CN115392013A - Novel noble metal electric contact device service life prediction method based on machine learning - Google Patents
Novel noble metal electric contact device service life prediction method based on machine learning Download PDFInfo
- Publication number
- CN115392013A CN115392013A CN202210992150.XA CN202210992150A CN115392013A CN 115392013 A CN115392013 A CN 115392013A CN 202210992150 A CN202210992150 A CN 202210992150A CN 115392013 A CN115392013 A CN 115392013A
- Authority
- CN
- China
- Prior art keywords
- electric contact
- noble metal
- mechanical property
- service life
- metal electric
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/04—Ageing analysis or optimisation against ageing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention relates to the technical field of noble metal electric contact materials, and provides a novel noble metal electric contact device service life prediction method based on machine learning, which comprises the following steps: the mechanical property parameters of every two mutually-matched existing noble metal electric contact materials and the service life of a corresponding electric contact device form an initial sample; extracting mechanical property characteristic vectors and life labels from the initial samples to form training samples; establishing and training a life prediction model of the noble metal electric contact device based on a machine learning algorithm by taking the mechanical property characteristic vector as input and the life label as output; and obtaining the mechanical property characteristic vectors of the novel noble metal electric contact material sample and the noble metal electric contact material sample matched with the novel noble metal electric contact material sample, inputting the mechanical property characteristic vectors into the trained model, and obtaining the predicted service life interval of the electric contact device which is respectively made into two electric contact elements by using the two materials. The invention can realize high-precision and high-efficiency prediction of the service life of the corresponding electric contact device in a small sample form, and saves time and capital cost.
Description
Technical Field
The invention relates to the technical field of noble metal electric contact materials, in particular to a novel noble metal electric contact device service life prediction method based on machine learning.
Background
The sliding electrical contact arrangement provides for the transmission of power and signals between the relatively rotating parts of the device through the mating of the brushes with the conductive slip rings. The noble metal electric contact material has the advantages of no need of lubrication, good temperature adaptability, no high-temperature creep, long service life and the like, and is widely applied to sliding electric contact devices in the fields of radars, spacecrafts and the like. With the development of modern technology, the requirements on military equipment and the on-orbit service cycle of a spacecraft are continuously improved, and the requirements on the reliability and the service life of the electric contact device are also continuously improved. Researchers in the relevant field are continuously developing high-performance noble metal electrical contact materials to continuously adapt to the requirements of use scenes. However, the service performance and service life of new materials can only be verified by making related finished products for reuse, and in this case, on one hand, the product development process is slowed down, on the other hand, a large amount of precious metal raw materials are added, and the research and development cost is increased.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a novel precious metal electric contact device service life prediction method based on machine learning, which can realize service life prediction of a corresponding electric contact device only in a small sample form without processing materials to finished products, greatly reduces the investment of precious metal raw materials in the new material research and development process, saves the time cost and the capital cost, and greatly improves the precision and the efficiency of service life prediction.
The technical scheme of the invention is as follows:
a service life prediction method of a novel noble metal electric contact device based on machine learning comprises the following steps:
step 1: obtaining an initial sample set
Obtaining mechanical property parameters of the existing noble metal electrical contact material, wherein the mechanical property parameters of every two existing noble metal electrical contact materials which can be matched with each other and the service life of an electrical contact device which is respectively manufactured into two electrical contact elements by using the two existing noble metal electrical contact materials form an initial sample, and obtaining an initial sample set;
and 2, step: constructing a training sample set
Preprocessing the mechanical property parameters in each initial sample, and extracting corresponding mechanical property characteristic vectors; dividing the service life data of the initial sample set into intervals, numbering each interval, and taking the interval number corresponding to the service life in each initial sample as a life label of the initial sample; each mechanical property feature vector and the corresponding life label form a training sample, and a training sample set is obtained;
and 3, step 3: building and training life prediction model
Establishing a service life prediction model of the noble metal electric contact device by taking the mechanical property characteristic vector as input and the service life label as output based on a machine learning algorithm, and training the service life prediction model of the noble metal electric contact device by utilizing a training sample set;
and 4, step 4: prediction of noble metal electrical contact material life
Preparing a novel noble metal electric contact material sample and a noble metal electric contact material sample matched with the novel noble metal electric contact material; respectively carrying out mechanical property test on the two samples to obtain mechanical property parameters of the two samples; preprocessing the mechanical property parameters of the two samples, and extracting corresponding mechanical property characteristic vectors; inputting the corresponding mechanical property characteristic vector into the trained life prediction model of the noble metal electric contact device to obtain the predicted life intervals of the electric contact device with two electric contact elements respectively made of the novel noble metal electric contact material and the noble metal electric contact material matched with the novel noble metal electric contact material.
Due to the strong data processing capacity of the machine learning, the cost of manpower and material resources in industrial development can be effectively reduced, and the research and development period is shortened. This is particularly suitable for noble metal electrical contact materials, but the research application in the related field is very few. In order to reduce the research and development cost brought by the life prediction in the research and development process of the noble metal electric contact material, the invention applies machine learning to the field of the development of the noble metal electric contact material, establishes a model for predicting the life of the noble metal electric contact device based on a machine learning algorithm, is used for verifying the service performance of a new material to be developed, improves the material development efficiency, and saves a large amount of time and cost in the development process of the new material.
Further, the electrical contact device is a sliding electrical contact device, and the two electrical contact elements are an electric brush and a conductive slip ring respectively. For the sliding electrical contact device, the service life is mainly influenced by the matching wear resistance of the materials of the electric brush and the conductive slip ring, the temperature, the humidity, the sliding rotation speed and other factors of the use environment. When a new material is developed to replace an old material, the use environment and the sliding rotation speed of the electric contact device are generally kept unchanged, and the matching wear resistance of the materials of the electric brush and the conductive slip ring is different due to the difference of the hardness and the tensile strength of the two materials. Therefore, the invention pays attention to two influencing factors of hardness and tensile strength for representing the performance of the material when the service life is predicted. Specifically, the mechanical property parameters comprise the hardness and tensile strength of a noble metal electric contact material for manufacturing the electric brush and the hardness of the noble metal electric contact material for manufacturing the conductive slip ring, and the service life is the service life of the electric contact device when any one of the electric brush and the conductive slip ring fails to cause the failure of the electric contact device; the mechanical property test comprises the steps of carrying out hardness test and tensile strength test on the noble metal electric contact material for manufacturing the electric brush and carrying out hardness test on the noble metal electric contact material for manufacturing the conductive slip ring.
Further, in step 2, preprocessing the mechanical property parameters in each initial sample, and extracting corresponding mechanical property feature vectors specifically includes: all the mechanical property parameters in each initial sample are subjected to normalization processing, and the mechanical property parameters after the normalization processing form corresponding mechanical property characteristic vectors; in the step 4, the mechanical property parameters of the two samples are preprocessed, and the corresponding mechanical property characteristic vectors are extracted, which specifically includes: and all the mechanical property parameters of the two samples are subjected to normalization treatment, and the mechanical property parameters after the normalization treatment form corresponding mechanical property characteristic vectors.
Further, in step 2, the interval division is performed on the service life data in the initial sample set, and the method specifically includes: the service life data in the initial sample set are equally divided into at least 10 intervals.
Further, in the step 3, the machine learning algorithm is one of linear regression, logistic regression, support vector machine regression, and decision tree regression.
Furthermore, the novel noble metal electric contact material sample is prepared by a method of vacuum induction melting, repeated rolling and heat treatment, and specifically comprises the following steps: and (3) smelting a novel noble metal electric contact material by vacuum induction, and repeatedly rolling and thermally treating the cast alloy ingot to obtain a plate or bar with a final deformation of 60-95%.
Furthermore, the novel noble metal electric contact material is one of Ag-based alloy, au-based alloy and Pd-based alloy.
Further, the testing temperature of the mechanical property test is 30-200 ℃.
Further, in the step 4, the sample is kept at the testing temperature for 1 hour before the mechanical property is tested.
The beneficial effects of the invention are as follows:
(1) The invention constructs an initial sample by the mechanical property parameters of every two existing noble metal electric contact materials which can be matched with each other and the service life of an electric contact device which is respectively made into two electric contact elements by using the two existing noble metal electric contact materials, preprocesses the mechanical property parameters in the initial sample to extract corresponding mechanical property characteristic vectors, divides the service life data in the initial sample set into intervals to take the interval number corresponding to the service life in each initial sample as the life label of the initial sample to construct a training sample set, constructs and trains a noble metal electric contact device service life prediction model by taking the mechanical property characteristic vectors as input and the life labels as output based on a machine learning algorithm, finally, the mechanical property characteristic vectors of the novel precious metal electric contact material sample and the precious metal electric contact material sample matched with the novel precious metal electric contact material are obtained and input into a life prediction model after training, the predicted life interval of the electric contact device with two electric contact elements respectively made of the novel precious metal electric contact material and the precious metal electric contact material matched with the novel precious metal electric contact material is obtained, the novel precious metal electric contact material can be researched and developed in the process of not processing the material to a finished product, the life prediction of the corresponding electric contact device can be realized only in a small sample form, the investment of precious metal raw materials in the research and development process is greatly reduced, the time cost and the capital cost are saved, the research and development efficiency of a new material is improved, and the precision and the efficiency of the life prediction of the novel precious metal electric contact device can be greatly improved.
(2) The invention takes the hardness and tensile strength of the noble metal electric contact material for manufacturing the electric brush and the hardness of the noble metal electric contact material for manufacturing the conductive slip ring into consideration, and the parameters representing the performance of the material per se are used for extracting the characteristic vector of the mechanical property, so that the precision of the service life prediction of the novel noble metal electric contact device can be greatly improved.
Drawings
Fig. 1 is a flow chart of a method for predicting the service life of a novel noble metal electric contact device based on machine learning according to the invention.
Detailed Description
The invention will be further described with reference to the drawings and the detailed description.
As shown in fig. 1, the method for predicting the service life of the novel noble metal electric contact device based on machine learning of the invention comprises the following steps:
step 1: obtaining an initial sample set
Obtaining mechanical property parameters of the existing noble metal electrical contact material, wherein the mechanical property parameters of every two existing noble metal electrical contact materials which can be matched with each other and the service life of an electrical contact device which is respectively manufactured into two electrical contact elements by using the two existing noble metal electrical contact materials form an initial sample to obtain an initial sample set;
and 2, step: constructing a training sample set
Preprocessing the mechanical property parameters in each initial sample, and extracting corresponding mechanical property characteristic vectors; dividing intervals of the service life data in the initial sample set, numbering each interval, and taking the interval number corresponding to the service life in each initial sample as a life label of the initial sample; each mechanical property characteristic vector and the corresponding life label form a training sample to obtain a training sample set;
and step 3: building and training life prediction model
Establishing a service life prediction model of the noble metal electric contact device by taking the mechanical property characteristic vector as input and the service life label as output based on a machine learning algorithm, and training the service life prediction model of the noble metal electric contact device by utilizing a training sample set;
and 4, step 4: prediction of noble metal electric contact material life
Preparing a novel noble metal electric contact material sample and a noble metal electric contact material sample matched with the novel noble metal electric contact material; respectively carrying out mechanical property test on the two samples to obtain mechanical property parameters of the two samples; preprocessing the mechanical property parameters of the two samples, and extracting corresponding mechanical property characteristic vectors; inputting the corresponding mechanical property characteristic vector into the trained life prediction model of the noble metal electric contact device to obtain the life prediction intervals of the electric contact device with two electric contact elements respectively made of the novel noble metal electric contact material and the noble metal electric contact material matched with the novel noble metal electric contact material.
In this embodiment, the electrical contact device is a sliding electrical contact device, and the two electrical contact elements are an electric brush and a conductive slip ring respectively; the mechanical property parameters comprise the hardness and tensile strength of a noble metal electric contact material for manufacturing the electric brush and the hardness of a noble metal electric contact material for manufacturing the conductive slip ring, and the service life of the electric contact device is the service time of the electric contact device when any one of the electric brush and the conductive slip ring fails to cause the failure of the electric contact device; the mechanical property test comprises the steps of carrying out hardness test and tensile strength test on the noble metal electric contact material for manufacturing the electric brush and carrying out hardness test on the noble metal electric contact material for manufacturing the conductive slip ring.
The matching of the noble metal electric contact material means that the two noble metal electric contact materials can be matched together to be respectively used for manufacturing an electric brush and a conductive slip ring of the same electric contact device. Specifically, in step 1, the mechanical property parameters of the existing noble metal electrical contact material are obtained, which means that the hardness and tensile strength of the existing noble metal electrical contact material for manufacturing the electric brush and the hardness of the existing noble metal electrical contact material for manufacturing the conductive slip ring are obtained; the mechanical property parameters in the initial sample are the hardness and tensile strength of the existing noble metal electric contact material used for manufacturing the electric brush and the hardness of the existing noble metal electric contact material used for manufacturing the conductive slip ring in the two existing noble metal electric contact materials which can be matched together and are respectively used for manufacturing the electric brush and the conductive slip ring of the same electric contact device. In step 4, the noble metal electrical contact material matched with the novel noble metal electrical contact material can be an existing noble metal electrical contact material or a novel noble metal electrical contact material, the noble metal electrical contact material used for manufacturing the electric brush is subjected to hardness test and tensile strength test, the noble metal electrical contact material used for manufacturing the conductive slip ring is subjected to hardness test, and the obtained mechanical property parameters of the two samples comprise the novel noble metal electrical contact material, the hardness and tensile strength of the noble metal electrical contact material used for manufacturing the electric brush in the noble metal electrical contact material matched with the novel noble metal electrical contact material, and the hardness of the noble metal electrical contact material used for manufacturing the conductive slip ring.
In step 1, the mechanical property parameters of the existing noble metal electrical contact material can be obtained by looking up documents or performing mechanical property tests on the material in a real production environment, and after classifying and summarizing the obtained mechanical property parameter data of the existing noble metal electrical contact material, the records are stored in a unified format, such as a csv file format, a (ID) meter, an existing noble metal electrical contact material A for manufacturing an electric brush, and a materialHardness h of A A Tensile Strength s of Material A A The hardness h of the existing noble metal electrical contact material B for manufacturing the conductive slip ring B Respectively manufacturing the service life t of an electric contact device of the electric brush and the conductive slip ring by using the material A and the material B A-B ). Each column is the name or the value of the mechanical property parameter of each existing noble metal electric contact material, each row is a group of mechanical property parameter data of the existing noble metal electric contact material which can be matched with each other and the service life data of the electric contact device made of the existing noble metal electric contact material, and each group of data (h) is A, s A, h B, t A-B ) An initial sample is constructed. In order to ensure the accuracy of the subsequent training of the machine learning model, the data set at least comprises 100 groups of mechanical property parameters and service life data of the existing noble metal electric contact material.
In step 2, the mechanical property parameter (h) in each initial sample is measured A, s A, h B ) And carrying out normalization processing to generate characteristic values between 0 and 1, wherein the characteristic values obtained after the normalization processing form corresponding mechanical property characteristic vectors. Meanwhile, dividing the service life data in the initial sample set into K intervals by adopting an equal time interval method, numbering each interval, taking the interval number corresponding to the service life in each initial sample as a life label of the initial sample, and forming a training sample by each mechanical property characteristic vector and the corresponding life label to obtain a training sample set. Therefore, the training sample set is divided into K training sub-sample sets, the service life label of the precious metal electric contact material corresponding to the ith (i is more than 0 and less than or equal to K) training sub-sample set is i, the service life is T- (i T/K), and T is the maximum value of the service life of the initial sample set. The service life data in the initial sample set are equally divided into at least 10 intervals (K is larger than or equal to 10) so as to ensure the precision of subsequent model training, the larger the K value is, the higher the division precision is, the higher the prediction precision of the training model is, but the difficulty of training the model is also higher.
In the step 3, a machine learning algorithm selected from linear regression, logistic regression, support vector machine regression and decision tree regression is adopted to construct the service life prediction model of the noble metal electric contact device, the service life prediction model of the noble metal electric contact device is trained by utilizing a training sample set, a loss function is calculated, and the model is adjusted optimally continuously.
In this embodiment, the acquired data set is divided into two parts according to the proportion of 80% and 20% by a random selection method, wherein 80% of the two parts are training data sets, 20% of the two parts are testing data sets, data in the testing data sets are input into a trained life prediction model of the noble metal electrical contact device, and whether the accuracy of the model meets preset requirements is verified. If the requirements are met, successfully training the life prediction model of the noble metal electric contact device; if the requirements are not met, the following steps are carried out: 1) Optimizing the regression algorithm, comprising: screening a regression algorithm and adjusting internal parameters of the regression algorithm; 2) Screening mechanical performance parameters for modeling; 3) And expanding the data set, and then retraining and testing until the prediction error of the test data set meets the requirement.
In step 4, the mechanical property parameters of the two samples are preprocessed, and corresponding mechanical property characteristic vectors are extracted, wherein the steps specifically comprise: and all the mechanical property parameters of the two samples are subjected to normalization treatment, and the mechanical property parameters subjected to normalization treatment form corresponding mechanical property characteristic vectors. The novel noble metal electric contact material sample is prepared by a method of vacuum induction melting, repeated rolling and heat treatment, and specifically comprises the following steps: and (3) smelting a novel noble metal electric contact material by vacuum induction, and repeatedly rolling and thermally treating the cast alloy ingot to obtain a plate or bar with a final deformation of 60-95%. The novel noble metal electric contact material is one of Ag-based alloy, au-based alloy and Pd-based alloy. The testing temperature of the mechanical property test is 30-200 ℃. The samples were held at the test temperature for 1h before mechanical property testing.
In the embodiment, a novel noble metal electrical contact material for manufacturing a conductive slip ring is designed, the component of the novel noble metal electrical contact material is AgCuNiAl20-2-0.5, high-purity Ag, agCu alloy, agNi alloy and AgAl alloy are used as raw materials, vacuum induction melting is carried out according to the component proportion, an alloy cast ingot is obtained after casting, multiple rolling-heat treatment is carried out on the alloy cast ingot, the final rolling deformation is 90%, and a 10 x 5mm sample is taken to carry out hardness test at 30 ℃ to obtain the hardness of the alloy. Meanwhile, a novel noble metal electric contact material AuNi9 for manufacturing the electric brush is designed, high-purity Au and AuNi alloy are used as raw materials, vacuum induction melting is carried out according to the component proportion, an alloy cast ingot is obtained after casting, rolling-heat treatment is carried out on the alloy cast ingot for multiple times, the rolling deformation of the final pass is 65%, and a sample is taken to carry out hardness test and tensile strength test at 30 ℃ to obtain the hardness and tensile strength of the alloy. The hardness of the novel noble metal electric contact material AgCuNiAl20-2-0.5, the hardness and the tensile strength of AuNi9 obtained by testing are normalized, mechanical property characteristic vectors are extracted, and then the mechanical property characteristic vectors are input into a trained model to obtain a predicted service life label of an electric contact device with an electric brush made of AuNi9 and a conductive slip ring made of AgCuNiAl20-2-0.5 at the temperature of 30 ℃, so that the predicted service life is obtained. The predicted result is consistent with the actual service life of the novel electric contact device well, and the prediction error is within 5%. Therefore, the method greatly improves the accuracy of service life prediction of the novel noble metal electric contact device.
It is to be understood that the above-described embodiments are only some of the embodiments of the present invention, and not all of the embodiments. The above examples are only for explaining the present invention and do not constitute a limitation to the scope of protection of the present invention. All other embodiments, which can be derived by those skilled in the art from the above-described embodiments without any creative effort, namely all modifications, equivalents, improvements and the like made within the spirit and principle of the present application, fall within the protection scope of the present invention claimed.
Claims (9)
1. A service life prediction method of a novel noble metal electric contact device based on machine learning is characterized by comprising the following steps:
step 1: obtaining an initial sample set
Obtaining mechanical property parameters of the existing noble metal electrical contact material, wherein the mechanical property parameters of every two existing noble metal electrical contact materials which can be matched with each other and the service life of an electrical contact device which is respectively manufactured into two electrical contact elements by using the two existing noble metal electrical contact materials form an initial sample, and obtaining an initial sample set;
step 2: constructing a training sample set
Preprocessing the mechanical property parameters in each initial sample, and extracting corresponding mechanical property characteristic vectors; dividing intervals of the service life data in the initial sample set, numbering each interval, and taking the interval number corresponding to the service life in each initial sample as a life label of the initial sample; each mechanical property characteristic vector and the corresponding life label form a training sample to obtain a training sample set;
and step 3: building and training life prediction model
Establishing a service life prediction model of the noble metal electric contact device by taking the mechanical property characteristic vector as input and the service life label as output based on a machine learning algorithm, and training the service life prediction model of the noble metal electric contact device by utilizing a training sample set;
and 4, step 4: prediction of noble metal electric contact material life
Preparing a novel noble metal electric contact material sample and a noble metal electric contact material sample matched with the novel noble metal electric contact material; respectively carrying out mechanical property test on the two samples to obtain mechanical property parameters of the two samples; preprocessing the mechanical property parameters of the two samples, and extracting corresponding mechanical property characteristic vectors; inputting the corresponding mechanical property characteristic vector into the trained life prediction model of the noble metal electric contact device to obtain the life prediction intervals of the electric contact device with two electric contact elements respectively made of the novel noble metal electric contact material and the noble metal electric contact material matched with the novel noble metal electric contact material.
2. The method for predicting the service life of the novel noble metal electric contact device based on the machine learning as claimed in claim 1, wherein the electric contact device is a sliding type electric contact device, and the two electric contact elements are a brush and a conductive slip ring respectively; the mechanical property parameters comprise the hardness and tensile strength of a noble metal electric contact material for manufacturing the electric brush and the hardness of the noble metal electric contact material for manufacturing the conductive slip ring, and the service life is the service time of the electric contact device when the electric contact device fails due to the failure of any material in the electric brush and the conductive slip ring; the mechanical property test comprises the steps of performing hardness test and tensile strength test on the noble metal electric contact material for manufacturing the electric brush and performing hardness test on the noble metal electric contact material for manufacturing the conductive slip ring.
3. The method for predicting the service life of a novel noble metal electric contact device based on machine learning according to claim 1, wherein in the step 2, the mechanical property parameters in each initial sample are preprocessed, and the extraction of the corresponding mechanical property feature vector specifically comprises: all the mechanical property parameters in each initial sample are subjected to normalization processing, and the mechanical property parameters after the normalization processing form corresponding mechanical property characteristic vectors; in the step 4, the mechanical property parameters of the two samples are preprocessed, and the corresponding mechanical property characteristic vectors are extracted, which specifically includes: and all the mechanical property parameters of the two samples are subjected to normalization treatment, and the mechanical property parameters after the normalization treatment form corresponding mechanical property characteristic vectors.
4. The method for predicting the service life of a novel noble metal electric contact device based on machine learning according to claim 1, wherein in the step 2, the service life data in the initial sample set is divided into intervals, and the method specifically comprises the following steps: the service life data in the initial sample set is equally divided into at least 10 intervals.
5. The method for predicting the service life of the novel noble metal electric contact device based on machine learning of claim 1, wherein in the step 3, the machine learning algorithm is one of linear regression, logistic regression, support vector machine regression and decision tree regression.
6. The method for predicting the service life of the novel noble metal electric contact device based on the machine learning as claimed in claim 1, wherein the novel noble metal electric contact material sample is prepared by a method of vacuum induction melting-repeated rolling-heat treatment, and specifically comprises the following steps: and (3) smelting a novel noble metal electric contact material by vacuum induction, and repeatedly rolling and thermally treating the cast alloy ingot to obtain a plate or bar with a final deformation of 60-95%.
7. The method of predicting life span of a novel noble metal electrical contact device based on machine learning of claim 1, wherein said novel noble metal electrical contact material is one of Ag-based alloy, au-based alloy, pd-based alloy.
8. The method for predicting the service life of the novel noble metal electric contact device based on machine learning according to claim 2, wherein the test temperature of the mechanical property test is 30-200 ℃.
9. The method for predicting the service life of the novel noble metal electric contact device based on machine learning as claimed in claim 1, wherein in the step 4, the sample is kept at the test temperature for 1h before the mechanical property test.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210992150.XA CN115392013A (en) | 2022-08-17 | 2022-08-17 | Novel noble metal electric contact device service life prediction method based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210992150.XA CN115392013A (en) | 2022-08-17 | 2022-08-17 | Novel noble metal electric contact device service life prediction method based on machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115392013A true CN115392013A (en) | 2022-11-25 |
Family
ID=84120127
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210992150.XA Pending CN115392013A (en) | 2022-08-17 | 2022-08-17 | Novel noble metal electric contact device service life prediction method based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115392013A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116592950A (en) * | 2023-07-14 | 2023-08-15 | 南通祥峰电子有限公司 | Electrical component design sample system based on intelligent model |
-
2022
- 2022-08-17 CN CN202210992150.XA patent/CN115392013A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116592950A (en) * | 2023-07-14 | 2023-08-15 | 南通祥峰电子有限公司 | Electrical component design sample system based on intelligent model |
| CN116592950B (en) * | 2023-07-14 | 2023-11-28 | 南通祥峰电子有限公司 | Electrical component design sample system based on intelligent model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111721535B (en) | Bearing fault detection method based on convolution multi-head self-attention mechanism | |
| CN110361176B (en) | Intelligent fault diagnosis method based on multitask feature sharing neural network | |
| CN106909933B (en) | A kind of stealing classification Forecasting Methodology of three stages various visual angles Fusion Features | |
| CN111914883B (en) | Spindle bearing state evaluation method and device based on deep fusion network | |
| CN106777606A (en) | A kind of gearbox of wind turbine failure predication diagnosis algorithm | |
| CN113076834B (en) | Rotating machine fault information processing method, processing system, processing terminal, and medium | |
| CN115392013A (en) | Novel noble metal electric contact device service life prediction method based on machine learning | |
| CN112305388B (en) | On-line monitoring and diagnosing method for insulation partial discharge faults of generator stator winding | |
| CN109308063A (en) | Fermentation process stage division method based on score matrix | |
| CN113051689B (en) | Bearing residual service life prediction method based on convolution gating circulation network | |
| CN115576293B (en) | Pressure-sensitive adhesive on-line production analysis method and system based on data monitoring | |
| CN113567159A (en) | Scraper conveyor state monitoring and fault diagnosis method based on edge cloud cooperation | |
| CN113343591A (en) | Product key part life end-to-end prediction method based on self-attention network | |
| CN111563622B (en) | Method for predicting insulation aging degree of stator bar | |
| CN117150333B (en) | Operation and maintenance supervision method and system for stamping forming equipment | |
| KR102523701B1 (en) | Solar module quality prediction apparatus and method using ai | |
| CN115146675B (en) | Rotary machine migration diagnosis method under variable working condition of depth multi-feature dynamic countermeasure | |
| CN111766513A (en) | Capsule network-based variable-working-condition multi-fault diagnosis method for three-phase induction motor | |
| CN112348700B (en) | Line capacity prediction method combining SOM clustering and IFOU equation | |
| CN115081514A (en) | Industrial equipment fault identification method under data imbalance condition | |
| CN115032500A (en) | Grid-connected fault identification method based on big data analysis | |
| CN117436848B (en) | Motor energy consumption monitoring method based on artificial intelligence | |
| CN113987951A (en) | Data sample screening and reconstruction method in modeling of high-nickel matte flotation process | |
| CN117634652B (en) | Dam deformation interpretable prediction method based on machine learning | |
| CN110032981B (en) | Rotary machine fault identification method based on improved support vector machine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20230511 Address after: 102299 01, floors 1-3, building 1, No. 33, Chaoqian Road, Changping District, Beijing Applicant after: GRIKIN ADVANCED MATERIALS Co.,Ltd. Applicant after: GRINM ADVANCED MATERIALS CO.,LTD. Address before: 102299 01, floors 1-3, building 1, No. 33, Chaoqian Road, Changping District, Beijing Applicant before: GRIKIN ADVANCED MATERIALS Co.,Ltd. |
|
| TA01 | Transfer of patent application right |