CN113447828B - Lithium battery temperature estimation method and system based on Bayesian neural network - Google Patents
Lithium battery temperature estimation method and system based on Bayesian neural network Download PDFInfo
- Publication number
- CN113447828B CN113447828B CN202110684729.5A CN202110684729A CN113447828B CN 113447828 B CN113447828 B CN 113447828B CN 202110684729 A CN202110684729 A CN 202110684729A CN 113447828 B CN113447828 B CN 113447828B
- Authority
- CN
- China
- Prior art keywords
- temperature
- data
- impedance
- battery
- neural network
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/374—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC] with means for correcting the measurement for temperature or ageing
Abstract
The invention discloses a lithium battery temperature estimation method and a system based on a Bayesian neural network, wherein the method comprises the following steps: collecting electrochemical impedance spectrum data and a temperature tag of a battery; processing the electrochemical impedance spectrum data of the battery based on an ARD algorithm to obtain temperature-related characteristics and temperature-related impedance frequency points; training a Bayesian neural network model based on temperature-related features and temperature labels, and acquiring impedance data at temperature-related impedance frequency points; and inputting the impedance imaginary part data into the temperature estimation model to obtain the battery internal estimation temperature and the confidence interval at the current moment. The system comprises: the device comprises an offline data acquisition module, a temperature-related data determination module, a model training module, an online data acquisition module and a temperature estimation module. By using the method and the device, accurate internal temperature estimation of the full life cycle of the power battery is realized. The lithium battery temperature estimation method and system based on the Bayesian neural network can be widely applied to the field of battery thermal management.
Description
Technical Field
The invention relates to the field of battery thermal management, in particular to a lithium battery temperature estimation method and system based on a Bayesian neural network.
Background
The lithium ion battery is widely used as a power battery of an electric automobile due to the outstanding characteristics of high energy density, low material price, good performance, no toxicity, no pollution, safety and the like. Temperature estimation of lithium ion batteries is crucial for safety and control purposes. For example, high temperatures accelerate battery aging, thereby reducing its life and performance, and may even lead to thermal runaway of the battery, further possibly leading to fire or explosion. A typical method for battery temperature monitoring includes installing a thermocouple on the surface of the battery, and this method has a problem in that the heat generation of the battery is caused by internal physicochemical reactions and there is a delay in heat transfer inside and outside the battery. Particularly in the event of thermal runaway of the cell (caused by cell failure or nearby heat sources) the temperature difference radially inside and outside may be as high as 40-50 c. Therefore, an additional diagnostic technique is required to monitor the internal temperature of the lithium ion battery, which is of great significance for early warning of battery thermal runaway. Current temperature estimation methods do not provide a measure of uncertainty and have limited interpretation of the results, resulting in estimates of these models that may mislead the decision maker.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide a lithium battery temperature estimation method and system based on a bayesian neural network, which can overcome the influence of unknown battery SOH and SOC and quantize the uncertainty of a model, so that a user can obtain more information representing the real state of the system.
The first technical scheme adopted by the invention is as follows: a lithium battery temperature estimation method based on a Bayesian neural network comprises the following steps:
acquiring electrochemical impedance spectrum data of the battery and a corresponding temperature tag off line;
processing the electrochemical impedance spectrum data of the battery based on an ARD algorithm to obtain temperature-related characteristics and temperature-related impedance frequency points;
training a Bayes neural network model based on the temperature-related features and the temperature labels to obtain a temperature estimation model;
acquiring impedance data under a temperature-related impedance frequency point on line to obtain actual impedance data;
and inputting the impedance imaginary part data in the actual impedance data into the temperature estimation model and performing calculation to obtain the battery internal estimated temperature and the confidence interval at the current moment.
Further, the step of acquiring the electrochemical impedance spectrum data of the battery and the corresponding temperature tag off-line specifically comprises:
carrying out aging cycle test on a plurality of batteries with the same model until the batteries are attenuated to be below 80% of the initial capacity;
measuring electrochemical impedance spectrum data of the battery at different temperatures, different states of charge (SOC) and different states of health (SOH) at fixed cycle intervals;
and acquiring the average internal temperature of the battery in the test process as a temperature label.
Further, the step of processing the electrochemical impedance spectrum data of the battery based on the ARD algorithm to obtain the temperature-related characteristic and the temperature-related impedance frequency point specifically comprises:
inputting the electrochemical impedance spectrum data of the battery as parameters into an ARD algorithm;
modeling and determining the width of zero mean Gaussian prior of the input parameters;
and optimizing the input parameters according to the width of Gaussian prior to obtain the impedance imaginary part characteristic with the most obvious temperature correlation and the alternating current impedance frequency point with the most obvious temperature correlation.
Further, the step of training a bayesian neural network model based on the temperature-related features and the temperature labels to obtain a temperature estimation model specifically includes:
constructing a multi-input single-output Bayes neural network model, wherein the Bayes neural network model comprises two hidden layers;
training a Bayesian neural network model based on the temperature-related features and the temperature labels;
and learning the predicted distribution through a Bayesian neural network model to obtain a temperature estimation model.
Further, the prediction distribution is learned through a bayesian neural network model, and the formula is as follows:
P(Y'|X',D)＝∫P(Y'|X',W)P(W|D)dW
in the above equation, W represents a network parameter, P (W) represents a prior distribution of the parameter, P (W | D) represents a posterior distribution, given observation data D = X, Y, X representing input data, and Y representing label data.
Further, the online acquisition of the impedance data at the temperature-dependent impedance frequency point is to generate a required excitation frequency by a signal generator and acquire the impedance data at the frequency point.
The second technical scheme adopted by the invention is as follows: a lithium battery temperature estimation system based on a Bayesian neural network comprises:
the off-line data acquisition module is used for off-line acquiring the electrochemical impedance spectrum data of the battery and the corresponding temperature label;
the temperature-related data determining module is used for processing the electrochemical impedance spectrum data of the battery based on an ARD algorithm to obtain temperature-related characteristics and temperature-related impedance frequency points;
the model training module trains a Bayes neural network model based on the temperature-related features and the temperature labels to obtain a temperature estimation model;
the online data acquisition module is used for acquiring impedance data under temperature-related impedance frequency points online to obtain actual impedance data;
and the temperature estimation module is used for inputting the impedance imaginary part data in the actual impedance data into the temperature estimation model and executing calculation to obtain the battery internal estimated temperature and the confidence interval at the current moment.
The method and the system have the beneficial effects that: the method realizes automatic feature selection of electrochemical impedance spectrum data by using an ARD method, and the temperature estimation model fully considers the influence of the SOH and the SOC state of the battery on the estimation of the battery temperature, so that the accurate internal temperature estimation of the full life cycle of the power battery is realized, and important support is provided for early warning of thermal runaway of the battery.
Drawings
FIG. 1 is a flow chart illustrating the steps of a method for estimating the temperature of a lithium battery based on a Bayesian neural network according to the present invention;
FIG. 2 is a schematic diagram of a Bayesian neural network model according to an embodiment of the present invention;
fig. 3 is a structural block diagram of a lithium battery temperature estimation system based on a bayesian neural network according to the present invention.
Detailed Description
The invention is described in further detail below with reference to the figures and the specific embodiments. The step numbers in the following embodiments are provided only for convenience of illustration, the order between the steps is not limited at all, and the execution order of each step in the embodiments can be adapted according to the understanding of those skilled in the art.
Referring to fig. 1, the invention provides a lithium battery temperature estimation method based on a bayesian neural network, which comprises the following steps:
s1, acquiring electrochemical impedance spectrum data of a battery and a corresponding temperature tag off line;
s2, processing the electrochemical impedance spectrum data of the battery based on an ARD algorithm to obtain temperature-related characteristics and temperature-related impedance frequency points;
s3, training a Bayes neural network model based on the temperature correlation characteristics and the temperature labels to obtain a temperature estimation model;
s4, acquiring impedance data under temperature-related impedance frequency points on line to obtain actual impedance data;
and S5, inputting impedance imaginary part data in the actual impedance data into the temperature estimation model and performing calculation to obtain the battery internal estimated temperature and the confidence interval at the current moment.
