Composition of powder high-speed steel and powder metallurgy construction method

Abstract

The invention relates to the technical field of alloy design, and discloses a powder high-speed steel composition and powder metallurgy construction method.The powder high-speed steel composition and powder metallurgy construction method comprises the steps that S1, powder high-speed steel data in literatures and production are collected, a machine learning model is trained, a relation model among components, an organization structure and mechanical properties is obtained, and an alloy formula with the needed mechanical properties is screened out; s2, inputting the alloy formula obtained in the step S1 into an alloy thermodynamic database for high-throughput calculation, screening out the alloy formula meeting the predicted organization structure in the step S1, and designing a corresponding heat treatment process; and S3, a preparation process of the powder high-speed steel is constructed, specifically, mixing is conducted according to the alloy formula obtained in the S2, and after pressing and activated sintering are conducted, heat treatment is conducted according to the heat treatment process obtained in the S2. According to the method, the design process of the powder high-speed steel composition and the preparation technology can be simplified, the research and development period is shortened, the trial and error cost is reduced, and flexible adjustment can be conducted according to different mechanical property requirements so as to meet different use requirements.

Description

Technical Field

[0001] The present invention relates to the technical field of alloy design, and particularly to the composition of a powder high-speed steel and a powder metallurgy construction method. Background Art

[0002] With the rapid development of the manufacturing industry, especially in the fields of automobiles, aerospace, semiconductor devices, etc., the demand for high-precision and high-reliability cutting tools has been continuously increasing, driving the expansion of the cutting tool market scale at home and abroad. Powder high-speed steel cutting tools have the advantages of high hardness, good toughness, strong thermal stability, etc., and can meet the usage requirements of complex parts in the above fields. However, at present, the grades of powder high-speed steel are single, and it is difficult to design alloy compositions and preparation processes that meet the usage requirements of complex working conditions for high-end powder high-speed steel for cutting tools.

[0003] Currently, powder high-speed steel is mainly prepared by the method of vacuum melting - gas atomization powder making - hot isostatic pressing. The products obtained by this method have high density and excellent comprehensive performance. However, this method has high energy consumption and large equipment investment. Therefore, it is of great significance to develop a preparation process with a short process flow, small equipment investment, and environmental friendliness. Using the powder metallurgy method, that is, the method of "powder pretreatment - pressing - activated sintering - heat treatment" to prepare powder high-speed steel (such as patent CN114561588A), provides an effective way to solve the above problems.

[0004] The processes that need to be adjusted in the traditional powder metallurgy method include multiple adjustable parameters, such as raw material types, ball-to-material ratio, pre-oxidation temperature, carbon-oxygen reaction temperature, quenching method, etc. The parameters to be adjusted are complex and numerous. It takes a huge amount of manpower and material resources to complete the adjustment and optimization of process parameters only through experimental means such as the "trial and error method". Summary of the Invention

[0005] In order to solve the above technical problems, that is, in the process of preparing powder high-speed steel by powder metallurgy, it is necessary to conduct multiple trials and errors for the adjustment of composition and process details, which is time-consuming and laborious. The present invention provides a composition of a powder high-speed steel and a powder metallurgy construction method. The method of the present invention can simplify the design process of the composition and preparation process of powder high-speed steel, shorten the R & D cycle, reduce the trial and error cost, and can be flexibly adjusted according to different mechanical property requirements to meet different usage requirements of powder high-speed steel.

[0006] The specific technical solution of the present invention is as follows: A composition of a powder high-speed steel and a powder metallurgy construction method, comprising: S1: Collect data of powder high-speed steel in literature and production, train a machine learning model to obtain a relationship model between composition - microstructure - mechanical properties, and screen out alloy formulas (including element types and content ranges of each element) with required mechanical properties; S2: Construct an alloy thermodynamics database, input the alloy formula obtained in S1 for high-throughput calculation; according to the calculation results, screen out the alloy formulas that meet the predicted organizational structure in S1, and design the corresponding heat treatment process; S3: The preparation process of powder high-speed steel is as follows: Mix materials according to the alloy formula obtained in S2, after pressing and activated sintering, perform heat treatment according to the heat treatment process obtained in S2 to obtain powder high-speed steel.

