CN104463327A - Method for predicting catalytic cracking coke yield - Google Patents
Method for predicting catalytic cracking coke yield Download PDFInfo
- Publication number
- CN104463327A CN104463327A CN201410585073.1A CN201410585073A CN104463327A CN 104463327 A CN104463327 A CN 104463327A CN 201410585073 A CN201410585073 A CN 201410585073A CN 104463327 A CN104463327 A CN 104463327A
- Authority
- CN
- China
- Prior art keywords
- data
- value
- coke yield
- normalized
- catalytic cracking
- 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
Landscapes
- Production Of Liquid Hydrocarbon Mixture For Refining Petroleum (AREA)
Abstract
The invention relates to a method for predicting the catalytic cracking coke yield. The method comprises the following steps that firstly, basic data related to the raw material characters, catalyst characters and operation of catalytic cracking and the actual value of the coke yield are obtained, and the data are preprocessed and normalized; secondly, model training is carried out through a BP neutral network, the normalized basic data in the first step serve as an input value, the actual value serves as expected output, and a model for predicting the coke yield is obtained; thirdly, the basic data, collected in the field, of the raw material characters, catalyst characters and operation of the catalytic cracking are preprocessed and serve as predicting data, the predicting data are substituted into the model acquired in the second step, and the coke yield predicting value is obtained. According to the method, the calculating speed is high, accuracy is good, and the method can be widely applied to industrial production.
Description
Technical field
The present invention relates to petrochemical industry, be specifically related to a kind of method predicting catalytic cracking coke yield.
Background technology
RFCC (RFCC) occupies critical role in petroleum refining industry of China, is the important means that refinery increases economic efficiency.But along with heaviness, the in poor quality of crude oil, the blending ratio of China RFCC raw material improves constantly, worse and worse, RFCC product is distributed and product quality variation, yield of light oil is low, and coke dry gas yied is high for feedstock property.Due to the parallel sequence reaction that catalytic cracking is extremely complicated, be that reaction is coupled with the height of flowing, mass-and heat-transfer.Its course of reaction and product slates are subject to the impact of feedstock property, catalyst property and operating conditions, are difficult to set up accurate mathematic(al) representation to describe its dynamic production run, are difficult to predict product slates.
The correlation of current calculating catalytic cracking process product yield generally draws according to practical operation data and medium-sized experimental data, is applicable to process program estimation or technical and economic evaluation, is difficult to use in and directs engineering design or field optimizing operation concretely.BP neural network, as a kind of Nonlinear Statistical data modeling tool, has unique Distribution parallel processing, adaptive ability and nonlinear prediction ability.Chinese patent CN 102737288 A proposes a kind of based on RBF neural parameter self-optimizing water quality multistep forecasting method, by the remote transmission of online water quality monitoring instrument, the data of each monitoring station are deposited in the database of home server, then water quality sample sequence is normalized, through RBF neural model training, prediction, final test is predicted water quality.Chinese patent CN 103559556 A proposes a kind of method of on-line prediction load capacity limit of electric power system, changes method and forms by based on the screening sample method of electric network state index of similarity, Lasso method and error back propagation type neural network three part.
Although utilize the widespread uses such as some character of neural network prediction commercial plant, in petroleum chemical industry, be especially still blank for the prediction of coke yield in catalytic cracking.
Summary of the invention
The object of the invention is to, what utilize BP analysis of neural network catalytic cracking unit to obtain affects coke yield data, thus the coke yield of fast prediction catalytic cracking unit, for further reasonably optimizing feedstock property, catalyst property and operating conditions etc., produce liquid fuel etc. to greatest extent and certain instruction is provided.
The invention provides a kind of method predicting catalytic cracking coke yield, said method comprising the steps of:
(1) basic data relevant with operation with the feedstock property of catalytic cracking, catalyst property is obtained, and the actual value of coke yield, pre-service is carried out to above-mentioned data, then is normalized;
(2) utilizing BP neural network to carry out model training, with the basic data after step (1) gained normalized for input value, take actual value as desired output, obtains the model of prediction coke yield;
(3) pre-service is carried out as predicted data to the basic data relevant with operation with the feedstock property of catalytic cracking, catalyst property of collection in worksite, predicted data is substituted in the model that step (2) obtains, obtain coke yield predicted value.
