CN109190202A - A kind of piezoelectric energy collecting device impedance modeling method based on artificial neural network - Google Patents
A kind of piezoelectric energy collecting device impedance modeling method based on artificial neural network Download PDFInfo
- Publication number
- CN109190202A CN109190202A CN201810931373.9A CN201810931373A CN109190202A CN 109190202 A CN109190202 A CN 109190202A CN 201810931373 A CN201810931373 A CN 201810931373A CN 109190202 A CN109190202 A CN 109190202A
- Authority
- CN
- China
- Prior art keywords
- neural network
- artificial neural
- meas
- esr
- esc
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/36—Circuit design at the analogue level
- G06F30/367—Design verification, e.g. using simulation, simulation program with integrated circuit emphasis [SPICE], direct methods or relaxation methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The piezoelectric energy collecting device impedance modeling method based on artificial neural network that the invention discloses a kind of: the N number of frequency point of impedance analyzer uniform scanning, measurement piezoelectric energy collecting device ESR in Freq frequency range are usedMEASAnd ESCMEAS；According to ESRMEASAnd ESCMEAS, obtain Qs、Ωs、Ωp、ρeff, calculate Lm、Rm、Cm、ESRBVD、ESCBVD；It to sum up obtains N group data and is divided into two groups of odd even, odd number group is artificial neural network training data, and even number set is artificial neural network test data；Artificial neural network training is completed, the biasing and weight of optimal hidden layer number, hidden layer neuron number, neuron are obtained；Trained artificial neural network is tested, ESR is exportedANNAnd ESCANN, so that ESRANNAnd ESRMEASAnd ESCANNAnd ESCMEASRelative error meet the requirements.The invention enables device resistance modeling accuracies to be increased dramatically.
Description
Technical field
The invention belongs to piezoelectric vibration energy assembling spheres, and more specifically, it relates to one kind to be based on artificial neural network
Piezoelectric energy collecting device impedance modeling method.
Background technique
Converting electric energy for the mechanical vibrational energy in natural environment by piezoelectric energy collecting device is microsensor
Node power supply is a hot research direction.Common piezoelectric energy collecting device has cantilever beam structure, in practical applications
Its main limitation is: when the vibration frequency in environment deviates the intrinsic frequency of piezoelectric energy collecting device, piezoelectric energy
The output power of collecting device can sharply decline, to limit the wideband application of piezoelectric energy collecting device.Biasing overturning electricity
Output electric current and output voltage phase relation of the road as a kind of adjustable piezoelectric energy collecting device of nonlinear technology, thus
Improve the output power outside device intrinsic frequency.
Voltage overturning efficiency has a major impact the collection power of piezoelectric energy collecting device in biasing reverse circuit,
In circuit simulation modeling, the EFFICIENCY PREDICTION deviation of very little, which will lead to predicted collection power, very big error.Therefore it is being based on
In the Earlier designs for biasing the energy collection circuit of turnover technology, voltage overturns the accurate prediction of efficiency for accurate estimating system
Collected energy is particularly significant.The calculating of voltage overturning efficiency depends primarily on the equivalent series electricity of piezoelectric energy collecting device
Resistance and series capacitance.And the equivalent series resistance and series capacitance of piezoelectric energy collecting device and biasing reverse circuit resonance frequency
It is related.Therefore voltage overturning efficiency accurately is predicted to depend on to piezoelectric energy collecting device equivalent series resistance and series electrical
The Accurate Model of appearance.
Traditional piezoelectric energy collecting device impedance modeling method has Mason model, Butterworth-van Dyke
(BVD) model and modified Butterworth-van Dyke (MBVD) model.These analytic modell analytical model complexities not once
There is closed solution.Mason model is made of and imaginary without actual physical meaning comprising one complicated transmission line structure
Negative resistance, so that solution equivalent series resistance and series capacitance are relatively difficult.Although and BVD and MBVD model is by simple lump
Parametric device composition, so that solution equivalent series resistance and series capacitance are easier, but is to solve for value and actual measured value
Differ larger.
Summary of the invention
Purpose of the invention is to overcome the shortcomings in the prior art, for existing traditional piezoelectric energy collecting device
Equivalent circuit impedance existing for model models inaccurate problem, proposes a kind of piezoelectric energy collector based on artificial neural network
Part impedance modeling method is crossed and combines artificial neural network and traditional BVD device equivalent-circuit model, so that device resistance is built
Mould precision is increased dramatically.
The purpose of the present invention is what is be achieved through the following technical solutions.
Piezoelectric energy collecting device impedance modeling method based on artificial neural network of the invention, comprising the following steps:
Step 1 measures each frequency point in biasing reverse circuit resonance frequency using the N number of frequency point of impedance analyzer uniform scanning
The actual measurement impedance data of piezoelectric energy collecting device within the scope of rate Freq: actual measurement equivalent series resistance ESRMEASWith actual measurement equivalent string
Join capacitor ESCMEAS；
Step 2 extracts the parameter of BVD equivalent-circuit model as artificial neural network input data: equivalent according to surveying
Series resistance ESRMEASWith actual measurement equivalent series capacitance ESCMEAS, obtain quality factor qs, series resonance frequency Ωs, parallel resonance
Frequency omegap, equivalent coupled coefficient ρeff, and then calculate inductance Lm, resistance Rm, resistance Cm, BVD model output equivalent series resistance
ESRBVD, BVD model output equivalent series capacitance ESCBVD；
Step 3 to sum up obtains N group data, including actual measurement impedance data and artificial neural network input data, by this N group
Data are divided into two groups of odd even, training data of the odd number group as artificial neural network, survey of the even number set as artificial neural network
Try data；
Step 4 completes the training of artificial neural network by MATLAB, obtains optimal hidden layer number, hidden layer mind
Biasing and weight through first number, neuron, obtain trained artificial neural network；
Step 5 tests trained artificial neural network, output equivalent series resistance ESRANNWith it is equivalent
Series capacitance ESCANN, calculate ESRANNAnd ESRMEASAnd ESCANNAnd ESCMEASRelative error, if relative error is met the requirements
Modeling process is then completed, step 5 is repeated if relative error cannot be met the requirements, until relative error meets the requirements and completes
Modeling process.
Inductance L described in step 2m, resistance Rm, resistance Cm, BVD model output equivalent series resistance ESRBVD, BVD model
Output equivalent series capacitance ESCBVDIt is calculated as follows:
Ωs=1/ (LmCm)
ρeff=Cm/C0
Qs=ΩsLm/Rm
Wherein, ΩsIt is series resonance frequency；QsIt is quality factor, according to actual measurement equivalent series resistance ESRMEASExperiment curv
Obtain Qs=f0/ Δ f, f0It is actual measurement equivalent series resistance ESRMEASThe centre frequency of experiment curv, Δ f are actual measurement equivalent series
Resistance ESRMEASThe three dB bandwidth of experiment curv；ρeffIt is equivalent coupled coefficient, passes throughIt is calculated, Ωp
It is parallel resonance frequency；C0It is the internal stationary capacitor of piezoelectric energy collecting device；ω is biasing reverse circuit resonance frequency
Angular frequency corresponding to Freq.
Compared with prior art, the beneficial effects brought by the technical solution of the present invention are as follows:
The present invention combines artificial neural network and traditional BVD device equivalent-circuit model, so that device resistance models
Precision is increased dramatically, and has very accurate impedance prediction result in wider frequency range.The artificial mind of comparison display
The result and experimental measurements predicted through network model are coincide substantially.
Detailed description of the invention
Fig. 1 is the piezoelectric energy collecting device impedance model schematic diagram the present invention is based on artificial neural network；
Fig. 2 is BVD equivalent-circuit model schematic diagram of the present invention；
Fig. 3 is artificial neural network structure's schematic diagram of the present invention；
Fig. 4 is training and the testing process schematic diagram of artificial nerve network model of the present invention；
Fig. 5 is that BVD model exports ESR, artificial neural network in 2600Hz-3600Hz resonant frequency range of the present invention
Export ESR and experiment measurement ESR curve graph；
Fig. 6 is that BVD model exports ESC, artificial neural network in 2600Hz-3600Hz resonant frequency range of the present invention
Export ESC and experiment measurement ESC curve graph.
Specific embodiment
The invention will be further described with reference to the accompanying drawing.
Piezoelectric energy collecting device impedance modeling method based on artificial neural network of the invention, as shown in Figure 1, packet
It includes: Conventional wisdom model and artificial neural network.The Conventional wisdom model provides rough piezoelectricity energy as priori knowledge
The impedance model of collecting device is measured, and artificial neural network can learn the non-linear input and output of any complexity by training
Relationship.Empirical model can be reduced into artificial neural network training time and required training data in conjunction with artificial neural network,
The modeling result of degree of precision is obtained simultaneously.
The Conventional wisdom model uses BVD equivalent-circuit model, as shown in Figure 2.In the BVD model, inductance LmGeneration
The effective mass of table device, resistance RmRepresent the mechanical damping of device, CmRepresent the mechanical stiffness of device.Capacitor C0Represent device
Internal stationary capacitor.The piezoelectric energy collecting device impedance expression are as follows:
Wherein, ω is angular frequency corresponding to biasing reverse circuit resonance frequency Freq.
Artificial neural network structure input data, output data and neuron as shown in figure 3, be made of.The nerve net
Network input data is biasing reverse circuit resonance frequency Freq, series resonance frequency Ωs, quality factor qs, equivalent coupled coefficient
ρeff, device internal stationary capacitor C0, BVD equivalent-circuit model output equivalent series resistance ESRBVDWith BVD equivalent circuit mould
Type output equivalent series capacitance ESCBVD.Neural network output data is equivalent series resistance ESRANNAnd equivalent series capacitance
ESCANN.The neural network is three layer perceptron structure, is made of input layer, output layer and hidden layer.The neuron of hidden layer
Quantity can obtain optimal setting by repeatedly training and test.Number needed for data needed for artificial neural network training or test
According to comprising artificial neural network input data and actual measurement impedance data: actual measurement equivalent series resistance ESRMEASWith capacitor ESCMEAS。
Excitation function used in neuron is Sigmoid function.
Piezoelectric energy collecting device impedance modeling method based on artificial neural network of the invention, as shown in figure 4, including
Following steps:
Step 1 is measured each frequency point and is overturn in a certain range of biasing using the N number of frequency point of impedance analyzer uniform scanning
The actual measurement impedance data of piezoelectric energy collecting device within the scope of circuit resonant frequencies Freq: actual measurement equivalent series resistance ESRMEASWith
Survey equivalent series capacitance ESCMEAS。
Step 2 extracts the parameter of BVD equivalent-circuit model as artificial neural network input data: equivalent according to surveying
Series resistance ESRMEASWith actual measurement equivalent series capacitance ESCMEAS, obtain quality factor qs, series resonance frequency Ωs, parallel resonance
Frequency omegap, equivalent coupled coefficient ρeff, and then calculate inductance Lm, resistance Rm, resistance Cm, BVD model output equivalent series resistance
ESRBVD, BVD model output equivalent series capacitance ESCBVD。
Wherein, inductance Lm, resistance Rm, resistance Cm, BVD model output equivalent series resistance ESRBVD, BVD model output equivalent
Series capacitance ESCBVDIt is calculated as follows:
Ωs=1/ (LmCm) (2)
ρeff=Cm/C0 (3)
Qs=ΩsLm/Rm (4)
Wherein, ΩsIt is series resonance frequency, the maximum value of corresponding piezoelectric energy collecting device admittance；QsIt is quality factor,
According to actual measurement equivalent series resistance ESRMEASExperiment curv obtains Qs=f0/ Δ f, f0It is actual measurement equivalent series resistance ESRMEASMeasurement
The centre frequency of curve, Δ f are actual measurement equivalent series resistance ESRMEASThe three dB bandwidth of experiment curv；ρeffIt is equivalent coupled system
Number, can pass throughIt is calculated, ΩpIt is parallel resonance frequency, corresponding piezoelectric energy collecting device impedance
Maximum value；C0It is the internal stationary capacitor of piezoelectric energy collecting device.
Step 4 to sum up obtains N group data, including actual measurement impedance data and artificial neural network input data, by this N group
Data are divided into two groups of odd even, training data of the odd number group as artificial neural network, survey of the even number set as artificial neural network
Try data；
Step 5 completes the training of artificial neural network by MATLAB, and training algorithm can use Quasi-Newton iterative method,
The biasing and weight of optimal hidden layer number, hidden layer neuron number, neuron are obtained, trained artificial neuron is obtained
Network；
Step 6 is different from the new data of training data by one group, carries out to trained artificial neural network
Test, output equivalent series resistance ESRANNWith equivalent series capacitance ESCANN, calculate ESRANNAnd ESRMEASAnd ESCANNWith
ESCMEASRelative error, complete modeling process if relative error is met the requirements, weighed if relative error cannot be met the requirements
Multiple step 5 completes modeling process until relative error is met the requirements.
Embodiment:
The experimental provision of specific embodiment includes: impedance analyzer and piezoelectric energy collecting device Mide V25W.It is specific real
Applying and biasing reverse circuit resonance frequency Freq range in example is 2600Hz-3600Hz.Use impedance analyzer uniform scanning 504
Frequency point measures and surveys equivalent series resistance ESR within the scope of 2600Hz-3600HzMEASWith actual measurement equivalent series capacitance ESCMEAS.Root
According to measured actual measurement equivalent series resistance ESRMEASThe available Q of experiment curvs=f0/ Δ f=26.9；According to measured
Survey equivalent series resistance ESRMEASWith actual measurement equivalent series capacitance ESCMEASCalculating finds the admittance of piezoelectric energy collecting device most
Parallel resonance frequency corresponding to the maximum value of big value corresponding series resonance frequency Ω s and piezoelectric energy collecting device impedance
Rate Ωp；Pass through formulaρ is calculatedeff=0.025.Low frequency Static Electro is measured using impedance analyzer
Hold C0=191nF.L is calculated according to formulam=7.20 × 10-1H、Cm=4.78 × 10-9F、Rm=456.6 Ω.Basis simultaneously
BVD model output equivalent series resistance ESR can be calculated in formulaBVDWith BVD model output equivalent series capacitance ESCBVD.It is comprehensive
On obtain 504 groups of data, including 1. artificial neural network input datas: biasing reverse circuit resonance frequency Freq, series resonance
Frequency omegas, quality factor qs, equivalent coupled coefficient ρeff, device internal stationary capacitor C0, BVD equivalent-circuit model output etc.
Imitate series resistance ESRBVDWith BVD equivalent-circuit model output equivalent series capacitance ESCBVD；2. experiment actual measurement impedance data: actual measurement
Equivalent series resistance ESRMEASWith actual measurement equivalent series capacitance ESCMEAS。
504 groups of data obtained by above-mentioned calculating are divided into two groups of odd even, training data of the odd number group as artificial neural network,
Test data of the even number set as artificial neural network.The training that artificial neural network is completed by MATLAB, obtains optimal
The biasing and weight of hidden layer number, hidden layer neuron number, neuron.Again to trained artificial neural network into
Row test, so that ESRANNAnd ESRMEASAnd ESCANNAnd ESCMEASRelative error it is sufficiently small.Fig. 5 show BVD model,
The equivalent series resistance ESR curve of artificial nerve network model and experiment measurement.Fig. 6 shows BVD model, artificial neural network
The equivalent series capacitance ESC curve of model and experiment measurement.The result and reality that comparison display artificial nerve network model is predicted
Test magnitude coincide substantially.
Although function and the course of work of the invention are described above in conjunction with attached drawing, the invention is not limited to
Above-mentioned concrete function and the course of work, the above mentioned embodiment is only schematical, rather than restrictive, ability
The those of ordinary skill in domain under the inspiration of the present invention, is not departing from present inventive concept and scope of the claimed protection situation
Under, many forms can be also made, all of these belong to the protection of the present invention.
Claims (2)
1. a kind of piezoelectric energy collecting device impedance modeling method based on artificial neural network, which is characterized in that including following
Step:
Step 1 measures each frequency point in biasing reverse circuit resonance frequency using the N number of frequency point of impedance analyzer uniform scanning
The actual measurement impedance data of piezoelectric energy collecting device within the scope of Freq: actual measurement equivalent series resistance ESRMEASWith actual measurement equivalent series
Capacitor ESCMEAS；
Step 2 extracts the parameter of BVD equivalent-circuit model as artificial neural network input data: according to actual measurement equivalent series
Resistance ESRMEASWith actual measurement equivalent series capacitance ESCMEAS, obtain quality factor qs, series resonance frequency Ωs, parallel resonance frequency
Ωp, equivalent coupled coefficient ρeff, and then calculate inductance Lm, resistance Rm, resistance Cm, BVD model output equivalent series resistance
ESRBVD, BVD model output equivalent series capacitance ESCBVD；
Step 3 to sum up obtains N group data, including actual measurement impedance data and artificial neural network input data, by this N group data
It is divided into two groups of odd even, training data of the odd number group as artificial neural network, test number of the even number set as artificial neural network
According to；
Step 4 is completed the training of artificial neural network by MATLAB, obtains optimal hidden layer number, hidden layer neuron
The biasing and weight of number, neuron, obtain trained artificial neural network；
Step 5 tests trained artificial neural network, output equivalent series resistance ESRANNAnd equivalent series
Capacitor ESCANN, calculate ESRANNAnd ESRMEASAnd ESCANNAnd ESCMEASRelative error, it is complete if relative error is met the requirements
At modeling process, step 5 is repeated if relative error cannot be met the requirements, until relative error meets the requirements and completes to model
Process.
2. the piezoelectric energy collecting device impedance modeling method according to claim 1 based on artificial neural network, special
Sign is, inductance L described in step 2m, resistance Rm, resistance Cm, BVD model output equivalent series resistance ESRBVD, BVD model it is defeated
Equivalent series capacitance ESC outBVDIt is calculated as follows:
Ωs=1/ (LmCm)
ρeff=Cm/C0
Qs=ΩsLm/Rm
Wherein, ΩsIt is series resonance frequency；QsIt is quality factor, according to actual measurement equivalent series resistance ESRMEASExperiment curv obtains
Qs=f0/ Δ f, f0It is actual measurement equivalent series resistance ESRMEASThe centre frequency of experiment curv, Δ f are actual measurement equivalent series resistances
ESRMEASThe three dB bandwidth of experiment curv；ρeffIt is equivalent coupled coefficient, passes throughIt is calculated, ΩpIt is simultaneously
Join resonance frequency；C0It is the internal stationary capacitor of piezoelectric energy collecting device；ω is biasing reverse circuit resonance frequency Freq institute
Corresponding angular frequency.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810931373.9A CN109190202A (en) | 2018-08-15 | 2018-08-15 | A kind of piezoelectric energy collecting device impedance modeling method based on artificial neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810931373.9A CN109190202A (en) | 2018-08-15 | 2018-08-15 | A kind of piezoelectric energy collecting device impedance modeling method based on artificial neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109190202A true CN109190202A (en) | 2019-01-11 |
Family
ID=64918007
