CN115490531A - Preparation method of hydrated magnesium silicate ceramic friction material - Google Patents
Preparation method of hydrated magnesium silicate ceramic friction material Download PDFInfo
- Publication number
- CN115490531A CN115490531A CN202110675324.5A CN202110675324A CN115490531A CN 115490531 A CN115490531 A CN 115490531A CN 202110675324 A CN202110675324 A CN 202110675324A CN 115490531 A CN115490531 A CN 115490531A
- Authority
- CN
- China
- Prior art keywords
- magnesium silicate
- friction material
- ceramic
- neural network
- training
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- C—CHEMISTRY; METALLURGY
- C04—CEMENTS; CONCRETE; ARTIFICIAL STONE; CERAMICS; REFRACTORIES
- C04B—LIME, MAGNESIA; SLAG; CEMENTS; COMPOSITIONS THEREOF, e.g. MORTARS, CONCRETE OR LIKE BUILDING MATERIALS; ARTIFICIAL STONE; CERAMICS; REFRACTORIES; TREATMENT OF NATURAL STONE
- C04B38/00—Porous mortars, concrete, artificial stone or ceramic ware; Preparation thereof
- C04B38/06—Porous mortars, concrete, artificial stone or ceramic ware; Preparation thereof by burning-out added substances by burning natural expanding materials or by sublimating or melting out added substances
- C04B38/063—Preparing or treating the raw materials individually or as batches
- C04B38/0635—Compounding ingredients
- C04B38/0645—Burnable, meltable, sublimable materials
-
- C—CHEMISTRY; METALLURGY
- C04—CEMENTS; CONCRETE; ARTIFICIAL STONE; CERAMICS; REFRACTORIES
- C04B—LIME, MAGNESIA; SLAG; CEMENTS; COMPOSITIONS THEREOF, e.g. MORTARS, CONCRETE OR LIKE BUILDING MATERIALS; ARTIFICIAL STONE; CERAMICS; REFRACTORIES; TREATMENT OF NATURAL STONE
- C04B35/00—Shaped ceramic products characterised by their composition; Ceramics compositions; Processing powders of inorganic compounds preparatory to the manufacturing of ceramic products
- C04B35/01—Shaped ceramic products characterised by their composition; Ceramics compositions; Processing powders of inorganic compounds preparatory to the manufacturing of ceramic products based on oxide ceramics
- C04B35/16—Shaped ceramic products characterised by their composition; Ceramics compositions; Processing powders of inorganic compounds preparatory to the manufacturing of ceramic products based on oxide ceramics based on silicates other than clay
- C04B35/20—Shaped ceramic products characterised by their composition; Ceramics compositions; Processing powders of inorganic compounds preparatory to the manufacturing of ceramic products based on oxide ceramics based on silicates other than clay rich in magnesium oxide, e.g. forsterite
-
- C—CHEMISTRY; METALLURGY
- C04—CEMENTS; CONCRETE; ARTIFICIAL STONE; CERAMICS; REFRACTORIES
- C04B—LIME, MAGNESIA; SLAG; CEMENTS; COMPOSITIONS THEREOF, e.g. MORTARS, CONCRETE OR LIKE BUILDING MATERIALS; ARTIFICIAL STONE; CERAMICS; REFRACTORIES; TREATMENT OF NATURAL STONE
- C04B35/00—Shaped ceramic products characterised by their composition; Ceramics compositions; Processing powders of inorganic compounds preparatory to the manufacturing of ceramic products
- C04B35/622—Forming processes; Processing powders of inorganic compounds preparatory to the manufacturing of ceramic products
- C04B35/62204—Forming processes; Processing powders of inorganic compounds preparatory to the manufacturing of ceramic products using waste materials or refuse
-
- C—CHEMISTRY; METALLURGY
- C04—CEMENTS; CONCRETE; ARTIFICIAL STONE; CERAMICS; REFRACTORIES
- C04B—LIME, MAGNESIA; SLAG; CEMENTS; COMPOSITIONS THEREOF, e.g. MORTARS, CONCRETE OR LIKE BUILDING MATERIALS; ARTIFICIAL STONE; CERAMICS; REFRACTORIES; TREATMENT OF NATURAL STONE
- C04B35/00—Shaped ceramic products characterised by their composition; Ceramics compositions; Processing powders of inorganic compounds preparatory to the manufacturing of ceramic products
- C04B35/622—Forming processes; Processing powders of inorganic compounds preparatory to the manufacturing of ceramic products
- C04B35/62227—Forming processes; Processing powders of inorganic compounds preparatory to the manufacturing of ceramic products obtaining fibres
- C04B35/62231—Forming processes; Processing powders of inorganic compounds preparatory to the manufacturing of ceramic products obtaining fibres based on oxide ceramics
