CN105854998A - Development method for intelligent grinding ball grading distribution controller - Google Patents

Abstract

The invention discloses a development method for an intelligent grinding ball grading distribution controller. The method comprises the steps of reasonably selecting the type and specification of a ball mill; determining a dynamic ball supplement parameter engineering model, carrying out programing on the dynamic ball supplement parameter engineering model, and inputting a storage model program, an operation control unit and known conditions and parameters; designing main and auxiliary control circuits and elements, and constituting an internal structure and an external structure of instruments of the developed intelligent grinding ball grading distribution controller; and conducting field operation performance verification and perfection on the developed intelligent grinding ball grading distribution controller. The intelligent grinding ball grading distribution controller is developed by synthesizing various factors, creating a mathematical model as the theoretical basis through long-term research and development, programing and arranging electrical and instrument devices. An operator of each ball mill can scientifically make grinding balls be made best use in the ball mill to the maximum extent through the intelligent grinding ball grading distribution controller, so that the indicators of energy conservation, consumption reduction and cost saving are realized to the maximum extent.

Description

A kind of method developing abrading-ball intelligence grating controller

Technical field

The present invention relates to abrading-ball field, being specifically related to is a kind of side developing abrading-ball intelligence grating controller Method.

Background technology

China's high-abrasive material industrial market scale quickly increases, and overall manufacture level improves constantly.Along with state The economic stable development of the people, the field such as China's thermal power generation, metallurgical mine, building material cement is to wear-resistant ball section Demand also keeps the situation of steady growth, the demand of the most domestic wear-resistant ball section to reach 208.74 in 2015 Ten thousand tons, following table lists 2010-2015, the growth pattern of this series products domestic demand amount:

2010-2015 domestic abrasion-resistant cast abrading-ball section market demand (ten thousand tons)

According to incompletely statistics, in recent years, the export volume of China's year all kinds of wear-resistant ball section is 1,000,000 ton-150 Ten thousand tons.It is visible, though this market demand total amount is bigger.

Ball mills quantifies the scientific research methods of grating technology can be divided into four rank by its development level Section:

1. the trial and error stage；

2. orthogonal experiment method (experimental period, relatively costly, and large sample cannot be processed)；

3. the exact test method that visual information processes；

4. use Intelligentized Control Model.

Abroad, have begun to grope the 3rd step and the 4th in abrading-ball manufacture and use field in recent years The research methods of step and use related art method and instrument to formulate, replenishment of process, manufacture, use and Monitoring embodies the exquisite device of technique " marrow " content, but very different, makes slow progress.

At home, remain in application orthogonal experiment method now and carry out the second of scientific research and production application On stage, a large amount of dynamically changing factor is caused to can not get effectively controlling and predicting, product manufacturing and product Using that long-standing waste is big, efficiency is low, the problem that cost is high can not get solving always；Can not be effective Utilize the information technology of high speed development, intellectually and automatically technology to accelerate the technological progress in this field, send out The pressure waving the science and technology effect of taking the lead in race is increasing.And for a long time just because of domestic abrading-ball manufacturing technology, Process equipment falls behind, and production technology many decades is constant, thinks little of refining, causes iron and steel liquid degree of purity poor, Modification effect is unstable, and process of setting exists a large amount of segregation, variation and unstable tissue, causes internal group Knit the finest and close and surface quality is poor；And abrading-ball heart portion is inconsistent to mechanical property, the serviceability on surface, Directly affects the service life of abrading-ball.Due to by grinding ball material, ball milling equipment, material characteristic, work The multifactor restriction such as environment and interactive effects, research abrading-ball problem is all stuck in qualitative grinding the most both at home and abroad Study carefully the stage, to using abrading-ball generation to save energy and reduce the cost and to reduce the effect of production cost the most notable, unstable, Plus too relying on experience, causing cannot " awkward situation " of popularization and application.

