Search

CN-122021300-A - Calculation method for pipe conveying resistance of full tailing filling slurry by considering unstable flow

CN122021300ACN 122021300 ACN122021300 ACN 122021300ACN-122021300-A

Abstract

The invention discloses a calculation method of pipe conveying resistance of full tailing filling slurry considering unstable flow, and belongs to the technical field of mine filling mining. According to the method, through a 9-step process of physical and chemical property analysis, rheological modeling, pipe transportation test, unstable flow identification, unstable flow resistance calculation, stable flow resistance calculation, local resistance calculation, total pipeline resistance integration and parameter optimization, the influence of unstable flow on resistance loss is fully considered, an accurate calculation model is built by combining a BP neural network, XGBoost and a genetic algorithm combined algorithm, and the on-site verification error is only 19.51%, so that the filling pipeline design and parameter optimization can be effectively guided, and the energy consumption and the production cost are reduced.

Inventors

  • WANG CHONGHAO
  • GONG YUHAN

Assignees

  • 华北理工大学

Dates

Publication Date
20260512
Application Date
20260128

Claims (9)

  1. 1. The calculation method of the pipe conveying resistance of the full tailing filling slurry considering unstable flow is characterized by comprising the following 9 steps of: s1, analyzing physical and chemical properties of full-tailing filling slurry, and obtaining a tailing grading, a crystalline structure, chemical components, a shape coefficient, an alkalinity coefficient and a slurry mass concentration application range, wherein the tailing grading comprises an uneven coefficient and a curvature coefficient; S2, constructing a rheological parameter calculation model of slurry yield stress tau 0 and plastic viscosity mu based on a multi-factor orthogonal test; s3, collecting pressure and flow rate data of each monitoring point of the pipeline under different working conditions through a variable-diameter pipeline transmission test system, wherein the working conditions comprise filling doubling lines, the filling doubling lines are defined as the ratio of the total length of the filling pipeline to the height difference of the starting point and the finishing point of the pipeline, and the ratio is recorded as N, N=L/H; S4, judging the flow state based on the Reynolds number, and establishing an unstable flow length calculation model through test data fitting; S5, constructing an unstable flow on-way resistance loss model by adopting a combined algorithm of a BP neural network, XGBoost and a genetic algorithm and taking the flow speed, the pipe diameter, the yield stress and the plastic viscosity as inputs, wherein the unstable flow on-way resistance loss is recorded as i 1 ; S6, calculating the stable flow on-way resistance loss by adopting a Bingham fluid formula, wherein the stable flow on-way resistance loss is recorded as i 2 ; S7, calculating local resistance loss by adopting a multiple method, and calibrating coefficients based on the on-site pipeline arrangement condition; s8, integrating stable flow, unstable flow and local resistance loss, and constructing a total pipeline resistance loss comprehensive calculation model, wherein I total= (n+1). I 2 ·(L-l 1 )+(n+1)·i 1 ·l 1 , L=N.H, H is the height difference between the starting point and the end point of a filling pipeline, and N is a filling doubling line defined in the step S3; and S9, optimizing the conveying parameters based on a genetic algorithm by taking the minimum resistance loss of the whole pipeline as a target, and realizing parameter input and optimization result output through an application program.
  2. 2. The calculation method according to claim 1, wherein in S1, a tailing non-uniformity coefficient Cu is more than or equal to 5, a curvature coefficient Cc=1-3, a shape coefficient Sr=0.60-0.99, an alkalinity coefficient Mtail is less than 1, an alkalinity coefficient Mtail =W(Fe 2 O 3 +CaO+MgO+K 2 O+Na 2 O+SrO+NiO+ZnO)/W(SiO 2 +Al 2 O 3 +P 2 O 5 +SO 3 +Cr 2 O 3 +ZrO 2 +R 2 O),W is an oxide mass fraction, and a slurry mass concentration application range is 70% -74%.
  3. 3. The calculation method of claim 1, wherein the factors of the multi-factor orthogonal test in S2 comprise 70% -74% of mass concentration, 20% -60 ℃ of temperature and 1:2-1:10 of ash-sand ratio, a rheological parameter calculation model is constructed by adopting a random forest method, a model determination coefficient R2 is more than or equal to 0.96, slurry density ρ is calculated according to the formula ρ=c.ρ s + (1-C) ρw, ρ s is the density of tailings, and ρw is the density of water.
  4. 4. The calculation method of the variable-diameter pipe conveying test system is characterized in that the variable-diameter pipe conveying test system in the step S3 comprises a preparation system, a pipe conveying system and an acquisition system, wherein the pipe conveying system is provided with four kinds of inner diameter horizontal straight pipes of 100 mm-250 mm and four kinds of height lifting vertical pipes of 800 mm-2600 mm, each horizontal straight pipe is provided with a monitoring point every 50cm, the acquisition system is connected with the acquisition module through an intelligent transmitter, the communication distance is less than or equal to 1000m, an RS485 interface is supported, the flow rate is obtained through conversion of pressure data and Bernoulli equation, and the value range of a filling doubling line N is 3-6.
  5. 5. The method according to claim 1, wherein in S4, reynolds number Re=ρvD/. Mu.Re >2000 is an unstable flow, re <2000 is a stable flow, the length model of the unstable flow is established by three polynomial fitting, the correlation coefficient R2 is not less than 0.93, the model expression is l 1 = -0.05 (lnc) 3-87.35 (lnc) 2-28.12lnc-0.78lnN+0.28lnD, c is mass concentration, N is a filling line defined in step S3, and D is pipe diameter.
  6. 6. The calculation method of claim 1, wherein the construction process of the combined algorithm in S5 is that ① is used for establishing a BP neural network, 4 neurons of an input layer, 10 neurons of a hidden layer and 1 neuron of an output layer are used for activating a function, learning rate is 0.01, iteration times are 1000, ② is used for establishing a XGBoost model, tree depth is 6, learning rate is 0.1, iteration times are 200, regularization parameters lambda=0.1 and gamma=0.1, ③ genetic algorithm is used for optimizing weight coefficients, ω1=0.3-0.5 and ω2=0.5-0.7, and the combined algorithm model determining coefficient R2 is more than or equal to 0.97.
  7. 7. The computing method of claim 1, wherein the bingham fluid formula in S6 is i 2 =16/(3D)τ 0 +32v/D2 μ, D is tube diameter, and v is flow rate.
  8. 8. The calculation method according to claim 1, wherein the coefficient n of the multiple method in S7 ranges from 0.1 to 0.3, n=0.2 to 0.3 when the number of pipe bends is equal to or greater than 3, and n=0.1 to 0.2 when the number of bends is less than 3.
  9. 9. The method according to claim 1, wherein the genetic algorithm parameters in S9 are 100 population sizes, 50 iteration times, 0.7 crossover probability, 0.05 mutation probability, and the fitness function is the least mean square error between the predicted value and the measured value.

Description

Calculation method for pipe conveying resistance of full tailing filling slurry by considering unstable flow Technical Field The invention relates to the technical field of mine filling mining, in particular to a method for calculating the pipe conveying resistance of full-tailing filling slurry by considering unstable flow, which is suitable for engineering scenes of metal mines, nonmetal mines and the like adopting a filling mining method, and can be used for filling pipeline design, conveying parameter optimization and energy consumption prediction. Background The filling mining method has become a mainstream technology in the mining industry because of being capable of realizing the recycling of solid wastes such as tailings and controlling surface subsidence. The pipeline transportation is a core link of filling mining, and the calculation accuracy of resistance loss directly influences pipeline pipe diameter selection, pump pressure matching and transportation energy consumption control. In the prior art, the calculation of the transportation resistance loss of the filling slurry pipeline is mostly based on the assumption of stable flow, and the influence of unstable flow generated by abrupt flow state change when slurry flows into a horizontal straight pipe from a vertical pipe is ignored. In practical engineering, the existence of unstable flow sections can lead to significant increases in the loss of resistance along the way, which is not considered by the traditional model, resulting in large calculation errors (typically more than 30%). Meanwhile, the traditional rheological parameter calculation mostly adopts a linear regression or extremely poor method, interaction of multiple factors such as mass concentration, temperature, ash-sand ratio and the like is not fully considered, calculation accuracy is not enough, the intelligent algorithm is applied to resistance loss calculation and lacks systematic integration, stability and accuracy are difficult to achieve, in addition, the traditional test system is mostly designed with fixed pipe diameters, pipe transmission characteristics under different working conditions cannot be simulated, data acquisition points are limited, and pressure and flow rate change rules of a middle section of a pipeline are difficult to capture. Therefore, development of a full-pipeline resistance loss calculation method based on system test data by fully considering the influence of unstable flow and integrating multi-factor coupling effect and intelligent algorithm is needed to solve the technical problems of low calculation precision and poor applicability in the prior art. Disclosure of Invention The invention aims to provide a calculation method for the pipe conveying resistance of full tailing filling slurry taking unstable flow into consideration, which aims to solve the problems that the influence of the unstable flow is ignored, the calculation precision is insufficient and the applicability is poor in the prior art. In order to achieve the aim, the invention adopts the following technical scheme that the method for calculating the pipeline conveying resistance loss of the full tailing filling slurry comprises the steps of, S1, analyzing physical and chemical properties of full-tailing filling slurry, and obtaining a tailing grading, a crystalline structure, chemical components, a shape coefficient, an alkalinity coefficient and a slurry mass concentration application range, wherein the tailing grading comprises an uneven coefficient and a curvature coefficient; S2, constructing a rheological parameter calculation model of slurry yield stress tau 0 and plastic viscosity mu based on a multi-factor orthogonal test; s3, collecting pressure and flow rate data of each monitoring point of the pipeline under different working conditions through a variable-diameter pipeline transmission test system, wherein the working conditions comprise filling doubling lines, the filling doubling lines are defined as the ratio of the total length of the filling pipeline to the height difference of the starting point and the finishing point of the pipeline, and the ratio is recorded as N, N=L/H; S4, judging the flow state based on the Reynolds number, and establishing an unstable flow length calculation model through test data fitting; S5, constructing an unstable flow on-way resistance loss model by adopting a combined algorithm of a BP neural network, XGBoost and a genetic algorithm and taking the flow speed, the pipe diameter, the yield stress and the plastic viscosity as inputs, wherein the unstable flow on-way resistance loss is recorded as i 1; S6, calculating the stable flow on-way resistance loss by adopting a Bingham fluid formula, wherein the stable flow on-way resistance loss is recorded as i 2; S7, calculating local resistance loss by adopting a multiple method, and calibrating coefficients based on the on-site pipeline arrangement condition; s8, integrating stable flow, un