- C04B35/6224—Fibres based on silica
-
- C—CHEMISTRY; METALLURGY
- C04—CEMENTS; CONCRETE; ARTIFICIAL STONE; CERAMICS; REFRACTORIES
- C04B—LIME, MAGNESIA; SLAG; CEMENTS; COMPOSITIONS THEREOF, e.g. MORTARS, CONCRETE OR LIKE BUILDING MATERIALS; ARTIFICIAL STONE; CERAMICS; REFRACTORIES; TREATMENT OF NATURAL STONE
- C04B35/00—Shaped ceramic products characterised by their composition; Ceramics compositions; Processing powders of inorganic compounds preparatory to the manufacturing of ceramic products
- C04B35/622—Forming processes; Processing powders of inorganic compounds preparatory to the manufacturing of ceramic products
- C04B35/62227—Forming processes; Processing powders of inorganic compounds preparatory to the manufacturing of ceramic products obtaining fibres
- C04B35/62231—Forming processes; Processing powders of inorganic compounds preparatory to the manufacturing of ceramic products obtaining fibres based on oxide ceramics
- C04B35/62263—Fibres based on magnesium oxide
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- C—CHEMISTRY; METALLURGY
- C04—CEMENTS; CONCRETE; ARTIFICIAL STONE; CERAMICS; REFRACTORIES
- C04B—LIME, MAGNESIA; SLAG; CEMENTS; COMPOSITIONS THEREOF, e.g. MORTARS, CONCRETE OR LIKE BUILDING MATERIALS; ARTIFICIAL STONE; CERAMICS; REFRACTORIES; TREATMENT OF NATURAL STONE
- C04B2235/00—Aspects relating to ceramic starting mixtures or sintered ceramic products
- C04B2235/02—Composition of constituents of the starting material or of secondary phases of the final product
- C04B2235/30—Constituents and secondary phases not being of a fibrous nature
- C04B2235/32—Metal oxides, mixed metal oxides, or oxide-forming salts thereof, e.g. carbonates, nitrates, (oxy)hydroxides, chlorides
- C04B2235/3217—Aluminum oxide or oxide forming salts thereof, e.g. bauxite, alpha-alumina
-
- C—CHEMISTRY; METALLURGY
- C04—CEMENTS; CONCRETE; ARTIFICIAL STONE; CERAMICS; REFRACTORIES
- C04B—LIME, MAGNESIA; SLAG; CEMENTS; COMPOSITIONS THEREOF, e.g. MORTARS, CONCRETE OR LIKE BUILDING MATERIALS; ARTIFICIAL STONE; CERAMICS; REFRACTORIES; TREATMENT OF NATURAL STONE
- C04B2235/00—Aspects relating to ceramic starting mixtures or sintered ceramic products
- C04B2235/02—Composition of constituents of the starting material or of secondary phases of the final product
- C04B2235/30—Constituents and secondary phases not being of a fibrous nature
- C04B2235/34—Non-metal oxides, non-metal mixed oxides, or salts thereof that form the non-metal oxides upon heating, e.g. carbonates, nitrates, (oxy)hydroxides, chlorides
- C04B2235/3427—Silicates other than clay, e.g. water glass
-
- C—CHEMISTRY; METALLURGY
- C04—CEMENTS; CONCRETE; ARTIFICIAL STONE; CERAMICS; REFRACTORIES
- C04B—LIME, MAGNESIA; SLAG; CEMENTS; COMPOSITIONS THEREOF, e.g. MORTARS, CONCRETE OR LIKE BUILDING MATERIALS; ARTIFICIAL STONE; CERAMICS; REFRACTORIES; TREATMENT OF NATURAL STONE
- C04B2235/00—Aspects relating to ceramic starting mixtures or sintered ceramic products
- C04B2235/02—Composition of constituents of the starting material or of secondary phases of the final product
- C04B2235/30—Constituents and secondary phases not being of a fibrous nature
- C04B2235/42—Non metallic elements added as constituents or additives, e.g. sulfur, phosphor, selenium or tellurium
- C04B2235/422—Carbon
- C04B2235/425—Graphite
-
- C—CHEMISTRY; METALLURGY
- C04—CEMENTS; CONCRETE; ARTIFICIAL STONE; CERAMICS; REFRACTORIES
- C04B—LIME, MAGNESIA; SLAG; CEMENTS; COMPOSITIONS THEREOF, e.g. MORTARS, CONCRETE OR LIKE BUILDING MATERIALS; ARTIFICIAL STONE; CERAMICS; REFRACTORIES; TREATMENT OF NATURAL STONE
- C04B2235/00—Aspects relating to ceramic starting mixtures or sintered ceramic products
- C04B2235/02—Composition of constituents of the starting material or of secondary phases of the final product
- C04B2235/30—Constituents and secondary phases not being of a fibrous nature
- C04B2235/44—Metal salt constituents or additives chosen for the nature of the anions, e.g. hydrides or acetylacetonate
- C04B2235/446—Sulfides, tellurides or selenides
-
- C—CHEMISTRY; METALLURGY
- C04—CEMENTS; CONCRETE; ARTIFICIAL STONE; CERAMICS; REFRACTORIES