At thermal power generation, metallurgical mine, the big industrial circle of building material cement three, numerous ball mill need to be used to use In pulverizing various mineral aggregates.For a long time, people are in order to extend the service life of abrading-ball, respectively in abrading-ball system Make and use abrading-ball the two field to carry out improvement and the mill of manufacture experience independently of one another with ball mill The improvement of the experience that ball uses, though effective, but DeGrain, to such an extent as to cause the great energy to disappear Consumption, environmental pollution greatly and higher cost, lack the competitiveness of product.In order to be supplied to user Produce the most economical purpose that can meet with performance and use the abrading-ball required, inventor, amass for many years to mill The design of ball and manufacturing experience, before, proposed entitled " a kind of for ball to State Patent Office Mill equipment abrading-ball quantifies the method for visualizing of grating technology ", sending out of Application No. " 201610161104X " Bright application, " design of a kind of high abrasion cast grinding ball composition and the method for visualizing of heat treating regime ", application Number it is patent application and " a kind of cast grinding ball level based on Visualization Platform of " 2016101611533 " Join Mathematical Modeling Methods ", give on the basis of the patent application of Application No. " 2016101794163 " and use Abrading-ball person proposes a kind of method developing abrading-ball intelligence grating controller and makes the operation of each table grinder Person can reach science by using abrading-ball intelligence grating controller, farthest makes abrading-ball at ball milling Machine reaches best application, thus realizes the most energy-saving and cost-reducing and cost-effective every finger Mark, thus to solve the problems referred to above.

Summary of the invention

It is an object of the invention to provide a kind of method developing abrading-ball intelligence grating controller, by the most all Multifactor, the Mathematical Modeling createed by long-term R & D is theoretical foundation, by program composition and electricity The setting of gas and metering device develops.It will enable the operator of each table grinder by making Reach science with abrading-ball intelligence grating controller, farthest make abrading-ball reach best in ball mill Application, thus realize farthest energy-saving and cost-reducing and cost-effective indices.

For reaching above-mentioned purpose, the technical scheme is that

A kind of method developing abrading-ball intelligence grating controller of the present invention, comprises the steps:

(1) ball is rationally selected according to the material behavior of abrasive material, the working environment of ball mill and job requirement Grinding machine kind, selects ball mill specification according to the production capacity demand of ball mill user；

(2) the most preferably different sphere diameter abrading-ball grading distribution scheme of ball mill, most economical material abrading-ball, abrading-ball are determined Optimal useful load and tip-in regimen cycle；

(3) dynamic tip-in parameter engineering model is determined；

(4) the dynamic tip-in parameter engineering model used is carried out program composition, and input storage mould Type program and arithmetic and control unit；

(5) program that described step (4) is worked out inputs known conditions and parameter；

(6) design the main and control circuit of auxiliary and element, and form developed abrading-ball intelligent level The art designing joining controller instrument external and internal compositions and appearance processes；

(7) to the model machine classifying type of the abrading-ball intelligence grating controller developed and different size series Abrading-ball intelligence grating controller carries out field operation performance verification and perfect.

Preferably, the described step (1) material behavior according to abrasive material, the working environment of ball mill, when Under conditions of when abrasive material is ferrous metal ore or nonferrous metals ore, ball mill selects metal ore mill；Work as mill Under conditions of when material is nonmetal soft ore deposit or nonmetal hard ore deposit, ball mill differential grinding machine；According to ball The job requirement of grinding machine, when ball mill ball mill under the conditions of requiring dry grinding or wet-milling selects coal pulverizer.

Preferably, the described production capacity demand according to ball mill user of described step (1) selects ball Grinding machine specification, specifically ball mill is according to treating the productivity ratio of ball milling product, production yields, scale and production The requirement of power is different, is divided into small grinder, medium-sized ball mill, large-size ball mill and Extra large ball mill, It is specially and rationally selects ball mill specification according to user's request.

Preferably, described in step (2), optimal different sphere diameter abrading-ball grading distribution schemes use four kinds of different balls Footpath abrading-ball, and different sphere diameter abrading-ball grading distribution scheme is specially Φ 60: Φ 50: Φ 40: Φ 30=11.65: 50.11:28.17:10.07, most economical material abrading-ball is the Cr grinding balls of chromium content 10.86%, and abrading-ball is Good useful load is 36.23t, and tip-in regimen cycle is 7 days.

Preferably, described in step (3), dynamic tip-in parameter engineering model is Ball wear matrix model Or neural network prediction abrading-ball tip-in scale model.

