High-speed-reduction brake pad based on neural network model matching optimization technology and preparation method of high-speed-reduction brake pad

Abstract

The invention discloses a high-speed-reduction brake pad based on a neural network model matching optimization technology and a preparation method of the high-speed-reduction brake pad. A friction material formula of the high-speed-reduction brake pad comprises steel fibers, friction powder, granular potassium hexatitanate, aramid fibers, mineral fibers, tin sulfide, graphite powder, coke powder, zirconium silicate / zirconium oxide / barium sulfate composite powder, calcium hydroxide and mica. The formula of the friction material is optimized through the neural network model, the friction coefficient, the high temperature resistance, the wear resistance and the comfort of the brake pad are remarkably improved, so that the braking performance of an automobile is improved, and the optimal proportion of the material parts required by the friction material is obtained by introducing the neural network model in the preparation process. Compared with a preparation method in the prior art, the scheme has the advantages that the formula optimization process is greatly shortened, required performance achievements can be obtained through accurate budgeting, the performance of the brake pad is remarkably improved, environmental protection is achieved, and the functions of friction plate replacement prompting and high-temperature alarming are provided.

Description

Technical Field

[0001] The invention relates to the technical field of automobile brake pads, and in particular to a high-deceleration brake pad based on a neural network model ratio optimization technology and a preparation method thereof. Background Art

[0002] With the rapid development of social economy and the improvement of people's living standards, the number of private cars has increased rapidly, and people have put forward higher requirements for the safety performance of cars. As an important part of the automobile braking system, the performance of automobile brake pads directly affects the driving safety of vehicles. Traditional brake pad materials are prone to thermal decay at high temperatures, which affects the braking effect and even endangers the lives of passengers. Therefore, the research and development of high-performance brake pad materials has become an important topic in the automotive industry.

[0003] At present, the widely used brake pad materials in the market mainly include cast iron, forged steel, etc. These materials have problems such as long manufacturing cycle, poor thermal conductivity and easy thermal cracking. With the development of the automobile industry, the requirements for brake pad materials are getting higher and higher. Not only do they need to have good friction coefficient, high temperature resistance, long life and other properties, but they also need to have good comfort and environmental protection.

[0004] The present invention aims to solve the above problems and proposes a high deceleration brake pad and a preparation method thereof based on a neural network model ratio optimization technology to improve the comprehensive performance of the brake pad. Summary of the invention

[0005] The present invention aims to solve the problems raised in the background technology and provides a high deceleration brake pad and a preparation method thereof based on a neural network model ratio optimization technology.

[0006] The specific technical solutions are as follows:

[0007] A high deceleration brake pad based on neural network model ratio optimization technology, comprising a high deceleration brake pad body, the high deceleration brake pad body comprising a steel back and a friction plate hot-pressed on one side of the steel back, the friction plate being made of a friction material, the friction material formula comprising steel fiber, friction powder, granular potassium hexatitanate, aramid fiber, mineral fiber, tin sulfide, graphite powder, coke powder, zirconium silicate / zirconium oxide / barium sulfate composite powder, calcium hydroxide and mica and modified resin, and the friction material formula of the high deceleration brake pad is obtained through the neural network model ratio optimization technology.

[0008] The above-mentioned high deceleration brake pad based on neural network model ratio optimization technology, wherein: the preparation method of the friction material includes the preparation of friction powder, zirconium silicate / zirconium oxide / barium sulfate composite powder and modified resin.

[0009] The high deceleration brake pad based on the neural network model ratio optimization technology, wherein: the preparation method of the friction powder comprises:

[0010] In parts by weight, 10-12 parts of cashew nut shell oil, 0.2-0.6 parts of concentrated sulfuric acid and 0.6-1.6 parts of acetic acid are added to the reactor, stirred at a speed of 50-100 r / min, heated to 105-135°C while stirring, reacted for 1-3 hours to increase the viscosity of the system, cooled to room temperature after the reaction is completed, 3-5 parts of tire powder are added, stirred at a speed of 500-1000 r / min for 30-60 minutes, the tire powder is evenly dispersed, hexamethylenetetramine is added, stirred at a speed of 50-100 r / min for 30-60 minutes, heated to 175-185°C while stirring, reacted for 1-2 hours, solidified after the reaction is completed, cooled to room temperature, and ground to obtain the homemade friction powder.

[0011] The high deceleration brake pad based on the neural network model ratio optimization technology, wherein: the preparation method of the zirconium silicate / zirconium oxide / barium sulfate composite powder comprises:

[0012] In parts by weight, 11.5-12.5 parts of barium sulfate powder and 1.2-2.1 parts of nonylphenol polyoxyethylene ether are added to 100-120 parts of distilled water, and the mixture is dispersed at a high speed of 2000-3000 r / min for 10-20 minutes to obtain a barium sulfate dispersion, which is then heated to 60-70°C and stirred at a speed of 200-300 r / min. While stirring, 22.8-24.1 parts of 50% ethyl orthosilicate ethanol solution are added, and the mixture is stirred evenly. Ammonia water with a concentration of 25% is added to adjust the pH. The pH value is increased to 9, and the hydrolysis reaction is carried out for 4-5 hours. Then, 24.2-25.0 parts of zirconium oxychloride, 30-35 parts of distilled water and 0.3-0.8 parts of lithium fluoride are added, and 25% ammonia water is added to adjust the pH value to 9. The mixture is stirred for 1-2 hours, and then dried at 110-130° C. to obtain a gel. The obtained gel is calcined at 800-1100° C. for 12-24 hours, and then crushed into a powder of 0.5-50 microns using a pulverizer to obtain a zirconium silicate / zirconium oxide / barium sulfate composite powder.

[0013] The high deceleration brake pad based on the neural network model ratio optimization technology, wherein: the preparation method of the modified resin comprises:

[0014] In parts by weight, 94-98 parts of phenol, 5.9-6.8 parts of ammonium molybdate and 900-1000 parts of distilled water are added to the reactor, stirred at a speed of 100-200 r / min, and then heated to 40-50°C and kept warm for 30-60 minutes to completely dissolve the phenol. Then 67-70 parts of 37% formaldehyde aqueous solution are added and stirred evenly. Hydrochloric acid is added to adjust the pH value to 1.8-2.3, and the mixture is reacted at 70-75°C. After the reaction is completed, continue to heat to 90-95°C and keep warm for 0.5-1.0 hour. After the insulation is completed, continue to heat to 95-100°C, add 5.9-6.3 parts of boric acid, 0.05-0.2 parts of 1-carboxyethyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide salt, 0.02-0.2 parts of 1,1'-ferrocenecarboxylic acid, react for 30-60 minutes, and distill under reduced pressure to remove water in the system to obtain the modified resin.

[0015] The above-mentioned high deceleration brake pad based on neural network model ratio optimization technology, wherein: the high deceleration brake pad body has an alarm device, the alarm device includes a first blind hole, a second blind hole, a temperature sensor and an infrared sensor, the first blind hole and the second blind hole both penetrate the steel back and extend to the inside of the friction plate, the temperature sensor and the infrared sensor are both fixedly mounted on a side of the steel back away from the friction plate, the detection part of the temperature sensor extends into the first blind hole to detect the temperature of the friction plate, the detection part of the infrared sensor extends into the second blind hole to detect the degree of wear of the friction plate, wherein:

[0016] The temperature sensor and the infrared sensor are both electrically connected to the vehicle's driving computer;

[0017] When the temperature sensor detects that the temperature of the friction plate exceeds a set value, the on-board computer controls the display screen in the vehicle to display a friction plate over-temperature alarm message, and at the same time controls the speaker in the vehicle to emit a friction plate over-temperature alarm sound;

[0018] When the friction plate is worn to the point that the second blind hole becomes a through hole, and the infrared sensor detects that the degree of wear of the friction plate has reached a set value, the on-board computer controls the display screen in the vehicle to display information that the friction plate has exceeded the limit of wear, and at the same time controls the speaker in the vehicle to sound a reminder that the friction plate needs to be replaced.

[0019] The high deceleration brake pad based on the neural network model ratio optimization technology mentioned above, wherein: a heat dissipation slot is provided on the side surface of the friction plate facing away from the steel back.

[0020] The above-mentioned high deceleration brake pad based on the neural network model ratio optimization technology, wherein: both ends of the steel back are integrally provided with mounting ears, and the mounting ears are provided with mounting holes.

[0021] The above-mentioned high deceleration brake pad based on neural network model ratio optimization technology, wherein: the specific steps of optimizing the friction material formula of the high deceleration brake pad through the neural network model ratio are as follows:

[0022] Step 1: Data preparation:

[0023] Collect 50 sets of friction material formula data, each set of data includes the ratio of steel fiber, friction powder, granular potassium hexatitanate, aramid fiber, mineral fiber, tin sulfide, graphite powder, coke powder, zirconium silicate / zirconium oxide / barium sulfate composite powder, calcium hydroxide and mica, as well as the corresponding deceleration and wear amount;

[0024] Step 2: Ratio optimization:

[0025] The deceleration, wear amount and speed and time involved in making friction powder obtained from 50 tests are input into the input layer of the neural network model, and the output layer outputs the optimal parameter change value Δ required for preparing homemade friction powder. According to the output results of the neural network model after optimization, the optimal ratio of the number of materials required for the friction material is determined, calculated in parts by weight. The specific formula is:

[0026] Steel fiber: 3.5-3.8 parts

[0027] Homemade friction powder: 9-10 parts

[0028] Granular potassium hexatitanate: 18-20 parts

[0029] Aramid fiber: 5-8 parts

[0030] Mineral fiber: 5-8 parts

[0031] Tin sulfide: 5-8 parts

[0032] Graphite powder: 2-5 parts

[0033] Coke powder: 3-6 portions

[0034] Zirconium silicate / zirconium oxide / barium sulfate composite powder: 31.5-34.5 parts

[0035] Calcium hydroxide: 5-8 parts

[0036] Mica: 5-8 parts

[0037] Modified resin: 8-12 parts.

[0038] The present invention also provides a method for preparing a high deceleration brake pad based on a neural network model ratio optimization technology, comprising the following steps:

[0039] S1. Steel back treatment: Test the size and flatness of the steel back. After the steel back is qualified, clean it with a cleaning agent, then perform shot blasting to remove impurities and oxide layer on the surface to obtain a steel back with a clean surface; then lay the cleaned steel back flat, spray glue on it to form a glue layer, the thickness of the glue layer is controlled at 15-20 microns, and the spraying area is 100%, to obtain a glued steel back;

[0040] S2. Mixed friction material: by weight, add 3.5-3.8 parts of steel fiber, 9-10 parts of friction powder, 18-20 parts of granular potassium hexatitanate, 5-8 parts of aramid fiber, 5-8 parts of mineral fiber, 5-8 parts of tin sulfide, 2-5 parts of graphite powder, 3-6 parts of coke powder, 31.5-34.5 parts of zirconium silicate / zirconium oxide / barium sulfate composite powder, 5-8 parts of calcium hydroxide and 5-8 parts of mica into a high-speed mixer, stir at 400-500 r / min for 30-60 min, then add 8-12 parts of modified resin, continue stirring at 400-500 r / min for 30-60 min to obtain a friction material mixture;

[0041] S3. Hot-pressed brake pads: The friction material mixture and the bonded steel back are placed in a mold for molding at a molding temperature of 155-185°C, a molding pressure of 25-30 MPa, and a molding time of 4-9 min to obtain a brake pad blank. The obtained brake pad blank is placed in an oven and heated at 185-215°C for 4.5-7.5 h, and then cooled to room temperature to obtain a brake pad.

[0042] The present invention has the following beneficial effects:

[0043] 1. Improve the performance of brake pads: The present invention optimizes the formula of friction materials through a neural network model, significantly improving the friction coefficient, high temperature resistance, wear resistance and comfort of brake pads, thereby improving the braking performance of automobiles. By introducing a neural network model in the preparation process, the optimal ratio of the required material portions of the friction material is obtained. Compared with the preparation method of the prior art, this solution greatly shortens the process of formula optimization and can accurately budget the required performance results.

[0044] 2. Environmental protection: The brake pad of the present invention does not contain asbestos, meets environmental protection requirements, and reduces pollution to the environment.

[0045] 3. Friction plate replacement reminder: When the wear degree of the friction plate reaches the set value, the on-board computer controls the display screen in the vehicle to display the friction plate wear exceeding limit information, and at the same time controls the speaker in the vehicle to issue a reminder sound that the friction plate needs to be replaced.

[0046] 4. High temperature alarm function: The brake pad of the present invention has a high temperature alarm function, which can promptly remind the driver when the temperature of the brake pad is too high, thereby improving driving safety.

[0047] In summary, the high deceleration brake pad and its preparation method based on neural network model ratio optimization technology provided in this embodiment have the advantages of significantly improving brake pad performance, environmental protection, providing friction pad replacement reminder and high temperature alarm function, and have important practical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A schematic diagram of the structure of a high deceleration brake pad based on a neural network model ratio optimization technology provided by an embodiment of the present invention;

[0049] Figure 2 A structural schematic diagram of a high deceleration brake pad based on a neural network model ratio optimization technology provided by an embodiment of the present invention from another perspective;

[0050] Figure 3 A schematic diagram of the exploded structure of a high deceleration brake pad based on a neural network model ratio optimization technology provided in an embodiment of the present invention.

[0051] 1. Steel back; 101. Mounting ear; 2. Friction plate; 201. Heat dissipation seam; 3. First blind hole; 4. Second blind hole; 5. Temperature sensor; 6. Infrared sensor. DETAILED DESCRIPTION

[0052] The technical solution of the present invention is further described below with reference to the accompanying drawings and through specific implementation methods.

[0053] Among them, the drawings are only used for illustrative explanations, and they only represent schematic diagrams rather than actual pictures, and should not be understood as limitations on this patent; in order to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product; for those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.

[0054] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if the terms "upper", "lower", "left", "right", "inner", "outer", etc. appear, the orientation or position relationship indicated is based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, the terms describing the position relationship in the drawings are only used for illustrative purposes and cannot be understood as a limitation on this patent. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0055] In the description of the present invention, unless otherwise clearly specified and limited, if the term "connection" or the like appears to indicate the connection relationship between components, the term should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two components or the interaction relationship between two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0056] Example

[0057] like Figure 1-3 As shown, the high deceleration brake pad based on the neural network model ratio optimization technology provided in this embodiment includes a high deceleration brake pad body, the high deceleration brake pad body includes a steel back 1 and a friction plate 2 hot-pressed on one side of the steel back 1, the friction plate 2 is made of friction material, the friction material formula includes steel fiber, friction powder, granular potassium hexatitanate, aramid fiber, mineral fiber, tin sulfide, graphite powder, coke powder, zirconium silicate / zirconium oxide / barium sulfate composite powder, calcium hydroxide and mica and modified resin, and the friction material formula of the high deceleration brake pad is obtained through the neural network model ratio optimization technology.

[0058] The high-deceleration brake pads using the above technical solution optimize the formula of the friction material through a neural network model, which significantly improves the friction coefficient, high temperature resistance, wear resistance and comfort of the brake pads, thereby improving the braking performance of the vehicle. It does not contain asbestos, meets environmental protection requirements, and reduces pollution to the environment.

[0059] Specifically, in this embodiment, the method for preparing the friction material includes preparing friction powder, zirconium silicate / zirconium oxide / barium sulfate composite powder and modified resin.

[0060] The preparation method of the friction powder comprises: adding 10-12 parts of cashew nut shell oil, 0.2-0.6 parts of concentrated sulfuric acid and 0.6-1.6 parts of acetic acid to a reactor by weight, stirring at a speed of 50-100 r / min, heating to 105-135°C while stirring, reacting for 1-3 hours to increase the viscosity of the system, cooling to room temperature after the reaction is completed, adding 3-5 parts of tire powder, stirring at a speed of 500-1000 r / min for 30-60 minutes, uniformly dispersing the tire powder, adding hexamethylenetetramine, stirring at a speed of 50-100 r / min for 30-60 minutes, heating to 175-185°C while stirring, reacting for 1-2 hours, solidifying after the reaction is completed, cooling to room temperature, and grinding to obtain the homemade friction powder. The quality and performance of the friction material are ensured through a precise preparation process, so that the homemade friction powder has excellent friction performance and thermal stability.

[0061] The preparation method of the zirconium silicate / zirconium oxide / barium sulfate composite powder comprises: adding 11.5-12.5 parts of barium sulfate powder and 1.2-2.1 parts of nonylphenol polyoxyethylene ether by weight to 100-120 parts of distilled water, dispersing at a high speed of 2000-3000 r / min for 10-20 minutes to obtain a barium sulfate dispersion, heating to 60-70°C, stirring at a speed of 200-300 r / min, adding 22.8-24.1 parts of 50% ethyl orthosilicate ethanol solution while stirring, stirring evenly, adding 25% ammonia water to adjust the pH value to 9. Hydrolyze for 4-5 hours, then add 24.2-25.0 parts of zirconium oxychloride, 30-35 parts of distilled water and 0.3-0.8 parts of lithium fluoride, continue to add 25% ammonia water to adjust the pH value to 9, stir for 1-2 hours, and then dry at 110-130° C. to obtain a gel; calcine the obtained gel at 800-1100° C. for 12-24 hours, and then grind it into a powder of 0.5-50 microns with a pulverizer to obtain a zirconium silicate / zirconium oxide / barium sulfate composite powder, which improves the high temperature resistance and wear resistance of the brake pad.

[0062] The preparation method of the modified resin comprises: adding 94-98 parts of phenol, 5.9-6.8 parts of ammonium molybdate and 900-1000 parts of distilled water to a reactor by weight, stirring evenly at a speed of 100-200 r / min, then heating to 40-50°C, keeping warm for 30-60 minutes to completely dissolve the phenol, then adding 67-70 parts of a 37% formaldehyde aqueous solution, stirring evenly, adding hydrochloric acid to adjust the pH value to 1.8-2.3, reacting at 70-75°C for 1- 2 hours, after the reaction is completed, continue to heat to 90-95 ° C, keep warm for 0.5-1.0 hours, continue to heat to 95-100 ° C after the insulation is completed, add 5.9-6.3 parts of boric acid, 0.05-0.2 parts of 1-carboxyethyl-3-methylimidazole bis(trifluoromethanesulfonyl)imide salt, 0.02-0.2 parts of 1,1'-dicarbocenecarboxylic acid, react for 30-60 minutes, and distill under reduced pressure to remove water in the system to obtain a modified resin, which enhances the overall strength and toughness of the brake pad.

[0063] Specifically, in this embodiment, the high-deceleration brake pad body has an alarm device, which includes a first blind hole 3, a second blind hole 4, a temperature sensor 5 and an infrared sensor 6. The first blind hole 3 and the second blind hole 4 both penetrate the steel back 1 and extend to the inside of the friction plate 2. The temperature sensor 5 and the infrared sensor 6 are both fixedly mounted on a side of the steel back 1 away from the friction plate 2. The detection part of the temperature sensor 5 extends into the first blind hole 3 to detect the temperature of the friction plate 2. The detection part of the infrared sensor 6 extends into the second blind hole 4 to detect the degree of wear of the friction plate 2, wherein:

[0064] The temperature sensor 5 and the infrared sensor 6 are both electrically connected to the vehicle's on-board computer;

[0065] When the temperature sensor 5 detects that the temperature of the friction plate 2 exceeds the set value, the on-board computer controls the display screen in the vehicle to display the friction plate over-temperature alarm information, and at the same time controls the speaker in the vehicle to emit a friction plate over-temperature alarm sound;

[0066] When the friction plate 2 is worn to the point that the second blind hole 4 becomes a through hole, and the infrared sensor 6 detects that the wear degree of the friction plate 2 reaches the set value, the on-board computer controls the display screen in the vehicle to display the friction plate wear exceeding limit information, and at the same time controls the speaker in the vehicle to sound a reminder that the friction plate needs to be replaced.

[0067] The high-deceleration brake pad adopting the above technical solution has a friction pad replacement reminder function. When the wear degree of the friction pad reaches the set value, the on-board computer controls the display screen in the vehicle to display the friction pad wear exceeding the limit information, and at the same time controls the speaker in the vehicle to issue a reminder sound that the friction pad needs to be replaced; it also has a high-temperature alarm function, which can promptly remind the driver when the brake pad temperature is too high, thereby improving driving safety.

[0068] The alarm device provides friction pad replacement reminder and high temperature alarm function, which can promptly remind the driver when the brake pad is excessively worn or the temperature is too high, thereby improving driving safety.

[0069] In order to improve the heat dissipation performance of the high deceleration brake pad, a heat dissipation slit 201 is provided on the side of the friction pad 2 facing away from the steel back 1. The design of the heat dissipation slit 201 improves the heat dissipation efficiency of the brake pad and prolongs the service life of the brake pad.

[0070] In order to facilitate the installation of the high deceleration brake pad, mounting ears 101 are integrally provided at both ends of the steel back 1, and mounting holes are opened on the mounting ears 101. The arrangement of the mounting ears 101 and the mounting holes facilitates the installation and removal of the brake pad and improves maintenance efficiency.

[0071] Specifically, in this embodiment, the specific steps of optimizing the friction material formula of the high deceleration brake pad by using the neural network model ratio are as follows:

[0072] Step 1: Data preparation:

[0073] Collect 50 sets of friction material formula data, each set of data includes the ratio of steel fiber, friction powder, granular potassium hexatitanate, aramid fiber, mineral fiber, tin sulfide, graphite powder, coke powder, zirconium silicate / zirconium oxide / barium sulfate composite powder, calcium hydroxide and mica, as well as the corresponding deceleration and wear amount;

[0074] Step 2: Build a neural network model:

[0075] The neural network model includes:

[0076] Input Layer:

[0077] Input characteristics: ratio of various components of friction material, deceleration, wear amount, etc.

[0078] Assume that the input feature vector is X and its dimension is n.

[0079] Hidden Layer:

[0080] The number of neurons in the hidden layer is assumed to be m.

[0081] Weight matrix W1: dimension is m×n.

[0082] Bias vector b1: dimension is m.

[0083] Output layer:

[0084] Output characteristics: optimal friction material formulation parameter change value Δ.

[0085] Weight matrix W2: dimension is 1×m.

[0086] Bias vector b2: dimension is 1.

[0087] Mathematical expressions

[0088] Forward propagation:

[0089] Hidden layer output: H = ReLU (W1X + b1);

[0090] Output layer output: Y = W2H + b2;

[0091] Activation function:

[0092] Use the ReLU activation function: ReLU(x)=max(0,x);

[0093] Loss function:

[0094] Use mean square error (MSE) as the loss function: Among them, Y i is the model prediction value; Ti is the true value, N is the number of samples; back propagation:

[0095] Calculate the gradient of the loss function with respect to each parameter:

[0096]

[0097] Parameter update:

[0098] Update the parameters using gradient descent:

[0099]

[0100] Where η is the learning rate;

[0101] X: input feature vector, dimension is n;

[0102] n: the dimension of input features;

[0103] m: number of neurons in the hidden layer;

[0104] W1: weight matrix from input layer to hidden layer, dimension is m×n;

[0105] b1: bias vector from the input layer to the hidden layer, dimension is m;

[0106] H: hidden layer output, dimension is mm;

[0107] W2: weight matrix from hidden layer to output layer, dimension is 1×m;

[0108] b2: bias vector from the hidden layer to the output layer, dimension is 1;

[0109] Y: model prediction value, dimension is 1;

[0110] N: sample size;

[0111] T i : The true value of the i-th sample;

[0112] L: loss function value;

[0113] The input layer inputs the deceleration, wear amount and rotation speed and time involved in making friction powder obtained from 50 tests, and the output layer outputs the optimal parameter change value Δ required for preparing homemade friction powder.

[0114] Model training:

[0115] Within the range specified by the above parameters, 50 samples are randomly initialized to prepare 50 brake pad samples, performance tests are performed on the 50 brake pad samples, and the 50 test results are input into the neural network model for training. After the model training is completed, the model is evaluated and tuned, and the final model is determined after the tuning is completed.

[0116] Recipe Optimization:

[0117] According to the output results of the neural network model after the optimization, the optimal ratio of the required materials of the friction material is determined. The specific formula is calculated by weight:

[0118] Steel fiber: 3.5-3.8 parts

[0119] Homemade friction powder: 9-10 parts

[0120] Granular potassium hexatitanate: 18-20 parts

[0121] Aramid fiber: 5-8 parts

[0122] Mineral fiber: 5-8 parts

[0123] Tin sulfide: 5-8 parts

[0124] Graphite powder: 2-5 parts

[0125] Coke powder: 3-6 portions

[0126] Zirconium silicate / zirconium oxide / barium sulfate composite powder: 31.5-34.5 parts

[0127] Calcium hydroxide: 5-8 parts

[0128] Mica: 5-8 parts

[0129] Modified resin: 8-12 parts.

[0130] Through the above specific scheme, the present invention optimizes the formula of friction materials by using a neural network model, significantly improves the friction coefficient, high temperature resistance, wear resistance and comfort of brake pads, thereby improving the braking performance of automobiles, and by introducing a neural network model in the preparation process, the optimal ratio of the number of materials required for the friction material is obtained. Compared with the preparation method of the prior art, this scheme greatly shortens the process of formula optimization, can accurately budget the required performance results, and improves production efficiency.

[0131] This embodiment also provides a method for preparing a high deceleration brake pad based on a neural network model ratio optimization technology, comprising the following steps:

[0132] S1. Steel back treatment: Test the size and flatness of the steel back 2. After the steel back 2 is qualified, clean the steel back 2 with a cleaning agent, and then perform shot blasting to remove surface impurities and oxide layers to obtain a steel back 2 with a clean surface; then lay the steel back 2 with a clean surface flat, spray glue on it to form a glue layer, the thickness of the glue layer is controlled to be 15-20 microns, and the spraying area is 100%, to obtain a glued steel back 2;

[0133] S2. Mixed friction material: by weight, add 3.5-3.8 parts of steel fiber, 9-10 parts of friction powder, 18-20 parts of granular potassium hexatitanate, 5-8 parts of aramid fiber, 5-8 parts of mineral fiber, 5-8 parts of tin sulfide, 2-5 parts of graphite powder, 3-6 parts of coke powder, 31.5-34.5 parts of zirconium silicate / zirconium oxide / barium sulfate composite powder, 5-8 parts of calcium hydroxide and 5-8 parts of mica into a high-speed mixer, stir at 400-500 r / min for 30-60 min, then add 8-12 parts of modified resin, continue stirring at 400-500 r / min for 30-60 min to obtain a friction material mixture;

[0134] S3. Hot-pressed brake pad: the friction material mixture and the glued steel back 2 are placed in a mold for molding at a molding temperature of 155-185°C, a molding pressure of 25-30 MPa, and a molding time of 4-9 min to obtain a brake pad blank. The obtained brake pad blank is placed in an oven and heated at 185-215°C for 4.5-7.5 h, and then cooled to room temperature to obtain a brake pad.

[0135] The method ensures the quality and performance of the brake pads by optimizing the optimal formula through precise preparation process and neural network model. At the same time, the method has the advantages of simple operation, high production efficiency and low cost, and is suitable for large-scale production.

[0136] In summary, the high deceleration brake pad and the preparation method thereof based on the neural network model ratio optimization technology provided in this embodiment have the following advantages:

[0137] 1. Improve brake pad performance: The present invention optimizes the formula of friction materials through a neural network model, significantly improving the friction coefficient, high temperature resistance, wear resistance and comfort of the brake pad, thereby improving the braking performance of the vehicle.

[0138] 2. Environmental protection: The brake pad of the present invention does not contain asbestos, meets environmental protection requirements, and reduces pollution to the environment.

[0139] 3. Friction plate replacement reminder: When the wear degree of the friction plate reaches the set value, the on-board computer controls the display screen in the vehicle to display the friction plate wear exceeding limit information, and at the same time controls the speaker in the vehicle to issue a reminder sound that the friction plate needs to be replaced.

[0140] 4. High temperature alarm function: The brake pad of the present invention has a high temperature alarm function, which can promptly remind the driver when the temperature of the brake pad is too high, thereby improving driving safety.

[0141] In summary, the high deceleration brake pad and its preparation method based on neural network model ratio optimization technology provided in this embodiment have the advantages of significantly improving brake pad performance, environmental protection, providing friction pad replacement reminder and high temperature alarm function, and have important practical application value.

[0142] The above are only preferred embodiments of the present invention, and are not intended to limit the implementation methods and protection scope of the present invention. Those skilled in the art should be aware that all solutions obtained by equivalent substitutions and obvious changes made using the description and illustrations of the present invention should be included in the protection scope of the present invention.

Claims

1. A high deceleration brake pad based on neural network model ratio optimization technology, characterized in that: The invention comprises a high-deceleration brake pad body, wherein the high-deceleration brake pad body comprises a steel back (1) and a friction plate (2) hot-pressed on one side of the steel back (1), wherein the friction plate (2) is made of a friction material, wherein the friction material formula comprises steel fiber, friction powder, granular potassium hexatitanate, aramid fiber, mineral fiber, tin sulfide, graphite powder, coke powder, zirconium silicate / zirconium oxide / barium sulfate composite powder, calcium hydroxide and mica and modified resin, and the friction material formula of the high-deceleration brake pad is obtained by using a neural network model ratio optimization technology.

2. The high deceleration brake pad based on neural network model ratio optimization technology according to claim 1 is characterized in that: The method for preparing the friction material comprises the steps of preparing friction powder, zirconium silicate / zirconium oxide / barium sulfate composite powder and modified resin.

3. The high deceleration brake pad based on neural network model ratio optimization technology according to claim 2 is characterized in that: The preparation method of the friction powder comprises: In parts by weight, 10-12 parts of cashew nut shell oil, 0.2-0.6 parts of concentrated sulfuric acid and 0.6-1.6 parts of acetic acid are added to the reactor, stirred at a speed of 50-100 r / min, heated to 105-135°C while stirring, reacted for 1-3 hours to increase the viscosity of the system, cooled to room temperature after the reaction is completed, 3-5 parts of tire powder are added, stirred at a speed of 500-1000 r / min for 30-60 minutes, the tire powder is evenly dispersed, hexamethylenetetramine is added, stirred at a speed of 50-100 r / min for 30-60 minutes, heated to 175-185°C while stirring, reacted for 1-2 hours, solidified after the reaction is completed, cooled to room temperature, and ground to obtain the homemade friction powder.

4. The high deceleration brake pad based on neural network model ratio optimization technology according to claim 3 is characterized in that: The preparation method of the zirconium silicate / zirconium oxide / barium sulfate composite powder comprises: In parts by weight, 11.5-12.5 parts of barium sulfate powder and 1.2-2.1 parts of nonylphenol polyoxyethylene ether are added to 100-120 parts of distilled water, and the mixture is dispersed at a high speed of 2000-3000 r / min for 10-20 minutes to obtain a barium sulfate dispersion, which is then heated to 60-70°C and stirred at a speed of 200-300 r / min. While stirring, 22.8-24.1 parts of 50% ethyl orthosilicate ethanol solution are added, and the mixture is stirred evenly. Ammonia water with a concentration of 25% is added to adjust the pH. The pH value is increased to 9, and the hydrolysis reaction is carried out for 4-5 hours. Then, 24.2-25.0 parts of zirconium oxychloride, 30-35 parts of distilled water and 0.3-0.8 parts of lithium fluoride are added, and 25% ammonia water is added to adjust the pH value to 9. The mixture is stirred for 1-2 hours, and then dried at 110-130° C. to obtain a gel. The obtained gel is calcined at 800-1100° C. for 12-24 hours, and then crushed into a powder of 0.5-50 microns using a pulverizer to obtain a zirconium silicate / zirconium oxide / barium sulfate composite powder.

5. The high deceleration brake pad based on neural network model ratio optimization technology according to claim 4 is characterized in that: The preparation method of the modified resin comprises: In parts by weight, 94-98 parts of phenol, 5.9-6.8 parts of ammonium molybdate and 900-1000 parts of distilled water are added to the reactor, stirred at a speed of 100-200 r / min, and then heated to 40-50°C and kept warm for 30-60 minutes to completely dissolve the phenol. Then 67-70 parts of 37% formaldehyde aqueous solution are added and stirred evenly. Hydrochloric acid is added to adjust the pH value to 1.8-2.3, and the mixture is reacted at 70-75°C. After the reaction is completed, continue to heat to 90-95°C and keep warm for 0.5-1.0 hour. After the insulation is completed, continue to heat to 95-100°C, add 5.9-6.3 parts of boric acid, 0.05-0.2 parts of 1-carboxyethyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide salt, 0.02-0.2 parts of 1,1'-ferrocenecarboxylic acid, react for 30-60 minutes, and distill under reduced pressure to remove water in the system to obtain the modified resin.

6. The high deceleration brake pad based on neural network model ratio optimization technology according to claim 5 is characterized in that: The high-deceleration brake pad body has an alarm device, which includes a first blind hole (3), a second blind hole (4), a temperature sensor (5) and an infrared sensor (6). The first blind hole (3) and the second blind hole (4) both penetrate the steel back (1) and extend to the interior of the friction plate (2). The temperature sensor (5) and the infrared sensor (6) are both fixedly mounted on a side of the steel back (1) away from the friction plate (2). The detection part of the temperature sensor (5) extends into the first blind hole (3) to detect the temperature of the friction plate (2). The detection part of the infrared sensor (6) extends into the second blind hole (4) to detect the degree of wear of the friction plate (2), wherein: The temperature sensor (5) and the infrared sensor (6) are both electrically connected to the vehicle's on-board computer; When the temperature sensor (5) detects that the temperature of the friction plate (2) exceeds a set value, the on-board computer controls the display screen in the vehicle to display a friction plate over-temperature alarm message, and at the same time controls the speaker in the vehicle to emit a friction plate over-temperature alarm sound; When the friction plate (2) is worn to the point where the second blind hole (4) becomes a through hole, and the infrared sensor (6) detects that the degree of wear of the friction plate (2) reaches a set value, the on-board computer controls the display screen in the vehicle to display information that the friction plate has exceeded the limit of wear, and at the same time controls the speaker in the vehicle to sound a reminder that the friction plate needs to be replaced.

7. The high deceleration brake pad based on neural network model ratio optimization technology according to claim 6 is characterized in that: A heat dissipation slit (201) is provided on a side surface of the friction plate (2) facing away from the steel back (1).

8. The high deceleration brake pad based on neural network model ratio optimization technology according to claim 7 is characterized in that: Both ends of the steel back (1) are integrally provided with mounting ears (101), and mounting holes are provided on the mounting ears (101).

9. The high deceleration brake pad based on neural network model ratio optimization technology according to claim 8, characterized in that: The specific steps of optimizing the friction material formula of the high deceleration brake pad by using the neural network model ratio are as follows: Step 1: Data preparation: Collect 50 sets of friction material formula data, each set of data includes the ratio of steel fiber, friction powder, granular potassium hexatitanate, aramid fiber, mineral fiber, tin sulfide, graphite powder, coke powder, zirconium silicate / zirconium oxide / barium sulfate composite powder, calcium hydroxide and mica, as well as the corresponding deceleration and wear amount; Step 2: Ratio optimization: The deceleration, wear amount and speed and time involved in making friction powder obtained from 50 tests are input into the input layer of the neural network model, and the output layer outputs the optimal parameter change value Δ required for preparing homemade friction powder. According to the output results of the neural network model after optimization, the optimal ratio of the number of materials required for the friction material is determined, calculated in parts by weight. The specific formula is: Steel fiber: 3.5-3.8 parts Homemade friction powder: 9-10 parts Granular potassium hexatitanate: 18-20 parts Aramid fiber: 5-8 parts Mineral fiber: 5-8 parts Tin sulfide: 5-8 parts Graphite powder: 2-5 parts Coke powder: 3-6 portions Zirconium silicate / zirconium oxide / barium sulfate composite powder: 31.5-34.5 parts Calcium hydroxide: 5-8 parts Mica: 5-8 parts Modified resin: 8-12 parts.

10. A method for preparing a high deceleration brake pad based on a neural network model ratio optimization technology as claimed in claim 9, characterized in that: The following steps are involved: S1. Steel back treatment: Test the size and flatness of the steel back. After the steel back is qualified, clean it with a cleaning agent, then perform shot blasting to remove impurities and oxide layer on the surface to obtain a steel back with a clean surface; then lay the cleaned steel back flat, spray glue on it to form a glue layer, the thickness of the glue layer is controlled at 15-20 microns, and the spraying area is 100%, to obtain a glued steel back; S2. Mixed friction material: by weight, add 3.5-3.8 parts of steel fiber, 9-10 parts of friction powder, 18-20 parts of granular potassium hexatitanate, 5-8 parts of aramid fiber, 5-8 parts of mineral fiber, 5-8 parts of tin sulfide, 2-5 parts of graphite powder, 3-6 parts of coke powder, 31.5-34.5 parts of zirconium silicate / zirconium oxide / barium sulfate composite powder, 5-8 parts of calcium hydroxide and 5-8 parts of mica into a high-speed mixer, stir at 400-500 r / min for 30-60 min, then add 8-12 parts of modified resin, continue stirring at 400-500 r / min for 30-60 min to obtain a friction material mixture; S3. Hot-pressed brake pads: The friction material mixture and the bonded steel back are placed in a mold for molding at a molding temperature of 155-185°C, a molding pressure of 25-30 MPa, and a molding time of 4-9 min to obtain a brake pad blank. The obtained brake pad blank is placed in an oven and heated at 185-215°C for 4.5-7.5 h, and then cooled to room temperature to obtain a brake pad.

Images