Further as a preferred embodiment of the method, the step of acquiring electrochemical impedance spectrum data of the battery and a corresponding temperature tag offline specifically includes:
carrying out aging cycle test on a plurality of batteries with the same type until the batteries attenuate to be below 80% of the initial capacity;
measuring electrochemical impedance spectrum data of the battery at different temperatures, different states of charge (SOC) and different states of health (SOH) at fixed cycle intervals;
and acquiring the average internal temperature of the battery in the test process as a temperature label.
In particular, offline battery Electrochemical Impedance Spectroscopy (EIS) data collection in this embodiment uses EIS data from pairs of electrochemical workstations measuring cells. In order to measure the data of the battery in the SOH state, the battery needs to be subjected to long-time charge-discharge aging test under a simulation working condition, and the capacity of the battery is calibrated by measuring at proper charge-discharge cycle intervals, wherein the capacity calibration of the lithium battery in the aging process can refer to national standards. And then, carrying out SOC calibration on the battery, standing the battery in an incubator for more than 2h, and measuring the alternating current impedance of the battery in a frequency band of 0.01Hz-5000Hz by using an electrochemical workstation. The battery aging is stopped below 80% of the initial capacity, for example, EIS data is collected every 50 cycles; the battery SOC state is from 0% to 100%, for example every 10% interval EIS data is collected; the cell temperature comprises collecting EIS data every 5 degrees from low to high temperature, e.g., 0 to 55 ℃. Therefore, a large amount of EIS data under different SOC and temperature states in the whole life cycle of the power battery are obtained through laboratory offline measurement, and the aim is to know the influence of the SOC and SOH states of the battery on temperature in EIS measurement.
Further, as a preferred embodiment of the method, the step of processing the electrochemical impedance spectrum data of the battery based on the ARD algorithm to obtain the temperature-related characteristic and the temperature-related impedance frequency point specifically includes:
inputting the electrochemical impedance spectrum data of the battery as parameters into an ARD algorithm;
modeling and determining the width of zero mean Gaussian prior of the input parameters;
and optimizing the input parameters according to the width of Gaussian prior to obtain the impedance imaginary part characteristic with the most obvious temperature correlation and the alternating current impedance frequency point with the most obvious temperature correlation.
In particular, if the width of the gaussian distribution is zero, these parameters are limited to zero, the corresponding inputs have no influence on the prediction and therefore become insignificant. The input data comprises all data of impedance imaginary parts of existing frequency points measured at different SOH, SOC and temperature of the battery, the output is the correlation degree of the imaginary part characteristics and the temperature at each frequency point, the weight value of the correlated frequency is large, and the weight value of the irrelevant frequency is close to zero.
As a preferred embodiment of the method, the step of training the bayesian neural network model based on the temperature-related features and the temperature labels to obtain the temperature estimation model specifically includes:
constructing a multi-input single-output Bayes neural network model, wherein the Bayes neural network model comprises two hidden layers;
training a Bayesian neural network model based on the temperature-related features and the temperature labels;
and learning the predicted distribution through a Bayesian neural network model to obtain a temperature estimation model.
Specifically, a schematic diagram of the bayesian neural network model is shown in fig. 2, h shows a hidden layer of the neural network, a number 1 shows a first hidden layer, μ shows an expectation of statistical distribution (for example, normal distribution), σ shows a standard deviation, impedance imaginary part data at certain frequency points extracted from the step S2 is input, and an output is a battery temperature, the network comprises two hidden layers, and a network structure (including a connection relation between neurons, the number of layers, the number of neurons in each layer, and a type of an activation function) and a learning rate and the like can be selected through hyper-parameter optimization.
As a preferred embodiment of the method, the prediction distribution is learned by a bayesian neural network model, and the formula is expressed as follows:
P(Y'|X',D)＝∫P(Y'|X',W)P(W|D)dW
in the above equation, W represents a network parameter, P (W) represents a prior distribution of the parameter, P (W | D) represents a posterior distribution, given observation data D = X, Y, X representing input data, and Y representing label data.
In addition:
here, P (W | D) is posterior distribution, P (D | W) is likelihood function, and P (D) is edge likelihood.
After a proper neural network architecture is constructed, a Bayesian neural network is trained based on variational inference, and a real posterior probability distribution P (w | D) is approximated through a probability model q (w | theta). The goal of the variational inference of the two distributions is to narrow the KL divergence (Kullback-Leibler divergence) between the two, i.e. the optimization:
here θ is not fixed but obeys a distribution θ = (μ, σ).
Further, as a preferred embodiment of the method, the online acquisition of the impedance data at the temperature-dependent impedance frequency point is implemented by generating a required excitation frequency through a signal generator and acquiring the impedance data at the frequency point.
By the method, the current battery temperature can be known in real time in a Battery Management System (BMS), the current battery temperature is compared with the normal working temperature interval of the battery, the early warning is timely carried out on the abnormal high temperature in the battery, and corresponding safety measures are taken to avoid thermal runaway.
As shown in fig. 3, a system for estimating the temperature of a lithium battery based on a bayesian neural network includes:
the off-line data acquisition module is used for off-line acquiring the electrochemical impedance spectrum data of the battery and the corresponding temperature label;
the temperature-related data determining module is used for processing the electrochemical impedance spectrum data of the battery based on an ARD algorithm to obtain temperature-related characteristics and temperature-related impedance frequency points;
the model training module is used for training a Bayesian neural network model based on the temperature-related characteristics and the temperature labels to obtain a temperature estimation model;
the online data acquisition module is used for acquiring impedance data under temperature-related impedance frequency points online to obtain actual impedance data;
and the temperature estimation module is used for inputting the impedance imaginary part data in the actual impedance data into the temperature estimation model and executing calculation to obtain the battery internal estimated temperature and the confidence interval at the current moment.
The contents in the above method embodiments are all applicable to the present system embodiment, the functions specifically implemented by the present system embodiment are the same as those in the above method embodiment, and the beneficial effects achieved by the present system embodiment are also the same as those achieved by the above method embodiment.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (6)
1. A lithium battery temperature estimation method based on a Bayesian neural network is characterized by comprising the following steps:
acquiring electrochemical impedance spectrum data of the battery and a corresponding temperature tag off line;
inputting the electrochemical impedance spectrum data of the battery as parameters into an ARD algorithm;
modeling and determining the width of zero mean Gaussian prior of the input parameters;
optimizing input parameters according to the width of Gaussian prior to obtain impedance imaginary part characteristics most significant to temperature correlation and alternating current impedance frequency points most significant to temperature correlation;
training a Bayes neural network model based on the temperature-related features and the temperature labels to obtain a temperature estimation model;
acquiring impedance data under temperature-related impedance frequency points on line to obtain actual impedance data;
and inputting the impedance imaginary part data in the actual impedance data into the temperature estimation model and performing calculation to obtain the battery internal estimated temperature and the confidence interval at the current moment.
2. The Bayesian neural network-based lithium battery temperature estimation method according to claim 1, wherein the step of collecting battery electrochemical impedance spectrum data and corresponding temperature labels offline specifically comprises:
carrying out aging cycle test on a plurality of batteries with the same type until the batteries attenuate to be below 80% of the initial capacity;
measuring electrochemical impedance spectrum data of the battery at different temperatures, different charge states and different health states at fixed cycle intervals;
and acquiring the average internal temperature of the battery in the test process as a temperature label.
3. The method for estimating the temperature of the lithium battery based on the bayesian neural network as recited in claim 2, wherein the step of training the bayesian neural network model based on the temperature-related features and the temperature labels to obtain the temperature estimation model specifically comprises:
constructing a multi-input single-output Bayes neural network model, wherein the Bayes neural network model comprises two hidden layers;
training a Bayesian neural network model based on the temperature-related features and the temperature labels;
and learning the predicted distribution through a Bayesian neural network model to obtain a temperature estimation model.
4. The Bayesian neural network-based lithium battery temperature estimation method according to claim 3, wherein the prediction distribution is learned through a Bayesian neural network model, and a formula is expressed as follows:
in the above formula, the first and second carbon atoms are,Wwhich is indicative of a parameter of the network,
a-priori distribution of the representative parameters,
representing posterior distribution, given observed data
，XWhich represents the input data, is,Yrepresenting the tag data.
5. The Bayesian neural network-based lithium battery temperature estimation method according to claim 4, wherein the online collection of the impedance data at the temperature-dependent impedance frequency point is specifically to generate a required excitation frequency by a signal generator and collect the impedance data at the frequency point.
6. A lithium battery temperature estimation system based on a Bayesian neural network is characterized by comprising:
the off-line data acquisition module is used for acquiring the electrochemical impedance spectrum data of the battery and the corresponding temperature label off line;
the temperature-related data determination module is used for inputting the electrochemical impedance spectrum data of the battery as parameters into an ARD algorithm; modeling and determining the width of zero mean Gaussian prior of the input parameters; optimizing input parameters according to the width of Gaussian prior to obtain impedance imaginary part characteristics with the most obvious temperature correlation and alternating current impedance frequency points with the most obvious temperature correlation;
the model training module trains a Bayes neural network model based on the temperature-related features and the temperature labels to obtain a temperature estimation model;
the online data acquisition module is used for acquiring impedance data under temperature-related impedance frequency points on line to obtain actual impedance data;
and the temperature estimation module is used for inputting the impedance imaginary part data in the actual impedance data into the temperature estimation model and executing calculation to obtain the battery internal estimated temperature and the confidence interval at the current moment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110684729.5A CN113447828B (en) | 2021-06-21 | 2021-06-21 | Lithium battery temperature estimation method and system based on Bayesian neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110684729.5A CN113447828B (en) | 2021-06-21 | 2021-06-21 | Lithium battery temperature estimation method and system based on Bayesian neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113447828A CN113447828A (en) | 2021-09-28 |
| CN113447828B true CN113447828B (en) | 2022-10-14 |
Family
ID=77811918
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110684729.5A Active CN113447828B (en) | 2021-06-21 | 2021-06-21 | Lithium battery temperature estimation method and system based on Bayesian neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113447828B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113945302B (en) * | 2021-12-20 | 2022-03-08 | 中创新航科技股份有限公司 | Method and device for determining internal temperature of battery |
| CN114280479A (en) * | 2021-12-23 | 2022-04-05 | 中山大学·深圳 | Electrochemical impedance spectrum-based rapid sorting method for retired batteries |
| CN114280491B (en) * | 2021-12-23 | 2024-01-05 | 中山大学 | Retired battery residual capacity estimation method based on active learning |
| CN114295983A (en) * | 2021-12-31 | 2022-04-08 | 上海洛轲智能科技有限公司 | Battery thermal runaway early warning method and device, vehicle, equipment and storage medium |
| CN114720890B (en) * | 2022-03-23 | 2023-04-21 | 四川大学 | Lithium battery internal temperature estimation method based on broadband impedance spectrum |
| CN116231165B (en) * | 2023-03-27 | 2023-10-20 | 深圳市南霸科技有限公司 | Heating method and system for high-capacity energy storage battery |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102494778B (en) * | 2011-11-14 | 2013-04-24 | 北京理工大学 | Artificial neural network-based highest surface temperature prediction method of secondary battery |
| US9267993B2 (en) * | 2012-05-23 | 2016-02-23 | Lawrence Livermore National Security, Llc | Battery management system with distributed wireless sensors |
| CN106289566B (en) * | 2016-07-19 | 2018-12-11 | 清华大学 | A method of secondary cell internal temperature is estimated based on electrochemical impedance |
| CN110244225A (en) * | 2019-06-27 | 2019-09-17 | 清华大学深圳研究生院 | A method of hot face temperature when prediction power battery electric discharge |
| CN112034347A (en) * | 2020-09-04 | 2020-12-04 | 国网陕西省电力公司汉中供电公司 | Lithium ion battery thermal runaway rapid monitoring method and system |
-
2021
- 2021-06-21 CN CN202110684729.5A patent/CN113447828B/en active Active
Non-Patent Citations (3)
| Title |
|---|
| Bayesian Deep-Learning-Based Health Prognostics Toward Prognostics Uncertainty;Weiwen Peng et al.;《IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS》;20200331;第67卷(第3期);第2283-2293页 * |
| 基因选择的快速Fisher优化模型;封举富 等;《北京大学学报（自然科学版）》;20050131;第41卷(第1期);第122-128页 * |
| 风冷车用锂离子动力电池包热管理研究;李康靖 等;《电源技术》;20191231;第43卷(第12期);第1975-1978、2035页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113447828A (en) | 2021-09-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113447828B (en) | Lithium battery temperature estimation method and system based on Bayesian neural network | |
| Tian et al. | Electrode ageing estimation and open circuit voltage reconstruction for lithium ion batteries | |
| CN112731159B (en) | Method for pre-judging and positioning battery faults of battery compartment of energy storage power station | |
| Zhang et al. | Aging characteristics-based health diagnosis and remaining useful life prognostics for lithium-ion batteries | |
| CN112763929B (en) | Method and device for predicting health of battery monomer of energy storage power station system | |
| Wei et al. | Remaining useful life prediction and state of health diagnosis for lithium-ion batteries using particle filter and support vector regression | |
| Widodo et al. | Intelligent prognostics for battery health monitoring based on sample entropy | |
| Wu et al. | Health prognosis with optimized feature selection for lithium-ion battery in electric vehicle applications | |
| CN111624494A (en) | Battery analysis method and system based on electrochemical parameters | |
| Li et al. | On-line estimation method of lithium-ion battery health status based on PSO-SVM | |
| Wang et al. | Perspectives and challenges for future lithium-ion battery control and management | |
| Ouyang et al. | A novel state of charge estimation method for lithium-ion batteries based on bias compensation | |
| Kang et al. | A comparative study of fault diagnostic methods for lithium-ion batteries based on a standardized fault feature comparison method | |
| Xu et al. | A vehicle-cloud collaborative method for multi-type fault diagnosis of lithium-ion batteries | |
| CN115032542A (en) | Hybrid model-based battery thermal runaway pre-judgment method for energy storage system | |
| CN114924193A (en) | Battery safety assessment method | |
| Chen et al. | State of health estimation for lithium-ion battery based on long short term memory networks | |
| Zhao et al. | State-of-health estimation with anomalous aging indicator detection of lithium-ion batteries using regression generative adversarial network | |
| CN115754724A (en) | Power battery state of health estimation method suitable for future uncertainty dynamic working condition discharge | |
| CN113687251A (en) | Dual-model-based lithium ion battery pack voltage abnormity fault diagnosis method | |
| Huang et al. | State of health estimation of lithium-ion batteries based on fine-tuning or rebuilding transfer learning strategies combined with new features mining | |
| Li et al. | State of Health Indicator Modeling of Lithium-ion Batteries Using Machine Learning Techniques | |
| Vatani et al. | Cycling lifetime prediction model for lithium-ion batteries based on artificial neural networks | |
| CN114720901A (en) | Method for estimating health state of lithium ion battery | |
| Yang et al. | Remaining Useful Life Prediction of Lithium-ion Batteries with Limited Degradation History Using Random Forest |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |