CN118032711B - Signal control method and system of laser gas sensor - Google Patents
Signal control method and system of laser gas sensor Download PDFInfo
- Publication number
- CN118032711B CN118032711B CN202410440868.7A CN202410440868A CN118032711B CN 118032711 B CN118032711 B CN 118032711B CN 202410440868 A CN202410440868 A CN 202410440868A CN 118032711 B CN118032711 B CN 118032711B
- Authority
- CN
- China
- Prior art keywords
- signal
- signals
- interference
- compensation
- network model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000001228 spectrum Methods 0.000 claims abstract description 66
- 238000006243 chemical reaction Methods 0.000 claims abstract description 62
- 238000001514 detection method Methods 0.000 claims abstract description 62
- 238000012545 processing Methods 0.000 claims abstract description 45
- 238000007781 pre-processing Methods 0.000 claims abstract description 43
- 238000012549 training Methods 0.000 claims abstract description 38
- 238000013507 mapping Methods 0.000 claims abstract description 32
- 230000008054 signal transmission Effects 0.000 claims abstract description 23
- 238000010494 dissociation reaction Methods 0.000 claims abstract description 14
- 230000005593 dissociations Effects 0.000 claims abstract description 14
- 238000012937 correction Methods 0.000 claims abstract description 8
- 239000011159 matrix material Substances 0.000 claims description 23
- 230000010354 integration Effects 0.000 claims description 16
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 claims description 16
- 238000010521 absorption reaction Methods 0.000 claims description 11
- 238000010276 construction Methods 0.000 claims description 9
- 230000005540 biological transmission Effects 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 5
- 238000000556 factor analysis Methods 0.000 claims description 5
- 230000007246 mechanism Effects 0.000 claims description 5
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000009826 distribution Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 4
- 230000002452 interceptive effect Effects 0.000 claims description 4
- 238000005457 optimization Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 19
- 238000005259 measurement Methods 0.000 abstract description 8
- 239000007789 gas Substances 0.000 description 84
- 230000000875 corresponding effect Effects 0.000 description 7
- 230000001276 controlling effect Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 230000003993 interaction Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 150000001335 aliphatic alkanes Chemical class 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000010219 correlation analysis Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 230000001915 proofreading effect Effects 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000002079 cooperative effect Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/39—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using tunable lasers
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
Landscapes
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Optics & Photonics (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The application discloses a signal control method and a system of a laser gas sensor, which relate to the technical field of signal processing, wherein the method comprises the following steps: constructing a disturbance factor library by interacting multiple application scenes; traversing a disturbance factor library, matching a compensation algorithm of factor interference, and supervising and training an anti-interference network model; carrying out multiple parallel acquisition of spectrum conversion signals and determining continuous spectrum conversion signals; transmitting the continuous spectrum conversion signals to an anti-interference network model, performing hierarchical anti-interference compensation processing, and determining a plurality of preprocessing signals; performing signal transmission attenuation compensation on the plurality of preprocessed signals; performing signal dissociation and mapping correction based on the plurality of compensation preprocessing signals, and determining homologous signal differences; and integrating the plurality of compensation preprocessing signals based on the signal difference standard, determining a target sensing signal, and extracting signal characteristics to obtain a gas detection result. Thereby achieving the technical effects of improving the detection efficiency and the measurement effect under the complex environment and reducing the detection error.
Description
Technical Field
The invention relates to the technical field of signal processing, in particular to a signal control method and system of a laser gas sensor.
Background
A laser gas sensor is a technique for detecting a specific gas concentration in air. Such sensors utilize the interaction of a laser beam with a gas to determine the concentration of the gas by measuring the absorption, scattering or interference of the beam. The sensor may have a certain response to various gases, is sensitive to the change of the ambient temperature and humidity, and has the technical problems of large detection error, low detection efficiency and poor measurement effect in a complex environment.
Disclosure of Invention
The application aims to provide a signal control method and system of a laser gas sensor. The method is used for solving the technical problems of large detection error, low detection efficiency and poor measurement effect in the complex environment in the prior art.
In view of the technical problems, the application provides a signal control method and a signal control system for a laser gas sensor.
In a first aspect, the present application provides a signal control method of a laser gas sensor, where the method includes:
constructing a disturbance factor library, wherein the disturbance factor library comprises disturbance factor matrixes mapped to all application scenes and has timeliness updating;
Traversing the disturbance factor library, matching a compensation algorithm of factor disturbance, and supervising and training an anti-disturbance network model, wherein the anti-disturbance network model is provided with a plurality of anti-disturbance processing layers;
Carrying out multiple parallel collection of spectrum conversion signals to determine continuous spectrum conversion signals, wherein the spectrum conversion signals are spectrum absorption signals of methane gas and specific laser wavelength;
Transmitting the continuous spectrum conversion signals to the anti-interference network model, performing hierarchical anti-interference compensation processing, and determining a plurality of preprocessing signals;
performing signal transmission attenuation compensation on the plurality of preprocessing signals to determine a plurality of compensated preprocessing signals;
performing signal dissociation and point-to-point mapping correction on the plurality of compensation preprocessing signals to determine homologous signal differences;
and integrating the plurality of compensation preprocessing signals based on a signal difference standard, determining a target sensing signal, and extracting signal characteristics to obtain a gas detection result.
In a second aspect, the present application also provides a signal control system of a laser gas sensor, wherein the system comprises:
the disturbance factor analysis module is used for interacting multiple application scenes and constructing a disturbance factor library, wherein the disturbance factor library comprises disturbance factor matrixes mapped to the application scenes and has timeliness updating;
The supervision construction module is used for traversing the disturbance factor library, matching a compensation algorithm of factor interference and supervising and training an anti-interference network model, wherein the anti-interference network model is provided with a plurality of anti-interference processing layers;
the signal acquisition module is used for carrying out multiple parallel acquisition of spectrum conversion signals and determining continuous spectrum conversion signals, wherein the spectrum conversion signals are spectrum absorption signals of methane gas and specific laser wavelengths;
The preprocessing module is used for transmitting the continuous spectrum conversion signals to the anti-interference network model, performing hierarchical anti-interference compensation processing and determining a plurality of preprocessing signals;
The transmission attenuation compensation module is used for carrying out signal transmission attenuation compensation on the plurality of preprocessed signals and determining a plurality of compensated preprocessed signals;
The mapping and checking module is used for carrying out signal dissociation and point-to-point mapping and checking on the plurality of compensation preprocessing signals and determining homologous signal differences;
And the integration output module is used for integrating the plurality of compensation preprocessing signals based on a signal difference standard, determining a target sensing signal and extracting signal characteristics to be used as a gas detection result.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
Constructing a disturbance factor library through interaction of multiple application scenes, wherein the disturbance factor library comprises disturbance factor matrixes mapped to the application scenes and has timeliness updating; traversing a disturbance factor library, matching a compensation algorithm of factor disturbance, and supervising and training an anti-interference network model, wherein the anti-interference network model is provided with a plurality of anti-interference processing layers; carrying out multiple parallel collection of spectrum conversion signals to determine continuous spectrum conversion signals, wherein the spectrum conversion signals are spectrum absorption signals of methane gas and specific laser wavelength; transmitting the continuous spectrum conversion signals to an anti-interference network model, performing hierarchical anti-interference compensation processing, and determining a plurality of preprocessing signals; performing signal transmission attenuation compensation on the plurality of preprocessed signals to determine a plurality of compensated preprocessed signals; performing signal dissociation and point-to-point mapping correction on the plurality of compensation preprocessing signals to determine homologous signal differences; and integrating the plurality of compensation preprocessing signals based on the signal difference standard, determining a target sensing signal, and extracting signal characteristics to obtain a gas detection result. Thereby achieving the technical effects of improving the detection efficiency and the measurement effect under the complex environment and reducing the detection error.
The foregoing description is only an overview of the present application, and is intended to more clearly illustrate the technical means of the present application, be implemented according to the content of the specification, and be more apparent in view of the above and other objects, features and advantages of the present application, as follows.
Drawings
Embodiments of the invention and the following brief description are described with reference to the drawings, in which:
FIG. 1 is a flow chart of a method for controlling signals of a laser gas sensor according to the present application;
FIG. 2 is a schematic flow chart of setting up a disturbance factor library according to the interactive multi-element application scene in the signal control method of the laser gas sensor;
fig. 3 is a schematic structural diagram of a signal control system of a laser gas sensor according to the present application.
Reference numerals illustrate: the system comprises a disturbance factor analysis module 11, a supervision construction module 12, a signal acquisition module 13, a preprocessing module 14, a transmission attenuation compensation module 15, a mapping correction module 16 and an integration output module 17.
Detailed Description
The application provides a signal control method and a signal control system for a laser gas sensor, which solve the technical problems of large detection error, low detection efficiency and poor measurement effect in a complex environment in the prior art.
In order to solve the above problems, the technical embodiment adopts the following overall concept:
Firstly, a disturbance factor library is established in a plurality of application scenes, wherein the disturbance factor library comprises disturbance factor matrixes mapped to each application scene, and the factor matrixes have timeliness and can be updated from time to time. And through traversing the disturbance factor library, applying compensation algorithm matching of factor disturbance to supervise and train an anti-disturbance network model, wherein the model comprises a plurality of anti-disturbance processing layers. Multiple parallel acquisitions of the spectrally transformed signal are performed to determine a continuous spectrally transformed signal. The spectrum conversion signal here refers to a spectrum absorption signal of the alkane gas with a specific laser wavelength. And transmitting the continuous spectrum conversion signals to an anti-interference network model, and performing hierarchical anti-interference compensation processing to obtain a plurality of preprocessing signals. And carrying out signal transmission attenuation compensation on the plurality of preprocessed signals to determine a plurality of compensated preprocessed signals. These compensated preprocessed signals are then signal-dissociated and point-to-point mapped against to determine homologous signal differences. And integrating the plurality of compensation preprocessing signals based on the signal difference standard to obtain a target sensing signal, and extracting signal characteristics. Finally, this extracted signal feature is used as a result of gas detection. Thereby achieving the technical effects of improving the detection efficiency and the measurement effect under the complex environment and reducing the detection error.
In order to better understand the above technical solutions, the following detailed description will be given with reference to the accompanying drawings and specific embodiments, and it should be noted that the described embodiments are only some embodiments of the present application, and not all embodiments of the present application, and it should be understood that the present application is not limited by the exemplary embodiments described herein. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to fall within the scope of the application. It should be further noted that, for convenience of description, only some, but not all of the drawings related to the present application are shown.
Examples
As shown in fig. 1, the present application provides a signal control method of a laser gas sensor, the method comprising:
s100: constructing a disturbance factor library, wherein the disturbance factor library comprises disturbance factor matrixes mapped to all application scenes and has timeliness updating;
The multi-element application scene refers to a plurality of application scenes preset by the target laser gas sensor, and the multi-element application scene comprises various factors which can influence the signals of the laser gas sensor. The method comprises the factors of weather conditions, temperature and humidity changes, illumination intensity and the like under different environments.
The disturbance factor matrix is a relation matrix of disturbance factors and application scenes, and provides association relations between a plurality of application scenes and the corresponding disturbance factors. The disturbance factor matrix maps different disturbance factors with each specific application scene so as to better understand the influence of each factor on the laser gas sensor signal in the specific scene.
Optionally, the disturbance factor library and the disturbance factor matrix contained in the disturbance factor library have timeliness updating. Exemplary, the method includes acquiring updated multi-element application scenes based on a preset update period or update frequency, and then adding, modifying, deleting and the like of the application scenes and disturbance factors to the disturbance factor library and the disturbance factor matrix contained in the disturbance factor library. The perturbation factor library and the perturbation factor matrix contained in the perturbation factor library are updated with the passage of time to adapt to the change of the environment.
Further, as shown in fig. 2, an interactive multi-element application scenario is constructed, and step S100 includes:
Reading specification configuration information of the laser gas sensor, and performing homologous search to determine gas detection records based on the multiple application scenes;
Identifying the gas detection record, positioning a scene interference source, and determining a disturbance factor set;
and performing attribution integration and matrix conversion on the disturbance factor set based on the multi-element application scene to generate the disturbance factor library.
The specification configuration information of the laser gas sensor reflects key parameters and performance characteristics of the target laser gas sensor. The data sources for acquiring the specification configuration information comprise: specification, description document, manufacturer, etc. of the target laser gas sensor. Illustratively, the specification configuration information includes specification parameters such as sensitivity, operating frequency, detection range, and the like of the sensor.
Optionally, the gas detection record related to the multi-component application scenario is determined by homologous retrieval based on specification configuration information of the laser gas sensor. And matching the specification configuration information with the real-world gas detection data to identify the historical gas detection conditions in the multiple application scenes in the actual application. The gas detection record comprises gas concentration, gas components, corresponding detection deviation degree, scene environment characteristics and the like.
Optionally, a scene disturbance source causing signal disturbance is located according to the gas detection record. The scene interference sources comprise other gases of non-target monitoring type, temperature and humidity change, illumination intensity, radiation, strong magnetic field and the like. Exemplary, scene disturbance source localization is performed to determine a disturbance factor set, and first, a gas detection record associated with a laser gas sensor signal is identified by correlation analysis. And acquiring correlations among different gases, environmental conditions and other possible disturbance factors and gas detection results in the gas detection records, so as to determine factors possibly influencing the laser gas sensor signals and acquire a disturbance factor set. The disturbance factor in the disturbance factor set may be a single gas component, an environmental characteristic, or the like, or may be a combination of a plurality of factors.
The source of the signal fluctuation of the laser gas sensor can be accurately identified through the scene interference source positioning of the correlation analysis method, and the signal control and compensation can be carried out in a targeted mode.
Optionally, the disturbance factor set is subjected to attribution integration according to the influence intensity and the characteristics. The resulting set of perturbation factors is then converted into a matrix form for better understanding and management of these perturbation factors. The attribution integration is to integrate disturbance factors in different scene states of the same application scene according to the influence intensity and characteristics of the disturbance factors, and comprises the steps of obtaining a union set of the disturbance factors in different scene states of the same application scene, obtaining the influence intensity and characteristics of a plurality of disturbance factors after integration, and realizing fusion division of the disturbance factor sets through attribution integration, so that matrix conversion is facilitated. The disturbance factor matrix is an m×n-dimensional matrix, where M is the number of multiple application scenarios, N is the number of disturbance factors after attribution integration, and elements in the matrix indicate the influence intensity and characteristics of the disturbance factors in the application scenarios. If a certain disturbance factor of the disturbance factor set has no influence on detection in a certain application scene, the element at the corresponding position is null.
Further, after the attribution integration and matrix conversion are performed on the disturbance factor set, the steps further include:
Identifying distribution characteristics and intensity characteristics of interference sources of each scene, and determining the correlation degree of the characteristic grade and the interference grade;
establishing an interference implication relation by combining the correlation, wherein the interference implication relation corresponds to the multiple application scenes one by one and comprises a single implication relation and a collaborative implication relation of each application scene;
And mapping and correlating the interference implication relation with the disturbance factor matrix.
Optionally, the disturbance factor matrix further includes a disturbance implication relationship, where the disturbance implication relationship reflects a correlation characteristic between a distribution characteristic and an intensity characteristic of each scene disturbance source and a disturbance level. To help understand the extent to which each factor affects the signal. The interference implication relationship comprises a single implication relationship and a cooperative implication relationship of each application scene, namely, under different scenes, the independent influence relationship and the cooperative action relationship of each disturbance factor on the laser gas sensor signal.
By mapping and correlating the interference implication relationship with the disturbance factor matrix. A comprehensive relation model is established to better understand the comprehensive influence of disturbance factors on laser gas sensor signals under different scenes.
S200: traversing the disturbance factor library, matching a compensation algorithm of factor disturbance, and supervising and training an anti-disturbance network model, wherein the anti-disturbance network model is provided with a plurality of anti-disturbance processing layers;
Optionally, firstly, traversing the built disturbance factor library, and checking disturbance factor matrixes one by one to acquire detailed factor disturbance information. And then, aiming at each disturbance factor, carrying out corresponding compensation algorithm matching. The compensation algorithm is obtained based on experiments and historical compensation record analysis, and comprises the steps of applying a filter, adjusting weights, specific activation functions and the like in signal processing so as to minimize the influence of factor interference on the laser gas sensor signal. The analysis method includes linear or nonlinear regression analysis. In addition, based on timeliness updating property of the disturbance factor library, corresponding real-time dynamic updating measurement is configured for the compensation algorithm obtained by matching. To maintain the adaptive capacity for the new perturbation factors.
Optionally, a compensation algorithm obtained by matching is integrated into the anti-interference network model. Ensuring that appropriate compensation algorithms are applied at different levels or nodes of the model to minimize the effects of perturbation factors. The anti-interference processing layers of the anti-interference network model are respectively used for processing various disturbance factors. And the number of layers of the anti-interference processing layer is consistent with the number of disturbance factors in the disturbance factor library. The anti-interference network model is ensured to have good generalization, so that the anti-interference network model is suitable for multiple application scenes, and the stability and the accuracy in the multiple application scenes are improved.
Further, the step S200 of supervised training the anti-interference network model includes:
Building a model framework of the anti-interference network model, wherein the model framework is of a multi-layer fully-connected neural network structure, and each network layer is configured with different compensation algorithms;
Reading a gas detection record, and determining a sample spectrum conversion signal, intermediate processing data and a sample pretreatment signal as training samples;
performing supervised training on the model framework based on the training sample to generate the anti-interference network model;
and verifying the anti-interference network model based on the training sample, and performing model convergence degree detection and retraining until convergence conditions are met, so as to obtain the constructed anti-interference network model.
Optionally, the model architecture of the anti-interference network model adopts a multi-layer fully-connected neural network structure, and each network layer is configured with a different compensation algorithm. This helps to increase the resistance of the network to interference.
Optionally, the gas detection record is read from the dataset, and the sample spectral conversion signal, the intermediate processing data and the sample pre-processing signal are used as training samples, which are basic data for training and verifying the anti-interference network model. Parameters of the anti-interference network model are adjusted through training samples, so that the anti-interference network model can be better adapted to various interference conditions, and spectrum conversion signals are correctly input into a plurality of corresponding anti-interference processing layers for processing.
Optionally, training samples are used to validate against the interfering network model. And (3) checking performance, including convergence degree checking, to ensure that the anti-interference network model is gradually stable in the training process. And if the anti-interference network model does not meet the convergence condition in the verification process, performing retraining. The training step is carried out again, and the anti-interference network model parameters are adjusted until the model meets the preset performance and convergence conditions.
Further, after the supervision training of the anti-interference network model, the steps further include:
Reading detection records of the anti-interference network model, and extracting record data of the homologous signal difference overrun;
counting the frequency of the recorded data, and generating an incremental learning instruction if the frequency threshold is met;
And receiving the incremental learning instruction, and combining the recorded data to train and update the anti-interference network model, wherein the training and updating mode comprises original model mechanism optimization and incremental processing layer construction.
Optionally, detection records are read from the anti-interference network model, and the detection records comprise the situation that the homologous signal difference detected in practical application exceeds the limit. And then, extracting specific data of the homologous signal difference overrun from the detection record, wherein the specific data comprise related input signals, output results of the model and the like. The homologous signal difference refers to the difference between the output results of the anti-interference network model in the homologous scene, and comprises the mean square error, standard deviation, absolute difference, average difference and the like of the output results.
Optionally, frequency statistics is performed on the extracted recorded data, and the occurrence frequency of the same type of interference is analyzed. If the frequency of the occurrence of a certain disturbance reaches a set frequency threshold, the influence of the disturbance is larger, and an incremental learning instruction is generated. The incremental learning instructions are used to control incremental learning of a particular type of disturbance to further improve the model's adaptability to such disturbances.
Optionally, the anti-interference network model receives the generated incremental learning instruction, and performs training update on the anti-interference network model by combining the incremental learning instruction and corresponding recorded data. Including the optimization of the original model mechanism and the build of the incremental processing layer. The mechanism of optimizing the prototype model is to improve overall performance, while the incremental processing layer is built to better handle the emerging disturbances. Through the steps, the anti-interference network model can be continuously learned and adapted in practical application, so that generalization of the anti-interference network model is improved, and recognition and resistance to new disturbance are improved.
S300: carrying out multiple parallel collection of spectrum conversion signals to determine continuous spectrum conversion signals, wherein the spectrum conversion signals are spectrum absorption signals of methane gas and specific laser wavelength;
Optionally, the target laser gas sensor is activated to perform multiple parallel acquisitions, that is, multiple identical spectrum conversion signal acquisitions are continuously performed under the same condition. The multiple parallel acquisition comprises a preset acquisition interval and acquisition duration. Signals acquired at shorter acquisition intervals have higher acquisition accuracy. The spectrum conversion signal is a spectrum absorption signal of a specific laser wavelength of the alkane gas and the target laser gas sensor. And in each acquisition process, recording spectrum signal data output by the spectrometer. These data include light intensities at different wavelengths.
Optionally, consistent acquisition parameters, such as sampling frequency, light intensity, temperature, etc., are maintained during parallel acquisition, ensuring that the acquired signals are comparable. In addition, the continuity of the multiple parallel acquisition results is verified, so that the continuously acquired spectrum conversion signals are continuous in spectrum, and no obvious interruption or jump exists.
Further, before performing multiple parallel acquisitions of the spectrum conversion signal, step S300 further includes:
detecting signal conditions of a target application scene, and evaluating and determining weak signal grades;
Based on the weak signal level, performing detection laser parameter regulation and control of the laser gas sensor, and determining compensation laser parameters;
and controlling the laser gas sensor to perform multiple parallel acquisition of spectrum conversion signals based on the compensation laser parameters.
Optionally, before multiple parallel acquisitions are performed, a specific scenario in which the laser gas sensor is to be used for monitoring, such as industrial production environment or indoor air quality monitoring, is first determined. Then, in the target scene, conditions such as signal strength, noise level, and interference sources that may exist in the environment are evaluated. Particular attention is paid to factors that may lead to weak signals. And the occurrence frequency and intensity of weak signals in the target application scene are evaluated according to the detection result, and particularly, the signal strength parameters such as signal-to-noise ratio, signal intensity and the like are related.
Optionally, the effect of laser parameters (e.g., wavelength, power, etc.) on the optical signal intensity and clarity is known. And designing a strategy for adjusting laser parameters according to signal conditions, wherein the strategy is used for adjusting the laser parameters of the laser gas sensor so as to improve the detection performance of weak signals.
Through the steps, the working parameters of the laser gas sensor are optimized in the target application scene so as to adapt to the detection requirement of the weak signal, and an accurate spectrum conversion signal is obtained in the acquisition process. Thereby improving the detection capability and accuracy of the sensor for weak signals.
S400: transmitting the continuous spectrum conversion signals to the anti-interference network model, performing hierarchical anti-interference compensation processing, and determining a plurality of preprocessing signals;
Optionally, first, a signal transmission channel is established, including setting a suitable communication protocol or interface, to ensure that the continuous spectrum conversion signal generated by the laser gas sensor can be transmitted to the anti-interference network model. And then, transmitting the continuous spectrum conversion signal to an input layer of the anti-interference network model through the signal transmission channel. Then, the anti-interference network model takes the continuous spectrum conversion signal as input, and the signals after the anti-interference compensation of a plurality of levels are obtained through the processing of each processing layer and stored as a plurality of preprocessing signals. The anti-interference compensation processing of the continuous spectrum conversion signal is realized. Providing a plurality of accurate, reliable and high-quality preprocessing signals for subsequent processing. The detection performance and the robustness of the laser gas sensor in a complex environment with disturbance factors are improved.
S500: performing signal transmission attenuation compensation on the plurality of preprocessing signals to determine a plurality of compensated preprocessing signals;
Optionally, transmission attenuation compensation of a plurality of pre-processed signals is performed, firstly, a signal transmission method is obtained based on a signal transmission path, and a signal transmission medium adopted by the signal transmission method is analyzed to generate attenuation effect and attenuation characteristics on signal transmission. Wherein the transmission medium comprises air, optical fibers, etc. Next, a signal transmission attenuation model is established based on the signal attenuation effect and the attenuation characteristics, the signal transmission attenuation model being used to compensate for the attenuation effect of the signal, including a mathematical model, or a model based on machine learning. Then, based on a plurality of signal transmission distances of a plurality of preprocessed signals, a proper compensation means is selected in combination with the signal transmission attenuation model to implement signal transmission attenuation compensation, wherein the compensation means comprises digital signal enhancement, signal power increase, signal amplifier use and the like.
Through the steps, attenuation effects in signal transmission can be dealt with, and stability and reliability of signals are improved, so that more accurate compensation pretreatment signals are obtained, and the method is helpful for improving performance of a laser gas sensor system.
S600: performing signal dissociation and point-to-point mapping correction on the plurality of compensation preprocessing signals to determine homologous signal differences;
Alternatively, signal dissociation refers to the decomposition of a signal into individual signals. Signal dissociation is implemented based on signal dissociation techniques, such as blind source separation (Blind Source Separation, BSS) algorithms, to extract individual components belonging to different gas components from the mixed signal. Point-to-point mapping calibration refers to performing differential analysis on homologous signals acquired through multiple parallel acquisitions to acquire differential features between multiple signals, and exemplary differential features of the homologous signals include: amplitude, frequency, phase, etc. Illustratively, the difference in homologous signals may be a mean square error, a standard deviation, a polar difference, etc. of the difference characteristics between the plurality of homologous signals.
S700: and integrating the plurality of compensation preprocessing signals based on a signal difference standard, determining a target sensing signal, and extracting signal characteristics to obtain a gas detection result.
Further, the integrating the plurality of compensation pre-processing signals based on the signal difference standard, step S700 includes:
Identifying the homologous signal difference, extracting a mapping signal group meeting the signal difference standard, and selecting any signal point in the group as a first local signal;
Extracting a mapping signal group which does not meet the signal difference standard, and carrying out intra-group signal point average value processing to serve as a second local signal;
And splicing the first local signal and the second local signal to serve as the target sensing signal.
Alternatively, the signal difference criterion refers to a threshold value of signal difference for controlling the quality of the plurality of compensated preprocessed signals. If the homologous signal difference of the compensation pre-processing signal meets the signal difference standard, a mapping signal group is extracted from the compensation pre-processing signal, and a signal point is selected from the mapping signal group and used as a first local signal.
Optionally, for the mapping signal group which does not meet the signal difference standard, performing intra-group signal point average processing. And the method is used for averaging random disturbance and obtaining a second local signal. And then, splicing the first local signal and the second local signal to form a final target sensing signal. The integration method can keep important information meeting the signal difference standard, reduce the influence of random disturbance of different source signals, further obtain sensing signals with robustness and anti-interference capability, and improve the performance of the laser gas sensor system.
In summary, the signal control method of the laser gas sensor provided by the invention has the following technical effects:
Constructing a disturbance factor library through interaction of multiple application scenes, wherein the disturbance factor library comprises disturbance factor matrixes mapped to the application scenes and has timeliness updating; traversing a disturbance factor library, matching a compensation algorithm of factor disturbance, and supervising and training an anti-interference network model, wherein the anti-interference network model is provided with a plurality of anti-interference processing layers; carrying out multiple parallel collection of spectrum conversion signals to determine continuous spectrum conversion signals, wherein the spectrum conversion signals are spectrum absorption signals of methane gas and specific laser wavelength; transmitting the continuous spectrum conversion signals to an anti-interference network model, performing hierarchical anti-interference compensation processing, and determining a plurality of preprocessing signals; performing signal transmission attenuation compensation on the plurality of preprocessed signals to determine a plurality of compensated preprocessed signals; performing signal dissociation and point-to-point mapping correction on the plurality of compensation preprocessing signals to determine homologous signal differences; and integrating the plurality of compensation preprocessing signals based on the signal difference standard, determining a target sensing signal, and extracting signal characteristics to obtain a gas detection result. Thereby achieving the technical effects of improving the detection efficiency and the measurement effect under the complex environment and reducing the detection error.
Examples
Based on the same conception as the signal control method of the laser gas sensor in the embodiment, as shown in fig. 3, the application also provides a signal control system of the laser gas sensor, which comprises:
the disturbance factor analysis module 11 is used for interacting multiple application scenes and building a disturbance factor library, wherein the disturbance factor library comprises disturbance factor matrixes mapped to the application scenes and has timeliness updating;
The supervision construction module 12 is used for traversing the disturbance factor library, matching a compensation algorithm of factor interference, and supervising and training an anti-interference network model, wherein a plurality of anti-interference processing layers exist in the anti-interference network model;
The signal acquisition module 13 is used for carrying out multiple parallel acquisition of spectrum conversion signals and determining continuous spectrum conversion signals, wherein the spectrum conversion signals are spectrum absorption signals of methane gas and specific laser wavelengths;
The preprocessing module 14 is configured to transmit the continuous spectrum conversion signal to the anti-interference network model, perform hierarchical anti-interference compensation processing, and determine a plurality of preprocessed signals;
A transmission attenuation compensation module 15, configured to perform signal transmission attenuation compensation on the plurality of preprocessed signals, and determine a plurality of compensated preprocessed signals;
a mapping and proofreading module 16, configured to perform signal dissociation and point-to-point mapping and proofreading on the plurality of compensated preprocessed signals, and determine a homologous signal difference;
And the integration output module 17 is used for integrating the plurality of compensation preprocessing signals based on a signal difference standard, determining a target sensing signal and extracting signal characteristics as a gas detection result.
Further, the disturbance factor analysis module 11 further includes:
the specification configuration unit is used for reading the specification configuration information of the laser gas sensor and carrying out homologous search to determine gas detection records based on the multiple application scenes;
the disturbance source positioning unit is used for identifying the gas detection record, positioning a scene disturbance source and determining a disturbance factor set;
and the integration and matrix conversion unit is used for carrying out attribution integration and matrix conversion on the disturbance factor set based on the multi-element application scene to generate the disturbance factor library.
Further, the integrating and matrix converting further includes:
the correlation unit is used for identifying the distribution characteristics and the intensity characteristics of the interference sources of each scene and determining the correlation between the characteristic level and the interference level;
The interference implication relation unit is used for establishing an interference implication relation by combining the correlation, wherein the interference implication relation corresponds to the multiple application scenes one by one and comprises a single implication relation and a collaborative implication relation of each application scene;
And the mapping association unit is used for mapping and associating the interference implicit relation with the disturbance factor matrix.
Further, the supervision construction module 12 further includes:
The architecture unit is used for building a model architecture of the anti-interference network model, wherein the model architecture is of a multi-layer fully-connected neural network structure, and each network layer is configured with different compensation algorithms;
The sample determining unit is used for reading the gas detection record and determining a sample spectrum conversion signal, intermediate processing data and a sample preprocessing signal to be used as a training sample;
The sample training unit is used for performing supervision training on the model framework based on the training sample to generate the anti-interference network model;
And the checking and retraining unit is used for verifying the anti-interference network model based on the training sample, and performing model convergence checking and retraining until convergence conditions are met, so as to obtain the constructed anti-interference network model.
Further, the supervision construction module 12 further includes:
The detection recording unit is used for reading the detection record of the anti-interference network model and extracting the record data of the homologous signal difference overrun;
the frequency statistics unit is used for carrying out frequency statistics on the recorded data, and if the frequency threshold is met, an incremental learning instruction is generated;
And the training updating unit is used for receiving the incremental learning instruction, combining the recorded data and carrying out training updating on the anti-interference network model, wherein the training updating mode comprises original model mechanism optimization and incremental processing layer construction.
Further, the signal acquisition module 13 further includes:
the signal grade evaluation unit is used for detecting signal conditions of the target application scene, evaluating and determining weak signal grade;
the laser parameter compensation unit is used for adjusting and controlling the detection laser parameters of the laser gas sensor based on the weak signal level and determining compensation laser parameters;
And the parallel acquisition unit is used for controlling the laser gas sensor to perform multiple parallel acquisitions of the spectrum conversion signals based on the compensation laser parameters.
Further, the integrated output module 17 further includes:
The first signal extraction unit is used for identifying the homologous signal difference, extracting a mapping signal group meeting the signal difference standard, and selecting any signal point in the group as a first local signal;
The second signal extraction unit is used for extracting a mapping signal group which does not meet the signal difference standard, and carrying out intra-group signal point average value processing to serve as a second local signal;
And the signal splicing unit is used for splicing the first local signal and the second local signal to be used as the target sensing signal.
It should be understood that the embodiments mentioned in this specification focus on the differences from other embodiments, and the specific embodiment in the first embodiment is equally applicable to the signal control system of a laser gas sensor described in the second embodiment, and is not further expanded herein for brevity of description.
It is to be understood that both the foregoing description and the embodiments of the present application enable one skilled in the art to utilize the present application. While the application is not limited to the above-mentioned embodiments, obvious modifications, combinations and substitutions of the above-mentioned embodiments are also within the scope of the application.
Claims (7)
1. A method of controlling a signal of a laser gas sensor, the method comprising:
constructing a disturbance factor library, wherein the disturbance factor library comprises disturbance factor matrixes mapped to all application scenes and has timeliness updating;
Traversing the disturbance factor library, matching a compensation algorithm of factor disturbance, and supervising and training an anti-disturbance network model, wherein the anti-disturbance network model is provided with a plurality of anti-disturbance processing layers;
Carrying out multiple parallel collection of spectrum conversion signals to determine continuous spectrum conversion signals, wherein the spectrum conversion signals are spectrum absorption signals of methane gas and specific laser wavelength;
Transmitting the continuous spectrum conversion signals to the anti-interference network model, performing hierarchical anti-interference compensation processing, and determining a plurality of preprocessing signals;
performing signal transmission attenuation compensation on the plurality of preprocessing signals to determine a plurality of compensated preprocessing signals;
performing signal dissociation and point-to-point mapping correction on the plurality of compensation preprocessing signals to determine homologous signal differences, wherein the signal dissociation refers to decomposing the signals into individual signals, the point-to-point mapping correction refers to performing differential analysis on homologous signals acquired through multiple parallel acquisition to acquire differential features among the plurality of homologous signals, and the homologous signal differences are mean square errors, standard deviations and extremely differences of the differential features among the plurality of homologous signals;
integrating the plurality of compensation preprocessing signals based on a signal difference standard, determining a target sensing signal, and extracting signal characteristics to serve as a gas detection result;
Wherein the integrating the plurality of compensated preprocessed signals based on the signal difference criteria comprises:
Identifying the homologous signal difference, extracting a mapping signal group meeting the signal difference standard, and selecting any signal point in the group as a first local signal;
Extracting a mapping signal group which does not meet the signal difference standard, and carrying out intra-group signal point average value processing to serve as a second local signal;
And splicing the first local signal and the second local signal to serve as the target sensing signal.
2. The method for controlling signals of a laser gas sensor according to claim 1, wherein the constructing a disturbance factor library from the interactive multi-element application scenario comprises:
Reading specification configuration information of the laser gas sensor, and performing homologous search to determine gas detection records based on the multiple application scenes;
Identifying the gas detection record, positioning a scene interference source, and determining a disturbance factor set;
and performing attribution integration and matrix conversion on the disturbance factor set based on the multi-element application scene to generate the disturbance factor library.
3. The method for controlling signals of a laser gas sensor according to claim 2, wherein after performing home integration and matrix conversion on the disturbance factor set, the method comprises:
Identifying distribution characteristics and intensity characteristics of interference sources of each scene, and determining the correlation degree of the characteristic grade and the interference grade;
establishing an interference implication relation by combining the correlation, wherein the interference implication relation corresponds to the multiple application scenes one by one and comprises a single implication relation and a collaborative implication relation of each application scene;
And mapping and correlating the interference implication relation with the disturbance factor matrix.
4. The method of claim 1, wherein the supervising training the antijamming network model comprises:
Building a model framework of the anti-interference network model, wherein the model framework is of a multi-layer fully-connected neural network structure, and each network layer is configured with different compensation algorithms;
Reading a gas detection record, and determining a sample spectrum conversion signal, intermediate processing data and a sample pretreatment signal as training samples;
performing supervised training on the model framework based on the training sample to generate the anti-interference network model;
and verifying the anti-interference network model based on the training sample, and performing model convergence degree detection and retraining until convergence conditions are met, so as to obtain the constructed anti-interference network model.
5. The method for controlling a signal of a laser gas sensor according to claim 4, wherein after said supervising training of said anti-interference network model, comprising:
Reading detection records of the anti-interference network model, and extracting record data of the homologous signal difference overrun;
counting the frequency of the recorded data, and generating an incremental learning instruction if the frequency threshold is met;
And receiving the incremental learning instruction, and combining the recorded data to train and update the anti-interference network model, wherein the training and updating mode comprises original model mechanism optimization and incremental processing layer construction.
6. The method for controlling a laser gas sensor signal according to claim 1, wherein before the performing the plurality of parallel acquisitions of the spectrum conversion signal, the method comprises:
detecting signal conditions of a target application scene, and evaluating and determining weak signal grades;
Based on the weak signal level, performing detection laser parameter regulation and control of the laser gas sensor, and determining compensation laser parameters;
and controlling the laser gas sensor to perform multiple parallel acquisition of spectrum conversion signals based on the compensation laser parameters.
7. A signal control system for a laser gas sensor, the system comprising:
the disturbance factor analysis module is used for interacting multiple application scenes and constructing a disturbance factor library, wherein the disturbance factor library comprises disturbance factor matrixes mapped to the application scenes and has timeliness updating;
The supervision construction module is used for traversing the disturbance factor library, matching a compensation algorithm of factor interference and supervising and training an anti-interference network model, wherein the anti-interference network model is provided with a plurality of anti-interference processing layers;
the signal acquisition module is used for carrying out multiple parallel acquisition of spectrum conversion signals and determining continuous spectrum conversion signals, wherein the spectrum conversion signals are spectrum absorption signals of methane gas and specific laser wavelengths;
The preprocessing module is used for transmitting the continuous spectrum conversion signals to the anti-interference network model, performing hierarchical anti-interference compensation processing and determining a plurality of preprocessing signals;
The transmission attenuation compensation module is used for carrying out signal transmission attenuation compensation on the plurality of preprocessed signals and determining a plurality of compensated preprocessed signals;
The mapping and calibrating module is used for carrying out signal dissociation and point-to-point mapping and calibrating on the plurality of compensation pretreatment signals to determine homologous signal differences, wherein the signal dissociation refers to decomposing the signals into individual signals, the point-to-point mapping and calibrating refers to carrying out differential analysis on the homologous signals acquired through multiple parallel acquisition to acquire differential characteristics among the plurality of homologous signals, and the homologous signal differences are mean square errors, standard deviations and extremely differences of the differential characteristics among the plurality of homologous signals;
The integrated output module is used for integrating the plurality of compensation preprocessing signals based on a signal difference standard, determining a target sensing signal and extracting signal characteristics to be used as a gas detection result;
Wherein, the integration output module further includes:
the first signal extraction unit is used for identifying the homologous signal difference, extracting a mapping signal group meeting the signal difference standard, and selecting any signal point in the group as a first local signal;
The second signal extraction unit is used for extracting a mapping signal group which does not meet the signal difference standard, and carrying out intra-group signal point average value processing to serve as a second local signal;
And the signal splicing unit is used for splicing the first local signal and the second local signal to be used as the target sensing signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410440868.7A CN118032711B (en) | 2024-04-12 | 2024-04-12 | Signal control method and system of laser gas sensor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410440868.7A CN118032711B (en) | 2024-04-12 | 2024-04-12 | Signal control method and system of laser gas sensor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118032711A CN118032711A (en) | 2024-05-14 |
| CN118032711B true CN118032711B (en) | 2024-07-02 |
Family
ID=90986204
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410440868.7A Active CN118032711B (en) | 2024-04-12 | 2024-04-12 | Signal control method and system of laser gas sensor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118032711B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118209491B (en) * | 2024-05-22 | 2024-08-06 | 华夏天信传感科技(大连)有限公司 | Gas concentration detection and evaluation method and system for optical gas chamber |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105987881A (en) * | 2015-03-05 | 2016-10-05 | 天津先阳科技发展有限公司 | Spectral-data interference suppression method, modeling method, forecasting method and processing device |
| CN107064021A (en) * | 2017-04-14 | 2017-08-18 | 华中农业大学 | A kind of online lossless detection method of blood cake egg changed based on spectrum amplitude area |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2002365125A1 (en) * | 2001-09-24 | 2003-07-09 | Kwong W. Au | Small spectral signal detection system |
| US20070229834A1 (en) * | 2004-10-22 | 2007-10-04 | Patel C Kumar N | System and method for high sensitivity optical detection of gases |
| JP5131211B2 (en) * | 2009-01-20 | 2013-01-30 | 株式会社島津製作所 | Optical emission spectrometer |
| DE102019208265A1 (en) * | 2019-06-06 | 2020-12-10 | Robert Bosch Gmbh | Adaptive cleaning of measurement data for different types of disturbances |
| CN112543411B (en) * | 2019-09-20 | 2023-07-11 | 中兴通讯股份有限公司 | Interference positioning method, device and system of wireless communication system |
| EP3940342B1 (en) * | 2020-07-13 | 2022-08-31 | Siemens Aktiengesellschaft | Measuring device |
| CN114878639A (en) * | 2022-05-17 | 2022-08-09 | 重庆文理学院 | Automobile exhaust concentration detection system based on neural network |
| CN117376717A (en) * | 2023-10-17 | 2024-01-09 | 智慧互通科技股份有限公司 | Camera disturbance correction method and system |
| CN117687533B (en) * | 2024-02-04 | 2024-04-26 | 广东省旭晟半导体股份有限公司 | Anti-interference data processing method of infrared touch control transmitting and receiving device |
-
2024
- 2024-04-12 CN CN202410440868.7A patent/CN118032711B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105987881A (en) * | 2015-03-05 | 2016-10-05 | 天津先阳科技发展有限公司 | Spectral-data interference suppression method, modeling method, forecasting method and processing device |
| CN107064021A (en) * | 2017-04-14 | 2017-08-18 | 华中农业大学 | A kind of online lossless detection method of blood cake egg changed based on spectrum amplitude area |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118032711A (en) | 2024-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN118032711B (en) | Signal control method and system of laser gas sensor | |
| CN109165670B (en) | TS-RBF fuzzy neural network robust fusion algorithm applied to infrared flame identification | |
| CN114994261B (en) | Remote calibration method for ambient air monitoring sensor | |
| KR20210014027A (en) | Apparatus and method for environmental data interpolation of smart farm | |
| CN118468608B (en) | Ocean ecosystem simulation and prediction method based on data assimilation technology | |
| CN108593557B (en) | Remote measurement error compensation method based on TE-ANN-AWF (transverse electric field analysis) -based mobile pollution source | |
| You et al. | Hidden Markov model based data-driven calibration of non-dispersive infrared gas sensor | |
| CN118297228A (en) | Optimization method and system for power transmission and distribution wire monitoring system | |
| Ingole et al. | Investigation of Different Regression Models For The Predictive Maintenance of Aircraft's Engine | |
| CN115266583A (en) | Ambient light filtering method, ambient light filtering system, computer device and computer readable storage medium | |
| CN118209491B (en) | Gas concentration detection and evaluation method and system for optical gas chamber | |
| CN118566173B (en) | Noise suppression effect inspection method and device | |
| CN118501721B (en) | Battery detection method based on multidimensional data analysis | |
| CN117527149B (en) | Agricultural Internet of things data transmission method | |
| CN118520313B (en) | High-speed train bogie fault identification and model training method and device | |
| CN113589311B (en) | Infrared differential detection method for dim and weak targets | |
| CN118030189B (en) | Method and system for monitoring natural ignition beam tube of coal mine | |
| CN118603156B (en) | Fiber bragg grating abrasion measurement method based on combination of multiple algorithms | |
| CN118332482B (en) | Soil data acquisition method and system based on noise cleaning | |
| CN115825344B (en) | Intelligent waste gas monitoring method and system for chemical enterprises | |
| CN118629542A (en) | Method for predicting water content of few-sample forest combustible based on transfer learning | |
| CN116209000B (en) | Carrier communication quality detection and analysis method and system for dual-mode module | |
| CN118090650B (en) | Reconstruction system and method for alkane gas background spectrum baseline | |
| CN118569445A (en) | Intelligent dynamic calibration multidimensional air quality monitoring data optimization system | |
| CN118449800A (en) | Intelligent voice recognition control system and method based on Internet of things |
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 |