A kind of method that designs single crystal super alloy solid solution system
Technical field
The present invention relates to design the nonlinear method of single crystal super alloy solid solution system, particularly a kind of method of the design single crystal super alloy solid solution system based on artificial neural network.
Background technology
Single crystal super alloy is owing to having higher warm ability, outstanding military service performance and the good antioxidant property of holding, and is the advanced engine preferred material of holding warm turbine blade at present and in the future.Because more and more higher to the requirement of use temperature, the alloying consumption of refractory element such as W, Mo, Re, Ru etc. is more and more large, the refractory element total amount is about 14w.t.% in the first-generation single crystal super alloy, brings up to approximately 20w.t.% during to the third generation such as CMSX-10.Simultaneously, the use of the second-phase forming elements such as Al, Ti, Ta, Nb is also near solid solubility limit.Also brought thus following shortcoming: solid solubility temperature is too high, and low melting point phase initial melting temperature is excessively low, and the solid solution window is narrow, and the solid solution process is slow, and the solid solution system design is difficult even can not realize solid solution.For example, the standard solid solution system of CMSX-10 is from 1315 ℃ of ladder-elevating temperatures to 1365 ℃, and 50 ℃ of temperature spans reach more than 40 hour when total.This shows, realize judging fast that can superalloy solid solution and Exact Design solid solution system, for superalloy design and industrial production important in inhibiting.
And traditional high temperature alloy research method, generally to pass through on alloy designs, mother alloy melting, single crystal casting, first fusing point metallographic, the solid solubility temperature roll off the production line, solid solution system design and adjustment, just can obtain the solid solution system.An entire flow generally needs half a year even time more of a specified duration.The situation that material can not be realized solid solution inevitably appears in the conventional alloys method of design simultaneously, causes unnecessary loss.
Simultaneously, some linear systems that are used for simulating heterogeneity superalloy performance are developed, but along with the alloy element range extension, Alloying Amount is near the superalloy design limit, and its composition and performance more show nonlinear feature.Major cause is that single crystal super alloy is γ-γ ' biphasic system, all alloy elements influence each other in two-phase, so that each element distribution constant all changes, cause linear system interalloy element prematrix to occur obviously to change and the increase simulation error.So for the wide variation of composition, the nonlinear system simulation result is closing to reality more, it is more accurate to judge.In non-linear system, the Back-Propagation artificial neural network technology is the most ripe, and compatible and error shows more outstanding.
Summary of the invention
The objective of the invention is to realize to judge fast that can superalloy solid solution and Exact Design solid solution system, propose a kind of method of the design single crystal super alloy solid solution system based on artificial neural network.
The invention provides a kind of novel method of the single crystal super alloy solid solution system design based on artificial neural network, pass through first the characteristic temperature database of heterogeneity single crystal super alloy, training Back-Propagation artificial neural network, afterwards, the artificial neural network of having trained is inputted alloying constituent to be measured, calculate characteristic temperature and design the solid solution system.
Here the characteristic temperature of selecting is solidus temperature, liquidus temperature, freezing range, second-phase solvent temperature, low melting point phase initial melting temperature and heat treatment state initial melting temperature.Wherein, solidus temperature is the temperature that begins to melt on the single crystal alloy equilibrium phase diagram; Liquidus temperature is the temperature that melts fully on the single crystal alloy equilibrium phase diagram; Dendrite form nuclear temperature was to the temperature head of interdendritic complete solidification temperature when the freezing range was single crystal casting; The second-phase solvent temperature is the solvent temperature of coherence γ ' in the single crystal super alloy; Low melting point phase initial melting temperature is the fusing point that the low melting point phase that forms latter stage is solidified in the interdendritic; The heat treatment state initial melting temperature is the initial melting temperature of alloy behind the elimination as cast condition microsegregation.