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810931373.9A Pending CN109190202A (en) | 2018-08-15 | 2018-08-15 | A kind of piezoelectric energy collecting device impedance modeling method based on artificial neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109190202A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110135010A (en) * | 2019-04-23 | 2019-08-16 | 天津大学 | The design method of RF power amplifier intervalve matching circuit is instructed using modeling |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101776862A (en) * | 2008-05-29 | 2010-07-14 | 通用电气公司 | System and method for advanced condition monitoring of an asset system |
| US20170368413A1 (en) * | 2016-03-12 | 2017-12-28 | Arie Shavit | Training system and methods for designing, monitoring and providing feedback of training |
-
2018
- 2018-08-15 CN CN201810931373.9A patent/CN109190202A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101776862A (en) * | 2008-05-29 | 2010-07-14 | 通用电气公司 | System and method for advanced condition monitoring of an asset system |
| US20170368413A1 (en) * | 2016-03-12 | 2017-12-28 | Arie Shavit | Training system and methods for designing, monitoring and providing feedback of training |
Non-Patent Citations (1)
| Title |
|---|
| SHENG ZHAO ET AL.: ""An artificial neural network-based approach for the impedance modeling of piezoelectric energy harvesting devices"", 《INTERNATIONAL JOURNAL OF NUMERICAL MODELLING: ELECTRONIC NETWORKS, DEVICES AND FIELDS》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110135010A (en) * | 2019-04-23 | 2019-08-16 | 天津大学 | The design method of RF power amplifier intervalve matching circuit is instructed using modeling |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Dunnmon et al. | Power extraction from aeroelastic limit cycle oscillations | |
| CN110245430A (en) | Improve Bouc-Wen model hysteresis modeling method | |
| CN108155648A (en) | Method for estimating state based on the infinite Extended Kalman filter of adaptive H | |
| Yang et al. | Modelling and analysis of an out-of-plane electret-based vibration energy harvester with AC and DC circuits | |
| Huang et al. | Experimental and computational simulation for wind effects on the Zhoushan transmission towers | |
| CN105138782A (en) | Non-stationary pulse wind speed high-precision prediction method based on EEMD-ELM | |
| Gu et al. | Back analysis of mechanical parameters of roller compacted concrete dam | |
| CN109472085A (en) | Band spring hinge arrangement multi-objective optimization design of power method based on Response surface meth od | |
| CN104317985A (en) | Fluid simulation method based on inter-belt finite element and Lagrange coordinate | |
| CN109800520A (en) | A kind of electric automobile charging station Harmonic Modeling method neural network based | |
| Wang et al. | Comparisons of bioinspired piezoelectric wind energy harvesters with different layout of stiffeners based on leaf venation prototypes | |
| CN110296833A (en) | A kind of flexible measurement method and system of Hydraulic Cylinder combined test stand | |
| CN107562990A (en) | BAW filter design methods, device and equipment based on BVD models | |
| CN105956242B (en) | The multiport impedance model construction method calculated based on organism electric shock impedance parameter | |
| CN106446469A (en) | Spatial mesh antenna electrical property dynamic response analysis method based on second-order approximate calculation formula | |
| CN112711902A (en) | Power grid voltage calculation method based on Monte Carlo sampling and deep learning | |
| CN109190202A (en) | A kind of piezoelectric energy collecting device impedance modeling method based on artificial neural network | |
| CN104217260B (en) | A kind of wind field measures the combination fill system of wind speed defect value adjacent to many typhoon motors | |
| CN106525368B (en) | A kind of cathead electric power pylon damping ratio recognition methods | |
| CN102324887A (en) | Wind driven generator stability control method based on hybrid neural network | |
| CN113872239A (en) | Knowledge and data combined drive-based grid-connected inverter impedance obtaining method | |
| CN108964125A (en) | Improve the large-scale wind electricity access point selection method of the safe robustness of major network generator rotor angle | |
| CN110245370A (en) | A kind of high CFRD multiple target mechanics parameter inversion method | |
| Ge et al. | A dynamic parameter model of harmonic source networks | |
| CN107453358A (en) | A kind of electric network natural resonance structure analysis method based on complex frequency domain lower node admittance matrix |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190111 |