CN111076758A - Automatic calibration method for high-altitude detection sensor based on Internet of things - Google Patents
Automatic calibration method for high-altitude detection sensor based on Internet of things Download PDFInfo
- Publication number
- CN111076758A CN111076758A CN201911175498.4A CN201911175498A CN111076758A CN 111076758 A CN111076758 A CN 111076758A CN 201911175498 A CN201911175498 A CN 201911175498A CN 111076758 A CN111076758 A CN 111076758A
- Authority
- CN
- China
- Prior art keywords
- data
- sensor
- calibration
- temperature
- humidity
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D18/00—Testing or calibrating apparatus or arrangements provided for in groups G01D1/00 - G01D15/00
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Abstract
The invention discloses an automatic calibration method for a high-altitude detection sensor based on the Internet of things, and belongs to the technical field of sensors. The invention comprises the following steps: designing a temperature, humidity and pressure sensor data acquisition mechanism; defining a data transmission communication protocol; constructing a sensor calibration algorithm based on a neural network; designing a storage database; and carrying out data analysis and visual display. The invention fully utilizes the advantages of the Internet of things in industrial production, combines the Internet of things technology with the neural network, realizes automation and intellectualization in the calibration process of the sensor, effectively liberates labor force and improves working efficiency.
Description
Technical Field
The invention relates to the technical field of sensors of the Internet of things, in particular to an automatic calibration method of a high-altitude detection sensor based on the Internet of things.
Background
At present, in the field of high-altitude detection, calibration of temperature, humidity and pressure sensors is mainly carried out manually, the distribution of characteristic points of the sensors is measured manually, and a polynomial is adopted to fit a characteristic curve of the sensors, so that calibration of the sensors is realized. However, the manual measurement method has low efficiency, high cost and low polynomial curve fitting precision, and is difficult to meet the requirements of yield and precision.
In recent years, the technology of internet of things has been widely applied to aspects of production and life, and has the new characteristics of interconnection of everything and intelligence. The industrial production mode is changed greatly based on the technology of the Internet of things, the productivity is greatly liberated, and the intelligent level and the working efficiency are improved. The internet of things technology is widely applied to complex scenes such as intelligent factories, intelligent homes and wearable medical systems, and therefore, the data of the high-altitude detection sensor can be collected through the internet of things technology.
For calibration of temperature, humidity and pressure sensors, the traditional modes are mainly a table look-up method and a curve fitting method, the table look-up method ignores the measurement error of calibration points, the fitting method only can reflect the integral trend of the sensor, and the fitting method is the approximation of a plurality of discrete measurement points to the integral model of the sensor, and cannot meet the calibration of the sensor under complex conditions. The neural network is used as a novel information processing method, a new solution is brought to industrial problems, data acquired by the sensor are learned by constructing a neural network unit, fitting of a characteristic curve of the sensor is realized, measurement accuracy of the sensor is improved, and correction of the sensor is completed.
Disclosure of Invention
The invention provides an automatic calibration method of an aerial detection sensor based on the Internet of things, aiming at the problems of high labor cost, long measurement time and low sensor calibration precision in aerial detection temperature measurement and correction of wet and pressure sensors. The technology of the Internet of things is combined with the neural network, so that automation and intellectualization of the calibration process of the sensor are realized, labor force is effectively liberated, and working efficiency is improved. By adopting the neural network technology, the measurement precision of the temperature, humidity and pressure sensors is further improved, and the requirement of high-altitude detection is better met.