Preferably, the foundation of described Ball wear matrix model comprises the steps:

A1: determine that Ball wear steady-state characteristic matrix is [p (t)],

$\[p\left(t\right)\]=\left[\begin{array}{cccc}0.9897& 0& 0& 0\\ 0.0095& 0.9853& 0& 0\\ 0.0006& 0.0135& 0.9888& 0\\ 0& 0.0008& 0.0109& 0.9885\end{array}\right];---S1$

A2: determine preferably dress ball grating vector { x₀And dress ball grating vector { x}, the 1 billiard ball mill of reality The specified ball load of equipment is W, and its Ball wear steady-state characteristic matrix is [p (t)], preferably dress ball level Orientation amount { x₀And actual dress ball grating vector x} respectively:

{x₀}={ x₁₀,x₂₀,…,x_k0,…,x_n0}^T, S2

{ x}={x₁,x₂,…,x_k,…,x_n}^T； S3

A3: after t days continuous coal-grindings, the kth rank abrading-ball resident amount in cylinder is with preferable The difference joining ball amount is △ W_k:

${\mathrm{\ΔW}}_k=W\·\[x_{k0}-\underset{j=1}{\overset{k}{\Σ}}{p}_{kj}\left(t\right)\·x_j\];---S4$

A4: the total amount that should add is △ W:

$\mathrm{\Δ}W=\underset{k=1}{\overset{n}{\Σ}}{\mathrm{\ΔW}}_k;---S5$

A5: j-th stage not rationally tip-in amount is △ W_BjCriterion, i.e. tip-in principle condition characterize Formula is as follows:

A1. whenTime, △ W_Bj=0；

A2. whenTime, take

A3. whenTime, take △ W_Bj=△ W_j；

A6: after tip-in, abrading-ball grating becomes:

x_Bj=x_j-△W_j/W+△W_Bj/ W, j=1,2 ..., n； S6

A7: utilize formula S 1-S6, utilizes MATLAB Programming with Pascal Language to calculate the ginseng of front 7 tip-in amounts Number.

Preferably, the foundation of described neural network prediction abrading-ball tip-in scale model comprises the steps:

B1: determine that neural network prediction abrading-ball tip-in scale model is [p (t)],

[p (t)]={ y} × { x}^-1

$\left\{\begin{array}{c}{P}_{60}={b_3}_1+\underset{j=1}{\overset{2}{\Σ}}\{{w_2}_{j1}\frac{2}{1+e^{-2_1{w}_{ij}{x}_i+{b_2}_j}}-1\}\\ P_{50}={b_3}_2+\underset{j=1}{\overset{2}{\Σ}}\{{w_2}_{j1}\frac{2}{1+e^{-2_1{w}_{ij}{x}_i+{b_2}_j}}-1\}\\ P_{40}={b_3}_3+\underset{j=1}{\overset{2}{\Σ}}\{{w_2}_{j1}\frac{2}{1+e^{-2_1{w}_{ij}{x}_i+{b_2}_j}}-1\}\\ P_{30=100-}{P}_{60-}{P}_{50-}{P}_{40}\end{array}\right.$

W1=

0.68284337499014

0.60406698796090

W2=

-0.20274877882385 0.20451645869560

-0.30974740916445 0.32924537190527

-0.18059072627341 0.10141961262160

B1=

-1.31928900327032

-1.29637427714563

B2=

0.99988918733054

0.99909388384959

1.00065821523216；

B2: j-th stage not rationally tip-in amount is △ W_BjConditional criterion:

WhenTime, △ W_Bj=0；

B3: use the neural network prediction abrading-ball tip-in scale model of described step B3 gained to combine artificial Neural network topology structure is identified training to the grading distribution scheme of optimal different sphere diameter abrading-balls, sets up optimal Quantify dynamic tip-in model scheme.

Preferably, known conditions and parameter described in step (5) are the most preferably different sphere diameter abrading-ball of ball mill Grading distribution scheme, the optimal useful load of abrading-ball and tip-in regimen cycle.

The beneficial effects of the present invention is,

(1) present invention is in ball mill running, and ball mill is whether in single hop or multistage workspace All can be adjusted by dress tip-in model and load under conditions of setting abrasive material and medium form preferable pack completeness Abrading-ball kind with change different sphere diameter abrading-ball additions ratio can make high-quality abrading-ball adapt to difference work Make environment, give full play to its excellent serviceability, embody the most energy-saving and cost-reducing and reduce ton product The effect of production cost；And effectively utilize the information technology of high speed development, intellectually and automatically technology to carry Go out a kind of to carry out to dynamic tip-in parameter engineering model with to the dynamic tip-in parameter engineering model used Combining of program composition accelerates the technological progress in this field；Controller of the present invention is with the most many Factor, the Mathematical Modeling createed by long-term R & D is theoretical foundation, by program composition with electric And the setting of metering device develops.It will enable the operator of each table grinder by using Abrading-ball intelligence grating controller reaches science, farthest makes abrading-ball reach best in ball mill Application, thus realize the most energy-saving and cost-reducing and cost-effective indices；

(2) present invention is in grinding machine running, and the abrading-ball of the various sphere diameters participating in grating all occurs Dynamic wear in various degree, just can maintain optimal at certain reasonable period by correct tip-in Abrading-ball grating, could control the effect amplitude at cyclic swing in minimum scope；Accordingly, propose (taking few experimental data, to calculate the abrading-ball in cycle instantly total as small sample to take Ball wear matrix model The actual wear amount that wear extent and every kind of sphere diameter abrading-ball produce), find to calculate after the 6th tip-in, dynamic State abrading-ball grating " is collapsed ", but front 5 grating characteristics are preferable, and obvious Ball wear matrix model forecasts Precision meets small sample requirement；Propose to set up neural network prediction abrading-ball tip-in scale model as full-page proof Originally, and it has the result of the dress tip-in effect without time integral error, and it can be directly at industrial equipment On carry out " test ", model data is directed to the situation of practical systems equipment, and model is reliable, knot Fruit is used directly for the industrial equipment that test uses.Need not do from lab scale.And combined with intelligent level Joining controller, the operator that can effectively realize each table grinder can be by using abrading-ball intelligence grating Controller reaches science, farthest makes abrading-ball reach best application in ball mill, reduces mill Ball cost depletions, improves abrading-ball service life, simplifies operating procedure, improves production efficiency；

(3) present invention is generally directed to (account for work for thermal power generation, mine and the big industrial circle of cement three Industry field uses more than the 90% of abrading-ball ball mill quantity), exploitation is for small grinder, medium-sized ball milling The abrading-ball grating intelligent controller of machine and large-size ball mill and Extra large ball mill, can realize different abrasive materials Material behavior, the working environment of ball mill, select ball according to the production capacity demand of ball mill user Grinding machine specification, design this three major types each tool its adapt to condition of work own, there is own characteristic , and by being programmed into row operation and realizing the abrading-ball intelligence grating controller controlled, give full play to abrading-ball Serviceability, thus energy-saving and cost-reducing and reduce production cost, there is the feature of applied range.

Detailed description of the invention

Below in conjunction with being embodied as example, the present invention is described in further details.

At thermal power generation, metallurgical mine, the big industrial circle of building material cement three, numerous ball mill need to be used to use In pulverizing various mineral aggregates.For a long time, people are in order to extend the service life of abrading-ball, respectively in abrading-ball system Make and use abrading-ball the two field to carry out improvement and the mill of manufacture experience independently of one another with ball mill The improvement of the experience that ball uses, though effective, but DeGrain, to such an extent as to cause the great energy to disappear Consumption, environmental pollution greatly and higher cost, lack the competitiveness of product.At aforementioned point of three major types On the premise of configuration abrading-ball intelligence grating controller, the problem that we it is also contemplated that be each big class due to Family is different for the requirement of production rate, production yields, scale and productivity, is generally divided into small-sized Ball mill, medium-sized ball mill, large-size ball mill and Extra large ball mill four kinds, ball mill diversified in specifications Sample, the various diameters mainly causing used abrading-ball sphere diameter size, drum's speed of rotation, grating to be used are big Little abrading-ball specification and ratio and addition are the most different, and this just determines us in certain kind series of design The merit that on the abrading-ball intelligence grating controller that ball mill is used, the type series specification to be considered is had Energy and the different requirements of operating characteristic.

The most economical purpose that can meet the abrading-ball using requirement with performance, this case is produced in order to be supplied to user Inventor makes the operator of each table grinder can reach by using abrading-ball intelligence grating controller Arrive science, farthest make abrading-ball reach best application in ball mill, thus realize maximum journey Energy-saving and cost-reducing and the cost-effective indices of degree, thus to solve the problems referred to above.