These solid solution system design method concrete steps are as follows:
Step 1, set up the database of the single crystal super alloy characteristic temperature of following composition range (w.t.%):
W |
0~8% |
Mo |
0~16% |
Ta |
0~9% |
Al |
4~8.5% |
Ti |
0~4% |
Nb |
0~2% |
Re |
0~7% |
Ru |
0~3% |
Co |
0~10% |
Cr |
0~10% |
Y |
0~0.1% |
Ni |
bal. |
Wherein, " bal. " is the weight percentage of Ni element, and its numerical value is 100% to deduct other element wt per-cent sums.
Select artificial neural network to be input as the single crystal super alloy composition, be output as solidus temperature, liquidus temperature, freezing range, second-phase solvent temperature, low melting point phase initial melting temperature and heat treatment state initial melting temperature.Setting up artificial neural network is the Back-Propagation neural network, adopts three-decker, comprises input layer, hidden layer and output layer; The input layer number is 12, and the output layer neuron number is 6,14 of hidden layer neuron numbers, and input layer and output layer neurone use the purelin function representation, and hidden layer uses the log-sigmoid function representation.
Step 2, from database, select 300 above data samples as learning sample, training of human artificial neural networks use error back-propagation algorithm, netinit condition are that (0,0.5) random assignment is to weight matrix, (0,1) random assignment is to the threshold values matrix.The definition error e is:
Wherein, t
kBe the value of learning sample, y
kBe the output of artificial neural network output layer, k is number of training.
Design requirements is that the training objective error is e<0.001.When output error satisfied the target error of setting requirement, training stopped, and preserved weight matrix and threshold values matrix.
Step 3: utilize the artificial neural network of having trained in the step 2, input alloying constituent to be measured, calculate the characteristic temperature of this alloy to be measured, and design superalloy solid solution system.Concrete grammar is as follows:
Situation one: when the second-phase solvent temperature〉during the heat treatment state initial melting temperature, single crystal super alloy can not solid solution;
Situation two: when second-phase solvent temperature≤heat treatment state initial melting temperature and low melting point phase initial melting temperature〉second-phase solvent temperature and freezing range〉25 ℃ times, single crystal super alloy solid solution system is: second-phase solvent temperature/16h;
Situation three: when second-phase solvent temperature≤heat treatment state initial melting temperature and low melting point phase initial melting temperature〉when second-phase solvent temperature and freezing range≤25 ℃, single crystal super alloy solid solution system is: second-phase solvent temperature/10h;
Situation four: when second-phase solvent temperature≤heat treatment state initial melting temperature and low melting point phase initial melting temperature≤second-phase solvent temperature and freezing range〉25 ℃ times, single crystal super alloy solid solution system is: low melting point phase initial melting temperature/4h+ second-phase solvent temperature/16h;
Situation five: when second-phase solvent temperature≤heat treatment state initial melting temperature and low melting point phase initial melting temperature≤second-phase solvent temperature and freezing range≤25 ℃, single crystal super alloy solid solution system is: low melting point phase initial melting temperature/4h+ second-phase solvent temperature/10h;
The invention has the beneficial effects as follows:
The artificial neural network input terminus of the present invention's training has been contained most alloying constituent and content range that superalloy adopts, realized that can quick judgement single crystal super alloy solid solution and Exact Design solid solution system, can be used for instructing single crystal super alloy design and industrial production.
Description of drawings
Fig. 1 is solid solution system design method flow diagram provided by the invention;
Fig. 2 is the scatter diagram that systematic error changes with the variation of hidden layer neuron number among the embodiment, and an error hour hidden layer neuron number is 14;
Fig. 3 a is the as-cast structure pattern of single crystal super alloy described in the embodiment 1;
Fig. 3 b is the microstructure morphology of the heat treatment state of single crystal super alloy described in the embodiment 1;
Fig. 3 c is the as-cast structure pattern of single crystal super alloy among the embodiment 2;
Fig. 3 d is the heat treatment state microstructure morphology of single crystal super alloy among the embodiment 2;
Fig. 3 e is the as-cast structure pattern of single crystal super alloy among the embodiment 3;
Fig. 3 f is the heat treatment state microstructure morphology of single crystal super alloy among the embodiment 3.
Embodiment
The present invention is further described below in conjunction with drawings and Examples.
Embodiment 1
Adopt method of design provided by the invention to carry out single crystal super alloy solid solution system design, concrete steps are as follows:
Step 1, set up the database of the single crystal super alloy characteristic temperature of following composition range (w.t.%):
W |
0~8% |
Mo |
0~16% |
Ta |
0~9% |
Al |
4~8.5% |
Ti |
0~4% |
Nb |
0~2% |
Re |
0~7% |
Ru |
0~3% |
Co |
0~10% |
Cr |
0~10% |
Y |
0~0.1% |
Ni |
bal. |
Select artificial neural network to be input as the single crystal super alloy composition, be output as solidus temperature, liquidus temperature, freezing range, second-phase solvent temperature, low melting point phase initial melting temperature and heat treatment state initial melting temperature.Setting up artificial neural network is the Back-Propagation neural network, adopts three-decker, comprises input layer, hidden layer and output layer; The input layer number is 12, and the output layer neuron number is 6,14 of hidden layer neuron numbers.Input layer and output layer neurone use the purelin function representation, and hidden layer uses the log-sigmoid function representation.
This hidden layer neuron number selection principle is: systematic error minimizes.Systematic error is that the artificial neural network frequency of training is greater than the error amount of 1000 back balance systems.As shown in Figure 2, the artificial neural network that obtains for single crystal super alloy characteristic temperature database training is when the hidden layer neuron number is 14() time, systematic error is minimum, and therefore selecting the hidden layer neuron number is 14.
Step 2, from database, select 311 data sample training artificial neural networks, training of human artificial neural networks use error back-propagation algorithm, the netinit condition is (0,0.5) random assignment is to weight matrix, (0,1) random assignment is to the threshold values matrix, and the training objective error is e<0.001.When the satisfied setting of output error required, training stopped, and preserves weight matrix and threshold values matrix.
Step 3: utilize the artificial neural network of having trained in the step 2, input following alloying constituent to be measured:
W |
0% |
Mo |
11% |
Ta |
6% |
Al |
7.6% |
Ti |
0% |
Nb |
0% |
Re |
3% |
Ru |
0% |
Co |
5% |
Cr |
0% |
Y |
0% |
Ni |
67.4% |
The characteristic temperature that calculates this composition is 1360 ℃ of 1389 ℃ of solidus temperatures, 1411.2 ℃ of liquidus temperatures, 22.2 ℃ of freezing ranges, 1371.2 ℃ of second-phase solvent temperatures, 1351 ℃ of low melting point phase initial melting temperatures and heat treatment state initial melting temperatures.Can obtain 1371.2 ℃ of second-phase solvent temperatures〉1360 ℃ of heat treatment state initial melting temperatures, namely meet situation one, alloy can't be realized solid solution.Experimental result as shown in Figure 3, as-cast structure is seen Fig. 3 a, 1360 ℃/50 hours microtextures of thermal treatment are seen Fig. 3 b, not solid solution.Experimental result conforms to design solid solution system.
Embodiment 2
Step 1, set up the database of the single crystal super alloy characteristic temperature of following composition range (w.t.%):
W |
0~8% |
Mo |
0~16% |
Ta |
0~9% |
Al |
4~8.5% |
Ti |
0~4% |
Nb |
0~2% |
Re |
0~7% |
Ru |
0~3% |
Co |
0~10% |
Cr |
0~10% |
Y |
0~0.1% |
Ni |
bal. |
Select artificial nerve network model to be input as the single crystal super alloy composition, be output as solidus temperature, liquidus temperature, freezing range, second-phase solvent temperature, low melting point phase initial melting temperature and heat treatment state initial melting temperature.Setting up artificial neural network is the Back-Propagation neural network, adopts three-decker, comprises input layer, hidden layer and output layer; The input layer number is 12, and the output layer neuron number is 6,14 of hidden layer neuron numbers, and input layer and output layer neurone use the purelin function representation, and hidden layer uses the log-sigmoid function representation.