In order to achieve the purpose, the invention adopts the following technical scheme:
an automatic calibration method for a high altitude detection sensor based on the Internet of things mainly comprises the following steps:
step 1: constructing a temperature, humidity and pressure sensor data acquisition system based on the Internet of things;
the architecture of the internet of things consists of 3 parts, namely a sensing layer, a transmission layer and an application layer. The sensing layer is arranged in the sensor calibration equipment and is used for temperature, humidity and pressure sensor equipment information and corresponding measurement data. The transmission layer is responsible for transmitting control instructions, sensor data information and other contents between the sensing layer and the application layer. And establishing connection between different devices by adopting a UDP communication protocol, and realizing data receiving, sending and storage based on Socket. The application layer realizes the control of equipment and the storage, analysis and display of data, and can control the related parameters of the temperature, humidity and pressure sensors in the acquisition process through different dialog boxes of the application layer so as to meet the acquisition requirements of the sensors of different types.
Step 2: designing an operation interface of the Internet of things;
the Internet of things architecture corresponds to a computer software system, in order to improve the man-machine interaction capacity, optimize the operation process and realize automation and intellectualization of the calibration process, the proposed method designs a relevant operation interface, the computer software system mainly comprises a navigation bar and a functional area, the navigation bar shows the main functions of the proposed method and adds hyperlinks to each function to enable the navigation bar to jump corresponding pages; the functional area is composed of a visual chart, a data dialog box, a data operation button and the like according to the specific operation requirements of the sensor calibration, so that the operation page has good operation feeling and display. The design and use operation interface comprises a home page, an equipment management interface, a sensor self-testing interface, a data acquisition interface and a data retest interface.
The home page mainly comprises a current calibration information overview and a historical data statistical function, the calibration quantity of temperature, humidity and air pressure sensors and a corresponding calibration total process are displayed in current acquisition, calibration historical records can be checked in historical information preview, and data retrieval is carried out according to the type of the sensors and the calibration time.
In the equipment management information table, the position of the acquisition board can be selected, the number of the sensor is input, the stable threshold condition of the calibration environment is set, and the calibration point distribution of different sensors is set.
In the self-testing information look-up table, the air pressure value and the temperature value of the sensor in the atmospheric environment are displayed, a primary air pressure distribution diagram and a primary temperature distribution diagram are drawn according to the measured values, a clustering center is calculated by using a K-mean method, and the condition of the sensor is judged according to a set threshold value. And displaying the acquired data information in real time on a data acquisition interface, drawing a temperature change information table, a humidity change information table and an air pressure change information table according to the acquired data, and monitoring the acquisition condition. And the data retest interface comprises sensor calibration data information, sensor retest data information, a corresponding data comparison graph and an error distribution graph, and is used for performing qualification judgment on the calibration data and the retest data and providing a data export function.
And step 3: calibration algorithm for designing temperature, humidity and pressure sensors
Adopting a BP neural network model to calibrate the temperature, humidity and pressure sensors, wherein the BP neural network model structurally comprises an input layer, a hidden layer and an output layer, and the update speed of BP network parameters is accelerated by introducing a Levenberg-Marquardt algorithm; the calibration accuracy of the sensor is further improved by designing the hidden layer activation function as a Morlet wavelet function. Different network structures are designed according to different sensor influence factors, and the measurement result of the temperature sensor is influenced by the ambient temperature, so that the input layer is 1 neuron, 5 neurons are selected as a network hiding layer, and 1 neuron is selected as a network output layer; the measuring result of the air pressure sensor is influenced by the temperature and the air pressure of the environment, so that the input layer is 2 neurons, 10 neurons are selected as a network hiding layer, and 1 neuron is selected as a network output layer; the measuring result of the humidity sensor is influenced by the environment humidity, so that the input layer is 1 neuron, 10 neurons are selected as a network hiding layer, and 1 neuron is selected as a network output layer; after BP neural network training, the final convergence is achieved, the fitting of the characteristic curves of the temperature, humidity and pressure sensors is realized, and the aim of correcting the sensors is fulfilled.
According to the characteristic curves of the temperature sensor, the humidity sensor and the pressure sensor, corresponding calibration point distribution is designed, and corresponding environment waiting time and data acquisition strategies are formulated by combining the actual conditions of calibration equipment.
And data transmission is carried out by adopting a UDP (user Datagram protocol), IP addresses and port numbers are distributed for each device, a communication protocol between the temperature, humidity and pressure sensors and the corresponding calibration device is specified, the control of the calibration device and the acquisition of the measurement data of the sensors are realized.
Dividing the collected calibration data into a training set and a testing set according to the ratio of 8:2, constructing corresponding BP network model structures for the temperature, humidity and pressure sensors, adopting a Morlet wavelet activation function for a network hiding layer, and updating network parameters by using a Levenberg-Marquardt algorithm.
The method comprises the steps of caching and storing sensor data by adopting MySQL + Redis, and designing a temperature, humidity and pressure sensor data table structure, wherein the temperature, humidity and pressure sensor data table structure mainly comprises acquisition time, a sequence number, standard temperature, standard humidity, standard air pressure, a temperature original value, a humidity original value and an air pressure original value.
Compared with the prior art, the invention has the following advantages:
1. the traditional manual sensor data acquisition method is abandoned, and the sensor data is acquired based on the Internet of things architecture, so that the labor force is greatly liberated, and the production efficiency is improved.
2. The neural network technology is adopted to calibrate the temperature, humidity and pressure sensors, and compared with the traditional calibration method, the generalization capability of the model is enhanced, and the measurement precision of the sensors is improved.
Drawings
FIG. 1 is a schematic diagram of the overall architecture of an automatic calibration method of a high altitude detection sensor based on the Internet of things;
FIG. 2 is a schematic diagram of a temperature sensor acquisition flow;
FIG. 3 is a schematic diagram of a humidity sensor acquisition flow;
FIG. 4 is a schematic view of an air pressure sensor acquisition flow;
FIG. 5 is a schematic view of a calibration model of the air pressure sensor;
FIG. 6 is a schematic diagram of a sensor calibration algorithm training flow;
FIG. 7 is a schematic view of a calibration model of a temperature sensor;
FIG. 8 is a schematic view of a calibration model of a humidity sensor;
FIG. 9 is a schematic diagram of a data storage flow;
FIG. 10 is a schematic view of a website home page;
FIG. 11 is a standby management information presentation intention;
FIG. 12 is a schematic diagram of sensor self-test information;
FIG. 13 is a schematic view of a barometric sensor data collection;
FIG. 14 is a schematic diagram of a sensor retest;
Detailed Description
The invention realizes automatic calibration of the high-altitude detection sensor based on the Internet of things. The specific method adopted by the invention will be described in detail below with reference to the accompanying drawings.
Fig. 1 is a diagram showing the overall architecture of an automatic calibration method for a high-altitude detection sensor based on the internet of things, and the method mainly comprises five parts, namely data acquisition, data transmission, sensor calibration, data storage and data display.
Data acquisition
For the temperature sensor, 6 calibration points are arranged according to the characteristic curve of the sensor, and the distribution is-80 ℃, 50 ℃, 30 ℃, 10 ℃, 30 ℃ and 50 ℃, and the distribution corresponds to 6 temperature control boxes. The time for each temperature control box to reach the temperature stabilization is 1 hour, and according to the actual condition, the temperature of the constant temperature box is kept unchanged, and the sensor is moved to different environmental temperatures to finish calibration. The calibration threshold is set to be 0.1 ℃, when the temperature of the incubator is stabilized at the calibration point +/-0.1 ℃, the data of the temperature sensor is acquired, and the specific acquisition flow is shown in figure 2.
For calibration of the humidity sensor, a constant-temperature humidity-changing tube is adopted for humidity control, and 4 humidity calibration points such as 8% RH, 25% RH, 75% RH, 95% RH and the like are selected according to a characteristic curve of the humidity sensor. The collection process is as shown in fig. 3, firstly setting a humidity value of the calibration and calibration device and a threshold range of 0.01RH when the humidity value is stable, then delaying for 5 minutes to wait for the device to become wet, then reading the current humidity value every 20 seconds to judge whether the humidity tube reaches a stable condition, and collecting the humidity sensor data when the humidity of the device is kept at the calibration point ± 0.01 RH.
For the air pressure sensor, because the sensor is influenced by two variables of temperature and air pressure, the control variable method is used for collecting sensor data, the collected temperature range is 40 ℃, 30 ℃, 20 ℃, 10 ℃, 0 ℃, 10 ℃, 20 ℃ and 30 ℃, and 8 temperature calibration points such as 1070hPa, 1050hPa, 1000hPa, 900hPa, 800hPa, 700hPa, 650hPa, 600hPa, 500hPa, 400 hPa, 300hPa, 200hPa, 150hPa, 100hPa, 70hPa, 50hPa, 30hPa, 10hPa, 50hPa and the like. The acquisition flow is as shown in fig. 4, and first, a calibration acquisition point threshold range is set, which includes a temperature threshold: ± 0.1 ℃, air pressure threshold: ± 0.1hPa, and then a calibration temperature value is set. After the setting is finished, the temperature value of the calibration equipment is read every 60 seconds to judge whether the temperature reaches a stable condition. And when the temperature is stable, setting the air pressure value of the calibration equipment, and reading data every 10 seconds to judge whether the air pressure is stable. And when the temperature and the air pressure meet the calibration environmental requirements, collecting the measurement data of the air pressure sensor. And after the collection is finished, changing the air pressure value in the calibration equipment, and entering the next air pressure calibration point.
(II) data transmission
The invention adopts UDP protocol to transmit data, and the IP address allocation condition of each device is shown in table 1. In the process of equipment control and data transmission, the communication protocol is specified as follows:
A. reading a temperature controller data instruction: 0xa0
Temperature controller data return: 0x000x180x330x02
Protocol analysis: byte 1 represents a positive or negative number, where 0x00 represents a negative number, 0x01 represents a positive number, and bytes 2, 3, and 4 represent temperature values: it is necessary to convert hexadecimal data into decimal from the lower order and perform the combination. The following expression:
0x02＝02
0x33＝51
0x18＝24
incubator data read was +025.124 deg.C
B. Setting a numerical value command of a temperature controller: 0xb0+ temperature value to be controlled
If the temperature of the thermostatic bath is set to be +015.112, the thermostatic bath is split 015112 and is converted into hexadecimal, the lower position is sent first, and the upper position is sent. The temperature is controlled to be positive, the first byte is 0x00, the last three bytes are firstly sequenced 125101 (decimal), and then converted into hexadecimal. It sends a control command for the thermostat at 0xb00x000x0c 0x330x 01.
C. Reading a data instruction of the air pressure controller: PS? The r/n air pressure controller data is returned to 1004.402/r/n, namely the current air pressure is 1004.402 hPa.
D. Setting a data instruction of an air pressure controller: if the control 1070hPa sends SET:1070000# \ r \ n
E. Reading humidity controller instructions: FETC? (@ Uw1) \ r \ n humidity controller returns data: uw 1-53.8 \ r \ n, namely the current humidity is 53.8%
F. Setting a humidity controller instruction: if the control humidity point is 92%, then send: TARGET UWA 92 humidity controller returns data: UWA 92.0\ r \ n, indicating successful setup.
(III) calibration of sensor
And constructing a sensor calibration model based on the BP network according to the measured sensor calibration data. As shown in fig. 5, the calibration model of the air pressure sensor is shown, wherein the input layer has 2 neurons, which are respectively an air pressure original value and a temperature original value, the hidden layer has 10 neurons, the wavelet function is used as an activation function to perform characteristic curve fitting of the air pressure sensor, and the output layer has 1 neuron, which represents the corrected air pressure value.
In order to enable the training result to correctly reflect the intrinsic rules of the sample and avoid overfitting, all collected data are divided into a training set and a testing set according to the ratio of 8: 2. And constructing a neural network model through the training set data, and detecting the correction effect of the model through the test set. The training flow chart is shown in fig. 6, and the main training process is as follows:
1) and carrying out normalization processing on input training sample data, eliminating abnormal values and missing values, completing network creation and initializing relevant parameters.
2) And calculating a weight w between the input layer and the hidden layer and an output value v of the hidden layer.
3) And calculating the weight w and the network output value y between the hidden layer and the output layer.
4) And calculating an error function J and adjusting the weight of each layer.
5) And judging whether the error meets the condition J and is not more than epsilon, and if not, repeating the training process until the condition is met.
In order to verify whether the established calibration model can accurately realize the calibration of the air pressure sensor, a network with the minimum error in the training model is selected for prediction, calculation is carried out by inputting test set data, and the error value is calculated by comparing the data with corresponding data under a calibration point. And if the error exceeds the allowable range, dividing the data again, performing network training, and if the error is within the allowable range, finishing the correction of the air pressure sensor.
The calibration models of the temperature sensor and the humidity sensor network are shown in fig. 7 and 8, wherein the calibration data division, the model training and the model testing processes are similar to those of the barometric sensor, and are not repeated.
(IV) data storage
In the method for automatically calibrating the high altitude detection sensor based on the internet of things, the MySQL database is adopted for data storage, in order to reduce the entry times of the MySQL database and improve the CPU utilization rate, the cache library Redis is introduced, the cache threshold is set to 15000 lines, the data exchange speed is accelerated, and the data storage flow is shown in figure 9. In order to facilitate reading and writing of the acquired data, corresponding sensor storage data table structures are designed, and as shown in tables 2 to 4, retrieval of different sensor types and different parameters is achieved.
(V) data presentation
The website home page interface is shown in fig. 10, and a current calibration information overview and historical data statistical function are designed, the calibration quantity of temperature, humidity and air pressure sensors and the corresponding calibration total progress are displayed in the current collection, the calibration historical records can be checked in the historical information preview, and data retrieval is performed according to the sensor type and the calibration time.
The equipment management interface mainly comprises an equipment management information table and a self-test information query table. In the device management information table shown in fig. 11, the number of the acquisition board is first selected to determine the position of the acquisition table, then the number of the corresponding channel sensor is entered, and after the number is input, the "save number" button is clicked, so that the input information can be saved in the database. Before calibration, a calibration temperature threshold, a calibration air pressure threshold and a calibration air pressure gradient are required to be set in a management form, and data conditions required to be met in a stable state are specified. The calibration points of the air pressure and the temperature can be clicked to carry out operations of storing calibration point values, storing calibration models, loading calibration models, clearing tables and the like.
The self-test information look-up table shown in fig. 12 shows the air pressure value and the temperature value of the sensor in the atmospheric environment, and the air pressure distribution map and the temperature distribution map are drawn according to the measured values, and the range threshold set by the user is loaded according to the clustering center of the initial measured values, so that the sensor quality can be judged. And when the initial value of the sensor is out of the threshold range, determining that the sensor is unqualified, and displaying the number and the position of the sensor.
The data collection interface of the air pressure sensor is shown in fig. 13, and includes a data collection information table, an air pressure change information table, and a temperature change information table. The data acquisition information table displays the real-time data of the current calibration point and supports the function of exporting the data. The air pressure and temperature change information tables respectively display the corresponding real-time acquisition value, calibration value and acquisition value in the coordinate system.
As shown in fig. 14, the data retest interface includes a data calibration information table and a data retest information table, and is used for visually displaying the respective content comparison graph and the error distribution graph, and the two tables provide a function of deriving data. The calibration air pressure comparison graph and the retest air pressure distribution graph respectively display the air pressure value, the calibration air pressure value and the initial measurement air pressure value in the coordinate system, the error distribution graph displays the errors of the calibration value and the initial measurement value along with the air pressure distribution, and the calibration precision and the calibration effect can be more clearly known through observing the graph distribution.
Table 1 IP address allocation table for each device
TABLE 2 temperature sensor data Structure Table
| Name of field | Type of field | byte/B | Description of the invention |
| time | datatime | 8 | Time of acquisition |
| num | int | 4 | Serial number |
| standard_temp | float | 8 | Standard temperature |
| temp | float | 8 | Actual temperature |
| org_temp | float | 16 | Original value of temperature |
Table 3 humidity sensor data structure table
| Name of field | Type of field | byte/B | Description of the invention |
| time | datatime | 8 | Time of acquisition |
| num | int | 4 | Serial number |
| standard_humidity | float | 8 | Standard humidity |
| humidity | float | 8 | Actual |
| org_humidity | float | ||
| 16 | Original value of humidity |
Table 4 baroceptor data structure table
Claims (8)
1. An automatic calibration method for a high-altitude detection sensor based on the Internet of things is characterized by comprising the following steps: comprises the following steps of (a) carrying out,
step 1: constructing a temperature, humidity and pressure sensor data acquisition system based on the Internet of things;
the Internet of things architecture consists of 3 parts, namely a sensing layer, a transmission layer and an application layer; the sensing layer is arranged in the sensor calibration equipment and is used for temperature, humidity and pressure sensor equipment information and corresponding measurement data; the transmission layer is responsible for transmitting control instructions and sensor data information contents between the sensing layer and the application layer; establishing connection between different devices by adopting a UDP communication protocol, and realizing data receiving, sending and storage based on Socket; the application layer realizes the control of equipment, the storage, the analysis and the display of data, and controls the related parameters of the temperature, humidity and pressure sensors in the acquisition process through different dialog boxes of the application layer;
step 2: designing an operation interface of the Internet of things;
the Internet of things architecture corresponds to a computer software system, the computer software system consists of a navigation bar and a functional area, the navigation bar shows the main functions of the proposed method, and hyperlinks are added to the functions to enable the functions to jump corresponding pages; the functional area is composed of a visual chart, a data dialog box and a data operation button according to the specific operation requirement of the sensor calibration; designing and using operation interfaces, wherein the operation interfaces comprise a home page, an equipment management interface, a sensor self-testing interface, a data acquisition interface and a data retest interface;
and step 3: calibration algorithm for designing temperature, humidity and pressure sensors
Adopting a BP neural network model to calibrate the temperature, humidity and pressure sensors, wherein the BP neural network model comprises an input layer, a hidden layer and an output layer, and the updating speed of BP network parameters is accelerated by introducing a Levenberg-Marquardt algorithm; the calibration precision of the sensor is further improved by designing the hidden layer activation function as a Morlet wavelet function; different network structures are designed according to different sensor influence factors, and the measurement result of the temperature sensor is influenced by the ambient temperature, so that the input layer is 1 neuron, 5 neurons are selected as a network hiding layer, and 1 neuron is selected as a network output layer; the measuring result of the air pressure sensor is influenced by the temperature and the air pressure of the environment, so that the input layer is 2 neurons, 10 neurons are selected as a network hiding layer, and 1 neuron is selected as a network output layer; the measuring result of the humidity sensor is influenced by the environment humidity, so that the input layer is 1 neuron, 10 neurons are selected as a network hiding layer, and 1 neuron is selected as a network output layer; after BP neural network training, the final convergence is achieved, the fitting of the characteristic curves of the temperature, humidity and pressure sensors is realized, and the aim of correcting the sensors is fulfilled.
2. The automatic calibration method for the high altitude detection sensor based on the Internet of things as claimed in claim 1, wherein the method comprises the following steps: the home page comprises a current calibration information overview and a historical data statistical function, the calibration quantity of temperature, humidity and air pressure sensors and a corresponding calibration total process are displayed in current acquisition, calibration historical records can be checked in historical information preview, and data retrieval is carried out according to the type of the sensors and the calibration time.
3. The automatic calibration method for the high altitude detection sensor based on the Internet of things as claimed in claim 1, wherein the method comprises the following steps: in the equipment management information table, selecting a collecting plate coding position, inputting a sensor number, setting a calibration environment stability threshold value condition, and setting different sensor calibration point distribution.
4. The automatic calibration method for the high altitude detection sensor based on the Internet of things as claimed in claim 1, wherein the method comprises the following steps: displaying the air pressure value and the temperature value of the sensor in the atmospheric environment in a self-test information look-up table, drawing a primary air pressure distribution diagram and a primary temperature distribution diagram according to the measured values, calculating a clustering center by using a K-mean method, and judging the quality of the sensor according to a set threshold value; displaying the acquired data information in real time on a data acquisition interface, drawing a temperature change information table, a humidity change information table and an air pressure change information table according to the acquired data, and monitoring the acquisition condition; and the data retest interface comprises sensor calibration data information, sensor retest data information, a corresponding data comparison graph and an error distribution graph, and is used for performing qualification judgment on the calibration data and the retest data and providing a data export function.
5. The automatic calibration method for the high altitude detection sensor based on the Internet of things as claimed in claim 1, wherein the method comprises the following steps: according to the characteristic curves of the temperature sensor, the humidity sensor and the pressure sensor, corresponding calibration point distribution is designed, and corresponding environment waiting time and data acquisition strategies are formulated by combining the actual conditions of calibration equipment.
6. The automatic calibration method for the high altitude detection sensor based on the Internet of things as claimed in claim 1, wherein the method comprises the following steps: and data transmission is carried out by adopting a UDP (user Datagram protocol), IP addresses and port numbers are distributed for each device, a communication protocol between the temperature, humidity and pressure sensors and the corresponding calibration device is specified, the control of the calibration device and the acquisition of the measurement data of the sensors are realized.
7. The automatic calibration method for the high altitude detection sensor based on the Internet of things as claimed in claim 1, wherein the method comprises the following steps: dividing the collected calibration data into a training set and a testing set according to the ratio of 8:2, constructing corresponding BP network model structures for the temperature, humidity and pressure sensors, adopting a Morlet wavelet activation function for a network hiding layer, and updating network parameters by using a Levenberg-Marquardt algorithm.
8. The automatic calibration method for the high altitude detection sensor based on the Internet of things as claimed in claim 1, wherein the method comprises the following steps: the method comprises the steps of caching and storing sensor data by adopting MySQL + Redis, and designing a temperature, humidity and pressure sensor data table structure which comprises acquisition time, a sequence number, standard temperature, standard humidity, standard air pressure, a temperature original value, a humidity original value and an air pressure original value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911175498.4A CN111076758A (en) | 2019-11-26 | 2019-11-26 | Automatic calibration method for high-altitude detection sensor based on Internet of things |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911175498.4A CN111076758A (en) | 2019-11-26 | 2019-11-26 | Automatic calibration method for high-altitude detection sensor based on Internet of things |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111076758A true CN111076758A (en) | 2020-04-28 |
Family
ID=70311804
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911175498.4A Pending CN111076758A (en) | 2019-11-26 | 2019-11-26 | Automatic calibration method for high-altitude detection sensor based on Internet of things |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111076758A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112990442A (en) * | 2021-04-21 | 2021-06-18 | 北京瑞莱智慧科技有限公司 | Data determination method and device based on spatial position and electronic equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104237457A (en) * | 2014-05-27 | 2014-12-24 | 李岩 | Air quality monitoring method and system |
| CN105258839A (en) * | 2015-10-30 | 2016-01-20 | 南京信息工程大学 | Array type air pressure measurement compensation device and method based on quantum particle swarm wavelet neural network |
| CN105987775A (en) * | 2016-07-20 | 2016-10-05 | 天津理工大学中环信息学院 | Temperature sensor nonlinearity correction method and system based on BP neural network |
| CN206684327U (en) * | 2017-04-06 | 2017-11-28 | 天津市滨海新区气象局 | Distributed low-power consumption weather station system based on Internet of Things |
| CN109724692A (en) * | 2018-12-29 | 2019-05-07 | 华南农业大学 | A kind of illuminance sensor bearing calibration, storage medium, calculates equipment at system |
-
2019
- 2019-11-26 CN CN201911175498.4A patent/CN111076758A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104237457A (en) * | 2014-05-27 | 2014-12-24 | 李岩 | Air quality monitoring method and system |
| CN105258839A (en) * | 2015-10-30 | 2016-01-20 | 南京信息工程大学 | Array type air pressure measurement compensation device and method based on quantum particle swarm wavelet neural network |
| CN105987775A (en) * | 2016-07-20 | 2016-10-05 | 天津理工大学中环信息学院 | Temperature sensor nonlinearity correction method and system based on BP neural network |
| CN206684327U (en) * | 2017-04-06 | 2017-11-28 | 天津市滨海新区气象局 | Distributed low-power consumption weather station system based on Internet of Things |
| CN109724692A (en) * | 2018-12-29 | 2019-05-07 | 华南农业大学 | A kind of illuminance sensor bearing calibration, storage medium, calculates equipment at system |
Non-Patent Citations (1)
| Title |
|---|
| 张晗: "基于物联网的传感器校正方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112990442A (en) * | 2021-04-21 | 2021-06-18 | 北京瑞莱智慧科技有限公司 | Data determination method and device based on spatial position and electronic equipment |
| CN112990442B (en) * | 2021-04-21 | 2021-08-06 | 北京瑞莱智慧科技有限公司 | Data determination method and device based on spatial position and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111126662B (en) | Irrigation decision making method, device, server and medium based on big data | |
| TWI738016B (en) | Intelligent learning mechanism and adaptive regulation system for liquid gas filling | |
| CN109856973A (en) | A kind of Technique for Controlling Greenhouse Environment and system based on fuzzy neural network | |
| CN115271000B (en) | State monitoring method and system for cable tunnel | |
| CN111076758A (en) | Automatic calibration method for high-altitude detection sensor based on Internet of things | |
| CN114994261A (en) | Remote calibration method for sensor for ambient air monitoring | |
| CN113065223A (en) | Multi-level probability correction method for digital twin model of tower mast cluster | |
| CN115809553A (en) | Temperature and humidity calculation method and terminal in substation protection cell | |
| CN112985692B (en) | Atmospheric pressure sensor error calibration method integrating polynomial and learning model | |
| CN114818888A (en) | Soil composition data fusion method and system based on multi-channel Kalman filtering | |
| CN115808211A (en) | Public building temperature and thermal comfort monitoring and predicting system | |
| CN111982299A (en) | Dynamic data quality scoring method and system for satellite-borne microwave radiometer | |
| CN114459523B (en) | Calibration early warning method of online quality detection instrument | |
| CN112649481B (en) | Park TVOC monitoring method using regional TVOC array monitoring system | |
| CN112507526B (en) | Performance prediction method and system for proton exchange fuel cell system | |
| EP3955587A1 (en) | Machine learning device | |
| CN113222268A (en) | Multi-mode reasoning-based tobacco baking quality prediction model establishing method | |
| CN113108949A (en) | Model fusion-based sonde temperature sensor error prediction method | |
| CN117664441A (en) | Aging detection method and system for pressure transmitter | |
| CN115459451B (en) | Intelligent communication method of switching value sensor based on power supply loop | |
| CN113349020B (en) | Method and device for accurately watering greenhouse vegetables and electronic equipment | |
| CN113344072B (en) | GPB algorithm-based operation and maintenance multi-mode decision method and system and cloud server | |
| CN109443449A (en) | A kind of green house of vegetables temperature and humidity monitoring system | |
| CN115022348B (en) | Intelligent factory cloud-level architecture data storage method for high-end battery | |
| CN113255117B (en) | Reliability digital twinning model of liquid level volatile product |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200428 |