CN112541569A - Sensor online training system and method based on machine learning - Google Patents
Sensor online training system and method based on machine learning Download PDFInfo
- Publication number
- CN112541569A CN112541569A CN202011328753.7A CN202011328753A CN112541569A CN 112541569 A CN112541569 A CN 112541569A CN 202011328753 A CN202011328753 A CN 202011328753A CN 112541569 A CN112541569 A CN 112541569A
- Authority
- CN
- China
- Prior art keywords
- data
- sensor
- cloud server
- acquisition module
- original
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012549 training Methods 0.000 title claims abstract description 43
- 238000010801 machine learning Methods 0.000 title claims abstract description 27
- 238000000034 method Methods 0.000 title claims abstract description 21
- 230000003993 interaction Effects 0.000 claims abstract description 8
- 239000000284 extract Substances 0.000 claims abstract description 5
- 238000004891 communication Methods 0.000 claims description 9
- 230000006870 function Effects 0.000 claims description 7
- FGUUSXIOTUKUDN-IBGZPJMESA-N C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 Chemical compound C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 FGUUSXIOTUKUDN-IBGZPJMESA-N 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 4
- 238000005457 optimization Methods 0.000 abstract description 4
- 239000013598 vector Substances 0.000 description 16
- 238000012544 monitoring process Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000005259 measurement Methods 0.000 description 5
- 238000013527 convolutional neural network Methods 0.000 description 4
- 229910002651 NO3 Inorganic materials 0.000 description 3
- NHNBFGGVMKEFGY-UHFFFAOYSA-N Nitrate Chemical compound [O-][N+]([O-])=O NHNBFGGVMKEFGY-UHFFFAOYSA-N 0.000 description 3
- 230000009471 action Effects 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 238000002835 absorbance Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 238000012856 packing Methods 0.000 description 1
- 238000010238 partial least squares regression Methods 0.000 description 1
- 239000000523 sample Substances 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 239000002351 wastewater Substances 0.000 description 1
Images
Classifications
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Telephonic Communication Services (AREA)
Abstract
The invention discloses a sensor on-line training system and method based on machine learning, which comprises a sensor, a data acquisition module, a man-machine interaction device, a smart phone application, data acquisition module software and cloud server software, wherein the data acquisition module acquires original data and actual measured values from the sensor and the man-machine interaction device respectively and uploads the original data and the actual measured values to a cloud server through a network, the cloud server extracts all data in a training database belonging to the same sensor to perform model training and downloads a trained new algorithm to the data acquisition module, and the data acquisition module updates the new algorithm to the sensor. A global optimization effect is achieved.
Description
Technical Field
The invention relates to the technical field of Internet of things, in particular to a system and a method for on-line training of a sensor based on machine learning.
Background
The machine learning technology has achieved great success in the fields of image recognition, natural language recognition and the like by utilizing big data and convolutional neural network algorithms, and the industry is constantly trying to apply the machine learning algorithm to online sensor equipment to cope with complex application environments, such as the field of wastewater monitoring and the like.
The following defects still exist in the traditional monitoring field: an engineer designs sensor software according to industry knowledge and own experience, but a software algorithm completed only by experience can only adapt to partial application scenes, when a complex object is monitored, the field environment and the monitored object are easy to change, and the engineer is limited by project time and cost, and often can only acquire original data of partial field working conditions to train the algorithm, so that the software designed by the traditional mode cannot stably and reliably complete monitoring work.
Based on the above problems, it is urgently needed to provide a machine learning-based sensor online training system and method, which utilize machine learning to collect a large amount of raw data of a sensor in a complex application scene and actual measurement values under corresponding working conditions, train a convolutional neural network model through mass data to obtain a set of algorithm matched with the current complex application scene, and then embed the algorithm into software of the sensor, so that the sensor algorithm is continuously improved in the whole life cycle, thereby enabling the sensor to adapt to different working conditions in the field to the maximum extent, stably and reliably completing monitoring work in the complex application scene, and achieving an overall optimization effect.
Disclosure of Invention
The present invention is directed to a system and a method for on-line sensor training based on machine learning, so as to solve the problems mentioned in the background art.
In order to solve the technical problems, the invention provides the following technical scheme:
a sensor online training system based on machine learning comprises hardware and software, wherein the hardware comprises a sensor, a data acquisition module and a man-machine interaction device, the software comprises a smart phone end application, data acquisition module software and cloud server software, the sensor is connected with the data acquisition module, the smart phone end application allows a user to input a measured value of a detected target, the data acquisition module software comprises an acquisition unit, an uploading unit and a deployment unit, the acquisition unit is responsible for acquiring original data from the sensor, the uploading unit is responsible for uploading the data to the cloud server, the deployment unit is responsible for updating algorithm parameters received from the cloud server to the sensor, the cloud server software comprises a data receiving unit, a model training unit and a data sending unit, the data receiving unit receives the original data through a network and stores the original data to a corresponding database, the model training unit trains model parameters by using continuously updated sensor raw data, and the data sending unit downloads the trained algorithm parameters to the data acquisition module through a network.
Furthermore, the connection mode of the sensor and the data acquisition module comprises an Ethernet and an RS-485 bus, and the data acquisition module is attached with a 4G/5G network communication function and a Wi-Fi function, so that the real-time property of data acquisition is ensured.
Further, the data acquisition module acquires the original data from the sensor, uploads the original data to the cloud server through the network, and finally updates the new model parameters to the online sensor as follows:
a: the data acquisition module acquires original data of the online sensor;
b: the data acquisition module adds timestamp information to the acquired original data;
c: judging whether the network is normal at the moment, if so, turning to the step d, otherwise, directly saving the data to a local database, and turning back to the step a;
d: b, uploading the original data added with the timestamp information in the step b to a cloud server through a network;
e: judging whether a new model parameter is received from the cloud server, if so, turning to the step f, otherwise, turning back to the step a;
f: the data acquisition module updates new model parameters to the online sensors and continues to acquire raw data from the sensors so that the sensor algorithm remains continuously improved over the full life cycle.
Furthermore, the acquisition of the original data of the online sensor in the step a and the updating of the new model parameters to the online sensor in the step f are performed through a Modbus RTU communication protocol, wherein Modbus is an industrial automation bus communication protocol, supports thousands of industrial intelligent instrument data communication, and supports multiple communication interfaces.
Further, the cloud server receives the original data, extracts the device serial number and the timestamp information, trains the algorithm to obtain new model parameters, and sends the new model parameters as follows:
(1) establishing a TCP Socket application programming interface service to wait for the on-site uploading of the original sensor data;
(2) the cloud server judges whether the original sensor data are received or not, if the original sensor data are received, the equipment serial number and the timestamp information are extracted, the original data are inserted into the database A, the step (3) is carried out, and if not, the cloud server continues to wait for uploading of the original sensor data on site;
(3) judging whether marking data sent by the smart phone end application are received or not, if yes, extracting the equipment serial number and the timestamp information, inserting the marking data into a database B, and turning to the step (4), otherwise, continuing to wait for uploading of the original sensor data on site;
(4) extracting all marking data from the database B, inquiring the database A, finding out a corresponding original data set through timestamp information, and retraining an algorithm by using the original data obtained through inquiry and the marking data to obtain new model parameters;
(5) and checking whether the equipment corresponding to the serial number has already established network connection, if so, sending new model parameters by the cloud server and downloading the new model parameters to the data acquisition module, otherwise, storing the new model parameters in the cloud server and sending the new model parameters again after the equipment is connected next time.
Further, the marking in the step (3) is a process that a user inputs a measured value of a current measured target through a smart phone end application and uploads the measured value to a cloud server.
Further, a machine learning-based sensor online training method comprises the following steps:
s1: acquiring original data from a sensor through an Ethernet or an RS-485 bus, and acquiring a measured value from a human-computer interaction device;
s2: inputting a measured value of a current measured target through the application of the smart phone end and uploading the measured value to the cloud server;
s3: uploading original data to a cloud server through a network;
s4: inserting the original data and the measured value into a training database, and indexing each piece of data of the training database by taking a time stamp as an index, wherein the data of the same time stamp comprises a primary original data sampling value and a primary target measured value;
s5: the cloud server extracts all data in a training database belonging to the same sensor and conducts model training again;
s6: downloading the trained algorithm, namely a group of new model parameters, to a data acquisition module;
s7: and the new model parameters are updated to the sensor through the Ethernet or the RS-485 bus, so that the adaptability of the sensor under different working conditions is greatly improved.
Further, the step S2 further includes the following steps:
s21: the user inputs an accurate value measured from third-party equipment or a laboratory through the application of the smart phone end;
s22: a user inputs a serial number of an online sensor through the application of a smart phone end;
s23: and the smart phone side application packs the measured value, the serial number and the current time information and sends the packed measured value, the serial number and the current time information to the cloud server.
Compared with the prior art, the invention has the following beneficial effects: the invention utilizes machine learning to collect a large amount of original data of the sensor under a complex application scene and actual measured values under corresponding working conditions, trains a convolutional neural network model through mass data to obtain a set of algorithm matched with the current complex application scene, and then embeds the algorithm into the software of the sensor, so that the algorithm of the sensor is continuously improved in the whole life cycle, thereby enabling the sensor to adapt to different working conditions on the spot to the maximum extent, stably and reliably completing the monitoring work under the complex application scene, and achieving the effect of global optimization.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a schematic diagram of a framework of a machine learning-based on-line sensor training system according to the present invention;
FIG. 2 is a schematic diagram of the steps of a machine learning-based on-line sensor training method of the present invention;
FIG. 3 is a schematic diagram of the working steps of a data acquisition module of the machine learning-based sensor online training system of the present invention;
FIG. 4 is a schematic diagram of the working steps of the cloud server of the machine learning-based on-line sensor training system according to the present invention;
fig. 5 is a schematic diagram of the working steps of the smartphone end application of the machine learning-based sensor online training system of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1-5, the present invention provides the following technical solutions:
the hardware comprises a sensor, a data acquisition module and a man-machine interaction device, the software comprises a smart phone end application, data acquisition module software and cloud server software, the sensor and the data acquisition module pass through an Ethernet or RS-485 bus, the data acquisition module is attached with a 4G/5G network communication function and a Wi-Fi function, the smart phone end application allows a user to input a measured value of a measured object, the data acquisition module software comprises an acquisition unit, an uploading unit and a deployment unit, the acquisition unit is responsible for acquiring original data from the sensor, the uploading unit is responsible for uploading the data to the cloud server, the deployment unit is responsible for updating algorithm parameters received from the cloud server to the sensor, and the cloud server software comprises a data receiving unit, a cloud server software, The system comprises a model training unit and a data sending unit, wherein the data receiving unit receives and stores original data into a corresponding database through a network, the model training unit trains model parameters by using continuously updated sensor original data, and the data sending unit downloads the trained algorithm parameters to a data acquisition module through the network.
The method comprises the following steps that a data acquisition module acquires original data from a sensor, uploads the original data to a cloud server through a network, and finally updates new model parameters to an online sensor:
a: the data acquisition module acquires original data of the online sensor;
b: the data acquisition module adds timestamp information to the acquired original data;
c: judging whether the network is normal at the moment, if so, turning to the step d, otherwise, directly saving the data to a local database, and turning back to the step a;
d: b, uploading the original data added with the timestamp information in the step b to a cloud server through a network;
e: judging whether a new model parameter is received from the cloud server, if so, turning to the step f, otherwise, turning back to the step a;
f: the data acquisition module updates the new model parameters to the on-line sensors and continues to acquire raw data from the sensors.
The acquisition of the raw data of the online sensor in the step a and the updating of the new model parameters to the online sensor in the step f are both realized through a Modbus RTU communication protocol.
The operation flow of the cloud server is as follows:
(1) establishing a TCP Socket application programming interface service to wait for the on-site uploading of the original sensor data;
(2) the cloud server judges whether the original sensor data are received or not, if the original sensor data are received, the equipment serial number and the timestamp information are extracted, the original data are inserted into the database A, the step (3) is carried out, and if not, the cloud server continues to wait for uploading of the original sensor data on site;
(3) judging whether marking data sent by the smart phone end application are received or not, if yes, extracting the equipment serial number and the timestamp information, inserting the marking data into a database B, and turning to the step (4), otherwise, continuing to wait for uploading of the original sensor data on site;
(4) extracting all marking data from the database B, inquiring the database A, finding out a corresponding original data set through timestamp information, and retraining an algorithm by using the original data obtained through inquiry and the marking data to obtain new model parameters;
(5) and checking whether the equipment corresponding to the serial number has already established network connection, if so, sending new model parameters by the cloud server and downloading the new model parameters to the data acquisition module, otherwise, storing the new model parameters in the cloud server and sending the new model parameters again after the equipment is connected next time.
A sensor online training method based on machine learning comprises the following steps:
s1: acquiring original data from a sensor through an Ethernet or an RS-485 bus, and acquiring a measured value from a human-computer interaction device;
s2: inputting a measured value of a current measured target through the application of the smart phone end and uploading the measured value to the cloud server;
s3: uploading original data to a cloud server through a network;
s4: inserting the original data and the measured value into a training database, and indexing each piece of data of the training database by taking a time stamp as an index, wherein the data of the same time stamp comprises a primary original data sampling value and a primary target measured value;
s5: the cloud server extracts all data in a training database belonging to the same sensor and conducts model training again;
s6: downloading the trained algorithm, namely a group of new model parameters, to a data acquisition module;
s7: the new model parameters are updated to the sensors via ethernet or RS-485 bus.
The step S2 further includes the following steps:
s21: the user inputs an accurate value measured from third-party equipment or a laboratory through the application of the smart phone end;
s22: a user inputs a serial number of an online sensor through the application of a smart phone end;
s23: and the smart phone side application packs the measured value, the serial number and the current time information and sends the packed measured value, the serial number and the current time information to the cloud server.
Taking the example of 'collecting data and predicting nitrate concentration by a spectrum probe', the method is realized by the following steps:
step 1: the on-line sensor collects the raw data corresponding to 256 discrete spectra, namely the input vector with 256 dimensions in mathematics
n sampling points, forming an input matrix X with n rows and 256 columns;
step 2: the user obtains the accurate nitrate concentration at that time according to the measurement of the third-party equipment or the laboratory, and the nitrate concentration is input through the application of the smart phone terminal, and the input vector is set according to the value
The target y of (1) is set as the ith sampling point, and the user set value is yiN sampling points, the setting value being the target value vector
Inputting the serial number of the online sensor through the application of the smart phone end, packaging the measured value, the serial number and the current time information and sending the packaged information to the cloud server;
and step 3: inputting the matrix X and the target value vector by a data acquisition module
Uploading to a cloud server;
and 4, step 4: approximating a real spectral input vector and a target value function by a machine learning algorithm, such as Partial Least Squares (PLS) or a neural network, on a cloud server
pls＝PLSRegression(n_components＝9)
pls.fit(x,y)
And 5: packing the multidimensional parameter theta trained by the cloud server, downloading the multidimensional parameter theta to a field data acquisition module through a network, downloading four groups of parameters including an average value vector X _ mean _, a standard deviation vector X _ std, a coefficient vector coef _andan average value y _ mean _ofa target value to a lower computer according to the PLS algorithm principle, and acquiring the parameters, wherein the method comprises the following steps:
Pls.x_mean_
Pls.x_std_.
Pls.coef_.
Pls.y_mean_[0]
step 6, the data acquisition module updates the average value vector, the standard deviation vector, the coefficient vector and the average value of the target value to an online sensor through a Modbus RTU protocol, and the sensor saves the 4 groups of vectors as parameters to Flash for storage;
and 7: the on-line sensor will use the new model parameters to calculate the measurement value in the next measurement cycle, such as the calculation formula:
wherein x is an absorbance input vector measured by the lower computer, xmeanFor the downloaded trained mean vector x _ mean _, xstdFor the downloaded trained standard deviation vector x _ std, C is the downloaded trained coefficient vector coef _, ymeanIs the average value y mean of the downloaded trained target values.
As shown in fig. 1, a data acquisition module acquires a large amount of raw data of a sensor in a complex application scene and actual measurement values under corresponding working conditions, a cloud server trains a convolutional neural network model through mass data to obtain a set of algorithm matched with the current complex application scene, and then the algorithm is embedded into software of the sensor, so that the algorithm of the sensor is continuously improved in the whole life cycle, the sensor can adapt to different working conditions on the site to the maximum extent, the monitoring work under the complex application scene is stably and reliably completed, and a global optimization effect is achieved.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that changes may be made in the embodiments and/or equivalents thereof without departing from the spirit and scope of the invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (8)
1. The utility model provides a sensor online training system based on machine learning, includes hardware and software, hardware includes sensor, data acquisition module and human-computer interaction device, software includes smart mobile phone end application, data acquisition module software and high in the clouds server software, its characterized in that:
the sensor is connected with the data acquisition module, the smart phone end application allows a user to input the measured value of the measured object, the data acquisition module software comprises an acquisition unit, an uploading unit and a deployment unit, wherein the acquisition unit is responsible for acquiring original data from the sensor, the uploading unit is responsible for uploading data to the cloud server, the deployment unit is responsible for updating algorithm parameters received from the cloud server to the sensors, the cloud server software comprises a data receiving unit, a model training unit and a data sending unit, the data receiving unit receives and stores raw data into a corresponding database through a network, the model training unit trains model parameters by using continuously updated sensor raw data, and the data sending unit downloads the trained algorithm parameters to the data acquisition module through a network.
2. The machine learning-based sensor online training system of claim 1, wherein: the connection mode of the sensor and the data acquisition module comprises an Ethernet and an RS-485 bus, and the data acquisition module is attached with a 4G/5G network communication function and a Wi-Fi function.
3. The machine learning-based sensor online training system of claim 1, wherein: the data acquisition module acquires original data from the sensor, uploads the original data to the cloud server through the network, and finally updates new model parameters to the online sensor, wherein the specific steps are as follows:
a: the data acquisition module acquires original data of the online sensor;
b: the data acquisition module adds timestamp information to the acquired original data;
c: judging whether the network is normal at the moment, if so, turning to the step d, otherwise, directly saving the data to a local database, and turning back to the step a;
d: b, uploading the original data added with the timestamp information in the step b to a cloud server through a network;
e: judging whether a new model parameter is received from the cloud server, if so, turning to the step f, otherwise, turning back to the step a;
f: the data acquisition module updates the new model parameters to the on-line sensors and continues to acquire raw data from the sensors.
4. The machine learning-based sensor online training system of claim 3, wherein: and (c) acquiring the raw data of the online sensor in the step a and updating the new model parameters to the online sensor in the step f both adopt a Modbus RTU communication protocol.
5. The machine learning-based sensor online training system of claim 1, wherein: the specific steps of the cloud server receiving the original data, extracting the equipment serial number and the timestamp information, training the algorithm to obtain new model parameters and sending the new model parameters are as follows:
(1) establishing a TCP Socket application programming interface service to wait for the on-site uploading of the original sensor data;
(2) the cloud server judges whether the original sensor data are received or not, if the original sensor data are received, the equipment serial number and the timestamp information are extracted, the original data are inserted into the database A, the step (3) is carried out, and if not, the cloud server continues to wait for uploading of the original sensor data on site;
(3) judging whether marking data sent by the smart phone end application are received or not, if yes, extracting the equipment serial number and the timestamp information, inserting the marking data into a database B, and turning to the step (4), otherwise, continuing to wait for uploading of the original sensor data on site;
(4) extracting all marking data from the database B, inquiring the database A, finding out a corresponding original data set through timestamp information, and retraining an algorithm by using the original data obtained through inquiry and the marking data to obtain new model parameters;
(5) and checking whether the equipment corresponding to the serial number has already established network connection, if so, sending new model parameters by the cloud server and downloading the new model parameters to the data acquisition module, otherwise, storing the new model parameters in the cloud server and sending the new model parameters again after the equipment is connected next time.
6. The machine learning-based sensor online training system of claim 5, wherein: and the marking in the step (3) is a process that a user inputs a measured value of a current measured target through the application of the smart phone end and uploads the measured value to the cloud server.
7. A sensor on-line training method based on machine learning is characterized in that: the method comprises the following steps:
s1: acquiring original data from a sensor through an Ethernet or an RS-485 bus, and acquiring a measured value from a human-computer interaction device;
s2: inputting a measured value of a current measured target through the application of the smart phone end and uploading the measured value to the cloud server;
s3: uploading original data to a cloud server through a network;
s4: inserting the original data and the measured value into a training database, and indexing each piece of data of the training database by taking a time stamp as an index, wherein the data of the same time stamp comprises a primary original data sampling value and a primary target measured value;
s5: the cloud server extracts all data in a training database belonging to the same sensor and conducts model training again;
s6: downloading the trained algorithm, namely a group of new model parameters, to a data acquisition module;
s7: the new model parameters are updated to the sensors via ethernet or RS-485 bus.
8. The machine learning-based online sensor training method according to claim 7, wherein: the step S2 further includes the steps of:
s21: the user inputs an accurate value measured from third-party equipment or a laboratory through the application of the smart phone end;
s22: a user inputs a serial number of an online sensor through the application of a smart phone end;
s23: and the smart phone side application packs the measured value, the serial number and the current time information and sends the packed measured value, the serial number and the current time information to the cloud server.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011328753.7A CN112541569A (en) | 2020-11-24 | 2020-11-24 | Sensor online training system and method based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011328753.7A CN112541569A (en) | 2020-11-24 | 2020-11-24 | Sensor online training system and method based on machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112541569A true CN112541569A (en) | 2021-03-23 |
Family
ID=75014707
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011328753.7A Pending CN112541569A (en) | 2020-11-24 | 2020-11-24 | Sensor online training system and method based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112541569A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113435126A (en) * | 2021-07-07 | 2021-09-24 | 魏天骐 | Knowledge sharing processing method, intelligent robot equipment, knowledge sharing system and task learning system |
| CN115052130A (en) * | 2022-05-27 | 2022-09-13 | 合肥富煌君达高科信息技术有限公司 | High low temperature experiment unmanned on duty bug information wireless transmission system |
| CN115908948A (en) * | 2023-01-05 | 2023-04-04 | 北京霍里思特科技有限公司 | Intelligent sorting system for online adjustment model and control method thereof |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120220835A1 (en) * | 2011-02-14 | 2012-08-30 | Wayne Chung | Wireless physiological sensor system and method |
| US20140081100A1 (en) * | 2012-09-20 | 2014-03-20 | Masimo Corporation | Physiological monitor with mobile computing device connectivity |
| CN104965503A (en) * | 2015-07-17 | 2015-10-07 | 江西洪都航空工业集团有限责任公司 | Intelligent household control system based on machine learning |
| CN105205436A (en) * | 2014-06-03 | 2015-12-30 | 北京创思博德科技有限公司 | Gesture identification system based on multiple forearm bioelectric sensors |
| CN106412111A (en) * | 2016-11-14 | 2017-02-15 | 南京物联传感技术有限公司 | Cloud-based whole house intelligent system |
| US20170083312A1 (en) * | 2015-09-22 | 2017-03-23 | Mc10, Inc. | Method and system for crowd-sourced algorithm development |
| CN107707657A (en) * | 2017-09-30 | 2018-02-16 | 苏州涟漪信息科技有限公司 | Safety custody system based on multisensor |
| CN107788994A (en) * | 2017-10-12 | 2018-03-13 | 微泰医疗器械(杭州)有限公司 | A kind of real-time dynamic blood sugar monitoring system of intelligence based on high in the clouds big data and method |
| CN108093030A (en) * | 2017-11-29 | 2018-05-29 | 杭州古北电子科技有限公司 | A kind of artificial intelligence model dispositions method based on Cloud Server |
| CN108614071A (en) * | 2018-03-21 | 2018-10-02 | 中国科学院自动化研究所 | Distributed outside atmosphere quality-monitoring accuracy correction system and parameter updating method |
| CN109631973A (en) * | 2018-11-30 | 2019-04-16 | 苏州数言信息技术有限公司 | A kind of automatic calibrating method and system of sensor |
| CN110216680A (en) * | 2019-07-05 | 2019-09-10 | 山东大学 | A kind of service robot cloud ground collaborative fault diagnosis system and method |
| CN210836549U (en) * | 2019-11-27 | 2020-06-23 | 肖世军 | Intelligent wearable sensing system for protecting distraction walking safety |
| CN111444848A (en) * | 2020-03-27 | 2020-07-24 | 广州英码信息科技有限公司 | Specific scene model upgrading method and system based on federal learning |
| CN111862165A (en) * | 2020-06-17 | 2020-10-30 | 南京理工大学 | Target tracking method for updating Kalman filter based on deep reinforcement learning |
-
2020
- 2020-11-24 CN CN202011328753.7A patent/CN112541569A/en active Pending
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120220835A1 (en) * | 2011-02-14 | 2012-08-30 | Wayne Chung | Wireless physiological sensor system and method |
| US20140081100A1 (en) * | 2012-09-20 | 2014-03-20 | Masimo Corporation | Physiological monitor with mobile computing device connectivity |
| CN105205436A (en) * | 2014-06-03 | 2015-12-30 | 北京创思博德科技有限公司 | Gesture identification system based on multiple forearm bioelectric sensors |
| CN104965503A (en) * | 2015-07-17 | 2015-10-07 | 江西洪都航空工业集团有限责任公司 | Intelligent household control system based on machine learning |
| CN108293174A (en) * | 2015-09-22 | 2018-07-17 | Mc10股份有限公司 | Method and system for crowdsourcing algorithm development |
| US20170083312A1 (en) * | 2015-09-22 | 2017-03-23 | Mc10, Inc. | Method and system for crowd-sourced algorithm development |
| CN106412111A (en) * | 2016-11-14 | 2017-02-15 | 南京物联传感技术有限公司 | Cloud-based whole house intelligent system |
| CN107707657A (en) * | 2017-09-30 | 2018-02-16 | 苏州涟漪信息科技有限公司 | Safety custody system based on multisensor |
| CN107788994A (en) * | 2017-10-12 | 2018-03-13 | 微泰医疗器械(杭州)有限公司 | A kind of real-time dynamic blood sugar monitoring system of intelligence based on high in the clouds big data and method |
| CN108093030A (en) * | 2017-11-29 | 2018-05-29 | 杭州古北电子科技有限公司 | A kind of artificial intelligence model dispositions method based on Cloud Server |
| CN108614071A (en) * | 2018-03-21 | 2018-10-02 | 中国科学院自动化研究所 | Distributed outside atmosphere quality-monitoring accuracy correction system and parameter updating method |
| CN109631973A (en) * | 2018-11-30 | 2019-04-16 | 苏州数言信息技术有限公司 | A kind of automatic calibrating method and system of sensor |
| CN110216680A (en) * | 2019-07-05 | 2019-09-10 | 山东大学 | A kind of service robot cloud ground collaborative fault diagnosis system and method |
| CN210836549U (en) * | 2019-11-27 | 2020-06-23 | 肖世军 | Intelligent wearable sensing system for protecting distraction walking safety |
| CN111444848A (en) * | 2020-03-27 | 2020-07-24 | 广州英码信息科技有限公司 | Specific scene model upgrading method and system based on federal learning |
| CN111862165A (en) * | 2020-06-17 | 2020-10-30 | 南京理工大学 | Target tracking method for updating Kalman filter based on deep reinforcement learning |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113435126A (en) * | 2021-07-07 | 2021-09-24 | 魏天骐 | Knowledge sharing processing method, intelligent robot equipment, knowledge sharing system and task learning system |
| CN113435126B (en) * | 2021-07-07 | 2024-02-02 | 魏天骐 | Knowledge sharing processing method, intelligent robot device, knowledge sharing system and task learning system |
| CN115052130A (en) * | 2022-05-27 | 2022-09-13 | 合肥富煌君达高科信息技术有限公司 | High low temperature experiment unmanned on duty bug information wireless transmission system |
| CN115908948A (en) * | 2023-01-05 | 2023-04-04 | 北京霍里思特科技有限公司 | Intelligent sorting system for online adjustment model and control method thereof |
| CN115908948B (en) * | 2023-01-05 | 2024-04-26 | 北京霍里思特科技有限公司 | Intelligent sorting system for online adjustment model and control method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112541569A (en) | Sensor online training system and method based on machine learning | |
| CN116129366B (en) | Digital twinning-based park monitoring method and related device | |
| CN107168294B (en) | Unmanned inspection monitoring method for thermal power water system equipment | |
| EP3361330B1 (en) | Event analyzing device, event analyzing system, event analyzing method, event analyzing program, and non-transitory computer readable storage medium | |
| CN108614071B (en) | Distributed outdoor air quality monitoring precision correction system and parameter updating method | |
| CN113872328A (en) | Transformer substation remote intelligent inspection method and system based on neural network | |
| CN107944005B (en) | Data display method and device | |
| CN116599857B (en) | Digital twin application system suitable for multiple scenes of Internet of things | |
| CN103401881A (en) | Data collection system and method based on intelligent instrument | |
| CN110910440A (en) | Power transmission line length determination method and system based on power image data | |
| CN108093419A (en) | A kind of base station information acquiring system and its control method based on Internet of Things | |
| CN117969789B (en) | Water regime prediction method and monitoring system for underground water resource | |
| CN116822115A (en) | Environment management method and system for intelligent park based on digital twin technology | |
| CN117639228A (en) | Power distribution network running state prediction method and system based on digital twin | |
| CN115224795A (en) | Intelligent substation equipment operation monitoring and early warning system and method | |
| CN102359879A (en) | Measurement and control system for test network and data acquisition control method | |
| CN117493498B (en) | Electric power data mining and analysis system based on industrial Internet | |
| CN117930781A (en) | Production management system based on intelligent equipment | |
| CN107843811A (en) | A kind of analysis method and system of grid equipment online monitoring data | |
| CN112132519A (en) | Bottled gas cross network system based on thing networking big data | |
| CN111108738B (en) | Data processing device, data analysis device, data processing system, and method for processing data | |
| CN116708084A (en) | Edge computing gateway based on industrial Internet | |
| CN115118578A (en) | SCADA system based on WEB | |
| CN115307810A (en) | Temperature compensation method and device of temperature and pressure sensor | |
| CN116150664A (en) | Data type differentiated acquisition frequency determining method and system |
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 |