[0007] Preferably, step S1 specifically includes: S1.1: Collect information on the composition, preparation temperature, organizational structure, and mechanical properties of powder high-speed steel in the literature and production. After screening and cleaning, obtain a data set; S1.2: Use part of the data in the data set obtained in S1.1 as input, and use the random forest algorithm to train a model to obtain a relationship model between composition-organizational structure-mechanical properties; S1.3: Use the remaining data in the data set obtained in S1.1 to verify the reliability of the relationship model obtained in S1.2; S1.4: Input the new powder high-speed steel composition parameters into the relationship model obtained in S1.2, predict the mechanical properties, and select the alloy formula with the required mechanical properties.

[0008] Preferably, in step S1.2, 80-85% of the data in the data set is used as input.

[0009] Preferably, in step S1.2, the process of training the model using the random forest algorithm includes: S1.2.1: Apply the Bagging random sampling method to randomly draw k sample sets from the original training set with replacement to form k decision trees; S1.2.2: Specify the value of m, randomly draw m variables from the samples, and at the same time, the selection of feature attributes during the splitting process satisfies the principle of minimum node impurity; S1.2.3: Repeat S1.2.2 until no further splitting is possible to form a forest composed of multiple decision trees; S1.2.4: Take the average of the predicted values output by N decision trees.

[0010] Preferably, in step S1, the mechanical properties include one or more of hardness, flexural strength, and impact toughness.

[0011] Preferably, step S2 specifically includes: S2.1: Collect the alloy system thermodynamics parameters related to the alloy formula obtained in S1 in the literature, and integrate them to construct an alloy thermodynamics database; S2.2: Input the low-order alloy system related to the alloy formula obtained in S1, compare the output results with the experimental data in the literature, and verify the reliability of the alloy thermodynamic database; S2.3: Use the alloy thermodynamic database to perform high-throughput calculations on the alloy formula obtained in S1, and screen out the alloy formula that meets the predicted microstructure in S1; S2.4: Design the alloy heat treatment process based on the equilibrium phase diagram and non-equilibrium solidification path of the alloy formula obtained in S2.3 calculated by the alloy thermodynamic database to obtain an alloy with a specific phase composition and solid solubility.

[0012] Preferably, in step S2.1, the alloy systems related to the alloy formula obtained in S1 include: a Fe-containing binary alloy system composed of two elements in the alloy formula obtained in S1, and a Fe-containing ternary alloy system composed of three elements in the alloy formula obtained in S1.

[0013] Preferably, in step S2.2, the low-order alloy system related to the alloy formula obtained in S1 is: a Fe-containing alloy system composed of three elements in the alloy formula obtained in S1.

[0014] Preferably, in step S2.1, after integrating the thermodynamic parameters of the alloy system, optimize the systems with conflicting parameters using the parameter optimization module of the thermodynamic calculation software Thermo-Calc or Pandat to obtain the alloy thermodynamic database.

[0015] Preferably, in step S2, the content of the high-throughput calculation includes one or more of the following a to d: a. The liquidus temperature of alloys with different formulas; b. The solidus temperature of alloys with different formulas; c. The composition and content of the precipitated phases under different formulas and temperature conditions; d. The solid solubility of elements in the matrix.

[0016] Preferably, in step S3, the preparation process of the powder high-speed steel is specifically as follows: S3.1: Prepare materials according to the alloy formula obtained in S2, mix and ball-mill them together with a binder and a dispersion medium, granulate after drying, and press them into a mold to obtain a green compact; S3.2: Perform debinding treatment, vacuum sintering, and hot press sintering on the green compact in sequence to obtain an alloy sample.

[0017] Preferably, in step S3.1, the dispersion medium is alcohol and / or n-butane; the conditions for the mixed ball-milling are as follows: the filling coefficient is 0.4 - 0.6, the ball-milling medium is cemented carbide balls and / or stainless steel balls, the ball-to-material ratio is 3 - 8:1, and the ball-milling time is 24 - 96 h.

[0018] Preferably, in step S3.1, the drying temperature is 60-90 °C.

[0019] Preferably, in step S3.1, the granulation process is carried out using a rubbing sieve granulator.

[0020] Preferably, in step S3.1, the method of pressing and forming is die pressing or cold isostatic pressing, and the pressing pressure is 300-500 MPa.

[0021] Preferably, in step S3.2, the temperature of the degreasing treatment is 500-550 °C, and the time is 5-6 h; the temperature of the vacuum sintering is 1100-1250 °C, and the time is 1-2 h; the temperature of the hot press sintering is 1050-1200 °C, the pressure is 5-20 Mpa, and the time is 1-2 h.

[0022] Preferably, the composition of the powder high-speed steel and the powder metallurgy construction method further include: S4: Prepare the powder high-speed steel according to the powder metallurgy process obtained in S3, and check whether the organizational structure and mechanical properties meet the expectations.

[0023] Compared with the prior art, the present invention has the following advantages: (1) Compared with the existing powder high-speed steel composition and powder metallurgy process development methods, the present invention designs the composition of the powder high-speed steel and the corresponding powder metallurgy process that meet the use requirements based on the existing literature and data collected in production, combined with machine learning and high-throughput phase diagram thermodynamics calculation methods, which can shorten the R & D cycle and save the cost of trial and error.

[0024] (2) The method of the present invention has a wide range of applications. It can not only be used to design the preparation method of high-strength and high-toughness powder high-speed steel, but also can be flexibly adjusted according to different mechanical property requirements to design powder high-speed steel with the required mechanical properties to meet different use requirements.

[0025] (2) The alloy formula and heat treatment process designed by the method of the present invention can be applied to the preparation of powder high-speed steel through the powder metallurgy process (mixing - pressing - activated sintering - heat treatment). The preparation process is short, which can avoid multiple melting processes, reduce energy consumption, and is close to net shaping with high material utilization rate. Description of the Drawings

[0026] Figure 1 It is a flow chart of designing the composition of the powder high-speed steel and the powder metallurgy process in the present invention.

[0027] Figure 2 It is a vertical cross-section of the multi-component alloy calculated in step S2.3 of Example 1.

[0028] Figure 3 It is the variation of the phase content of the multi - element alloy calculated in step S2.3 of Example 1 with temperature.

[0029] Figure 4 It is the scanning electron microscope image of the powder high - speed steel sample prepared in Example 1.

[0030] Figure 5 It is the scanning electron microscope image of the powder high - speed steel sample prepared in Example 2.

[0031] Figure 6 It is the mechanical properties of the powder high - speed steel samples prepared in Example 1 and Example 2. Specific embodiments

[0032] The present invention will be further described below in conjunction with embodiments.

[0033] General embodiment A composition and powder metallurgy construction method of a powder high - speed steel, comprising: S1: Collect data of powder high - speed steel in literature and production, train a machine learning model, obtain a relationship model between composition - microstructure - mechanical properties, and screen out alloy formulas (including element types and content ranges of each element) with required mechanical properties; S2: Construct an alloy thermodynamics database, input the alloy formula obtained in S1 for high - throughput calculation; according to the calculation results, screen out alloy formulas that meet the microstructure predicted in S1, and design corresponding heat treatment processes; S3: Construct the preparation process of the powder high - speed steel as follows: Mix materials according to the alloy formula obtained in S2, after pressing and activated sintering, perform heat treatment according to the heat treatment process obtained in S2 to obtain the powder high - speed steel.

[0034] In some specific embodiments, in step S1, the mechanical properties include one or more of hardness, flexural strength, and impact toughness.

[0035] In some specific embodiments, step S1 specifically includes: S1.1: Collect information on the composition, preparation temperature, microstructure, and mechanical properties of powder high - speed steel in literature and production, and obtain a data set after screening and cleaning; S1.2: Use part of the data in the data set obtained in S1.1 as input, and train a model using the random forest algorithm to obtain a relationship model between composition - microstructure - mechanical properties; S1.3: Use the remaining data in the data set obtained in S1.1 to verify the reliability of the relationship model obtained in S1.2; S1.4: Input the composition parameters of the new powder high-speed steel into the relationship model obtained in S1.2, predict the mechanical properties, and select the alloy formula with the required mechanical properties.

[0036] In the above specific embodiments: Optionally or preferably, in step S1.2, 80-85% of the data in the dataset is used as input.

[0037] Optionally or preferably, in step S1.2, the process of training the model using the random forest algorithm includes: S1.2.1: Apply the Bagging random sampling method to randomly draw k sample sets from the original training set with replacement to form k decision trees; S1.2.2: Specify the value of m, randomly draw m variables from the samples, and at the same time, the selection of feature attributes in the splitting process satisfies the principle of minimum node impurity; S1.2.3: Repeat S1.2.2 until no further splitting is possible to form a forest composed of multiple decision trees; S1.2.4: Take the average value of the predicted values output by N decision trees.

[0038] In some specific embodiments, step S2 specifically includes: S2.1: Collect the thermodynamic parameters of the alloy systems related to the alloy formula obtained in S1 from the literature, and construct an alloy thermodynamic database after integration; S2.2: Input the low-order alloy systems related to the alloy formula obtained in S1, compare the output results with the experimental data in the literature, and verify the reliability of the alloy thermodynamic database; S2.3: Perform high-throughput calculations on the alloy formula obtained in S1 using the alloy thermodynamic database, and screen out the alloy formulas that meet the predicted organizational structure in S1; S2.4: Design the alloy heat treatment process according to the equilibrium phase diagram and non-equilibrium solidification path of the alloy formula obtained in S2.3 calculated by the alloy thermodynamic database to obtain an alloy with a specific phase composition and solid solubility.

[0039] In the above specific embodiments: Optionally or preferably, in step S2.1, the alloy systems related to the alloy formula obtained in S1 include: a binary Fe-containing alloy system composed of two elements in the alloy formula obtained in S1, and a ternary Fe-containing alloy system composed of three elements in the alloy formula obtained in S1.

[0040] Optionally or preferably, in step S2.1, after integrating the thermodynamic parameters of the alloy systems, optimize the systems with conflicting parameters using the parameter optimization module of the thermodynamic calculation software Thermo-Calc or Pandat to obtain the alloy thermodynamic database.

[0041] Optionally or preferably, in step S2.2, the low-order alloy system related to the alloy formula obtained in S1 is: an Fe-containing alloy system composed of three elements in the alloy formula obtained in S1.

[0042] In some specific embodiments, in step S2, the content of high-throughput calculation includes one or more of the following a to d: a. Liquidus temperature of alloys with different formulas; b. Solidus temperature of alloys with different formulas; c. Composition and content of precipitated phases under different formulas and temperature conditions; d. Solubility of elements in the matrix.

[0043] In some specific embodiments, in step S3, the preparation process of powder high-speed steel is specifically as follows: S3.1: Prepare materials according to the alloy formula obtained in S2, mix and ball-mill them together with a binder and a dispersion medium, granulate after drying, and press them into shape in a mold to obtain a green compact; S3.2: Perform degreasing treatment, vacuum sintering, and hot press sintering on the green compact in sequence to obtain an alloy sample.

[0044] In the above specific embodiments: Optionally or preferably, in step S3.1, the dispersion medium is alcohol and / or n-butane; the conditions for the mixed ball-milling are as follows: the filling coefficient is 0.4 - 0.6, the ball-milling medium is cemented carbide balls and / or stainless steel balls, the ball-to-material ratio is 3 - 8:1, and the ball-milling time is 24 - 96 h.

[0045] Optionally or preferably, in step S3.1, the drying temperature is 60 - 90 °C.

[0046] Optionally or preferably, in step S3.1, the granulation process is carried out using a rubbing sieve granulator.

[0047] Optionally or preferably, in step S3.1, the method of pressing into shape is die pressing or cold isostatic pressing, and the pressing pressure is 300 - 500 MPa.

[0048] Optionally or preferably, in step S3.2, the temperature of the degreasing treatment is 500 - 550 °C, and the time is 5 - 6 h; the temperature of the vacuum sintering is 1100 - 1250 °C, and the time is 1 - 2 h; the temperature of the hot press sintering is 1050 - 1200 °C, the pressure is 5 - 20 Mpa, and the time is 1 - 2 h.

[0049] In some specific embodiments, the composition of the powder high-speed steel and the powder metallurgy construction method further include: S4: Prepare powder high-speed steel according to the powder metallurgy process obtained in S3, and check whether the microstructure and mechanical properties meet the expectations.

[0050] The present invention will be described below through specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and not to limit the scope of the present invention. Without departing from the spirit and scope of the inventive concept, changes and advantages that can be conceived by those skilled in the art are included in the present invention, and the scope of protection of the present invention is the appended claims and any equivalents thereof.

[0051] Embodiment 1 A composition and powder metallurgy construction method of powder high-speed steel, the process is as Figure 1 shown, and the specific process is as follows: S1: Through machine learning, screen alloy formulas that meet the required mechanical properties, and the specific steps are as follows: S1.1: Collect information on the composition, preparation temperature, microstructure and mechanical properties of powder high-speed steel in the literature and production.

[0052] S1.2: Screen and clean the data collected in step S1.1 to obtain a dataset containing composition, microstructure and mechanical properties.