The actual value of the described basic data relevant with feedstock property, catalyst property and operation of step of the present invention (1) and coke yield all obtains from production scene by normal experiment collection.
Wherein, the described basic data relevant with feedstock property of step (1) comprises following parameter: feedstock oil saturated hydrocarbon content, feedstock oil arene content, feedstock oil bituminous matter+gum level, feedstock oil 10% recovered (distilled) temperature, feedstock oil 50% recovered (distilled) temperature, feedstock oil 90% recovered (distilled) temperature and feedstock oil sodium content.
The described basic data relevant with catalyst property of step (1) comprises following parameter: catalyst activity, regenerant micro anti-active index, regenerant determine carbon content, leveler activity, regenerant sodium content, regenerant nickel and content of vanadium.
The described basic data relevant with operation of step (1) comprises following parameter: catalyst temperature, material temperature, temperature of reaction, reaction pressure, raw material oil mass, pre-lift steam flow, feedstock oil atomizing steam flow, pre-lift dry gas flow, recycle ratio, oil ratio, two anti-material levels and raw gasoline are to riser reactor Flow Control.
Step of the present invention (1) or (3) described pre-service are specially: get the data in the common period of parameters, parameters is carried out to the rejecting of exceptional value, described exceptional value comprises zero, negative value, empty data and be greater than the data of 3 times of standard deviations with the difference of mean value.
Step of the present invention (1) described normalized is: be decided to be { X (n) } by sample sequence, is normalized data according to minimax method, and described minimax method formula is X
k=(X
i-X
min)/(X
max-X
min), wherein, X
maxand X
minmaximal value and the minimum value of { X (n) } respectively.
BP neural network structure described in step of the present invention (2) as shown in Figure 1; Having good non-linear quality, high fitting precision and extensive function as a kind of Nonlinear Modeling and Forecasting Methodology, is a kind of multilayer feedforward neural network of one way propagation, and the principal feature of this network is signal propagated forward, error back propagation.X
1, X
2..., X
nthe input value of BP neural network, Y
1, Y
2..., Y
mthe predicted value of BP neural network, ω
ijand ω
jkfor BP neural network weight.Input signal successively processes from input layer through hidden layer, until output layer.The neuron state of every one deck only affects lower one deck neuron state.If output layer can not get desired output, then proceed to backpropagation, according to predicated error adjustment network weight and threshold value, thus make the output of BP neural network prediction constantly approach desired output.
Described BP neural network comprises each one deck of input layer, output layer and hidden layer; Wherein, the basic data after step (1) gained normalized is as input layer, and coke yield is as output layer, and node in hidden layer selects reference formula
in formula, m is input number of nodes, and being preferably 25, n is output node number, is preferably 1, and the square error that under more different node, training pattern obtains respectively, finds out best the number of hidden nodes H; Use tangent S type function tansig as transport function between input layer and hidden layer, with selectively acting function Lin as the transport function between output layer and hidden layer, by function inputoutput data training BP neural network.
Step of the present invention (2) also comprises BP neural network initialization step, preferably the parameter of this step is: in the data of step (1) gained, Stochastic choice 99% group of data are as training sample, BP neural network is utilized to carry out model training to training sample, the iterations arranged is 100 times, learning rate is 0.2, and desired value is 0.00004; When the error of iteration result is less than permissible error 0.001 ~ 0.00001, system finishing iterative computation, model construction completes.
The algorithm flow of BP neural network of the present invention as shown in Figure 2.
The present invention is analyzed by the coke yield of BP neural network to refinery catalytic cracking device, establishes coke yield model, can on-line prediction catalytic cracking unit coke yield.Method computing velocity provided by the invention is fast, and accuracy is good, can be widely used in commercial production.
Accompanying drawing explanation
Fig. 1 is the structural drawing of BP neural network.
Fig. 2 is the algorithm flow schematic diagram of BP neural network.
Fig. 3 is BP neural network prediction catalytic cracking coke yield and true productive rate comparison diagram.
Fig. 4 is BP neural network prediction catalytic cracking coke yield prediction absolute error curve map.
Embodiment
Following examples for illustration of the present invention, but are not used for limiting the scope of the invention.
Embodiment 1
With an annual data of the refinery catalytic cracking device of A for research object, BP neural network is utilized to carry out pre-service, normalization, model training, modelling verification to the 130000 groups of data gathered, thus the prediction of canbe used on line coke yield, concrete implementation step is as follows:
(1) from production scene by the normal experiment collection basic data relevant with feedstock property, catalyst property and operation and coke yield actual value; Wherein, relevant with feedstock property basic data comprises following parameter: feedstock oil saturated hydrocarbon content, feedstock oil arene content, feedstock oil bituminous matter+gum level, feedstock oil 10% recovered (distilled) temperature, feedstock oil 50% recovered (distilled) temperature, feedstock oil 90% recovered (distilled) temperature and feedstock oil sodium content; The basis relevant with catalyst property comprises following parameter: catalyst activity, regenerant micro anti-active index, regenerant determine carbon content, leveler activity, regenerant sodium content, regenerant nickel and content of vanadium; The basis relevant with operation comprises following parameter: catalyst temperature, material temperature, temperature of reaction, reaction pressure, raw material oil mass, pre-lift steam flow, feedstock oil atomizing steam flow, pre-lift dry gas flow, recycle ratio, oil ratio, two anti-material levels and raw gasoline are to riser reactor Flow Control;
Pre-service is carried out to data, described pre-treatment step is: get the data in the common period of parameters, parameters is carried out to the rejecting of exceptional value, described exceptional value comprises zero, negative value, empty data and be greater than the data of 3 times of standard deviations with the difference of mean value, obtain 10,000 groups of data;
Be normalized, described normalization processing method is again: be decided to be { X (n) } by sample sequence, is normalized data according to minimax method, and described minimax method formula is X
k=(X
i-X
min)/(X
max-X
min), wherein, X
kfor normalized value, X
ifor the data of collection in worksite, X
maxand X
minmaximal value and the minimum value of { X (n) } respectively;
(2) utilize BP neural network to carry out model training to the data after step (1) gained normalized, obtain coke yield model;
Wherein, the basic data after step (1) gained normalized is as input layer, and coke yield is as output layer, and coke yield actual value is as desired output; Node in hidden layer selects reference formula:
in formula: m=25 is input number of nodes, n=1 is output node number, and the square error that under more different node, training and verification model obtain respectively, finds out best the number of hidden nodes H=7; Use tangent S type function tansig as transport function between input layer and hidden layer, with selectively acting function Lin as the transport function between output layer and hidden layer;
In step (1) gained 10,000 groups of data, Stochastic choice 9900 groups of data are as training sample, utilize BP neural network to carry out model training to training sample, and the iterations of setting is 100 times, and learning rate is 0.2, and desired value is 0.00004; When the error of iteration result is less than permissible error 0.001 ~ 0.00001, system finishing iterative computation, model construction completes.
(3) in step (1) gained 10,000 groups of data Stochastic choice 1 group of basic data as forecast sample, in the coke yield model that substitution step (2) obtains, obtain coke yield predicted value, described data and gained predicted value are in table 1.
Table 1: the numerical value and the coke yield numerical value that affect the parameter of coke yield
As shown in Table 1, output valve and predicted value are 6.84%, and the actual yield that scene records coke is 6.96%; Compared with actual value, predicated error absolute value is 1.75%.
Continue Stochastic choice 100 groups of basic datas as forecast sample at embodiment 1 step (1) gained 10,000 groups of data relays, in the coke yield model that substitution embodiment 1 step (2) obtains, obtain 100 groups of coke yield predicted values; Predicted value is shown in Fig. 3 with comparing of actual value, and the absolute error of prediction is shown in Fig. 4.
From embodiment 1 acquired results, the prediction absolute error of method provided by the invention is within 10%, and accuracy rate is higher, can be applicable in actual industrial production.
Although above with general explanation, embodiment and test, the present invention is described in detail, and on basis of the present invention, can make some modifications or improvements it, this will be apparent to those skilled in the art.Therefore, these modifications or improvements without departing from theon the basis of the spirit of the present invention, all belong to the scope of protection of present invention.
Claims (8)
1. predict a method for catalytic cracking coke yield, said method comprising the steps of:
(1) basic data relevant with operation with the feedstock property of catalytic cracking, catalyst property is obtained, and the actual value of coke yield, pre-service is carried out to above-mentioned data, then is normalized;
(2) utilizing BP neural network to carry out model training, with the basic data after step (1) gained normalized for input value, take actual value as desired output, obtains the model of prediction coke yield;
(3) pre-service is carried out as predicted data to the basic data relevant with operation with the feedstock property of catalytic cracking, catalyst property of collection in worksite, predicted data is substituted in the model that step (2) obtains, obtain coke yield predicted value.
2. method according to claim 1, it is characterized in that, the described basic data relevant with feedstock property of step (1) comprises following parameter: feedstock oil saturated hydrocarbon content, feedstock oil arene content, feedstock oil bituminous matter+gum level, feedstock oil 10% recovered (distilled) temperature, feedstock oil 50% recovered (distilled) temperature, feedstock oil 90% recovered (distilled) temperature and feedstock oil sodium content.
3. method according to claim 1, it is characterized in that, the described basis relevant with catalyst property of step (1) comprises following parameter: catalyst activity, regenerant micro anti-active index, regenerant determine carbon content, leveler activity, regenerant sodium content, regenerant nickel and content of vanadium.
4. method according to claim 1, it is characterized in that, the described basis relevant with operation of step (1) comprises following parameter: catalyst temperature, material temperature, temperature of reaction, reaction pressure, raw material oil mass, pre-lift steam flow, feedstock oil atomizing steam flow, pre-lift dry gas flow, recycle ratio, oil ratio, two anti-material levels and raw gasoline are to riser reactor Flow Control.
5. the method according to Claims 1 to 4 any one, it is characterized in that, step (1) or (3) described pre-service are specially: get the data in the common period of parameters, parameters is carried out to the rejecting of exceptional value, described exceptional value comprises zero, negative value, empty data and be greater than the data of 3 times of standard deviations with the difference of mean value.
6. the method according to Claims 1 to 4 any one, it is characterized in that, step (1) described normalized is: be decided to be { X (n) } by sample sequence, is normalized data according to minimax method, and described minimax method formula is X
k=(X
i-X
min)/(X
max-X
min), wherein, X
kfor normalized value, X
ifor the data of collection in worksite, X
maxand X
minmaximal value and the minimum value of { X (n) } respectively.
7. method according to claim 1, is characterized in that, step (2) described BP neural network comprises each one deck of input layer, output layer and hidden layer; Wherein, the basic data after step (1) gained normalized is as input layer, and coke yield is as output layer, and coke yield actual value is as desired output; Node in hidden layer selects reference formula
in formula, m is input number of nodes, and n is output node number, and the square error that under more different node, training pattern obtains respectively, finds out best the number of hidden nodes H; Use tangent S type function tansig as transport function between input layer and hidden layer, use selectively acting function Lin as transport function between hidden layer and output layer, utilize above-mentioned function inputoutput data to train BP neural network.