Embodiment

A kind of method developing abrading-ball intelligence grating controller, comprises the steps:

(1) ball is rationally selected according to the material behavior of abrasive material, the working environment of ball mill and job requirement Grinding machine kind, selects ball mill specification according to the production capacity demand of ball mill user；When abrasive material is black Under conditions of when non-ferrous metal ore deposit or nonferrous metals ore, ball mill selects metal ore mill；Described according to abrasive material Material behavior, the working environment of ball mill, particularly as follows: when abrasive material be nonmetal soft ore deposit or nonmetal Under conditions of during hard ore deposit, ball mill differential grinding machine；According to the job requirement of ball mill, work as ball mill Ball mill under the conditions of requiring dry grinding or wet-milling selects coal pulverizer；The described life according to ball mill user Producing ability need and select ball mill specification, specifically ball mill is according to treating the productivity ratio of ball milling product, production Yield, scale are different with the requirement of productivity, are divided into small grinder, medium-sized ball mill, large-scale ball milling Machine and Extra large ball mill, be specially and rationally select ball mill specification according to user's request.

(2) the most preferably different sphere diameter abrading-ball grading distribution scheme of ball mill, most economical material abrading-ball, abrading-ball are determined Optimal useful load and tip-in regimen cycle；

Before, proposed to State Patent Office and entitled " a kind of quantified grating for ball-grinding machine abrading-ball The method for visualizing of technology " patent application, the patent application of Application No. " 201610161104X " The grinding machine dress tip-in technical solution that the system that proposes is complete, uses method for visualizing to obtain abrading-ball Good useful load, most economical Material quality of grinding balls and optimal abrading-ball grading distribution scheme, by optimal different sphere diameter abrading-balls The determination of grading distribution scheme high-quality abrading-ball can be made to adapt to different operating environment, give full play to its excellent making Use performance；Table 1 is visualization optimum results system, as shown in table 1, by four kinds of different sphere diameters Influence factor (armoured material, abrasives, abrading-ball useful load, abrading-ball sphere diameter proportioning, the abrading-ball of abrading-ball Alloying component and heat treating regime) to grating performance assessment criteria (abrasion of energy consumption of mill, abrading-ball, flour extraction, Production cost) map visual analyzing, optimization, consider further that some technological factor, it is determined that M2 Group (Φ 60: Φ 50: Φ 40: Φ 30=11.65:50.11:28.17:10.07) can be as optimum gradation Scheme.So that it is determined that abrading-ball optimal useful load is: 36.23t；Most economical preferred Material quality of grinding balls is: High-Cr grinding balls containing about Cr10.86%, optimal tip-in regimen cycle is 7 days.It is achieved thereby that to chromium system Material quality of grinding balls, with grating performance assessment criteria as test stone, becomes with the influence factor of performance assessment criteria for experiment Amount, finally achieves and is using 4 kinds of sphere diameter abrading-balls of chromium material at optimal abrading-ball useful load and most economical abrading-ball The visualized operation of the grading distribution scheme of the optimal different sphere diameter abrading-balls under the conditions of material.

Table 1 visualizes optimum results system

(3) dynamic tip-in parameter engineering model is determined；

Dynamically tip-in parameter engineering model includes that Ball wear matrix model and neural network prediction abrading-ball are mended Ball scale model；The Cr grinding balls being chromium content 10.86% with most economical material abrading-ball, optimal four kinds of different balls Footpath abrading-ball grading distribution scheme is Φ 60: Φ 50: Φ 40: Φ 30=11.65:50.11:28.17:10.07, The optimal useful load of abrading-ball is 36.23t, based on tip-in regimen cycle is 7 days.

1) foundation of described Ball wear matrix model comprises the steps:

A1: determine that Ball wear steady-state characteristic matrix is [p (t)],

$\[p\left(t\right)\]=\left[\begin{array}{cccc}0.9897& 0& 0& 0\\ 0.0095& 0.9853& 0& 0\\ 0.0006& 0.0135& 0.9888& 0\\ 0& 0.0008& 0.0109& 0.9885\end{array}\right];---S1$

A2: determine preferably dress ball grating vector { x₀And dress ball grating vector { x}, the 1 billiard ball mill of reality The specified ball load of equipment is W, and its Ball wear steady-state characteristic matrix is [p (t)], preferably dress ball level Orientation amount { x₀And actual dress ball grating vector x} respectively:

{x₀}={ x₁₀,x₂₀,L,x_k0,…,x_n0}^T, S2

{ x}={x₁,x₂,…,x_k,…,x_n}^T； S3

A3: after t days continuous coal-grindings, the kth rank abrading-ball resident amount in cylinder is with preferable The difference joining ball amount is △ W_k:

${\mathrm{\ΔW}}_k=W\·\[x_{k0}-\underset{j=1}{\overset{k}{\Σ}}{p}_{kj}\left(t\right)\·x_j\];---S4$

A4: the total amount that should add is △ W:

$\mathrm{\Δ}W=\underset{k=1}{\overset{n}{\Σ}}{\mathrm{\ΔW}}_k;---S5$

A5: j-th stage not rationally tip-in amount is △ W_BjCriterion, i.e. tip-in principle condition characterize Formula is as follows:

A1. whenTime, △ W_Bj=0；

A2. whenTime, take

A3. whenTime, take △ W_Bj=△ W_j；

A6: after tip-in, abrading-ball grating becomes:

x_Bj=x_j-△W_j/W+△W_Bj/ W, j=1,2 ..., n； S6

A7: utilize formula S 1-S6, utilizes MATLAB Programming with Pascal Language to calculate the ginseng of front 7 tip-in amounts Number.

2) foundation of described neural network prediction abrading-ball tip-in scale model comprises the steps:

B1: determine that neural network prediction abrading-ball tip-in scale model is [p (t)],

[p (t)]={ y} × { x}^-1

$\left\{\begin{array}{c}{P}_{60}={b_3}_1+\underset{j=1}{\overset{2}{\Σ}}\{{w_2}_{j1}\frac{2}{1+e^{-2_1{w}_{ij}{x}_i+{b_2}_j}}-1\}\\ P_{50}={b_3}_2+\underset{j=1}{\overset{2}{\Σ}}\{{w_2}_{j1}\frac{2}{1+e^{-2_1{w}_{ij}{x}_i+{b_2}_j}}-1\}\\ P_{40}={b_3}_3+\underset{j=1}{\overset{2}{\Σ}}\{{w_2}_{j1}\frac{2}{1+e^{-2_1{w}_{ij}{x}_i+{b_2}_j}}-1\}\\ P_{30=100-}{P}_{60-}{P}_{50-}{P}_{40}\end{array}\right.$

W1=

0.68284337499014

0.60406698796090

W2=

-0.20274877882385 0.20451645869560

-0.30974740916445 0.32924537190527

-0.18059072627341 0.10141961262160

B1=

-1.31928900327032

-1.29637427714563

B2=

0.99988918733054

0.99909388384959

1.00065821523216；

B2: j-th stage not rationally tip-in amount is △ W_BjConditional criterion:

WhenTime, △ W_Bj=0；

B3: use the neural network prediction abrading-ball tip-in scale model of described step B3 gained to combine artificial Neural network topology structure is identified training to the grading distribution scheme of optimal different sphere diameter abrading-balls, sets up optimal Quantify dynamic tip-in scheme.

(4) the dynamic tip-in parameter engineering model used is carried out program composition, and input storage mould Type program and arithmetic and control unit；

(5) program that described step (4) is worked out inputs known conditions and parameter, the most most preferably Four kinds of different sphere diameter abrading-ball grading distribution schemes are Φ 60: Φ 50: Φ 40: Φ 30=11.65:50.11:28.17: 10.07, the optimal useful load of abrading-ball is 36.23t, and tip-in regimen cycle is 7 days；

(6) design the main and control circuit of auxiliary and element, and form developed abrading-ball intelligent level The art designing joining controller instrument external and internal compositions and appearance processes；

(7) to the model machine classifying type of the abrading-ball intelligence grating controller developed and different size series Abrading-ball intelligence grating controller carries out field operation performance verification and perfect.

Based on above-mentioned, the present invention is in ball mill running, and ball mill is whether at single hop or multistage All can be under conditions of setting abrasive material and medium form preferable pack completeness in workspace, by dress tip-in model Adjusting the abrading-ball kind loaded can make high-quality abrading-ball to fit with the ratio changing different sphere diameter abrading-ball additions Answer different operating environment, give full play to its excellent serviceability, embody the most energy-saving and cost-reducing and fall The effect of low ton production cost；And effectively utilize the information technology of high speed development, automation and intelligence Change technology proposes a kind of to dynamic tip-in parameter engineering model with to the dynamic tip-in parameter engineering used Model carries out combining of program composition and accelerates the technological progress in this field；Controller of the present invention with Comprehensive factors, the Mathematical Modeling createed by long-term R & D is theoretical foundation, is compiled by program System and electric and metering device setting develop.It will enable the operator of each table grinder By using abrading-ball intelligence grating controller to reach science, abrading-ball is farthest made to reach in ball mill To best application, thus realize the most energy-saving and cost-reducing and cost-effective indices；And it is main (to account for industrial circle for for thermal power generation, mine and the big industrial circle of cement three and to use abrading-ball ball milling More than the 90% of machine quantity), exploitation is for small grinder, medium-sized ball mill and large-size ball mill and spy The abrading-ball grating intelligent controller of large-size ball mill, can realize the different material behavior of abrasive material, ball mills Working environment, select ball mill specification according to the production capacity demand of ball mill user, design this Each tool of three major types its adapt to condition of work own, there is own characteristic, and carried out by programming The abrading-ball intelligence grating controller that computing and realization control, gives full play to the serviceability of abrading-ball, thus saves Can lower consumption and reduce production cost, there is the feature of applied range.

As known by the technical knowledge, the present invention can be by other essence without departing from its spirit or essential feature Embodiment realize.Therefore, embodiment disclosed above, for each side, is all to lift Example illustrates, is not only.All within the scope of the present invention or within being equal to the scope of the present invention Change and all comprised by the present invention.

Claims (8)

1. the method developing abrading-ball intelligence grating controller, it is characterised in that comprise the steps:

(1) ball mill is rationally selected according to the material behavior of abrasive material, the working environment of ball mill and job requirement Kind, selects ball mill specification according to the production capacity demand of ball mill user；

(2) determining the most preferably different sphere diameter abrading-ball grading distribution scheme of ball mill, most economical material abrading-ball, abrading-ball is optimal Useful load and tip-in regimen cycle；

(3) dynamic tip-in parameter engineering model is determined；

(4) the dynamic tip-in parameter engineering model used is carried out program composition, and input storage model journey Sequence and arithmetic and control unit；

(5) program that described step (4) is worked out inputs known conditions and parameter；

(6) design the main and control circuit of auxiliary and element, and form developed abrading-ball intelligence grating control The art designing of instruments and meters external and internal compositions processed and appearance processes；

(7) model machine classifying type and the abrading-ball of different size series to the abrading-ball intelligence grating controller developed Intelligence grating controller carries out field operation performance verification and perfect.

A kind of method developing abrading-ball intelligence grating controller the most according to claim 1, its feature exists In material behavior according to abrasive material of, described step (1), the working environment of ball mill, when abrasive material is black gold Under conditions of when genus ore deposit or nonferrous metals ore, ball mill selects metal ore mill；When abrasive material is nonmetal soft ore deposit Or under conditions of during nonmetal hard ore deposit, ball mill differential grinding machine；According to the job requirement of ball mill, work as ball Grinding machine ball mill under the conditions of requiring dry grinding or wet-milling selects coal pulverizer.

A kind of method developing abrading-ball intelligence grating controller the most according to claim 1, its feature exists In, the described production capacity demand according to ball mill user of described step (1) selects ball mill specification, tool Body ground ball mill, according to treating that the productivity ratio of ball milling product, production yields, scale are different with the requirement of productivity, divides For small grinder, medium-sized ball mill, large-size ball mill and Extra large ball mill, it is specially according to user's request Rationally select ball mill specification.

A kind of method developing abrading-ball intelligence grating controller the most according to claim 1, its feature exists In, optimal different sphere diameter abrading-ball grading distribution schemes four kinds of different sphere diameter abrading-balls of employing described in step (2), and not It is specially Φ 60: Φ 50: Φ 40: Φ 30=11.65:50.11:28.17:10.07 with sphere diameter abrading-ball grading distribution scheme, Most economical material abrading-ball is the Cr grinding balls of chromium content 10.86%, and the optimal useful load of abrading-ball is 36.23t, tip-in side The case cycle is 7 days.

A kind of method developing abrading-ball intelligence grating controller the most according to claim 1, its feature exists In, dynamic tip-in parameter engineering model described in step (3) be Ball wear matrix model or neutral net pre- Report abrading-ball tip-in scale model.

A kind of method developing abrading-ball intelligence grating controller the most according to claim 5, its feature exists In, the foundation of described Ball wear matrix model comprises the steps:

A1: determine that Ball wear steady-state characteristic matrix is [p (t)],

$\[p\left(t\right)\]=\left[\begin{array}{cccc}0.9897& 0& 0& 0\\ 0.0095& 0.9853& 0& 0\\ 0.0006& 0.0135& 0.9888& 0\\ 0& 0.0008& 0.0109& 0.9885\end{array}\right];---S1$

A2: determine preferably dress ball grating vector { x₀And dress ball grating vector { x}, 1 ball-grinding machine of reality Specified ball load be W, its Ball wear steady-state characteristic matrix is [p (t)], preferably dress ball grating vector { x₀} With actual dress ball grating vector x} respectively:

{x₀}={ x₁₀,x₂₀,…,x_k0,…,x_n0}^T, S2

{ x}={x₁,x₂,…,x_k,…,x_n}^T； S3

A3: after t days continuous coal-grindings, the kth rank abrading-ball resident amount in cylinder and ideal join ball The difference of amount is Δ W_k:

${\mathrm{\ΔW}}_k=W\·\[x_{k0}-\underset{j=1}{\overset{k}{\mathrm{\Σ}}}{p}_{kj}\left(t\right)\·x_j\];---S4$

A4: the total amount that should add is Δ W:

$\mathrm{\Δ}W=\underset{k=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\ΔW}}_k;---S5$

A5: j-th stage not rationally tip-in amount is Δ W_BjCriterion, i.e. tip-in principle condition characterize formula As follows:

A1. whenTime, Δ W_Bj=0；

A2. whenTime, take

A3. whenTime, take Δ W_Bj=Δ W_j；

A6: after tip-in, abrading-ball grating becomes:

x_Bj=x_j-ΔW_j/W+ΔW_Bj/ W, j=1,2 ..., n； S6

A7: utilize formula S 1-S6, utilizes MATLAB Programming with Pascal Language to calculate the parameter of front 7 tip-in amounts.

A kind of method developing abrading-ball intelligence grating controller the most according to claim 5, its feature exists In, the foundation of described neural network prediction abrading-ball tip-in scale model comprises the steps:

B1: determine that neural network prediction abrading-ball tip-in scale model is [p (t)],

[p (t)]={ y} × { x}^-1

$\left\{\begin{array}{c}{P}_{60}={b_3}_1+\underset{j=1}{\overset{2}{\mathrm{\Σ}}}\{{w_2}_{j1}\frac{2}{1+e^{-2_1{w}_{ij}{x}_i+{b_2}_j}}-1\}\\ P_{50}={b_3}_2+\underset{j=1}{\overset{2}{\mathrm{\Σ}}}\{{w_2}_{j1}\frac{2}{1+e^{-2_1{w}_{ij}{x}_i+{b_2}_j}}-1\}\\ P_{40}={b_3}_3+\underset{j=1}{\overset{2}{\mathrm{\Σ}}}\{{w_2}_{j1}\frac{2}{1+e^{-2_1{w}_{ij}{x}_i+{b_2}_j}}-1\}\\ P_{30}=100-P_{60}-P_{50}-P_{40}\end{array}\right.$

W1=

0.68284337499014

0.60406698796090

W2=

-0.20274877882385 0.20451645869560

-0.30974740916445 0.32924537190527

-0.18059072627341 0.10141961262160

B1=

-1.31928900327032

-1.29637427714563

B2=

0.99988918733054

0.99909388384959

1.00065821523216；

B2: j-th stage not rationally tip-in amount is Δ W_BjConditional criterion:

WhenTime, Δ W_Bj=0；

B3: use the neural network prediction abrading-ball tip-in scale model of described step B3 gained to combine artificial neuron Network topology structure is identified training to the grading distribution scheme of optimal different sphere diameter abrading-balls, sets up optimal quantization dynamic Tip-in model scheme.

A kind of method developing abrading-ball intelligence grating controller the most according to claim 1, its feature exists In, known conditions and parameter described in step (5) they are the most preferably different sphere diameter abrading-ball grading distribution scheme of ball mill, mill The optimal useful load of ball and tip-in regimen cycle.