[0053] S1.3: Use 80% of the data in the dataset obtained in step S1.2 as input, and use the random forest algorithm to train the model to obtain a relationship model between composition - microstructure - mechanical properties; the process of the random forest algorithm is as follows: S1.3.1: Apply the Bagging random sampling method to randomly draw k sample sets from the original training set with replacement to form k decision trees; S1.3.2: Specify the value of m, randomly draw m variables from the samples, and at the same time, the selection of feature attributes during the splitting process satisfies the principle of minimum node impurity; S1.3.3: Repeat the above step S1.3.2 until no further splitting is possible to form a forest composed of multiple decision trees; S1.3.4: Take the average value of the predicted values output by N decision trees.

[0054] S1.4: Input the remaining 20% of the data in the dataset obtained in step S1.2 into the relationship model obtained in step S1.3 to verify the reliability of the model.

[0055] S1.5: Input new powder high-speed steel composition parameters into the relationship model obtained in step S1.3 to predict its mechanical properties (including hardness, flexural strength and impact toughness) and microstructure.

[0056] S1.6: Statistically analyze the mechanical properties predicted in step S1.5. According to the requirements of high hardness and high toughness, the types of additive elements selected are Co, Mo, W, Cr, V, and C. The powder high-speed steel formula is as follows: 0 - 15 wt.% Co, 0 - 10 wt.% Mo, 0 - 10 wt.% W, 0 - 10 wt.% Cr, 0 - 10 wt.% V, 0 - 5 wt.% C, with the balance being iron.

[0057] S2: Through high-throughput thermodynamic calculations, screen the alloy formula and design the heat treatment process. The specific steps are as follows: S2.1: Consult and collect the thermodynamic parameters of the binary systems (i.e., Fe-containing binary alloy systems composed of two elements in Fe-Co-Mo-W-Cr-V-C) and ternary systems (i.e., Fe-containing ternary alloy systems composed of three elements in Fe-Co-Mo-W-Cr-V-C) in the literature. Integrate the above parameters, and use the parameter optimization module of the thermodynamic calculation software Pandat to further optimize the systems with conflicting parameters to obtain a multi-component Fe-based thermodynamic database.

[0058] S2.2: Input the low-order alloy systems (Fe-Co-Mo alloy system, Fe-Cr-C alloy system) into the multi-component Fe-based thermodynamic database obtained in step S2.1 for calculation, and compare the calculation results with the experimental data in the literature to verify the reliability of the multi-component database.

[0059] S2.3: Use the multi-component Fe-based thermodynamic database constructed in step S2.1 to perform high-throughput calculations on the formula obtained in step S1.6. The calculation contents include: the liquidus temperature of alloys with different formulas, the types and contents of carbide phases under different formulas and temperature conditions, and the solid solubility of elements in the matrix. Representative results of the above high-throughput calculations are as Figure 2 and Figure 3 shown. The calculation results show that the liquidus temperature corresponding to the phase region where "austenite + carbide" exists in each alloy is between 1100 - 1300 °C. The optimal alloy formula that can meet the predicted organizational structure in step S1 is: 7 - 12 wt.% Co, 6 - 13 wt.% Mo, 5 - 10 wt.% W, 5 - 12 wt.% Cr, 3 - 10 wt.% V, 1.8 - 2.5 wt.% C, with the balance being Fe.

[0060] S2.4: According to the equilibrium phase diagram and non-equilibrium solidification path under the alloy formula obtained in S2.3 calculated using the multi-component Fe-based thermodynamic database, design the heat treatment process as follows: First, perform primary preheating at 950 °C for 1 hour, then hold at 1134 °C for 30 minutes, and then temper at 550 °C three times, each time for 2 hours.

[0061] S3: Prepare a powder high-speed steel sample and conduct microstructure and mechanical property inspections. The specific steps are as follows: S3.1: According to the formula obtained in step S2.3, weigh the following raw materials by weight percentage (the particle size of each raw material is 5 - 15 μm, and the oxygen content ≤ 0.1%): 11 wt.% Co powder, 6 wt.% Mo powder, 5 wt.% W powder, 6 wt.% Cr powder, 8 wt.% V powder, 1.8 wt.% C powder, and the balance is Fe powder. Put all the above raw materials into a three-dimensional mixer for sufficient mixing to obtain mixed powder materials.

[0062] S3.2: Put the mixed powder materials prepared in step S3.1 and a binder into a planetary ball mill. The dispersion medium is alcohol, the filling coefficient is 0.6, the grinding ball medium is cemented carbide balls, the ball-to-material ratio is 7:1, and the ball milling time is 96 hours to obtain a mixed slurry after ball milling.

[0063] S3.3: Put the mixed slurry prepared in step S3.2 into a vacuum drying oven for drying. The drying temperature is 90 °C, and the drying time is 3 hours. Put the dried powder into a sieve granulator for granulation.

[0064] S3.4: Put the granulated powder in step S3.3 into a hydraulic press and press it with a pressure of 350 Mpa for a holding time of 60 s to obtain a green compact.

[0065] S3.5: Put the green compact prepared in step S3.4 into a tube furnace for degreasing treatment at 500 °C for 6 hours. After cooling, transfer it to a vacuum sintering furnace for sintering treatment. The sintering temperature is 1240 °C, and the holding time is 2 hours; after cooling, transfer the sample to a hot press sintering furnace for hot press sintering. The sintering temperature is 1100 °C, the sintering pressure is 10 Mpa, and the holding and pressure maintaining time is 1 hour to further improve the density of the product and obtain a sintered sample.

[0066] S3.6: According to the heat treatment process designed in step S2.4, conduct heat treatment on the sintered sample prepared in step S3.5 to obtain a powder high-speed steel sample, and conduct microscopic structure observation and mechanical property testing on it. The results are as Figure 4 and Figure 6 shown.

[0067] Example 2 A composition and powder metallurgy construction method of a powder high-speed steel, the process is as Figure 1 shown, and the specific process is as follows: S1: Through machine learning, screen alloy formulas that meet the required mechanical properties. The specific steps are as follows: S1.1: Collect information on the composition, preparation temperature, microstructure, and mechanical properties of powder high-speed steel in the literature and production.

[0068] S1.2: Screen and clean the data collected in step S1.1 to obtain a dataset containing composition, microstructure, and mechanical properties.

[0069] S1.3: Use 80% of the data in the dataset obtained in step S1.2 as input, and train a model using the random forest algorithm to obtain a relationship model between composition - microstructure - mechanical properties; the process of the random forest algorithm is as follows: S1.3.1: Apply the Bagging random sampling method to randomly draw k sample sets from the original training set with replacement to form k decision trees. S1.3.2: Specify the value of m, randomly draw m variables from the samples, and at the same time, the selection of feature attributes during the splitting process satisfies the principle of minimum node impurity. S1.3.3: Repeat the above step S1.3.2 until no further splitting is possible to form a forest composed of multiple decision trees. S1.3.4: Take the average of the predicted values output by N decision trees.

[0070] S1.4: Input the remaining 20% of the data in the dataset obtained in step S1.2 into the relationship model obtained in step S1.3 to verify the reliability of the model.

[0071] S1.5: Input new powder high - speed steel composition parameters into the relationship model obtained in step S1.3 to predict its mechanical properties (including hardness, flexural strength, and impact toughness) and microstructure.

[0072] S1.6: Statistically analyze the mechanical properties predicted in step S1.5. According to the requirements of high hardness and high toughness, the types of added elements selected are Co, Mo, W, Cr, V, and C. The powder high - speed steel formula is as follows: 0 - 15wt.% Co, 0 - 10wt.% Mo, 0 - 10wt.% W, 0 - 10wt.% Cr, 0 - 10wt.% V, 0 - 5wt.% C, and the balance is iron.

[0073] S2: Screen alloy formulas and design heat treatment processes through high - throughput thermodynamic calculations. The specific steps are as follows: S2.1: Consult and collect the thermodynamic parameters of the binary systems (i.e., Fe - containing binary alloy systems composed of two elements in Fe - Co - Mo - W - Cr - V - C) and ternary systems (i.e., Fe - containing ternary alloy systems composed of three elements in Fe - Co - Mo - W - Cr - V - C) in the literature, integrate the above parameters, and further optimize the systems with conflicting parameters using the parameter optimization module of the thermodynamic calculation software Pandat to obtain a multi - component Fe - based thermodynamic database.

[0074] S2.2: Input the low-order alloy system (Fe-Co-Mo alloy system, Fe-Cr-C alloy system) into the multivariate Fe-based thermodynamic database obtained in step S2.1 for calculation, and compare the calculation results with the experimental data in the literature to verify the reliability of the multivariate database.

[0075] S2.3: Use the multi-element Fe-based thermodynamic database constructed in step S2.1 to perform high-throughput calculations on the formula obtained in step S1.6. The calculation contents include: liquidus temperature of alloys with different formulas, type and content of carbide phase under different formulas and temperature conditions, and solid solubility of elements in the matrix. The calculation results show that the liquidus temperature corresponding to the phase region where "austenite + carbide" exists in each alloy is between 1100 and 1300°C, and the optimal alloy formula that can meet the organizational structure predicted in step S1 is: 7-12wt.%Co, 6-13wt.%Mo, 5-10wt.%W, 5-12wt.%Cr, 3-10wt.%V, 1.8-2.5wt.%C, and the balance is Fe.

[0076] S2.4: Based on the equilibrium phase diagram and non-equilibrium solidification path of the alloy formula obtained in S2.3 calculated using the multi-element Fe-based thermodynamic database, the heat treatment process is designed as follows: first, preheat at 950°C for 1 hour, then keep at 1134°C for 30 minutes, and then temper at 550°C three times, each time for 2 hours.

[0077] S3: Prepare powder high speed steel samples and test their structure and mechanical properties. The specific steps are as follows: S3.1: According to the formula obtained in step S2.3, weigh the following raw materials in weight percentage (the particle size of each raw material is 5-15 μm, and the oxygen content is ≤0.1%): 11wt.% Co powder, 6wt.% Mo powder, 5wt.% W powder, 6wt.% Cr powder, 8wt.% V powder, 2.5wt.% C powder, and the balance is Fe powder. Put all the above raw materials into a three-dimensional mixer and mix them thoroughly to obtain a mixed powder.

[0078] S3.2: Put the mixed powder and binder prepared in step S3.1 into a planetary ball mill. The dispersion medium is alcohol, the filling coefficient is 0.6, the grinding ball medium is carbide balls, the ball-to-material ratio is 5:1, and the ball milling time is 48 hours. After ball milling, a mixed slurry is obtained.

[0079] S3.3: Place the mixed slurry obtained in step S3.2 in a vacuum drying oven for drying at 90°C for 3 hours, and place the dried powder in a sieving granulator for granulation.

[0080] S3.4: Put the granulated powder obtained in step S3.3 into a hydraulic press, press it under a pressure of 350 Mpa for a holding time of 60 s to obtain a green compact.

[0081] S3.5: Put the green compact obtained in step S3.4 into a tube furnace, carry out debinding treatment at 500 °C for 6 hours, transfer it to a vacuum sintering furnace for sintering treatment after cooling, the sintering temperature is 1190 °C, and keep the temperature for 2 hours; transfer the sample to a hot press sintering furnace for hot press sintering after cooling, the sintering temperature is 1050 °C, the sintering pressure is 10 Mpa, and the holding and pressurizing time is 1 hour to further improve the density of the product and obtain a sintered sample.

[0082] S3.6: According to the heat treatment process designed in step S2.4, perform heat treatment on the sintered sample obtained in step S3.5 to obtain a powder high-speed steel sample, observe its microstructure and test its mechanical properties, and the results are as Figure 5 and Figure 6 shown.

[0083] Unless otherwise defined, all technical terms and scientific terms used in the present invention have the same meanings as those commonly understood by those of ordinary skill in the technical field to which the present disclosure belongs. The raw materials and equipment used in the present invention are conventional raw materials and equipment in the field and can be obtained from conventional commercial channels without special instructions; the methods used in the present invention are conventional methods in the field without special instructions.

[0084] The above are only the preferred embodiments of the present invention, and do not impose any limitations on the present invention. Any simple modifications, changes and equivalent transformations made to the above embodiments according to the technical essence of the present invention still fall within the protection scope of the technical solutions of the present invention.

Claims

1. A composition of a powder high-speed steel and a powder metallurgy building method, characterized in that, Including: S1: Collect data of powder high-speed steel in literature and production, train a machine learning model, obtain a relationship model between composition - microstructure - mechanical properties, and screen out alloy formulations with desired mechanical properties; S2: Construct an alloy thermodynamics database, input the alloy formulations obtained in S1 for high-throughput calculation; according to the calculation results, screen out alloy formulations that meet the microstructures predicted in S1, and design corresponding heat treatment processes; S3: Construct the preparation process of powder high-speed steel as follows: Mix materials according to the alloy formulations obtained in S2, after pressing and activated sintering, perform heat treatment according to the heat treatment process obtained in S2 to obtain powder high-speed steel.

2. The composition and powder metallurgy building method of the powder high-speed steel according to claim 1, characterized in that Step S1 specifically includes: S1.1: Collect information on the composition, preparation temperature, microstructure, and mechanical properties of powder high-speed steel in literature and production. After screening and cleaning, obtain a data set; S1.2: Use part of the data in the data set obtained in S1.1 as input, and use the random forest algorithm to train the model to obtain a relationship model between composition - microstructure - mechanical properties; S1.3: Use the remaining data in the data set obtained in S1.1 to verify the reliability of the relationship model obtained in S1.2; S1.4: Input the composition parameters of the new powder high-speed steel into the relationship model obtained in S1.2, predict the mechanical properties, and select the alloy formulations with desired mechanical properties.

3. The composition and powder metallurgy building method of the powder high-speed steel according to claim 2, characterized in that, In step S1.2, the process of training the model using the random forest algorithm includes: S1.2.1: Apply the Bagging random sampling method to randomly draw k sample sets from the original training set with replacement to form k decision trees; S1.2.2: Specify the value of m, randomly draw m variables from the samples, and at the same time, the selection of characteristic attributes during the splitting process satisfies the principle of minimum node impurity; S1.2.3: Repeat S1.2.2 until no further splitting is possible to form a forest composed of multiple decision trees; S1.2.4: Take the average value of the predicted values output by N decision trees.

4. The composition and powder metallurgy building method of the powder high-speed steel according to claim 1, characterized in that, Step S2 specifically includes: S2.1: Collect the alloy system thermodynamics parameters related to the alloy formulations obtained in S1 in the literature, and integrate them to construct an alloy thermodynamics database; S2.2: Input the low-order alloy systems related to the alloy formulations obtained in S1, compare the output results with the experimental data in the literature, and verify the reliability of the alloy thermodynamics database; S2.3: Use the alloy thermodynamics database to perform high-throughput calculation on the alloy formulations obtained in S1, and screen out alloy formulations that meet the microstructures predicted in S1; S2.4: According to the equilibrium phase diagram and non-equilibrium solidification path of the alloy formulations obtained in S2.3 calculated by the alloy thermodynamics database, design the alloy heat treatment process to obtain an alloy with a specific phase composition and solid solubility.

5. According to the composition of the powder high-speed steel and the powder metallurgy construction method described in claim 4, characterized in that: In step S2.1, the alloy systems related to the alloy formulations obtained in S1 include: a Fe-containing binary alloy system composed of two elements in the alloy formulations obtained in S1, a Fe-containing ternary alloy system composed of three elements in the alloy formulations obtained in S1; and / or In step S2.2, the low-order alloy system related to the alloy formula obtained in S1 is: an Fe-containing alloy system composed of three elements in the alloy formula obtained in S1.

6. The composition and powder metallurgy building method of the powder high-speed steel according to claim 1 or 4, characterized in that, In step S2, the content of high-throughput calculation includes one or more of the following a to d: a. The liquidus temperature of alloys with different formulas; b. The solidus temperature of alloys with different formulas; c. The composition and content of precipitated phases under different formulas and temperature conditions; d. The solid solubility of elements in the matrix.

7. The composition and powder metallurgy building method of the powder high-speed steel according to claim 1, characterized in that, In step S3, the preparation process of powder high-speed steel is specifically as follows: S3.1: Prepare materials according to the alloy formula obtained in S2, mix and ball-mill them together with a binder and a dispersing medium, granulate after drying, and press them into a mold to obtain a green compact; S3.2: Perform degreasing treatment, vacuum sintering, and hot-press sintering on the green compact in sequence to obtain an alloy sample.

8. The composition and powder metallurgy building method of the powder high-speed steel according to claim 7, characterized in that, In step S3.1, the dispersing medium is alcohol and / or n-butane; the conditions for the mixed ball-milling are as follows: the filling coefficient is 0.4 - 0.6, the ball-milling medium is cemented carbide balls and / or stainless steel balls, the ball-to-material ratio is 3 - 8:1, and the ball-milling time is 24 - 96 h.

9. The composition and powder metallurgy building method of the powder high-speed steel according to claim 7, characterized in that, In step S3.2, the temperature of the degreasing treatment is 500 - 550 °C, and the time is 5 - 6 h; the temperature of the vacuum sintering is 1100 - 1250 °C, and the time is 1 - 2 h; the temperature of the hot-press sintering is 1050 - 1200 °C, the pressure is 5 - 20 Mpa, and the time is 1 - 2 h.

10. The composition and powder metallurgy building method of the powder high-speed steel according to claim 1 or 7, characterized in that, It also includes: S4: Prepare powder high-speed steel according to the powder metallurgy process obtained in S3, and check whether the organizational structure and mechanical properties meet the expectations.

Images