8. method according to claim 7, it is characterized in that, described step (2) comprises BP neural network initialization step, the design parameter of this step is: in step (1) the data obtained, Stochastic choice 99% group of data are as training sample, BP neural network is utilized to carry out model training to training sample, the iterations arranged is 100 times, and learning rate is 0.2, and desired value is 0.00004; When the error of iteration result is less than permissible error 0.001 ~ 0.00001, system finishing iterative computation, model construction completes.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410585073.1A CN104463327A (en) | 2014-10-27 | 2014-10-27 | Method for predicting catalytic cracking coke yield |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410585073.1A CN104463327A (en) | 2014-10-27 | 2014-10-27 | Method for predicting catalytic cracking coke yield |
Publications (1)
Publication Number | Publication Date |
---|---|
CN104463327A true CN104463327A (en) | 2015-03-25 |
Family
ID=52909332
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410585073.1A Pending CN104463327A (en) | 2014-10-27 | 2014-10-27 | Method for predicting catalytic cracking coke yield |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104463327A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105631107A (en) * | 2015-12-23 | 2016-06-01 | 浙江中控软件技术有限公司 | Prediction method and apparatus for fluidized catalytic cracking reaction |
CN108219826A (en) * | 2017-12-27 | 2018-06-29 | 浙江大学宁波理工学院 | A kind of control method of catalytic cracking unit |
CN108416439A (en) * | 2018-02-09 | 2018-08-17 | 中南大学 | Oil refining process Method For Predicting The Products and system based on variable weighting deep learning |
CN108717506A (en) * | 2018-06-25 | 2018-10-30 | 华北理工大学 | A method of prediction coke hot strength |
CN109022019A (en) * | 2018-08-06 | 2018-12-18 | 宁波大红鹰学院 | A kind of feedback of catalytic cracking unit |
CN109658988A (en) * | 2018-12-20 | 2019-04-19 | 中海石油炼化有限责任公司 | A kind of Hydrobon catalyst performance prediction method |
CN109993358A (en) * | 2019-03-25 | 2019-07-09 | 联想(北京)有限公司 | The method and apparatus of training yield prediction model |
CN110931085A (en) * | 2019-12-10 | 2020-03-27 | 中海石油宁波大榭石化有限公司 | Continuous reforming catalyst coking carbon content online prediction and optimization method |
CN111950854A (en) * | 2020-07-14 | 2020-11-17 | 包头钢铁(集团)有限责任公司 | Coke quality index prediction method based on multilayer neural network |
CN112151123A (en) * | 2020-08-25 | 2020-12-29 | 华南理工大学 | Method for predicting yield of baked biomass pyrolytic coke |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102663235A (en) * | 2012-03-20 | 2012-09-12 | 浙江大学 | Modeling method for catalytic cracking main fractionator with varying-population-size DNA genetic algorithm |
-
2014
- 2014-10-27 CN CN201410585073.1A patent/CN104463327A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102663235A (en) * | 2012-03-20 | 2012-09-12 | 浙江大学 | Modeling method for catalytic cracking main fractionator with varying-population-size DNA genetic algorithm |
Non-Patent Citations (5)
Title |
---|
J.MICHALOPOULOS: "Modelling of an Industrial Fluid Catalytic Cracking Unit Using Neural Networks", 《CHEMICAL ENGINEERING RESEARCH & DESIGN》 * |
张忠洋等: "GA辅助BP神经网络预测催化裂化装置汽油产率", 《石油炼制与化工》 * |
王学武等: "基于PSO-BP神经网络的催化裂化C3含量软测量模型", 《***仿真学报》 * |
邓威: "催化裂化气体烃收率神经网络模型", 《抚顺石油学院学报》 * |
马法书等: "基于神经网络的催化裂解产品产率模型", 《炼油设计》 * |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105631107B (en) * | 2015-12-23 | 2019-01-29 | 浙江中控软件技术有限公司 | A kind of prediction technique and device of fluid catalytic cracking reaction |
CN105631107A (en) * | 2015-12-23 | 2016-06-01 | 浙江中控软件技术有限公司 | Prediction method and apparatus for fluidized catalytic cracking reaction |
CN108219826B (en) * | 2017-12-27 | 2020-06-02 | 浙江大学宁波理工学院 | Control method of catalytic cracking device |
CN108219826A (en) * | 2017-12-27 | 2018-06-29 | 浙江大学宁波理工学院 | A kind of control method of catalytic cracking unit |
CN108416439A (en) * | 2018-02-09 | 2018-08-17 | 中南大学 | Oil refining process Method For Predicting The Products and system based on variable weighting deep learning |
CN108717506A (en) * | 2018-06-25 | 2018-10-30 | 华北理工大学 | A method of prediction coke hot strength |
CN108717506B (en) * | 2018-06-25 | 2022-03-18 | 华北理工大学 | Method for predicting hot strength of coke |
CN109022019B (en) * | 2018-08-06 | 2021-04-16 | 宁波大红鹰学院 | Feedback control method of catalytic cracking device |
CN109022019A (en) * | 2018-08-06 | 2018-12-18 | 宁波大红鹰学院 | A kind of feedback of catalytic cracking unit |
CN109658988A (en) * | 2018-12-20 | 2019-04-19 | 中海石油炼化有限责任公司 | A kind of Hydrobon catalyst performance prediction method |
CN109993358A (en) * | 2019-03-25 | 2019-07-09 | 联想(北京)有限公司 | The method and apparatus of training yield prediction model |
CN109993358B (en) * | 2019-03-25 | 2021-07-16 | 联想(北京)有限公司 | Method and device for training yield prediction model |
CN110931085A (en) * | 2019-12-10 | 2020-03-27 | 中海石油宁波大榭石化有限公司 | Continuous reforming catalyst coking carbon content online prediction and optimization method |
CN110931085B (en) * | 2019-12-10 | 2022-07-01 | 中海石油宁波大榭石化有限公司 | Continuous reforming catalyst coking carbon content online prediction and optimization method |
CN111950854A (en) * | 2020-07-14 | 2020-11-17 | 包头钢铁(集团)有限责任公司 | Coke quality index prediction method based on multilayer neural network |
CN111950854B (en) * | 2020-07-14 | 2022-04-19 | 包头钢铁(集团)有限责任公司 | Coke quality index prediction method based on multilayer neural network |
CN112151123A (en) * | 2020-08-25 | 2020-12-29 | 华南理工大学 | Method for predicting yield of baked biomass pyrolytic coke |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104463327A (en) | Method for predicting catalytic cracking coke yield | |
CN104463343A (en) | Method for predicting catalytic cracking light oil yield | |
CN104392098A (en) | Method for predicting yield of catalytically cracked gasoline | |
Zhang et al. | A weighted auto regressive LSTM based approach for chemical processes modeling | |
Ochoa-Estopier et al. | Operational optimization of crude oil distillation systems using artificial neural networks | |
CN103927412B (en) | Instant learning debutanizing tower soft-measuring modeling method based on gauss hybrid models | |
CN104789256B (en) | A kind of yield real-time predicting method of catalytic cracking unit | |
CN104765346B (en) | A kind of oil refining process whole process modeling method | |
Yang et al. | A hybrid deep learning and mechanistic kinetics model for the prediction of fluid catalytic cracking performance | |
CN101863088B (en) | Method for forecasting Mooney viscosity in rubber mixing process | |
CN102213949B (en) | Method for optimizing value of ethylene device | |
CN104361153A (en) | Method for predicting coking amount of heavy oil catalytic cracking settler | |
CN101414158A (en) | Method for optimizing cracking reaction operating condition of ethylene cracking furnace | |
CN105868164B (en) | A kind of soft-measuring modeling method based on the linear dynamic system model for having supervision | |
CN104484714A (en) | Real-time prediction method for catalytic reforming device | |
CN106709654A (en) | Global operating condition evaluating and quality tracing method for hydrocracking process | |
CN105425583A (en) | Control method of penicillin production process based on cooperative training local weighted partial least squares (LWPLS) | |
WO2023131257A1 (en) | Big data-based method for identifying and optimizing mode in oil refining process | |
CN105740960A (en) | Optimization method of industrial hydrocracking reaction condition | |
CN104765347A (en) | Yield real-time prediction method in residual oil delayed coking process | |
Yuan et al. | A comparative study of adaptive soft sensors for quality prediction in an industrial refining hydrocracking process | |
WO2007009322A1 (en) | Real-time operating optimized method of multi-input and multi-output continuous manufacture procedure | |
CN105389442A (en) | Reverse design method for coupling genetic algorithm, neural network and numerical simulation | |
Ghosh et al. | Application of data-driven modeling approaches to industrial hydroprocessing units | |
Wang et al. | Soft-sensing modeling and intelligent optimal control strategy for distillation yield rate of atmospheric distillation oil refining process |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150325 |
|
RJ01 | Rejection of invention patent application after publication |