CN112799328A - Air injection control system - Google Patents
Air injection control system Download PDFInfo
- Publication number
- CN112799328A CN112799328A CN202110023054.XA CN202110023054A CN112799328A CN 112799328 A CN112799328 A CN 112799328A CN 202110023054 A CN202110023054 A CN 202110023054A CN 112799328 A CN112799328 A CN 112799328A
- Authority
- CN
- China
- Prior art keywords
- neural network
- network model
- gas flow
- gas
- output
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
- G05B19/042—Programme control other than numerical control, i.e. in sequence controllers or logic controllers using digital processors
- G05B19/0423—Input/output
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- 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
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/24—Pc safety
- G05B2219/24215—Scada supervisory control and data acquisition
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
Abstract
The invention relates to the field of automatic production, and discloses a jet control system, wherein an NARX neural network model 1 and an NARX neural network model 2 are used for respectively controlling gas flow and predicting a gas flow value, because the NARX neural network establishes a dynamic recursive network of the model by introducing a delay module and outputting feedback, input and output vector delay feedback is introduced into network training to form a new input vector, and the NARX neural network has good nonlinear mapping capability, the input of the NARX neural network not only comprises the input data of the error, the controlled quantity and the actual gas flow of the original gas flow, but also comprises corresponding output data after training, the generalization capability of the network is improved, so that the NARX neural network has better prediction precision and self-adaption capability in gas flow corresponding parameter prediction compared with the traditional static neural network, and the singlechip controller improves the precision, the gas flow value and the self-adaption capability of the control system, Robustness and reliability of the system.
Description
Technical Field
The invention relates to the technical field of automatic production, in particular to an air injection control system.
Background
At present, most of scanners used for recording important materials of existing files are in contact type scanning, paper is conveyed by using a combined roller and a conveyor belt, because the physical characteristics of the material paper are very sensitive, one part of the paper is likely to be adsorbed on a pressing plate above the paper during scanning, and the other part of the paper is flatly laid on a transparent placing plate below the paper, so that the flatness of the paper during scanning is not uniform, and the problems of clear partial scanning and unclear partial scanning occur; in addition, incomplete scanning may be caused by paper scrap blocking the paper, or uneven rolling of the paper.
In addition, in the existing scanning equipment, the controller has poor control precision, stability, timeliness and adaptability to all parts during scanning, and the scanning effect is influenced.
Disclosure of Invention
The purpose of the invention is as follows: aiming at the problems in the prior art, the invention provides the air injection control system which can improve the control precision, stability, timeliness and adaptability of a controller to an air injection device during scanning and improve the scanning effect.
The technical scheme is as follows: the invention provides an air injection control system, which comprises an object conveying device, an air injection device and a scanning device, wherein the object conveying device, the air injection device and the scanning device are arranged on a rack; the object conveying device is used for conveying an object to be scanned into a scanning area, the air injection device is used for blowing the object to be scanned flat, and the scanning device is used for scanning the object to be scanned; the gas injection device comprises a single chip microcomputer controller, a gas injection box and a gas injection shell communicated with the bottom of the gas injection box through a gas injection pipe provided with an electromagnetic valve, and a gas flow sensor is arranged on the gas injection pipe; the single chip microcomputer controller controls the gas in the gas spraying box to be sprayed out of the gas spraying hole of the gas spraying shell through the gas spraying pipe by adjusting the opening degree of the electromagnetic valve, and the single chip microcomputer controller ensures that the gas flow sensor detects that the gas flow of the gas spraying pipe reaches a set value by adjusting the opening degree of the electromagnetic valve; the single chip microcomputer controller comprises an STM32 single chip microcomputer and an intelligent controller in an STM32 single chip microcomputer, wherein the intelligent controller comprises 2 NARX neural network models, 3 LSTM neural network models, a dynamic recursive wavelet neural network controller, an Empirical Mode Decomposition (EMD) model and a plurality of ESN neural network models; the gas flow regulation platform is formed by the STM32 single chip microcomputer, the jet box, the electromagnetic valve, the jet pipe, the jet shell and the gas flow sensor, and intelligent regulation is realized on the gas flow by the intelligent controller in the STM32 single chip microcomputer.
Further, in the intelligent controller, 2 NARX neural network models are a NARX neural network model 1 for controlling the gas flow and a NARX neural network model 2 for predicting the gas flow value, and 3 LSTM neural network models are an LSTM neural network model 1, an LSTM neural network model 2 and an LSTM neural network model 3, respectively.
Further, the output value of the LSTM neural network model 1 is respectively used as the input of the dynamic recursive wavelet neural network controller and the corresponding input of the NARX neural network model 1, the sum of the output value of the dynamic recursive wavelet neural network controller and the output value of the NARX neural network model 1 is used as the input of the LSTM neural network model 3, the output of the LSTM neural network model 3 is respectively used as the control value for adjusting the opening of the solenoid valve and the corresponding input of the NARX neural network model 1, the gas flow sensor for detecting the flow of the gas nozzle outputs the gas flow value for a period of time as the input of the LSTM neural network model 2, the output of the LSTM neural network model 2 is respectively used as the input of the empirical mode decomposition EMD model and the corresponding input of the NARX neural network model 1, the empirical mode decomposition EMD model outputs the low-frequency trend part and the high-frequency fluctuation parts of the LSTM neural network model 2 as the inputs of the multiple ESN neural network, the output values of the plurality of ESN neural network models are used as the input of the NARX neural network model 2, the output value of the NARX neural network model 2 is used as the gas flow feedback value, and the error change rate of the gas flow set value of the gas nozzle and the output value of the NARX neural network model 2 are used as the input of the LSTM neural network model 1.
Further, the LSTM neural network model 3 realizes the prediction of the sum of the output value of the dynamic recursive wavelet neural network controller and the output of the NARX neural network model 1 and the prediction control of the gas flow rate again, and the NARX neural network model 2 realizes the fusion of the output values of the multiple ESN neural network models and the accurate prediction of the gas flow rate again.
Further, in the gas flow regulation, the output of an LSTM neural network model 3 of an intelligent controller in an STM32 singlechip is used as the input of a solenoid valve opening degree control value as a solenoid valve control signal, the LSTM neural network model 3 outputs the solenoid valve opening degree control value to ensure that the gas flow entering the gas spraying pipe from the gas spraying box through the solenoid valve reaches the gas flow set value of the gas spraying pipe, a gas flow sensor detects the gas flow value of the gas spraying pipe as the input of the LSTM neural network model 2, and the gas flowing out of the gas spraying pipe is sprayed out of the gas spraying shell.
Furthermore, in the object conveying device, a conveying guide rail is horizontally fixed on the rack, a transparent placing plate is movably connected with the conveying guide rail, and a placing plate driving mechanism arranged on the rack is used for driving the transparent placing plate to horizontally move in or out of a scanning area along the conveying guide rail.
Preferably, in the placing plate driving mechanism, a first motor is fixed on the frame, a rotating shaft of a gear is fixed on an output shaft of the first motor, one side of a rack is fixed with the transparent placing plate, and the other side of the rack is meshed with the gear. The first motor drives the gear to rotate, the transparent placing plate is driven by the gear and the rack to enter or exit from a scanning area along the conveying guide rail, and the defects that short-distance conveying is not stable and unreliable, precision is not well controlled and the like caused by the transmission of a conveying belt on the traditional market are overcome.
Furthermore, the scanning device further comprises a lower scanning probe which is arranged on the machine frame and on the lower side of the scanning area, and the lower scanning probe is positioned below the transparent placing plate. The cooperation of the upper scanning probe and the lower scanning probe realizes the simultaneous scanning of the front and back surfaces of the object to be scanned.
Furthermore, in the air injection device, the side surface of the air injection shell is also provided with a plurality of air injection holes. The side of the air injection shell is also provided with the air injection hole, so that the edge of an object to be scanned below the air injection shell can be effectively prevented from tilting.
Preferably, a plurality of air injection holes on the side surface of the air injection shell form an included angle of 45 degrees with the horizontal plane. The design of such an angle can effectively avoid the edge tilting of the object to be scanned below the air injection shell.
Further, the gas injection control system further comprises a scanning rendering lamp installed on the rack, and the scanning rendering lamp is located between the upper scanning probe and the transparent placing plate.
Has the advantages that: compared with the traditional controller, the intelligent controller in the single chip microcomputer has the advantages
1. The invention utilizes the NARX neural network model 1 and the NARX neural network model 2 to predict the gas flow control and the gas flow value respectively, because the NARX neural network establishes the dynamic recursive network of the model by introducing the delay module and the output feedback, the NARX neural network introduces the input and output vector delay feedback into the network training to form a new input vector, and has good nonlinear mapping capability, the input of the NARX neural network not only comprises the input data of the error, the control quantity and the actual gas flow of the original gas flow, but also comprises the corresponding output data after training, the generalization capability of the network is improved, and the NARX neural network has better prediction precision and self-adaptive capability in the prediction of the corresponding parameters of the gas flow compared with the traditional static neural network.
2. The LSTM neural network model is similar to a standard network containing a recursion hidden layer, the only change is to use a memory module to replace an original hidden layer unit, the problems of gradient disappearance and sharp increase are solved by self-feedback of the internal state of a memory cell and truncation of errors of input and output, compared with a BP neural network and a common RNN, the LSTM is added with 1 state unit c and 3 control gates, the feature inclusion capacity and the memory capacity of the model are greatly increased, and under-fitting and gradient disappearance are avoided. The function of the LSTM aims at a plurality of gas flow errors and error changes, gas flow values and the correlation relationship existing in the sum data output by the NARX neural network model 1 and the dynamic recursive wavelet neural network controller, and remembers the relationship and the change of the relationship in time, so that a more accurate result of the control quantity of the output control gas flow of the LSTM neural network model 3, the output gas flow of the LSTM neural network model 2 and the output gas flow error of the LSTM neural network model 1 are obtained, and the accuracy of the output control gas flow is improved.
3. The LSTM neural network model has a chain-like repeating network structure similar to a standard RNN, the repeating network in the standard RNN is very simple, and the repeating network in the LSTM neural network model has 4 interaction layers including 3 gate layers and 1 tanh layer. The processor state is a key variable in the LSTM neural network model, and the 3 LSTM neural network models respectively carry the information of the previous NARX neural network model 1 and the sum of the dynamic recursive wavelet neural network controller outputs, the multiple ESN neural network model outputs and the gas flow sensor output steps and gradually pass through the whole LSTM neural network model. The gates in the interaction layer of the 3 LSTM neural network models can partially delete the processor state of the previous step and add the sum of the NARX neural network model 1 and the dynamic recursive wavelet neural network controller output, the multiple ESN neural network model outputs, and the gas flow sensor output new information into the processor state of the current step based on the hidden state of the previous step and the input of the current step. The inputs to each repeating network include the hidden state and processor state of the previous step and the input of the current step. The processor state is updated according to the calculation results of the 4 interaction layers. The updated processor state and hidden state constitute the output and are passed on to the next step.
4. The 3 LSTM neural network model is a recurrent neural network with 4 interaction layers in a repeating network. It can not only extract information from the sum of the NARX neural network model 1 and the dynamic recursive wavelet neural network controller output, multiple ESN neural network model outputs and gas flow sensor output sequence data, like a standard recurrent neural network, but also retain information with long-term correlation from previous distant steps. In addition, since the sampling interval between the sum of the NARX neural network model 1 and the dynamic recursive wavelet neural network controller output, the multiple ESN neural network model outputs, and the gas flow sensor output is relatively small, there is a long-term spatial correlation between the sum of the NARX neural network model 1 and the dynamic recursive wavelet neural network controller output, the multiple ESN neural network model outputs, and the gas flow sensor output, while the 3 LSTM neural network models have sufficient long-term memory to handle this problem.
5. The ESN neural network model designs a network hidden layer into a sparse network consisting of a plurality of neurons, achieves the function of memorizing the output data of the EMD model by a plurality of empirical modes by adjusting the characteristic of the internal weight of the network, the internal dynamic reserve pool contains a large number of sparsely connected neurons, contains the running state of the system, has the function of memorizing the output value of the EMD model by a plurality of empirical modes in a short term, ensures the stability of the internal recursive network of the reserve pool by presetting the spectrum radius of the internal connection weight matrix of the ESN neural network model, and improves the stability and the accuracy of the output gas flow feedback value of the ESN neural network model.
6. The ESN neural network model inherits the current time of the state of the reserve pool to the previous time of the state of the reserve pool and has transient memory characteristics to historical data output by the EMD model, and research results show that the ESN neural network model with the historical memory has the characteristics of high precision, high accuracy, high timeliness and stability for predicting the output value of the EMD model; as a novel dynamic recurrent neural network, the ESN neural network model is established by adopting a linear regression method, the problems that the traditional neural network is low in convergence speed and easy to fall into local minimum are solved, the complexity of the training process is simplified, and the purpose of efficiently predicting the EMD model output value by empirical mode decomposition is realized.
7. The difference between the dynamic recursive wavelet neural network controller and the common static wavelet neural network lies in that the dynamic recursive wavelet neural network controller has two function associated layer nodes which play the role of storing the internal state of the network, a self-feedback loop with fixed gain is added on the two associated layer nodes, and the memory performance of time series characteristic information is enhanced, so that the accuracy of controlling gas flow is enhanced to ensure the accuracy and stability of better gas flow control.
8. According to the invention, an output sequence of an original LSTM neural network model 2 is decomposed into components of different frequency bands through an EMD model, and each component displays different characteristic information of gas flow hidden in the original sequence so as to reduce the non-stationarity of the sequence. The data relevance of the high-frequency fluctuation part is not strong, the frequency is higher, the high-frequency fluctuation part represents the fluctuation component of the original sequence, and the high-frequency fluctuation part has certain periodicity and randomness, and the periodicity and the randomness are consistent with the periodic variation of the gas flow; the low-frequency trend component represents the variation trend of the original sequence. The EMD model can gradually decompose fluctuation components, periodic components and trend components of the gas flow, each decomposed component contains the same deformation information, mutual interference among different gas flow characteristic information is reduced to a certain degree, and the decomposed component change curve is smoother than the original gas flow deformation sequence curve. Therefore, the EMD model can effectively analyze gas flow deformation data under the combined action of multiple factors, and each component obtained by decomposition is favorable for building and better predicting the ESN neural network model. A plurality of ESN neural network models are input to each component, in order to avoid the problems of randomness of selection of input dimension of the extreme learning machine, component information loss and the like, the phase space is reconstructed for each component, and finally the prediction results are fused through the NARX neural network model 2, so that the fused prediction results have high prediction precision.
Drawings
FIG. 1 is a schematic view of the overall configuration of the jet control system of the present invention;
FIG. 2 is a schematic view of a partial configuration of the air injection control system of the present invention;
FIG. 3 is a schematic view of a partial configuration of the air injection control system of the present invention;
FIG. 4 is a schematic structural view of an air sparging shell;
fig. 5 is a flow chart of the gas flow regulating platform and the intelligent controller.
Detailed Description
The present invention will be described in detail with reference to the accompanying drawings.
Embodiment 1:
the present embodiment provides an air jet control system, which is mainly composed of an object conveying device 4, an air jet device 2, and a scanning device 1, as shown in fig. 1 to 4. The object conveying device 4 is used for conveying an object to be scanned into a scanning area, the air injection device 2 is used for blowing the object to be scanned flat, and the scanning device 1 is used for scanning the object to be scanned.
In the object conveying device 4, a conveying guide rail 41 is horizontally fixed on a machine frame 10, a transparent placing plate 42 is movably connected with the conveying guide rail 41, and a placing plate driving mechanism arranged on the machine frame 10 is used for driving the transparent placing plate 42 to horizontally move into or out of a scanning area along the conveying guide rail 41; in the placing plate driving mechanism, a first motor 45 is fixed on the frame 10, a rotating shaft of a gear 44 is fixed on an output shaft of the first motor 45, one side of a rack 43 is fixed with the transparent placing plate 42, and the other side is meshed with the gear 44.
The gas injection device 2 comprises a single chip microcomputer controller, a gas injection box 22 and a gas injection shell 21 communicated with the bottom of the gas injection box 22 through a gas injection pipe 23 provided with an electromagnetic valve, wherein a gas flow sensor (the preferred model is HHF-SD1) is arranged on the gas injection pipe 23. The single chip microcomputer controller controls the gas in the gas spraying box 22 to be sprayed out of the gas spraying holes of the gas spraying shell 21 through the gas spraying pipe 23 by adjusting the opening degree of the electromagnetic valve, and the single chip microcomputer controller ensures that the gas flow sensor detects that the gas flow of the gas spraying pipe 23 reaches a set value by adjusting the opening degree of the electromagnetic valve. The air injection box 22 is fixed on the frame 10, and the bottom surface and the side surface of the air injection shell 21 are both provided with a plurality of air injection holes, and the bottom surface of the air injection shell 21 is arranged in parallel with the transparent placing plate 42 and is positioned above the scanning area; a plurality of gas injection holes on the side surface of the gas injection shell 21 form an included angle of 45 degrees with the horizontal plane.
In the scanning device 1, the upper scanning probe 11 and the lower scanning probe 12 are respectively installed on the rack 10 above and below the scanning area, and the upper scanning probe 11 is located above the transparent placing plate 42, the lower scanning probe 12 is located below the transparent placing plate 42, and after the transparent placing plate 42 enters the scanning area, the upper scanning probe 11 and the lower scanning probe 12 are respectively located above and below the transparent placing plate 42.
A scanning rendering lamp is further installed on the gantry 10 between the upper scanning probe 11 and the transparent placing plate 42.
The operating principle of the air injection control system in the present embodiment is as follows:
after a user places paper to be scanned on the transparent placing plate 42 on the object conveying device 4, the first motor 45 drives the gear 44 to rotate forwards, the transparent placing plate 42 is meshed with the rack 43 through the gear 44 to drive the transparent placing plate 42 to enter a scanning area along the conveying guide rail 41, after the paper enters the scanning area, the paper to be scanned on the transparent placing plate 42 is just positioned under the air injection shell 21 in the air injection device 2 and is also positioned under the upper scanning probe 11, and the first motor 45 stops rotating forwards.
Then the single chip microcomputer controller controls the opening of the electromagnetic valve to increase, after the gas in the gas spraying box 22 enters the gas spraying shell 21 through the gas spraying pipe 23, the gas is sprayed out through a plurality of gas spraying holes on the bottom surface and the side surface of the gas spraying shell 21, and then the upper scanning probe 11 and the lower scanning probe 12 in the scanning device 1 start to perform non-contact scanning on the front surface and the back surface of the paper to be scanned; when the gas flow detected by the gas flow sensor through the gas injection pipe 23 reaches a set value, the opening of the electromagnetic valve is controlled to be reduced, the gas injection device 2 stops injecting gas, and the upper scanning probe 11 and the lower scanning probe 12 stop scanning. The first motor 45 drives the gear 44 to rotate reversely, the transparent placing plate 42 is driven to move out of the scanning area along the conveying guide rail 41 through the meshing of the gear 44 and the rack 43, and the first motor 45 stops rotating reversely. And after the scanning is finished, taking away the scanned paper to perform the next scanning.
Aiming at old or wrinkled paper, the scanning rendering lamp 13 can be turned on in the scanning process to adjust different brightness and light positions, so that the problems of old, wrinkling and the like are solved, and the scanning effect is better.
This scanning equipment can realize non-contact scanning, and object conveyer can realize carrying out automatic scanning to the object of different specifications and irregular shape, and the user only need will wait to scan the object put the transparent board of placing on can. The air jet device can blow the object to be scanned flat when scanning, so that the object to be scanned is completely laid on the transparent placing plate below, and sundries such as paper scraps on the object to be scanned are blown away, so that the scanning integrity, uniformity and definition are improved while non-contact scanning is realized.
The single-chip microcomputer controller comprises an STM32 single-chip microcomputer and an intelligent controller in an STM32 single-chip microcomputer, wherein the intelligent controller comprises 2 NARX neural network models, 3 LSTM neural network models, a dynamic recursive wavelet neural network controller, an Empirical Mode Decomposition (EMD) model and a plurality of ESN neural network models; the gas flow adjusting platform consists of an STM32 singlechip, a gas spraying box, an electromagnetic valve, a gas spraying pipe, a gas spraying shell and a gas flow sensor, and an intelligent controller in the STM32 singlechip realizes intelligent adjustment of the gas flow; the gas flow regulating platform and intelligent controller are shown in fig. 5.
The specific design idea of the single chip microcomputer controller is as follows:
1. dynamic recursive wavelet neural network controller design
The output of the LSTM neural network model 1 is used as the input of a dynamic recursive wavelet neural network controller, and the sum of the output value of the dynamic recursive wavelet neural network controller and the output value of the NARX neural network model 1 is used as the input of an LSTM neural network model 3; the current value output by the LSTM neural network model 1, the previous time value output by the LSTM neural network model 1 and the previous two time values output by the LSTM neural network model 1 are used as 3 inputs of a dynamic recursive wavelet neural network controller, and the output value of the dynamic recursive wavelet neural network controller is used as a gas flow control value; 3 continuous LSTM neural network model 1 output values are used as the input of a dynamic recursive wavelet neural network controller, and the output value of the dynamic recursive wavelet neural network controller is used as a gas flow control value; wavelet Neural network WNN (wavelet Neural networks) theoretical basis is a feedforward network provided by taking a wavelet function as an excitation function of a neuron and combining an artificial Neural network, wherein the expansion and contraction, the translation factor and the connection weight of wavelets in the wavelet Neural network are adaptively adjusted in the optimization process of an error energy function. An input signal of the dynamic recursive wavelet neural network controller can be represented as an input one-dimensional vector xi(i ═ 1,2, …, n), the output signal is denoted yk(k ═ 1,2, …, m), the calculation formula of the dynamic recursive wavelet neural network controller output value is:
in the formula omegaijInputting the connection weight between the i node of the layer and the j node of the hidden layer,
as wavelet basis functions, bjIs a shift factor of the wavelet basis function, ajScale factor, omega, of wavelet basis functionsjkThe connection weight between the node of the hidden layer j and the node of the output layer k. The difference between the dynamic recursive wavelet neural network controller and the ordinary static wavelet neural network is that the dynamic recursive wavelet neural network controller has two storage networksThe action of the part state is related to the layer nodes, a self-feedback loop with fixed gain is added on the two related layer nodes, and the memory performance of time sequence characteristic information is enhanced, so that the tracking precision of the evolution track control of the gas flow value is enhanced to ensure better control precision; the first associated layer node is used for storing the state of the phase point of the hidden layer node at the previous moment and transmitting the state to the hidden layer node at the next moment; the second correlation layer node is used for storing the state of the phase point of the output layer node at the previous moment and transmitting the state to the hidden layer node at the next moment; feedback information of neurons of the hidden layer and the output layer can influence the dynamic processing capacity of a gas flow value output by the dynamic recursive wavelet neural network controller, and two related layers belong to state feedback inside the dynamic recursive wavelet neural network controller, so that the dynamic memory performance specific to the recursion of the dynamic recursive wavelet neural network controller is formed, and the accuracy and the dynamic performance of controlling the gas flow by the dynamic recursive wavelet neural network controller are improved; a group of connection weights are added between the first association layer node and the output layer node of the dynamic recursive wavelet neural network controller to enhance the dynamic approximation capability of the dynamic recursive wavelet neural network controller in controlling the gas flow value and improve the precision of predicting the gas flow value. The weight and threshold correction algorithm of the dynamic recursive wavelet neural network controller in the patent adopts a gradient correction method to update the network weight and the wavelet basis function parameters, so that the output of the dynamic recursive wavelet neural network is continuously close to the expected output.
2. NARX neural network model design
The output of the LSTM neural network model 1 is respectively the input of a dynamic recursive wavelet neural network controller and the corresponding input of the NARX neural network model 1, the output of the LSTM neural network model 3 is respectively used as the control value of the opening of the regulating solenoid valve and the corresponding input of the NARX neural network model 1, the gas flow sensor for detecting the flow of the gas ejector pipe outputs a gas flow value for a period of time as the input of the LSTM neural network model 2, the output of the LSTM neural network model 2 is respectively used as the input of an empirical mode decomposition EMD model and the corresponding input of the NARX neural network model 1, and the sum of the output value of the dynamic recursive wavelet neural network controller and the output value of the NARX neural network model 1 is used as the input of the LSTM neural network model 3; the multiple ESN neural network model outputs as inputs to the NARX neural network model 2, the NARX neural network model 2 output values as the gas flow feedback values, and the error and error rate of change of the gas flow setpoint and the NARX neural network model 2 output values as inputs to the LSTM neural network model 1. The NARX neural network model (Nonlinear Auto-Regression with External input neural network) is a dynamic feedforward neural network, the NARX neural network model is a Nonlinear autoregressive network with External input, the NARX neural network model has a dynamic characteristic of multistep time delay and is connected with a plurality of layers of closed networks through feedback, and the recurrent neural network of the NARX neural network model is a dynamic neural network which is widely applied in a Nonlinear dynamic system and has the performance generally superior to that of a full recurrent neural network. Before application, the delay order and the number of hidden layer neurons of the input and output are generally determined in advance, the current output of the NARX neural network model not only depends on the past output y (t-n), but also depends on the current input prediction vector X (t), the delay order of the input prediction vector and the like, wherein an input signal is transmitted to the hidden layer through an epitaxial layer, the hidden layer processes the input signal and then transmits the processed signal to the output layer, the output layer linearly weights the output signal of the hidden layer to obtain a final neural network output signal, and the epitaxial layer delays a signal fed back by the network and a signal output by the input layer and then transmits the final neural network output signal to the hidden layer. The NARX neural network model has the characteristics of nonlinear mapping capability, good robustness, adaptability and the like, and is suitable for predicting nonlinear change parameters. x (t) represents the external input of the NARX neural network model, and m represents the delay order of the external input; y (t) is the output of the NARX neural network model, n is the output delay order; s is the number of hidden layer neurons; the output of the jth implicit element can be found as:
in the above formula，wjiAs a connection weight between the ith input and the jth implicit neuron, bjIs the bias value of the jth implicit neuron, and the output y (t +1) of the network has the value:
y(t+1)＝f[y(t),y(t-1),…,y(t-n),x(t),x(t-1),…,x(t-m+1)；W] (3)
3. LSTM neural network model design
The output of the LSTM neural network model 1 is respectively the input of a dynamic recursive wavelet neural network controller and the corresponding input of the NARX neural network model 1, the sum of the output value of the dynamic recursive wavelet neural network controller and the output value of the NARX neural network model 1 is used as the input of the LSTM neural network model 3, the output of the LSTM neural network model 3 is respectively used as the control value of the opening of the regulating solenoid valve and the corresponding input of the NARX neural network model 1, a gas flow sensor for detecting the flow of the gas ejector pipe outputs a gas flow value for a period of time as the input of the LSTM neural network model 2, and the output of the LSTM neural network model 2 is respectively used as the input of the EMD model and the corresponding input of the NARX neural network model 1. The temporal Recurrent Neural Network (RNN) model, which consists of Long Short Term Memory (LSTM) elements, is called the LSTM temporal recurrent neural network, also commonly referred to as the LSTM network. The LSTM neural network model introduces mechanisms of Memory cells (Memory cells) and hidden layer states (Cell states) to control the transfer of information between hidden layers. The memory unit of an LSTM neural network has 3 Gates (Gates) as Input Gate, forgetting Gate and Output Gate. Wherein, the input gate can control the adding or filtering of new information; the forgetting door can forget the information to be lost and keep the useful information in the past; the output gate enables the memory unit to output only information related to the current time step. The 3 gate structures carry out operations such as matrix multiplication, nonlinear summation and the like in the memory unit, so that the memory still cannot be attenuated in continuous iteration. The long-short term memory unit (LSTM) structure unit is composed of a unit (Cell), an Input Gate (Input Gate), an Output Gate (Output Gate) and a forgetting Gate (Forget Gate). The unit is responsible for remembering values at arbitrary time intervals, and all three gates can be considered as conventional artificial neurons for computing the addition of activation functionsThe sum of the weights. The LSTM neural network model is a model which can last for a long time and has short-term memory, is suitable for work such as prediction of time sequences and the like, effectively prevents gradient disappearance during RNN training by the LSTM, and is a special RNN by a long-short-term memory (LSTM) network. The model can learn long-term dependency information while avoiding the gradient vanishing problem. LSTM adds a structure called a Memory Cell (Memory Cell) to a neural node of a hidden layer of an internal structure RNN of a neuron to memorize past information, and adds three kinds of gate structures (Input, form, Output) to control use of history information. Let the number sequence of the input LSTM neural network model be (x)1,x2,…xT) The hidden layer state is (h)1,h2,…hT) Then, time t has:
it＝sigmoid(Whiht-1+WxiXt) (4)
ft＝sigmoid(Whfht-1+WhfXt) (5)
ct＝ft⊙ct-1+it⊙tanh(Whcht-1+WxcXt) (6)
ot＝sigmoid(Whoht-1+WhcXt+Wcoct) (7)
ht＝ot⊙tanh(ct) (8)
wherein it、ft、otRepresenting input, forget and output doors, CtRepresenting cell units, Wh represents weight of recursive connections, Wx represents weight of input layer to hidden layer, and sigmoid and tanh are two activation functions. The method comprises the steps of firstly establishing an LSTM time recurrent neural network model, establishing a training set by using gas flow value data and training the model, wherein the LSTM neural network model considers the time sequence and nonlinearity of the output data of the gas flow sensor, and improves the prediction accuracy of the gas flow value.
4. EMD model design
A gas flow sensor for detecting the flow of the gas ejector pipe outputs a gas flow value for a period of time as the input of an LSTM neural network model 2, the output of the LSTM neural network model 2 is respectively used as the input of an empirical mode decomposition EMD model and the corresponding input of an NARX neural network model 1, and the empirical mode decomposition EMD model outputs a low-frequency trend part and a plurality of high-frequency fluctuation parts of the LSTM neural network model 2 and is respectively used as the input of a plurality of ESN neural network models; an Empirical Mode Decomposition (EMD) model is a self-adaptive signal screening method and has the characteristics of simplicity in calculation, intuition, experience-based and self-adaption. It can screen the trends of different characteristics existing in the gas flow signal step by step to obtain a plurality of high frequency fluctuation parts (IMF) and low frequency trend parts. The IMF component decomposed by EMD contains components of different frequency bands of the gas flow signal from high to low, and the frequency resolution contained in each frequency band changes along with the signal, so that the self-adaptive multi-resolution analysis characteristic is realized. The purpose of EMD decomposition is to extract gas flow value information more accurately. The IMF component must satisfy two conditions simultaneously: in a gas flow signal to be decomposed, the number of extreme points is equal to the number of zero-crossing points, or the difference is one at most; at any one time, the envelope mean defined by the local maxima and the local minima is zero. The empirical mode decomposition method aims at the steps of a screening process of an LSTM neural network model output value 2 signal as follows:
(1) all local extreme points of the output value signals of the LSTM neural network model 2 are determined, and then the left and right local extreme points are connected by three spline lines to form an upper envelope line.
(2) When the local minimum value points of the output values of the LSTM neural network model 2 are connected by the cubic spline lines to form a lower envelope line, the upper envelope line and the lower envelope line should envelop all data points.
(3) The average of the upper and lower envelope lines is denoted as m1(t), obtaining:
x(t)-m1(t)＝h1(t) (9)
x (t) is the original signal of the output value of LSTM neural network model 2, if h1(t) is aIMF, then h1(t) is the first IMF component of x (t). Note c1(t)＝h1k(t), then c1(t) is the first component of signal x (t) that satisfies the IMF condition.
(4) C is to1(t) separating from x (t) to obtain:
r1(t)＝x(t)-c1(t) (10)
will r is1(t) repeating the steps (1) to (3) as the original data to obtain the 2 nd component c satisfying the IMF condition of x (t)2. The cycle is repeated n times to obtain n components of the signal x (t) satisfying the IMF condition. Thus, the output value of the LSTM neural network model can be decomposed into a low-frequency trend part and a plurality of high-frequency fluctuation parts of the gas flow.
5. ESN neural network model design
The gas flow sensor for detecting the flow of the gas ejector pipe outputs a gas flow value for a period of time as the input of the LSTM neural network model 2, the output of the LSTM neural network model 2 is respectively used as the input of the empirical mode decomposition EMD model and the corresponding input of the NARX neural network model 1, the empirical mode decomposition EMD model outputs the low-frequency trend part and the high-frequency fluctuation parts of the LSTM neural network model 2 and is respectively used as the input of a plurality of ESN neural network models, the output of the ESN neural network models is used as the input of the NARX neural network model 2, the output value of the NARX neural network model 2 is used as a gas flow feedback value, and the error change rate of a gas flow set value and the output value of the NARX neural network model 2 are used as the input of the LSTM neural network. An ESN (Echo state network, ESN) is a novel dynamic neural network, has all the advantages of the dynamic neural network, and can better adapt to nonlinear system identification compared with a common dynamic neural network because the Echo state network introduces a reserve pool concept. The reserve pool is a randomly connected reserve pool which is formed by converting a part connected among traditional dynamic neural networks, and the whole learning process is a process of learning how to connect the reserve pool. The "pool" is actually a randomly generated large-scale recursive structure in which the interconnection of neurons is sparse, usually denoted SD as the percentage of interconnected neurons in the total number of neurons N. The state equation of the ESN neural network model is as follows:
wherein W is the state variable of the neural network, WinInput variables of the ESN neural network model; wbackConnecting a weight matrix for an output state variable of the ESN neural network model; x (n) represents the internal state of the ESN neural network model; woutA connection weight matrix among a nuclear reserve pool of the ESN neural network model, the input of the neural network and the output of the neural network;
is the output deviation of the ESN neural network model or may represent noise; f ═ f [ f1,f2,…,fn]N activation functions for neurons within the "pool of stores"; f. ofiIs a hyperbolic tangent function; f. ofoutIs the epsilon output functions of the ESN neural network model.
The above embodiments are merely illustrative of the technical concepts and features of the present invention, and the purpose of the embodiments is to enable those skilled in the art to understand the contents of the present invention and implement the present invention, and not to limit the protection scope of the present invention. All equivalent changes and modifications made according to the spirit of the present invention should be covered within the protection scope of the present invention.
Claims (10)
1. An air injection control system is characterized by comprising an object conveying device (4), an air injection device (2) and a scanning device (1) which are arranged on a rack (10); the object conveying device is used for conveying an object to be scanned into a scanning area, the air injection device is used for blowing the object to be scanned flat, and the scanning device is used for scanning the object to be scanned;
the gas injection device (2) comprises a single chip microcomputer controller, a gas injection box (22) and a gas injection shell (21) communicated with the bottom of the gas injection box (22) through a gas injection pipe (23) provided with an electromagnetic valve, wherein a gas flow sensor is arranged on the gas injection pipe (23); the single chip microcomputer controller controls the gas in the gas spraying box (22) to be sprayed out of the gas spraying hole of the gas spraying (21) shell through the gas spraying pipe (23) by adjusting the opening degree of the electromagnetic valve, and the single chip microcomputer controller ensures that the gas flow sensor detects that the gas flow of the gas spraying pipe (23) reaches a set value by adjusting the opening degree of the electromagnetic valve;
the single chip microcomputer controller comprises an STM32 single chip microcomputer and an intelligent controller in an STM32 single chip microcomputer, wherein the intelligent controller comprises 2 NARX neural network models, 3 LSTM neural network models, a dynamic recursive wavelet neural network controller, an Empirical Mode Decomposition (EMD) model and a plurality of ESN neural network models; the gas flow regulation platform is formed by the STM32 single chip microcomputer, the jet box (22), the electromagnetic valve, the jet pipe (23), the jet shell (21) and the gas flow sensor, and the intelligent controller in the STM32 single chip microcomputer realizes intelligent regulation of gas flow.
2. The jet control system of claim 1, wherein in the intelligent controller, 2 NARX neural network models are NARX neural network model 1 that controls gas flow and NARX neural network model 2 that predicts gas flow values, and 3 LSTM neural network models are LSTM neural network model 1, LSTM neural network model 2, and LSTM neural network model 3, respectively.
3. The jet control system of claim 2, wherein the sum of the LSTM neural network model 1 output value and the NARX neural network model 1 output value is used as the input of the LSTM neural network model 3, the output of the LSTM neural network model 3 is used as the input of the adjusting solenoid valve opening control value and the NARX neural network model 1, the gas flow sensor for detecting the flow of the jet pipe outputs the gas flow value for a period of time as the input of the LSTM neural network model 2, the LSTM neural network model 2 output is used as the input of the EMD model and the input of the NARX neural network model 1, the EMD model outputs the low frequency trend part and the high frequency fluctuation parts of the LSTM neural network model 2 as the inputs of the ESN neural network models, the output values of the plurality of ESN neural network models are used as the input of the NARX neural network model 2, the output value of the NARX neural network model 2 is used as the gas flow feedback value, and the error change rate of the gas flow set value of the gas nozzle and the output value of the NARX neural network model 2 are used as the input of the LSTM neural network model 1.
4. The jet control system of claim 3, wherein the LSTM neural network model 3 enables prediction of the sum of the dynamic recursive wavelet neural network controller output and the NARX neural network model 1 output and a further predictive control of the gas flow, and the NARX neural network model 2 enables fusion of the multiple ESN neural network model outputs and a further accurate prediction of the gas flow.
5. The gas injection control system according to claim 3, wherein in the gas flow regulation, the output of the LSTM neural network model 3 of the intelligent controller in the STM32 singlechip is used as the input of the electromagnetic valve opening control value as the control signal of the electromagnetic valve, the LSTM neural network model 3 outputs the control value of the opening of the electromagnetic valve to ensure that the gas flow entering the gas injection pipe from the gas injection box through the electromagnetic valve reaches the gas flow set value of the gas injection pipe, the gas flow sensor detects the gas flow value of the gas injection pipe as the input of the LSTM neural network model 2, and the gas flowing out of the gas injection pipe is injected from the gas injection shell.
6. The air blast control system according to claim 1, wherein in said object transfer device (4), a conveying guide rail (41) is horizontally fixed on said frame (10), a transparent placing plate (42) is movably connected with said conveying guide rail (41), and a placing plate driving mechanism provided on said frame (10) is used for driving said transparent placing plate (42) to horizontally move into or out of a scanning area along said conveying guide rail (41).
7. The jet control system according to claim 2, wherein in the placing plate driving mechanism, a first motor (45) is fixed on the frame (10), a rotating shaft of a gear (44) is fixed on an output shaft of the first motor (45), one side of a rack (43) is fixed with the transparent placing plate (42), and the other side is meshed with the gear (44), (44).
8. The gas injection control system according to claim 1, wherein in the scanning device (1), an upper scanning probe (11) and a lower scanning probe (12) are respectively mounted on the frame (10) above and below the scanning area and respectively located above and below the transparent placement plate (42).
9. The air injection control system according to claim 1, characterized in that the air injection device (2) is provided with a plurality of air injection holes on the bottom surface and the side surface of the air injection shell (21); and a plurality of air injection holes on the side surface of the air injection shell (21) form an included angle of 45 degrees with the horizontal plane.
10. The gas injection control system according to any one of claims 1 to 7, further comprising a scanning rendering lamp (13) mounted on the gantry (10), the scanning rendering lamp (13) being located between the overhead scanning probe (11) and the transparent placement plate (42).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110023054.XA CN112799328B (en) | 2021-01-08 | 2021-01-08 | Air injection control system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110023054.XA CN112799328B (en) | 2021-01-08 | 2021-01-08 | Air injection control system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112799328A true CN112799328A (en) | 2021-05-14 |
| CN112799328B CN112799328B (en) | 2021-12-17 |
Family
ID=75809175
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110023054.XA Active CN112799328B (en) | 2021-01-08 | 2021-01-08 | Air injection control system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112799328B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114839869A (en) * | 2022-04-11 | 2022-08-02 | 哈尔滨工程大学 | ADRC gas injection amount closed-loop control method based on high-pressure natural gas circulating gas injection amount real-time detection |
| CN116755477A (en) * | 2023-08-16 | 2023-09-15 | 西安倍得新数据科技有限公司 | Automatic flow control and regulation method and system for fluid channel |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005117415A (en) * | 2003-10-08 | 2005-04-28 | Oki Electric Ind Co Ltd | Scanner device |
| CN108521533A (en) * | 2018-04-16 | 2018-09-11 | 安徽宝葫芦信息科技集团股份有限公司 | A kind of books archives scan instrument |
| CN207978025U (en) * | 2018-03-06 | 2018-10-16 | 绍兴市中凯电脑有限公司 | A kind of intelligent scanning device |
| JP2018182379A (en) * | 2017-04-03 | 2018-11-15 | 京セラドキュメントソリューションズ株式会社 | Image reading device and image forming apparatus including the image reading device |
| CN209170484U (en) * | 2018-11-28 | 2019-07-26 | 福州锐创工业设计有限公司 | Full-automatic books apparatus for scanning archives |
| CN111885279A (en) * | 2020-07-29 | 2020-11-03 | 福建省赛维工程勘测有限公司 | Archive scanner control device and control method thereof |
-
2021
- 2021-01-08 CN CN202110023054.XA patent/CN112799328B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005117415A (en) * | 2003-10-08 | 2005-04-28 | Oki Electric Ind Co Ltd | Scanner device |
| JP2018182379A (en) * | 2017-04-03 | 2018-11-15 | 京セラドキュメントソリューションズ株式会社 | Image reading device and image forming apparatus including the image reading device |
| CN207978025U (en) * | 2018-03-06 | 2018-10-16 | 绍兴市中凯电脑有限公司 | A kind of intelligent scanning device |
| CN108521533A (en) * | 2018-04-16 | 2018-09-11 | 安徽宝葫芦信息科技集团股份有限公司 | A kind of books archives scan instrument |
| CN209170484U (en) * | 2018-11-28 | 2019-07-26 | 福州锐创工业设计有限公司 | Full-automatic books apparatus for scanning archives |
| CN111885279A (en) * | 2020-07-29 | 2020-11-03 | 福建省赛维工程勘测有限公司 | Archive scanner control device and control method thereof |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114839869A (en) * | 2022-04-11 | 2022-08-02 | 哈尔滨工程大学 | ADRC gas injection amount closed-loop control method based on high-pressure natural gas circulating gas injection amount real-time detection |
| CN114839869B (en) * | 2022-04-11 | 2022-12-27 | 哈尔滨工程大学 | ADRC gas injection amount closed-loop control method based on high-pressure natural gas circulating gas injection amount real-time detection |
| CN116755477A (en) * | 2023-08-16 | 2023-09-15 | 西安倍得新数据科技有限公司 | Automatic flow control and regulation method and system for fluid channel |
| CN116755477B (en) * | 2023-08-16 | 2023-11-03 | 西安倍得新数据科技有限公司 | Automatic flow control and regulation method and system for fluid channel |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112799328B (en) | 2021-12-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112799328B (en) | Air injection control system | |
| CN111899510B (en) | Intelligent traffic system flow short-term prediction method and system based on divergent convolution and GAT | |
| CN109685252B (en) | Building energy consumption prediction method based on cyclic neural network and multi-task learning model | |
| Kim et al. | HyFIS: adaptive neuro-fuzzy inference systems and their application to nonlinear dynamical systems | |
| CN112488415A (en) | Power load prediction method based on empirical mode decomposition and long-and-short-term memory network | |
| CN108280551A (en) | A kind of photovoltaic power generation power prediction method using shot and long term memory network | |
| Waterhouse | Classification and regression using mixtures of experts | |
| Yang et al. | Mini-batch algorithms with online step size | |
| CN103971163A (en) | Adaptive learning rate wavelet neural network control method based on normalization lowest mean square adaptive filtering | |
| Theocharis | A high-order recurrent neuro-fuzzy system with internal dynamics: Application to the adaptive noise cancellation | |
| CN106920014A (en) | A kind of short-term load forecasting method and device | |
| Yang et al. | Prediction model of energy market by long short term memory with random system and complexity evaluation | |
| CN111709550A (en) | Junctor planning method and system based on deep learning | |
| CN113281465A (en) | Livestock and poultry house breeding environment harmful gas detection system | |
| CN116562908A (en) | Electric price prediction method based on double-layer VMD decomposition and SSA-LSTM | |
| Shene et al. | Neural networks for the prediction of the state of Zymomonas mobilis CP4 batch fermentations | |
| Lucas et al. | Methodology based on spiking neural networks for univariate time-series forecasting | |
| CN112859599B (en) | Object transfer cam control system | |
| Cohen | Training synaptic delays in a recurrent neural network | |
| Su et al. | Self-organizing pipelined recurrent wavelet neural network for time series prediction | |
| CN115775156A (en) | Mine economy evaluation raw material price prediction method based on hybrid learning model | |
| Normandin et al. | Optimizing control of a continuous stirred tank fermenter using a neural network | |
| CN112866500A (en) | Folding platform control system | |
| CN112839142A (en) | Shading control system | |
| Linko et al. | Neural network programming in bioprocess estimation and control |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20221207 Address after: Room 2506, Floor 25, Building 1, No. 188, Section 2, Renmin North Road, Jinniu District, Chengdu, Sichuan 610,000 Patentee after: Sichuan Chaoyihong Technology Co.,Ltd. Address before: 223005 Jiangsu Huaian economic and Technological Development Zone, 1 East Road. Patentee before: HUAIYIN INSTITUTE OF TECHNOLOGY |