into a numerical model;

4) Solving the numerical model by utilizing a differential equation numerical solution to obtain a budget result of the research area;

5) And iteratively adjusting the parameter values in the generalization model to be converged based on the difference between the budget result and the actual observation result to obtain a salt lake brine numerical model.

In a second aspect, the invention also provides a salt lake brine numerical model constructed by the construction method.

In a third aspect, the invention also provides a method for calculating the water-replenishing and mineral-dissolving efficiency of salt lake brine, which comprises the following steps:

providing the salt lake brine numerical model as a standard model, and only reserving a pumping well in the standard model as a unique outflow item to construct a general model;

calculating standard inflow and outflow items in each time step based on the standard model, and calculating general inflow and outflow items in each time step based on the general model;

and calculating the water replenishing and mineral dissolving efficiency changing with time according to the standard inflow and outflow items and the general inflow and outflow items.

Based on the technical scheme, compared with the prior art, the invention has the beneficial effects that:

the numerical model and the construction method thereof fully consider the characteristics of the salt lake brine, and avoid the conditions of non-consideration of variable density, variable permeability and the like in the traditional modeling; and the evaluation and calculation method for the water-replenishing and ore-dissolving efficiency based on the numerical simulation of the salt lake brine is designed based on the numerical model, compared with the prior method, the evaluation and calculation method provided by the invention is more visual, is realized by a geochemical technical means from the perspective of resource reserves, and avoids the interference of other factors such as a production end and the like.

The technical scheme provided by the invention is based on geological data, a computer and mathematics are used as theoretical basis, the degree of freedom is higher, the water replenishing plan can be continuously adjusted before water replenishing and ore dissolving, the water replenishing and ore dissolving scheme with maximized efficiency is obtained through simulation and then is applied to reality, and the risks of low efficiency and resource waste caused by improper water replenishing without pre-simulation are obviously reduced.

The above description is only an overview of the technical solutions of the present invention, and in order to enable those skilled in the art to more clearly understand the technical means of the present application, the present invention may be implemented according to the content of the specification, the following description is given of the preferred embodiments of the present invention with reference to the accompanying drawings.

Drawings

FIG. 1 is a schematic diagram of the structure and flow of a numerical model provided by an exemplary embodiment of the present invention;

FIG. 2 is a graph showing calculation of water-replenishing and mineral-dissolving efficiency according to an exemplary embodiment of the present invention;

FIG. 3 is a graph showing permeability coefficient profiles of a three-layer model according to an exemplary embodiment of the present invention;

FIG. 4 is a graph showing the trend of the water level of the lake on the south side according to an exemplary embodiment of the present invention;

FIG. 5 is a spatial distribution diagram of head values on days 119, 239 and 359 of a model provided by an exemplary embodiment of the present invention;

FIG. 6 is a spatial distribution diagram of solute concentration on days 119, 239 and 359 of a model provided by an exemplary embodiment of the present invention;

FIG. 7 is a graph showing the variation trend of the water head value observed by 27 holes according to an exemplary embodiment of the present invention;

FIG. 8 is a graph showing the trend of the observed solute concentration values for 27-mouth holes according to an exemplary embodiment of the present invention.

Detailed Description

In view of the shortcomings in the prior art, the inventor of the present invention has long studied and practiced in a large number of ways to propose the technical scheme of the present invention. Specifically, along with continuous extraction of brine, the groundwater level of the salt lake is severely reduced. In order to fully utilize the solid salt minerals above the diving surface and to alleviate the environmental damage problem caused by the water level drop, salt lake enterprises supplement water and dissolve the minerals into the salt lake underground space in various modes. Because of the lack of fine characterization of underground conditions, the water-replenishing and mineral-dissolving effects are often not ideal. There is an urgent need for an integrated solution that can be used to evaluate the effectiveness of water-replenishing and mineral-dissolving processes. Therefore, the invention fully considers the characteristics of the salt lake brine, designs a salt lake brine numerical model, and establishes a water-replenishing and mineral-dissolving efficiency calculation method based on a numerical simulation result.

The technical scheme, the implementation process, the principle and the like are further explained as follows. The technical scheme provided by the embodiment of the invention comprises two parts, wherein one part is used for constructing a numerical model of salt lake brine, and the other part is used for calculating the water supplementing and mineral dissolving efficiency based on the model. Wherein, constructing the brine numerical model mainly comprises: discretizing a research area, generalizing simulation conditions, converting a conceptual model into a numerical model, solving by using a finite difference method and calibrating the model. The evaluation of the mine dissolution efficiency part may mainly include: and constructing a standard model, calculating the resource amount and calculating the ore dissolution efficiency.

In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced otherwise than as described herein, and therefore the scope of the present invention is not limited to the specific embodiments disclosed below.

Moreover, relational terms such as "first" and "second", and the like, may be used solely to distinguish one from another component or method step having the same name, without necessarily requiring or implying any actual such relationship or order between such components or method steps.

Furthermore, some of the technical terms involved in the present invention have the meanings as follows:

salt lake brine: liquid mineral resources with extremely high salt content are extracted from the underground space of the salt lake region.

Old brine: in the extracted salt lake brine, the residual liquid after the target salt mineral is extracted.

Supplementing water and dissolving ore: and (3) artificially filling a certain amount of fresh water or old brine with low concentration of target salt minerals into the underground space of the salt lake, so that solid minerals originally in an underground mineral layer are dissolved.

Referring to fig. 1, the method for constructing a salt lake brine numerical model provided by the embodiment of the invention comprises the following steps:

1) The study area is spatially discretized and time discretized to generate a plurality of discrete cells and divided into a plurality of time steps on a time scale.

2) Constructing a generalization model comprising a seepage sub-model and a solute transport sub-model, wherein information exchange exists between the seepage sub-model and the solute transport sub-model, and the simulation conditions in the generalization model are generalized, and the simulation conditions comprise inherent characteristics of the research area and boundary conditions; the simulation conditions are assigned to the discrete cells, and the density and permeability coefficient of the salt lake brine in the discrete cells are set as variables based on different time steps.

3) And writing and converting the generalized model into a numerical model.

4) And solving the numerical model by utilizing a differential equation numerical solution to obtain a budget result of the research area.

Among them, the study area discretization includes spatial Discretization (DIS) and Time Discretization (TDIS). The space discretization refers to splitting the original continuous space of the research area into a limited number of discrete cells according to a certain rule, so as to realize the calculation of the numerical solution of each cell to replace the differential equation analysis solution which can not be obtained. The number of discrete cell settings depends on the individual's need for model accuracy, the computer's own computing power, and the study of the spatially diverse characteristics. The most common discretization is to split the investigation region into several typical cubes of equal length, width and height. The time discretization means that the whole simulation time is divided into a plurality of time steps, and the simulation condition can be modified at the initial stage of each time step; if the latter time step does not modify the settings, the simulation conditions for the previous time step will be automatically followed.

The discretization of the research area can use other modes, such as triangle units, diamond units or hexagon units, and the like, and can also adopt a local encryption function.

Further in some embodiments, a dominant channel may be included in the generalized model, the dominant channel representing a continuous region in discrete cells having a salt lake brine permeability coefficient greater than a preset threshold, and the continuous region extending beyond the investigation region; the boundary between the dominant channel and the research area is set as a drain.

The characteristics of the dominant channel and the like are used as loss indexes for calculating the water supplementing and ore dissolving efficiency, which is an important technical means of the technical scheme provided by the embodiment of the invention, and the obtained simulation result is closer to the actual situation of the salt lake area, so that the referenceability of the simulation result is stronger.

In some embodiments, in step 2), the content of the information exchange may include that the seepage sub-model provides driving information of solute flow in salt lake brine for the solute transport sub-model, and that the solute transport sub-model provides density change information of salt lake brine for the seepage sub-model.

In some embodiments, there may also be an exchange of chemical reaction information between a plurality of the solute transport sub-models.

In the above technical solution, as shown in fig. 1, a salt lake brine simulation model needs to include a seepage sub-model (GWF) and one or more solute transport sub-models (GWT). There is an exchange of information (GWF-GWT) between the osmotic sub-model and the solute transport sub-model, such as the osmotic sub-model providing driving information (SPC, SSM) of the solute flow for the solute transport sub-model and the solute transport sub-model providing density change information (BUY) for the osmotic sub-model. Meanwhile, exchange of chemical reaction information can exist between solute transport sub-models, and of course, the solute transport sub-models can be adaptively set based on the solute composition and content in salt lake brine.

As some typical application examples of the above technical solutions, the primary task of performing numerical simulation is to identify and generalize complex geological environments. First, the inherent characteristics of the investigation region (e.g., NPF, STO) should be defined, and the seepage submodel includes the elevation of the roof and floor of each aquifer, the permeability coefficients in different directions, the water supply or storage rates, etc., which are assigned to each discrete cell. Solute transport sub-models include porosity (MST), convection mode (ADV), and hydrodynamic diffusion coefficient (DSP), among others. Boundary conditions of the investigation region, including well boundary (WEL), river boundary (RIV), drainage canal boundary (DRN), designated head boundary (CHD), and General Head Boundary (GHB), etc., are then considered, while there may be advanced boundaries such as lake boundary (LAK), unsaturated Zone (UZF), multi-aquifer well boundary (MAW), etc. The boundary conditions of the solute transport sub-model are mostly determined by the solubility of the percolation boundary conditions (SPC) except for the solute loading boundary (SRC) and the assigned concentration boundary (CNC), and further, the high-order concentration boundary corresponds to the boundary of the percolation sub-model one by one (for example, SFT, UZT, LKT, MWT).

The boundary conditions should be selected according to the actual situation of the investigation region. For example, the well boundary may require coordinates of the mating well, water extraction or injection per time step. In the case of a water injection well, the solute concentration of the injected water is also required. River boundaries require the elevation of the river bed, the water level, the permeability of the river, etc. Other boundary conditions are similar, and in practice, those skilled in the art can see, for example, the U.S. local agency' S seepage submodel (LANGEVIN C D, HUGHES J D, BANTA E R, et al 2017.Documentati on for the MODFLOW 6Groundwater Flow Model:6-A55[ R ]. Reston, va: U.S. logical surface: 197.) and solute transport submodel report (LANGEVIN C D, PROVOST A M, PANDAY S, et al 2022.Docume ntation for the MODFLOW 6Groundwater Transport Model:6-A61[ R ]. Reston, va: U.S. logical surface: 56.).

The inherent characteristics of these models, the head value of the boundary, the flow of replenishment or evacuation, and the concentration of solute are assigned to discrete cells at the location of the boundary. If these boundaries belong to an unstable flow, i.e. parameters such as flow, head and concentration change with time, it is also necessary to assign different values to the initial phase of each time step. Finally, considering the own characteristics of salt lake brine, the three problems of variable density, variable permeability coefficient and dominant channels are mainly included. Too high a solute concentration in brine can change the fluid density and the effect on the seepage submodel cannot be simply ignored.

With respect to the variable parameter problem described above and the loss of dominant pathways, in some embodiments, the density of the salt lake brine may be calculated by determining by reference to fluid density and brine concentration:

wherein ρ is the brine density to be determined; ρ _ref For reference fluid density, typically the fluid density of fresh or old brine; n is the number of brine solute types participating in simulation; d (D) _i Is a slope parameter describing the change in density with solute concentration; c (C) _i For the concentration of the i-th solute in the fluid, ri is the concentration of the i-th solute in the reference fluid.

The variable permeability problem is determined primarily by providing each discrete cell with a different permeability coefficient value (TVK) at different time nodes (in multiple time steps).

In addition, the dominant channel can be understood as a continuous region of extremely high permeability, such as permeability greater than a threshold, or a certain multiple, such as 2 or 5 times, of the average permeability of the study area, while the dominant channel is designed as a drain at the boundary of the dominant channel and the model, along with the characteristic that the brine flow direction will leave the scope of the study area. Finally, the whole salt lake brine model conceptualization process is shown in fig. 1.

Based on example technical implementations, in some embodiments of the invention, the intrinsic properties may include a first set of factors in the percolation sub-model and a second set of factors in the solute transport sub-model.

In some embodiments, the first set of factors may include a top-bottom elevation of each aquifer in the generalized model, permeability coefficients of different directions of salt lake brine in the discrete cells, water supply and water storage rate.

In some embodiments, the second set of factors may include porosity, convection mode, and hydrodynamic dispersion coefficient.

In some embodiments, the inherent properties may also include aquifer seepage properties as well as groundwater storage properties.

In some embodiments, the boundary conditions may include well boundaries, river boundaries, drainage canal boundaries, designated head boundaries, and universal head boundaries.

In some embodiments, advanced boundaries may also be included, including lake boundaries, unsaturated zone, multiple aquifer well boundaries.

In some embodiments, when the boundary condition is a non-constant value, the respective parameter values may be set in a plurality of time steps based on the change in the boundary condition.

After solving the generalized model, the variable parameter problem and the dominant channel loss problem, the conceptual model needs to be converted into a numerical model:

the step of generalizing the simulation conditions only generalizes the research area, and the step here mainly comprises writing the conceptual model into different text files according to a specific format. Still referring to fig. 1, the simulated text control requires writing of initial control conditions (IC) and output control conditions (OC). In particular experimental China, the simulation itself may be a calculation engine that invokes MODIFW 6 (an open source calculation program developed by the United states geological survey) with different format requirements for each boundary or property file, see MODIFW 6 document (U.S. device of the Interor, U.S. logical Survey. MODIFW 6-Description of Input and Output [ Z ]. 2021.). Meanwhile, different suffix names are needed for program identification, for example, discretized cell file suffix is · dis, well file is · wel, stream file is · a dv, etc., and specific writing formats and suffix requirements are found in the above documents. These files may be written by manually creating text files and changing suffix names, or by means of corresponding packages in programming languages such as Python or Matlab (e.g. the Flopy module of Python, see BAKKER M, POST V, LANGEVIN C D, et al 2016.Scripting MODFLOW Model Development Using Python and FloPy[J, groundwater,54 (5): 733-739).

Numerical simulation involves the grasping and writing formats and requirements of various simulation conditions in the MODIFLOW 6 program, and needs to refer to the latest document of MODIFL OW 6. In addition, the writing process requires a third party open source module Flopy using Python. The pattern of migration and modeling theory of brine is described in the prior art (MAJID HASSANIZADEH S. Development of basic equations of mass tran sport in porous media, part 1.Macroscopic balance laws[J ]. Advances in Water Resources,1986,9 (4): 196-206. And MAJID HASSANIZADEH S. Development of basic equations of mass transport in porous media, part 2.Generalized Darcy's and Fick's laws[J ]. Advances in Water Resources,1986,9 (4): 207-222. And HASANIZADH S M, LEIJNSE T.On the modeling of brine transport in porous media [ J ]. Water Resources Research,1988, 24 (3): 321-330.).

The names and abbreviations (suffixes) of the modules of MODFLOW6 used in the above embodiment are shown in table 1 below.

TABLE 1 MODIFLOW6 names of the modules and their meanings

Of course, in the specific implementation, the choice of the calculation engine is not limited thereto, and other calculation engines may be used instead of the mode LOW6, such as using a flow or other hydrodynamic scheme. Any equivalent substitution or replacement based on the characteristic technical idea of the present invention falls within the protection scope of the present invention.

After the conversion of the numerical model, the solution and calibration of the model can be realized.

The solving process may be, for example:

after the numerical simulation is constructed, model solving is needed. The solution may be, for example, a finite difference method of controlling the volume, the parameters of which are adjustable by the IMS module. And calling an engine of MODIflow 6 at a command line interface (such as a CMD of a Windows system or a terminal interface of a Linu x system) of the computer, and calculating the text file to obtain the water head and concentration values of the research area and the budget of the water head concentration.

Of course, other differential equation numerical solutions may be used instead of finite difference methods, such as finite element methods and finite volume methods, or may be solved using depth mathematical methods commonly used in recent years, such as embedded Physical Information Neural Networks (PINNs). The specific solving mode is provided by a plurality of referent examples in the prior art, and can realize the solving function, and is not limited in scope by the specific examples disclosed by the embodiment of the invention.

The model is calibrated by continuously adjusting parameters in the model generalization step. Due to the complex nature, only approximate ranges or rough values can be given to most boundaries or characteristics in the model generalization process, and the true values are not known. These conditions are often difficult or costly to observe. The output value of the model is usually the water head and concentration distribution of the whole area. In practical application, the actual observation value of part of discrete points can be obtained by monitoring the water level of the observation hole and collecting a sample test mode. In this way, an evaluation function between the observation and its corresponding simulated Observation (OBS) can be established to measure the accuracy of the model, and the model can be iteratively updated.

Also, other algorithms may be used for modeling, such as simulated annealing, particle swarm optimization, etc., and probabilistic modeling, such as Markov chain Monte Carlo modeling, etc., may be used. The scope of the specific disclosed examples is not intended to limit the scope of the invention.

Thus, in some embodiments, the construction method may specifically include:

and obtaining the water head and solute concentration distribution of the whole area in the research area by solving the numerical model, and taking the water head and solute concentration distribution as the budget result.

In the actual salt lake area, the water level of the observation holes is monitored, and the sample test mode is collected to obtain the actual observation results corresponding to part of the discrete cells in the generalized model.

And calculating a model loss value according to the difference between the budget result and the actual observation result, and iteratively updating the generalized model based on the model loss value.

In some embodiments, the calculation method of the model loss value may be expressed as:

wherein R is ² For the model loss value, f _i The simulated value of the brine water head or concentration of the salt lake of the discrete unit lattice corresponding to each observation point; yi is the actual observed value of the brine water head or concentration of the salt lake at each observation point; The average value of the brine water head or concentration of the salt lake at all observation points.

At this time, only some optimization algorithms, such as genetic algorithm, are adopted, and the evaluation function is converged to the maximum value by continuously adjusting the generalization parameters of the iterative model, so that accurate simulation is realized

The second aspect of the embodiment of the invention also provides a salt lake brine numerical model constructed by the construction method.

The third aspect of the embodiment of the invention also provides a calculation method of salt lake brine water-replenishing and mineral-dissolving efficiency, which comprises the following steps:

providing the salt lake brine numerical model as a standard model, and only retaining a pumping well in the standard model as a unique outflow item to construct a general model.

A standard inflow and outflow item in each time step is calculated based on the standard model, and a general inflow and outflow item in each time step is calculated based on the general model.

In the implementation, regarding the construction of the standard model and the general model, a numerical model can be firstly constructed based on the model construction flow as the standard model. In the standard model, conditions such as a dominant channel, constant water supplementing channel concentration and the like need to be removed, and the pumping well is ensured to be the only outflow item of the model. And then loading loss items such as dominant channels and the like, and constructing a general model.

In some embodiments, the calculation of the standard inflow and outflow items and the general inflow and outflow items can be expressed as:

wherein P is _IN Representing standard inflow items; p (P) _OUT Represents a standard outflow item; p (P) _{CHD_IN} Representing a standard head boundary inflow item; p (P) _{RIV_IN} Represents a standard river boundary inflow item; p (P) _{WEL_OUT} Representing a standard well effluent; q (Q) _IN Represents a general inflow item; QOUT represents a general outflow term; q (Q) _{CHD_IN} Represents a general head boundary inflow term; q (Q) _{RIV_IN} Represents a general river boundary inflow item; q (Q) _{WEL_OUT} Represents a general pump well outflow item; m represents the total number of time steps.

In some embodiments, the general model may also be loaded with a loss term comprising salt lake brine losses corresponding to dominant channels in the salt lake brine numerical model.

As some typical application examples of the above technical solutions, after the standard model and the general model are run, the budget file of the seepage sub-model and the solute transport sub-model can be obtained in addition to the water head distribution and the concentration distribution of the research area. For convenience of description, the standard model inflow and outflow items are replaced by P, and the general model inflow and outflow items are replaced by Q.

If a specified head boundary (CHD), river boundary (RIV) and pumping well (wil) are used in the model, then in the standard model the inflow term includes P _{CHD_IN} ，P _{CHD_IN} The outflow item has only a unique P _{WEL_OUT} The method comprises the steps of carrying out a first treatment on the surface of the While the inflow term of the general model includes Q _{CHD_IN} ，Q _{CHD_IN} The outflow term may include Q _{WEL_OUT} ，Q _{CHD_OUT} ，Q _{RIv_OUT} . It is noted that at this point all streaming entries have been specific to each time step, so both P and Q are sequence values based on multiple time steps. Assuming that m time steps are taken, summing the inflow item and the outflow item of all the time steps of the standard model to obtain P _IN And P _OUT The method comprises the steps of carrying out a first treatment on the surface of the Summing each time step of the general model forward to obtain a time sequence value Q _IN And Q _OUT 。

Wherein, although the outflow term of the general model includes a plurality of terms, Q _OUT Still only include Q _{WEL_OUT} The term, the remaining effluent terms are all considered loss terms.

In some embodiments, a water replenishment efficiency coefficient may be calculated based on the standard inflow and outflow term and the general inflow and outflow term, and then a water replenishment efficiency may be calculated based on the water replenishment efficiency coefficient.

In some embodiments, the water replenishment efficiency factor is calculated by combining P _IN And P _OUT And Q is equal to _IN And Q _OUT The quotient is calculated and then averaged, which can be expressed as:

wherein z represents the water replenishment efficiency coefficient.

In some embodiments, the calculation process of the water replenishment efficiency uses z as an argument, and constructs a Logistic function, which can be expressed as:

wherein R represents the water replenishing efficiency; n is a fixed value; alpha and beta are undetermined parameters, and alpha represents the efficiency of salt solid dissolution; beta represents the amount of resources of the soluble solid salts in the aquifer.

In some embodiments, N may have a value of 1, for example.

The relationship between the water replenishing efficiency and the water replenishing efficiency coefficient (assuming that alpha is 25.0, beta is 10.0, and z is obtained in a assumed case) thus obtained is shown in FIG. 2, and based on the relationship, the water replenishing efficiency can be directly calculated

The foregoing is illustrative of the technical solution of the present invention, and in the foregoing examples, the key technical means of the technical solution of the present invention are:

1. the numerical simulation of the salt lake brine is realized on the premise of fully considering the properties of variable density, variable permeability and the like, and the simulation of the salt lake brine is set as the information exchange of a seepage sub-model and the solute migration bidirectional feedback.

2. The numerical simulation mode is adopted for the first time to calculate the water supplementing and ore dissolving efficiency of the salt lake brine.

3. And taking the characteristics of the dominant channel and the like as loss indexes for calculating the water supplementing and ore dissolving efficiency.

4. The water replenishing and ore dissolving efficiency is regarded as an index which changes along with time, and the time variable is coupled.

Based on the key technical means, the technical scheme of the embodiment of the invention can achieve the following technical effects:

1. the numerical model designed by the embodiment of the invention and the construction method thereof fully consider the characteristics of the salt lake brine, and avoid the conditions of non-consideration of variable density, variable permeability and the like in the traditional modeling.

2. The embodiment of the invention designs a water supplementing and mineral dissolving efficiency evaluation method based on salt lake brine numerical simulation. Compared with the prior art, the method is more visual, is realized by a geoscience technical means from the perspective of resource reserves, and avoids the interference of other factors such as a production end and the like.

3. The embodiment of the invention takes geological data as a basis, takes computers and mathematics as theoretical basis, and has stronger degree of freedom. Before water supplementing and ore dissolving, a water supplementing plan can be continuously adjusted to obtain a water supplementing scheme with maximized efficiency, and then the water supplementing scheme is applied to practice, so that the inefficiency and the resource waste caused by improper direct water supplementing are prevented.

On the basis of the above-mentioned exemplary technical solution, the technical effects of the present invention are more intuitively illustrated by some specific application cases as follows:

example 1

The process of performing simulation calculation in the exemplary region of the salt lake in the north of the Dabson lake is illustrated in the embodiment, and is specifically as follows:

1. a salt lake brine numerical model is constructed according to the figure 1, the water level is observed by using a long observation hole arranged in Qinghai salt lake research institute of China academy of sciences, and a coefficient calibration simulation value is determined by adopting a genetic algorithm and linear regression.

2. The water-replenishing and mineral-dissolving efficiency was calculated using the method of the specific example described above.

3. And repeatedly adjusting water supplementing parameters or adopting a multi-objective optimization algorithm on the basis of simulation, and trying to find an optimal water supplementing and mineral dissolving scheme.

Specifically, the outline of the model adopted in this embodiment is as follows:

the model is a simulation process of the whole extraction process and the water supplementing and ore dissolving efficiency of a salt lake enterprise in a well extraction implementation mode.

Related concepts:

and (3) ditch collection: the canal is a way for salt lake enterprises to collect brine, specifically, a canal with depth of several meters to more than ten meters is dug on the ground, and a proper water level difference is built in the canal, so that underground brine around the canal is converged into the canal and runs towards the downstream of the canal. And (5) constructing a water suction pump at the downstream of the ditch, and pumping brine. This is trench mining.

Well production: well production is relative to trench production. I.e. by well drilling, brine is directly pumped underground.

The model has 4 stress cycles, wherein the 1 st stress cycle is 1 day (used for adjusting the initial state), the last 3 stress cycles are all 120 days, and the whole simulation time is 361 days. In the simulation, we performed the simulation in ten days as one time step.

The model top elevation (i.e., altitude) was 2500m. Comprising 3 water-bearing layers, wherein the 1 st layer is diving, the back 2 layers are pressurized water. Each aquifer was 10m thick, for a total of 30m. The plane of the model is a matrix area with east-west length of 1200m and north-south width of 600 m. The computational grid is discretized according to a 30m by 30m specification.

Related concepts:

diving: groundwater buried in a first stable water barrier below the surface. In brine aquifers, it can be considered simply the first layer. During the simulation, diving is set such that when the head is below the aquifer top elevation, the saturation thickness automatically changes with the calculated head.

Pressurized water: groundwater blocked by the two water-proof roof and floor boards. Can be simply considered groundwater below the diving layer.

The simulated horizontal permeability coefficient is presented as a gaussian distribution with a mean of 300.0m/d and a standard deviation of 20.0 m/d. The vertical permeability coefficient is 0.6 times the horizontal permeability coefficient. The permeability coefficient distribution of the three-layer model is shown in fig. 3. The circle mark is set as a dominant channel. It can be seen that the dominant channels are mainly concentrated in layer 3 of the model, with a permeability coefficient of 20000m/d.

The porosity of the model was a gaussian distribution with a mean of 0.18 and standard deviation of 0.001. The longitudinal hydrodynamic dispersion coefficient is 3.3, and the transverse and vertical dispersion coefficients are 0.1 times of the longitudinal dispersion coefficient. The diffusion coefficient of the model was 0.0176.

Related concepts: diffusion coefficient and hydrodynamic diffusion coefficient are hydrogeologic parameters of solute transport in groundwater.

The initial water head of all the water-bearing layers is set to be Gaussian distribution below 0-2m of the ground, the initial concentration is set to be Gaussian distribution with 300kg/m3 as the mean value and 1.0kg/m3 as the standard deviation. The simulation is substantially unaffected by the initial value.

Regarding boundary conditions:

boundary conditions of the model comprise a lake on the south side, a water supplementing channel in the middle of the model, two pumping wells and an dominant channel. In addition, there are dissolution boundaries on both sides of the make-up channel. The water level of the lake is easily observed, and in the simulation period, the water level of the lake gradually returns after gradually decreasing, as shown in fig. 4. The solute concentrations of the lake water at simulated days 0,1, 121, 241 and 361 are shown in table 1 below. The concentrations at other times were interpolated over the days.

TABLE 1 lake water concentration

There are two wells in the model, where well a is at the (3, 11, 13) coordinates and well B is at the (2, 11, 26) coordinates. The water extraction of the two wells at simulated days 0,1, 121, 241 and 361 is shown in table 2. The amount of water drawn at other times was interpolated over the days.

Table 2 two well water extraction

The depth of the water supplementing channel is 5m. The water level in the water replenishment channel remained stable, at the beginning of four stress cycles 2497.9m,2496.9m,2498.1m and 2499.5m, respectively, with each time step being linearly interpolated over the respective stress cycle. The water supplementing channel is arranged on the 8 th row of the simulation first layer. The dissolution boundaries of the model were concentrated on both sides of the water replenishment canal, see table 3, and the remaining time was interpolated in the dissolution amount of time in the table.

TABLE 3 distribution of dissolution boundary and dissolution amount

During dissolution, the osmotic coefficient of the model changes. The permeability coefficient changes at various times for the above dissolution boundaries are shown in table 4, with other times being linearly interpolated over these times.

TABLE 4 osmotic coefficient variation of dissolution boundary

The construction thought of the model is as follows:

the simulation constructs two models, a seepage model GWF and a solute transport model GWT. Wherein the solute is total dissolved solids. For ease of calculation, we calculated the water replenishment efficiency directly for total dissolved solids. In other cases, if it is specific to ions, such as potassium ions, then it is necessary to additionally construct a model of solute transport of potassium ions.

Related concepts: total dissolved solids (Total Dissolved Solids, TDS) refers to the sum of total solids dissolved in water.

Because the TDS concentration of salt lake brine is far higher than that of seawater, the migration of fluid is influenced, so that the simulation is that a seepage model influences a solute migration model, and the solute migration model in turn influences the evolution of the seepage model.

The model is discretized into 3×20×40 units altogether. In this model, the changes in the properties such as various boundary conditions of the model are no longer in the smallest units of stress cycles, but may be arbitrarily changed (TS) in each time step.

The lake adopts a CHD boundary, the water supplementing channel adopts a RIV boundary, the pumping well adopts a WEL boundary, and the concentration dynamic change of the three boundaries uses SPC. The solute dissolution process around the water supplementing channel adopts SRC boundary, and the osmotic coefficient change caused by dissolution adopts TVK. In addition, BUY was used to control the density of the permeate fluid.

The maximum number of external iterations of the model is 100, and the maximum number of internal iterations in each external iteration is 300. The absolute and relative level values of solver convergence were 1e-6, with a relaxation factor of 1.0, and were solved using a standard PCG algorithm. The time step adjustment parameter is 1.0, and no time step adjustment is performed.

The entire simulation process is performed entirely using the Python programming language. Scientific calculation modules such as Os, numpy, pandas and the like are called, and MODIOW 6 is called by using a Flopy module to carry out model solving. All images were drawn using Matplotlib modules.

Because the parameters of the present model are known, the model parameters are not inverted using an optimization algorithm such as a genetic algorithm.

Simulation results concerning the above simulation

Fig. 5 represents the spatial distribution of the head values of the model on days 119, 239 and 359. Each column in the figure represents the same layer model, and each row represents the same day's three-layer model. FIG. 6 represents the spatial distribution of solute concentrations of the model on days 119, 239 and 359.

In addition, 9 sets of 27-port observation holes were deployed in the model, each set observing the head and concentration of the three-layer model simultaneously. The observation hole coordinates are shown in table 5 below.

TABLE 5 observation hole coordinates

The observed water head value and solute solubility change for the 27-port holes are shown in fig. 7 and 8, respectively.

And (3) calculating the water supplementing and ore dissolving efficiency based on the model:

in this model, boundaries include WEL, RIV, CHD and SRC together with respect to solutes. Therefore, first, a standard model needs to be considered. In the standard model, the above are input terms except for WEL, which is the only output term. Accumulating the input and output items at each moment to obtain P _IN And P _{WEL_OUT} 。

The simulated input and output terms are then calculated:

calculated according to the above equation, the z value is 0.66338055, approximately 0.66. The simulation cycle was then normalized to 1, with α being 25 and β being 10, and a Logistic curve was established as shown in fig. 2 according to the following equation. The R value is 0.74, and the water-replenishing ore-dissolving efficiency of the model is 74%.

Based on the above embodiment, it can be clear that the numerical model and the construction method thereof provided by the embodiment of the invention fully consider the characteristics of salt lake brine, and avoid the conditions of non-consideration of variable density, variable permeability and the like in the traditional modeling; and the evaluation and calculation method for the water-replenishing and ore-dissolving efficiency based on the numerical simulation of the salt lake brine is designed based on the numerical model, compared with the prior method, the evaluation and calculation method provided by the invention is more visual, is realized by a geochemical technical means from the perspective of resource reserves, and avoids the interference of other factors such as a production end and the like.

The technical scheme provided by the embodiment of the invention is based on geological data, a computer and mathematics are used as theoretical basis, the degree of freedom is higher, the water replenishing plan can be continuously adjusted before water replenishing and ore dissolving, the water replenishing and ore dissolving scheme with maximized efficiency is obtained through simulation and then is applied to reality, and the risks of low efficiency and resource waste caused by improper water replenishing without pre-simulation are obviously reduced.

Meanwhile, in the model construction process of the concrete example of the invention, a construction system of a salt lake brine numerical model is also provided, and the system is used for executing the steps of the construction method; and a readable storage medium having stored therein a computer program or the above salt lake brine numerical model, which when executed, performs the steps of the above construction method.

It should be understood that the above embodiments are merely for illustrating the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the present invention and implement the same according to the present invention without limiting the scope of the present invention. All equivalent changes or modifications made in accordance with the spirit of the present invention should be construed to be included in the scope of the present invention.

Claims

1. The construction method of the salt lake brine numerical model is characterized by comprising the following steps of:

2) Constructing a generalization model comprising a seepage sub-model and a solute transport sub-model, wherein information exchange exists between the seepage sub-model and the solute transport sub-model, and the simulation conditions in the generalization model are generalized, and the simulation conditions comprise inherent characteristics of the research area and boundary conditions; assigning the simulation conditions to the discrete cells, the density and permeability coefficient of the salt lake brine in the discrete cells being set as variables based on different time steps;

3) Writing and converting the generalized model into a numerical model;

2. The method of claim 1, wherein the generalized model further comprises a dominant channel, the dominant channel represents a continuous region of discrete cells having a salt lake brine permeability coefficient greater than a predetermined threshold, and the continuous region extends beyond the investigation region;

The boundary between the dominant channel and the research area is set as a drain.

3. The construction method according to claim 1, wherein in step 2), the content of the information exchange includes that the seepage sub-model provides driving information of solute flow in salt lake brine for the solute transport sub-model, and that the solute transport sub-model provides density change information of salt lake brine for the seepage sub-model;

preferably, chemical reaction information is exchanged among a plurality of solute transport sub-models;

preferably, the density of the brine in the salt lake is calculated by the following method:

wherein ρ is the brine density to be determined; ρ _rer Is the reference fluid density; n is the number of brine solute types participating in simulation; d (D) _i Is a slope parameter describing the change in density with solute concentration; c (C) _i Is the concentration of the i-th solute in the fluid, R _i Is the concentration of the i-th solute in the reference fluid.

4. The method of claim 1, wherein the intrinsic characteristics comprise a first set of factors in the percolation sub-model and a second set of factors in a solute transport sub-model;

preferably, the first factor group comprises the top-bottom plate elevation of each aquifer in the generalized model, the permeability coefficients of salt lake brine in different directions in the discrete unit cells, the water supply degree and the water storage rate;

Preferably, the second set of factors includes porosity, convection mode, and hydrodynamic dispersion coefficient;

preferably, the inherent properties further include aquifer seepage properties and groundwater storage properties.

5. The method of construction of claim 1, wherein the boundary conditions include well boundaries, river boundaries, drainage canal boundaries, specified head boundaries, and universal head boundaries;

preferably, the boundary conditions further include high-level boundaries including lake boundaries, unsaturated zones, and multi-aquifer well boundaries.

6. The construction method according to claim 5, wherein when the boundary condition is a non-constant value, respective parameter values are set in a plurality of time steps based on a change in the boundary condition.

7. The construction method according to claim 1, characterized in that it comprises in particular:

obtaining the water head and solute concentration distribution of the whole area in the research area by solving the numerical model, and taking the water head and solute concentration distribution as the budget result;

monitoring the water level of the observation holes in the actual salt lake area, and acquiring actual observation results corresponding to part of discrete cells in the generalized model in a sample test mode;

Calculating a model loss value according to the difference between the budget result and the actual observation result, and iteratively updating the generalized model based on the model loss value;

preferably, the calculation method of the model loss value is expressed as:

wherein R is ² For the model loss value, f _i The simulated value of the brine water head or concentration of the salt lake of the discrete unit lattice corresponding to each observation point; y is _i The actual observed value of the brine water head or concentration of the salt lake at each observation point is obtained;the average value of the brine water head or concentration of the salt lake at all observation points.

8. A salt lake brine numerical model constructed by the construction method of any one of claims 1 to 7.

9. The calculation method of the salt lake brine water supplementing and mineral dissolving efficiency is characterized by comprising the following steps of:

10. The computing method according to claim 9, wherein the computing process of the standard inflow and outflow items and the general inflow and outflow items is expressed as:

wherein P is _IN Representing standard inflow items; p (P) _OUT Represents a standard outflow item; p (P) _{CHD_IN} Representing a standard head boundary inflow item; p (P) _{RIV_IN} Represents a standard river boundary inflow item; p (P) _{WEL_OUT} Representing a standard well effluent; q (Q) _IN Represents a general inflow item; QOUT represents a general outflow term; q (Q) _{CHD_IN} Represents a general head boundary inflow term; q (Q) _{RIV_IN} Represents a general river boundary inflow item; q (Q) _{WEL_OUT} Represents a general pump well outflow item; m represents the total number of time steps;

preferably, the general model is further loaded with a loss term, and the loss term comprises salt lake brine loss corresponding to a dominant channel in the salt lake brine numerical model;

preferably, a water replenishing efficiency coefficient is calculated based on the standard inflow and outflow item and the general inflow and outflow item, and then a water replenishing efficiency is calculated based on the water replenishing efficiency coefficient;

preferably, the calculation process of the water replenishing efficiency coefficient is expressed as follows:

wherein z represents the water replenishing efficiency coefficient;

preferably, the calculation process of the water replenishing efficiency is expressed as:

wherein R represents the water replenishing efficiency; n is a fixed value; alpha represents the efficiency of the salt solid in dissolving; beta represents the amount of resources of the soluble solid salts in the aquifer;

Preferably, N has a value of 1.