Step 2, from database, select 311 data samples, training of human artificial neural networks use error back-propagation algorithm, the netinit condition is (0,0.5) random assignment is to weight matrix, (0,1) random assignment is to the threshold values matrix, and the training objective error is e<0.001.When the satisfied setting of output error required, training stopped, and preserves weight matrix and threshold values matrix.
Step 3: utilize the artificial neural network of having trained in the step 2, input following alloying constituent to be measured:
W |
5% |
Mo |
2% |
Ta |
6% |
Al |
7.6% |
Ti |
1% |
Nb |
0% |
Re |
1.5% |
Ru |
0% |
Co |
0% |
Cr |
0% |
Y |
0% |
Ni |
76.9% |
The characteristic temperature that calculates this alloy to be measured is 1345 ℃ of 1355.3 ℃ of solidus temperatures, 1391.2 ℃ of liquidus temperatures, 35 ℃ of freezing ranges, 1336.6 ℃ of second-phase solvent temperatures, 1333 ℃ of low melting point phase initial melting temperatures and heat treatment state initial melting temperatures.There is following relation:
1345 ℃ of 1336.6 ℃≤heat treatment state of second-phase solvent temperature initial melting temperatures;
1336.6 ℃ of 1333 ℃≤second-phase of low melting point phase initial melting temperature solvent temperatures;
The freezing range〉25 ℃
Namely meet situation four, single crystal super alloy solid solution system to be measured is: 1333 ℃/4h+1336.6/16h.As-cast structure is seen Fig. 3 c, and 1333 ℃/4h+1336.6/16h thermal treatment microtexture is seen Fig. 3 d, realizes solid solution.Experimental result conforms to the design result of solid solution system of the present invention.
Embodiment 3
The present embodiment and embodiment 1,2 difference only exist: step 3: utilize the artificial neural network of having trained in the step 2, input following target component:
W |
0% |
Mo |
9.5% |
Ta |
3% |
Al |
7.8% |
Ti |
0% |
Nb |
0% |
Re |
1.5% |
Ru |
0% |
Co |
0% |
Cr |
1.5% |
Y |
0.05% |
Ni |
76.65% |
The characteristic temperature that calculates this composition is 1352 ℃ of 1371.7 ℃ of solidus temperatures, 1401.8 ℃ of liquidus temperatures, 30 ℃ of freezing ranges, 1326.1 ℃ of second-phase solvent temperatures, 1330 ℃ of low melting point phase initial melting temperatures and heat treatment state initial melting temperatures.There is following relation:
1352 ℃ of 1326.1 ℃≤heat treatment state of second-phase solvent temperature initial melting temperatures;
1330 ℃ of low melting point phase initial melting temperatures〉1326.1 ℃ of second-phase solvent temperatures;
The freezing range〉25 ℃,
Namely meet situation two, superalloy solid solution system is: 1326.1 ℃/16h.As-cast structure is seen Fig. 3 e, and 1327 ℃/16h thermal treatment microtexture is seen Fig. 3 f, realizes solid solution.Experimental result conforms to the design result of solid solution system of the present invention.
The A that relates in the specification sheets ℃/Bh is illustrated under the A temperature and is incubated B hour.
The single crystal super alloy characteristic temperature that is used for the training of human artificial neural networks derives from published or certified single crystal super alloy characteristic temperature data, and the predictable single crystal super alloy composition range of artificial neural network is the published or certified high temperature alloy composition scope of typing learning sample.For example, Re element published or confirmed that superalloy addition (w.t.%) is 0(trade mark PWA1480), 0(trade mark SRR99), 0(trade mark CMSX-6), 1.5(IC27 confirms), 1.5(IC21 confirms), 3(trade mark PWA1484), 3(trade mark ReneN5), 3(trade mark CMSX-4), 4(trade mark MC-NG), 5(trade mark ReneN6), 5(trade mark TMS-75), 5(trade mark TMS-162), 7(trade mark CMSX-10) etc., then the predictable Re elemental range of artificial neural network is 0~7(w.t.%).In like manner obtain the predictable scope of other elements, this predictable range has contained most alloying constituent and content range that superalloy adopts.