- C04B—LIME, MAGNESIA; SLAG; CEMENTS; COMPOSITIONS THEREOF, e.g. MORTARS, CONCRETE OR LIKE BUILDING MATERIALS; ARTIFICIAL STONE; CERAMICS; REFRACTORIES; TREATMENT OF NATURAL STONE
- C04B2235/00—Aspects relating to ceramic starting mixtures or sintered ceramic products
- C04B2235/02—Composition of constituents of the starting material or of secondary phases of the final product
- C04B2235/30—Constituents and secondary phases not being of a fibrous nature
- C04B2235/44—Metal salt constituents or additives chosen for the nature of the anions, e.g. hydrides or acetylacetonate
- C04B2235/448—Sulphates or sulphites
-
- C—CHEMISTRY; METALLURGY
- C04—CEMENTS; CONCRETE; ARTIFICIAL STONE; CERAMICS; REFRACTORIES
- C04B—LIME, MAGNESIA; SLAG; CEMENTS; COMPOSITIONS THEREOF, e.g. MORTARS, CONCRETE OR LIKE BUILDING MATERIALS; ARTIFICIAL STONE; CERAMICS; REFRACTORIES; TREATMENT OF NATURAL STONE
- C04B2235/00—Aspects relating to ceramic starting mixtures or sintered ceramic products
- C04B2235/02—Composition of constituents of the starting material or of secondary phases of the final product
- C04B2235/50—Constituents or additives of the starting mixture chosen for their shape or used because of their shape or their physical appearance
- C04B2235/52—Constituents or additives characterised by their shapes
- C04B2235/5208—Fibers
- C04B2235/5212—Organic
-
- C—CHEMISTRY; METALLURGY
- C04—CEMENTS; CONCRETE; ARTIFICIAL STONE; CERAMICS; REFRACTORIES
- C04B—LIME, MAGNESIA; SLAG; CEMENTS; COMPOSITIONS THEREOF, e.g. MORTARS, CONCRETE OR LIKE BUILDING MATERIALS; ARTIFICIAL STONE; CERAMICS; REFRACTORIES; TREATMENT OF NATURAL STONE
- C04B2235/00—Aspects relating to ceramic starting mixtures or sintered ceramic products
- C04B2235/02—Composition of constituents of the starting material or of secondary phases of the final product
- C04B2235/50—Constituents or additives of the starting mixture chosen for their shape or used because of their shape or their physical appearance
- C04B2235/52—Constituents or additives characterised by their shapes
- C04B2235/5208—Fibers
- C04B2235/5216—Inorganic
-
- C—CHEMISTRY; METALLURGY
- C04—CEMENTS; CONCRETE; ARTIFICIAL STONE; CERAMICS; REFRACTORIES
- C04B—LIME, MAGNESIA; SLAG; CEMENTS; COMPOSITIONS THEREOF, e.g. MORTARS, CONCRETE OR LIKE BUILDING MATERIALS; ARTIFICIAL STONE; CERAMICS; REFRACTORIES; TREATMENT OF NATURAL STONE
- C04B2235/00—Aspects relating to ceramic starting mixtures or sintered ceramic products
- C04B2235/02—Composition of constituents of the starting material or of secondary phases of the final product
- C04B2235/50—Constituents or additives of the starting mixture chosen for their shape or used because of their shape or their physical appearance
- C04B2235/52—Constituents or additives characterised by their shapes
- C04B2235/5276—Whiskers, spindles, needles or pins
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Ceramic Engineering (AREA)
- Manufacturing & Machinery (AREA)
- Materials Engineering (AREA)
- Structural Engineering (AREA)
- Organic Chemistry (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Inorganic Chemistry (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Braking Arrangements (AREA)
Abstract
The invention belongs to the technical field of friction material preparation, in particular to a preparation method of a hydrated magnesium silicate ceramic friction material, which solves the problems of high preparation cost, large brake noise of the obtained friction material in the use process and the like of a high-performance friction material in the prior art, and comprises the following steps: preparing microporous ceramic particles; preparing magnesium silicate ceramic fibers; mixing the microporous ceramic particles, the magnesium silicate ceramic fibers and other raw materials; pressing and forming; training a neural network; preparing the hydrated magnesium silicate ceramic friction material. The hydrous magnesium silicate ceramic friction material prepared by the invention further reduces the brake noise in the use process while ensuring the excellent performance of the material, and greatly reduces the production cost and the weight of the product.
Description
Technical Field
The invention relates to the technical field of friction material preparation, in particular to a preparation method of a hydrated magnesium silicate ceramic friction material.
Background
The improvement of scientific technology enables automobile manufacturing to develop towards light weight and long service life, the development of automobiles in the future is bound to change day by day along with the improvement of automobile technology, and meanwhile, the automobile brake system and the automobile speed are required to be higher and higher. The friction material is a key material of the automobile braking friction transmission braking device, and the performance of the friction material is directly related to the safety and the stability of an automobile, the life safety of a driver and other major problems. The ideal friction material has the advantages of better stability, higher friction coefficient, smaller wear rate, higher friction coefficient, better wear resistance, low noise, long service life and the like.
Currently, the primary materials meeting the above criteria are low metal and non-metallic ceramic friction materials. Among them, the braking performance of less cermet friction materials is undoubtedly high, but the friction materials have the defects of high preparation cost, large braking noise in the use process and the like, and the comprehensive performance needs to be improved. Based on the statement, the invention provides a preparation method of a hydrated magnesium silicate ceramic friction material.
Disclosure of Invention
The invention aims to solve the problems of high preparation cost, high braking noise of the obtained friction material in the use process and the like of a high-performance friction material in the prior art, and provides a preparation method of a hydrated magnesium silicate ceramic friction material.
A preparation method of a hydrated magnesium silicate ceramic friction material comprises the following steps:
s1, preparing microporous ceramic particles:
mixing magnesium silicate powder with the premixed solution to prepare ceramic slurry, adding corn starch and a dispersing agent, carrying out ball milling, adding a catalyst and an initiator, casting into a blank, demolding, sintering to form magnesium silicate porous ceramic, and crushing into microporous ceramic particles;
s2, preparing magnesium silicate ceramic fibers:
mixing 10-30% of talcum powder, 40-60% of quartz sand powder and 10-40% of industrial soda ash by mass percent, melting and spinning to obtain ceramic fiber, putting the ceramic fiber into hydrochloric acid, heating to 80-90 ℃, keeping the temperature for 5-8h, washing with water, carrying out heat treatment, and crushing to obtain magnesium silicate ceramic fiber;
s3, mixing materials:
weighing 5-8% of potassium titanate whisker, 6-9% of aramid fiber, 2-3% of copper fiber, 4-7% of magnesium silicate ceramic fiber, 1-3% of alumina powder, 2-3% of zirconium silicate, 8-12% of adhesive, 5-10% of graphite, 3-5% of mica, 2-3% of fly ash, 3-5% of barite, 5-8% of metal sulfide and the balance of microporous ceramic particles according to mass percentage, adding the raw materials into a mixer together, and fully and uniformly mixing;
s4, press forming:
pressing and molding the uniformly mixed materials, and then carrying out heat treatment and surface treatment to obtain a friction material primary product;
s5, training a neural network:
randomly initializing 50 samples within the range limited by each parameter in the steps S1-S4, repeating the steps S1-S4 to prepare 50 friction material primary samples, carrying out performance test on the 50 friction material primary samples, inputting 50 test results into a neural network model, and training according to a neural network model training method, wherein the output of the neural network model is a vector formed by all undetermined parameters;
s6, preparing a hydrated magnesium silicate ceramic friction material:
and repeating the steps S1 to S4 according to the output result of the neural network model to obtain the low-noise, low-cost and low-weight hydrous magnesium silicate ceramic friction material.
Preferably, the premixed solution in the step S1 is prepared by compounding acrylamide and N, -methylene bisacrylamide in a mass ratio of 10-18.
Preferably, in the step S1, the solid content of the ceramic slurry is greater than 50% by mass, the addition amount of the corn starch is 10-15% of the magnesium silicate powder, and the addition amount of the dispersing agent is 5-7% of the magnesium silicate powder.
Preferably, the sintering conditions in step S1 are controlled as follows: firstly heating to 600 ℃, then controlling the heating rate to be 3-5 ℃/min, heating to 1300-1320 ℃, and preserving heat for 1.5h.
Preferably, in the step S4, the pressing pressure is 26-29MPa, the pressing temperature is 155-165 ℃, and the pressing time is 8-12min.
Preferably, the heat treatment temperature in the step S4 is 230-260 ℃, and the heat treatment time is 1.2h.
Preferably, the neural network model training method in step S5 specifically includes the following steps:
a. constructing a model network layer and randomly initializing parameters W, U and V;
b. forward propagation calculation:
hidden states for neurons of the first layer
From the input layer X t And last hidden state
Collectively, the calculation can be expressed as:
wherein U is (i) ，W (i) ，V (i) The parameters to be trained are required for the ith layer.
When from a hidden state
Enter the next hidden state
When, it is calculated as:
the final output result C t Comprises the following steps:
c. and (3) back propagation:
defining a loss value e over the entire sequence t ：
Wherein N is the length of the sequence,
is C t The mean value of (a);
the loss function E over the entire sequence is:
the gradient of the parameter is calculated with a gradient descent algorithm:
updating parameters:
according to the chain derivation rule, the partial derivative when the tth input can be obtained is:
the update of the parameter W is calculated as:
the update of the parameter U is calculated as:
the update of parameter V is calculated as:
where eta is the adaptive learning rate
d. And fusing output results through an attention mechanism:
inputting a result C obtained by passing through a three-layer neural network 1 To C t The output result after weighted mean value is O t The calculation formula is as follows:
wherein a is k As input features of the classifier, are composed ofThe formula is calculated to obtain:
a k ＝softmax(V T tanh(Wh k-1 +Uh k ))
e. dividing the measured data into a training set and a test set, inputting the training set into a model for training, and testing the training effect by using the test set.
The preparation method of the hydrous magnesium silicate ceramic friction material provided by the invention has the following beneficial effects:
1. the invention provides a preparation method of a hydrated magnesium silicate ceramic friction material, which comprises the steps of preparing microporous ceramic particles and magnesium silicate ceramic fibers, blending the microporous ceramic particles and the magnesium silicate ceramic fibers with other raw materials, then performing compression molding to obtain a friction material primary product, training a neural network of the friction material primary product by introducing a neural network model to obtain an output result, and finally producing and processing the friction material according to the output result to obtain the hydrated magnesium silicate ceramic friction material; the hydrous magnesium silicate ceramic friction material prepared by the invention further reduces the brake noise in the use process while ensuring the excellent performance of the material, and greatly reduces the production cost and the weight of the product.
2. According to the invention, the neural network model is introduced in the preparation process, the optimization process of the formula is greatly shortened, the optimal proportion of the parts of the materials required by the friction material under the condition of accurately budgeting the performance achievement required by the invention can be obtained, and compared with the preparation method in the prior art, the preparation method is more accurate and reasonable in formula design.
Drawings
FIG. 1 is a neural network model in the preparation method of a hydrous magnesium silicate ceramic friction material provided by the invention.
Detailed Description
The present invention will be further illustrated with reference to the following specific examples.
Example one
The invention provides a preparation method of a hydrated magnesium silicate ceramic friction material, which comprises the following steps:
s1, preparing microporous ceramic particles:
mixing magnesium silicate powder with the premixed solution to prepare ceramic slurry, adding corn starch and a dispersing agent, carrying out ball milling, adding a catalyst and an initiator, casting into a blank, demolding and sintering to form the magnesium silicate porous ceramic, wherein the sintering conditions are controlled as follows: firstly heating to 600 ℃, then controlling the heating rate to be 3-5 ℃/min, heating to 1300-1320 ℃, preserving heat for 1.5h, and crushing into microporous ceramic particles;
the premixed solution is prepared by compounding acrylamide and N, N-methylene bisacrylamide in a mass ratio of 10-18; the solid content of the ceramic slurry is more than 50 percent by mass percent, the adding amount of the corn starch is 10 to 15 percent of the magnesium silicate powder, and the adding amount of the dispersing agent is 5 to 7 percent of the magnesium silicate powder;
s2, preparing magnesium silicate ceramic fibers:
mixing 10-30% of talcum powder, 40-60% of quartz sand powder and 10-40% of industrial soda ash by mass percent, melting and spinning to obtain ceramic fiber, putting the ceramic fiber into hydrochloric acid, heating to 80-90 ℃, keeping the temperature for 5-8h, washing with water, carrying out heat treatment, and crushing to obtain magnesium silicate ceramic fiber;
s3, mixing materials:
weighing 5-8% of potassium titanate whisker, 6-9% of aramid fiber, 2-3% of copper fiber, 4-7% of magnesium silicate ceramic fiber, 1-3% of alumina powder, 2-3% of zirconium silicate, 8-12% of adhesive, 5-10% of graphite, 3-5% of mica, 2-3% of fly ash, 3-5% of barite, 5-8% of metal sulfide and the balance of microporous ceramic particles according to mass percentage, adding the raw materials into a mixer together, and fully and uniformly mixing;
s4, press forming:
pressing and molding the uniformly mixed material, wherein the pressing pressure is 26-29MPa, the pressing temperature is 155-165 ℃, the pressing time is 8-12min, then carrying out heat treatment and surface treatment, the heat treatment temperature is 230-260 ℃, the heat treatment time is 1.2h, and obtaining a friction material primary product after the treatment;
s5, training a neural network:
randomly initializing 50 samples within the range limited by each parameter in the steps S1-S4, repeating the steps S1-S4 to prepare 50 friction material primary samples, carrying out performance test on the 50 friction material primary samples, inputting 50 test results into a neural network model, and training according to a neural network model training method, wherein the output of the neural network model is a vector formed by all undetermined parameters;
the neural network model training method specifically comprises the following steps:
a. constructing a model network layer and randomly initializing parameters W, U and V;
b. forward propagation calculation:
hidden states for neurons of the first layer
From the input layer X t And last hidden state
Collectively, the calculation can be expressed as:
wherein U is (i) ，W (i) ，V (i) The parameters to be trained are required for the ith layer.
When from a hidden state
Enter the next hidden state
Then, it is calculated as:
the final output result C t Comprises the following steps:
c. and (3) back propagation:
defining a loss value e over the entire sequence t ：
Wherein N is the length of the sequence,
is C t The mean value of (a);
the loss function E over the entire sequence is:
the gradient of the parameter is calculated with a gradient descent algorithm:
updating parameters:
according to the chain derivation rule, the partial derivative at the t-th input can be obtained as follows:
the update of the parameter W is calculated as:
the update of the parameter U is calculated as:
the update of parameter V is calculated as:
where eta is the adaptive learning rate
d. And (5) fusing output results through an attention mechanism:
inputting a result C finally obtained after passing through three layers of neural networks 1 To C t The output result after weighted average is O t The calculation formula is as follows:
wherein a is k The input characteristic of the classifier is calculated by the following formula:
a k ＝softmax(V T tanh(Wh k-1 +Uh k ))
e. dividing the measured data into a training set and a test set, inputting the training set into a model for training, and testing the training effect by using the test set;
s6, preparing a hydrated magnesium silicate ceramic friction material:
and repeating the steps S1 to S4 according to the output result of the neural network model to obtain the low-noise, low-cost and low-weight hydrous magnesium silicate ceramic friction material.
Compared with the common friction materials in the market, the friction material prepared in the first embodiment of the invention has basically the same performances, wherein the friction material has slightly larger abrasion at the temperature of above 560 ℃ due to the existence of the fly ash, but the weight of the friction material is reduced by about 17% and the cost is reduced by about 13% under the same volume, and the friction material has huge competitive advantages.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.
Claims (7)
1. The preparation method of the hydrous magnesium silicate ceramic friction material is characterized by comprising the following steps of:
s1, preparing microporous ceramic particles:
mixing magnesium silicate powder with the premixed solution to prepare ceramic slurry, adding corn starch and a dispersing agent, carrying out ball milling, adding a catalyst and an initiator, casting into a blank, demolding, sintering to form magnesium silicate porous ceramic, and crushing into microporous ceramic particles;
s2, preparing magnesium silicate ceramic fibers:
mixing 10-30% of talcum powder, 40-60% of quartz sand powder and 10-40% of industrial soda ash by mass percent, melting and spinning to obtain ceramic fiber, putting the ceramic fiber into hydrochloric acid, heating to 80-90 ℃, keeping the temperature for 5-8h, washing with water, carrying out heat treatment, and crushing to obtain magnesium silicate ceramic fiber;
s3, mixing materials:
weighing 5-8% of potassium titanate whisker, 6-9% of aramid fiber, 2-3% of copper fiber, 4-7% of magnesium silicate ceramic fiber, 1-3% of alumina powder, 2-3% of zirconium silicate, 8-12% of adhesive, 5-10% of graphite, 3-5% of mica, 2-3% of fly ash, 3-5% of barite, 5-8% of metal sulfide and the balance of microporous ceramic particles according to mass percentage, adding the raw materials into a mixer together, and fully and uniformly mixing;
s4, press forming:
pressing and molding the uniformly mixed material, and then carrying out heat treatment and surface treatment to obtain a friction material primary product;
s5, training a neural network:
randomly initializing 50 samples within the range limited by each parameter in the steps S1-S4, repeating the steps S1-S4 to prepare 50 friction material primary samples, carrying out performance test on the 50 friction material primary samples, inputting 50 test results into a neural network model, and training according to a neural network model training method, wherein the output of the neural network model is a vector formed by all undetermined parameters;
s6, preparing a hydrated magnesium silicate ceramic friction material:
and repeating the steps S1 to S4 according to the output result of the neural network model to obtain the low-noise, low-cost and low-weight hydrous magnesium silicate ceramic friction material.
2. The method for preparing a hydrous magnesium silicate ceramic friction material as claimed in claim 1, wherein the premixed solution in step S1 is prepared by compounding acrylamide and N, -methylene bisacrylamide in a mass ratio of 10-18.
3. The method of claim 1, wherein the ceramic slurry of step S1 contains more than 50% by weight of solids, the corn starch is added in an amount of 10-15% by weight of the magnesium silicate powder, and the dispersant is added in an amount of 5-7% by weight of the magnesium silicate powder.
4. The method of claim 1, wherein the sintering conditions in step S1 are controlled as follows: firstly heating to 600 ℃, then controlling the heating rate to be 3-5 ℃/min, heating to 1300-1320 ℃, and preserving heat for 1.5h.
5. The method of claim 1, wherein the pressing pressure in step S4 is 26-29MPa, the pressing temperature is 155-165 ℃, and the pressing time is 8-12min.
6. The method of claim 1, wherein the heat treatment temperature in step S4 is 230-260 ℃ and the heat treatment time is 1.2h.
7. The method for preparing a hydrous magnesium silicate ceramic friction material as claimed in claim 1, wherein the neural network model training method in step S5 specifically comprises the steps of:
a. constructing a model network layer and randomly initializing parameters W, U and V;
b. forward propagation calculation:
hidden states for neurons of the first layer
From the input layer X t And last hidden state
Collectively, the calculation can be expressed as:
wherein U is (i) ,W (i) ，V (i) The parameters to be trained are required for the ith layer.
When from a hidden state
Enter the next hidden state
When, it is calculated as:
the final output result C t Comprises the following steps:
c. and (3) back propagation:
defining a loss value e over the entire sequence t ：
Wherein N is the length of the sequence,
is C t The mean value of (a);
the loss function E over the entire sequence is:
the gradient of the parameter is calculated with a gradient descent algorithm:
updating parameters:
according to the chain derivation rule, the partial derivative at the t-th input can be obtained as follows:
the update of the parameter W is calculated as:
the update calculation for parameter U is:
the update of parameter V is calculated as:
where eta is the adaptive learning rate
d. And fusing output results through an attention mechanism:
inputting a result C obtained by passing through a three-layer neural network 1 To C t The output result after weighted average is O t The calculation formula is as follows:
wherein a is k The input features of the classifier are calculated by the following formula:
a k ＝softmax(V T tanh(Wh k-1 +Uh k ))
e. dividing the measured data into a training set and a test set, inputting the training set into a model for training, and testing the training effect by using the test set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110675324.5A CN115490531A (en) | 2021-06-17 | 2021-06-17 | Preparation method of hydrated magnesium silicate ceramic friction material |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110675324.5A CN115490531A (en) | 2021-06-17 | 2021-06-17 | Preparation method of hydrated magnesium silicate ceramic friction material |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115490531A true CN115490531A (en) | 2022-12-20 |
Family
ID=84464042
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110675324.5A Pending CN115490531A (en) | 2021-06-17 | 2021-06-17 | Preparation method of hydrated magnesium silicate ceramic friction material |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115490531A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101591177A (en) * | 2009-06-19 | 2009-12-02 | 厦门大学 | A kind of preparation method of magnesium silicate porous ceramic |
| CN102515731A (en) * | 2011-12-08 | 2012-06-27 | 武汉科技大学 | Magnesium silicate ceramic fibre and preparation method therof |
| JP2012197352A (en) * | 2011-03-22 | 2012-10-18 | Nisshinbo Brake Inc | Friction material |
| CN103059807A (en) * | 2012-10-31 | 2013-04-24 | 瑞阳汽车零部件（仙桃）有限公司 | Friction material composition for high-performance brake pad |
| CN103077288A (en) * | 2013-01-23 | 2013-05-01 | 重庆科技学院 | Small sample test data-oriented method for soft measurement and formula decision of multielement alloy material |
| CN106589802A (en) * | 2017-02-20 | 2017-04-26 | 马乐群 | Disc brake pad |
| CN107502292A (en) * | 2017-08-10 | 2017-12-22 | 甘肃澜驰汽车配件有限公司 | One kind enhancing organic ceramic friction material and its compound method |
| CN110255303A (en) * | 2019-06-13 | 2019-09-20 | 天津大学 | A kind of elevator waiting floor optimization method |
| CN112413020A (en) * | 2020-10-19 | 2021-02-26 | 浙江万赛汽车零部件股份有限公司 | Preparation method of hydrated magnesium silicate ceramic friction material |
-
2021
- 2021-06-17 CN CN202110675324.5A patent/CN115490531A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101591177A (en) * | 2009-06-19 | 2009-12-02 | 厦门大学 | A kind of preparation method of magnesium silicate porous ceramic |
| JP2012197352A (en) * | 2011-03-22 | 2012-10-18 | Nisshinbo Brake Inc | Friction material |
| CN102515731A (en) * | 2011-12-08 | 2012-06-27 | 武汉科技大学 | Magnesium silicate ceramic fibre and preparation method therof |
| CN103059807A (en) * | 2012-10-31 | 2013-04-24 | 瑞阳汽车零部件（仙桃）有限公司 | Friction material composition for high-performance brake pad |
| CN103077288A (en) * | 2013-01-23 | 2013-05-01 | 重庆科技学院 | Small sample test data-oriented method for soft measurement and formula decision of multielement alloy material |
| CN106589802A (en) * | 2017-02-20 | 2017-04-26 | 马乐群 | Disc brake pad |
| CN107502292A (en) * | 2017-08-10 | 2017-12-22 | 甘肃澜驰汽车配件有限公司 | One kind enhancing organic ceramic friction material and its compound method |
| CN110255303A (en) * | 2019-06-13 | 2019-09-20 | 天津大学 | A kind of elevator waiting floor optimization method |
| CN112413020A (en) * | 2020-10-19 | 2021-02-26 | 浙江万赛汽车零部件股份有限公司 | Preparation method of hydrated magnesium silicate ceramic friction material |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5701744B2 (en) | Steel wire manufacturing method | |
| CN103252492B (en) | Preparation method for powder metallurgy for automotive brake valve block | |
| CN109626917A (en) | A kind of C70 concrete and preparation method thereof | |
| CN103252491B (en) | Method for preparing powder metallurgy of engine piston pin | |
| Yeddula et al. | Experimental investigations and prediction of thermal behaviour of ferrosialate-based geopolymer mortars | |
| CN115490531A (en) | Preparation method of hydrated magnesium silicate ceramic friction material | |
| CN113214597A (en) | Basalt fiber-based high-performance environment-friendly friction material and preparation method thereof | |
| CN102635654B (en) | Brake pad for motor vehicles and preparation method thereof | |
| CN112209680A (en) | High-strength light-transmitting concrete and preparation method thereof | |
| CN106223121B (en) | A kind of durable antibiotic glass fibre air filter paper and preparation method thereof | |
| CN110343977A (en) | A kind of preparation method of continuous carbon fibre reinforced aluminum matrix composites | |
| CN103899694A (en) | Heavy truck brake pad and manufacturing method thereof | |
| CN117172116A (en) | Concrete compressive strength prediction method based on PSO-BP combined model | |
| CN111308894B (en) | Parameter adjusting method for coupling ore blending and control process in sintering process | |
| CN103213230A (en) | Fracturing ball manufacturing method implemented by taking carbon fiber and polyether-ether-ketone as raw materials | |
| Abdellahi et al. | Modeling effect of SiO 2 nanoparticles on the mechanical properties of the concretes | |
| CN116597919A (en) | 3D printing cement-based material design method based on artificial intelligence | |
| CN106277748B (en) | A kind of antibacterial glass fibre reinforced plasterboard and its manufacturing method | |
| CN106277747B (en) | A kind of load silver/titanium dioxide compound glass is fine and its ties up preparation method | |
| Zhang et al. | Performance optimization of geopolymer mortar blending in nano-SiO2 and PVA fiber based on set pair analysis | |
| CN101497504B (en) | Method for improving gypsum gel strength gray model optimized grain composition | |
| CN106868354A (en) | A kind of Anti-corrosion composite metal material and preparation method thereof | |
| Surekha et al. | Forward and reverse mappings of the cement-bonded sand mould system using fuzzy logic | |
| CN105733140A (en) | High-strength and high-wear-resisting plastic valve element and preparation method thereof | |
| CN114656748B (en) | Phenolic resin composition, preparation method thereof and application thereof in preparation of stamping die |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |