CN109631973B - Automatic calibration method and system for sensor - Google Patents
Automatic calibration method and system for sensor Download PDFInfo
- Publication number
- CN109631973B CN109631973B CN201811459123.6A CN201811459123A CN109631973B CN 109631973 B CN109631973 B CN 109631973B CN 201811459123 A CN201811459123 A CN 201811459123A CN 109631973 B CN109631973 B CN 109631973B
- Authority
- CN
- China
- Prior art keywords
- sensor
- data
- calibration
- module
- sensors
- 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
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
Abstract
The invention provides an automatic calibration system and method of a sensor, wherein the sensor is used for collecting sensor data and uploading the data to a gateway; the gateway transmits the sensor data uploaded by the sensor to the cloud platform; the standard metering module is used for acquiring real value data of sensor data corresponding to each sensor and transmitting the real value data to the cloud platform; the cloud platform is used for transmitting the obtained sensor data and real value data to the big data platform; the big data platform is used for storing and transmitting the obtained sensor data, real value data and other condition data which are automatically counted and influence the accuracy of the sensor to the calibration module; the calibration module is used for analyzing the obtained data to obtain a calibration model corresponding to the sensor calibrated by the target or obtaining the calibration model by machine learning; the calibration model is used for calibrating data during sensor calibration and outputting a calibrated result.
Description
Technical Field
The invention relates to the field of automatic calibration, in particular to a method and a system for automatically calibrating a sensor.
Background
With the rapid development of the field of internet of things, the application of sensors is more extensive and goes deep into various industries, a large number of sensors are applied to various different environments, different environments can affect electronic devices of the sensors to a certain degree, and the temperature, the humidity, the electromagnetic field, the service life of the devices and the like can affect the sensors, so that the data of the sensors are inaccurate, and even deviate from actual values more and more. Calibration of the sensor is therefore also increasingly important. However, even the same sensor devices are lack of consistency, and for sensors with poor consistency, each batch or even each finished sensor needs to be verified, and at present, most sensor calibration methods require field manual verification through a certain device system, which is troublesome in operation and labor-consuming. In actual use, due to the change of environment, the sensor needs to be recalibrated due to the change of each external condition, which is very inconvenient.
The following describes a sensor calibration method and apparatus commonly used in the prior art by means of several specific schemes.
One is to perform a simple value calibration. For example, patent application No. 201310722409.X, "sensor calibration apparatus and method", the apparatus comprising: an I2C interface connected to the sensor via an I2C signal line; and a calibration module in communication with the sensor via the I2C interface, the calibration module comprising: a calibration kernel capable of providing calibration values to the sensor to calibrate the output of the sensor; a start value register that stores an initialization calibration value, which the calibration core writes into the sensor as a calibration value via the I2C interface; a target value register storing a desired output value of the sensor; and the tolerance register is used for storing a tolerance value, wherein the calibration kernel reads the output value of the sensor after calibration based on the calibration value and judges whether the difference between the output value and the expected output value is within the tolerance value, if so, the calibration of the sensor is successfully completed, otherwise, the calibration value is modified and a new calibration value is written into the sensor until the difference between the output value of the sensor and the expected output value is within the tolerance value.
This type of method belongs to a simple value adjustment calibration method, and has the limitations that:
1. the calibration value needs to be continuously adjusted, the effect on a simple linear sensor is good, but for a nonlinear sensor, the sensor which does not meet the linear rule cannot be calibrated
2. The adaptability is poor and recalibration is required for the environment.
The other is calibration by using a generation coefficient table. For example, patent application No. 201710258680.0, "sensor calibration method": numbering the M sensors; calculating a calibration coefficient of each sensor and generating a sensor offset calibration coefficient table including a number of each sensor and the calibration coefficient of each sensor; calculating a calibration coefficient of the combined module when the combined module is combined with each sensor, and generating a combined module deviation calibration coefficient table comprising the serial number of each sensor and the calibration coefficient of the combined module when the combined module is combined with each sensor; generating a comprehensive deviation calibration coefficient table of M sensors and the combined module according to the sensor deviation calibration coefficient table and the combined module deviation calibration coefficient table; and the client side which is provided with the sensor, the coupling module and the comprehensive deviation calibration coefficient table calls the corresponding comprehensive calibration coefficient from the comprehensive deviation calibration coefficient table according to the serial number of the sensor to calibrate the sensor.
This type of method requires the generation of a calibration coefficient for each sensor, with the limitations that:
1. for a complex nonlinear sensor, a calibration formula is complicated and changeable with the environment, and the sensor is difficult to be accurately calibrated only through adjustment of coefficients.
2. Without the concept of an automated system, calibrating the numbers for each sensor one-by-one is inefficient.
In summary, the disadvantages of the calibration method and apparatus for the sensor in the prior art include the following aspects:
firstly, one-sidedness of the method: only the sensor calibration meeting a certain specific formula rule is supported, and the sensor with a complex and changeable calibration formula along with the change of external conditions cannot be calibrated.
Secondly, the sensor has no self-adaptability, and needs to be recalibrated after being used for a period of time or external conditions are changed.
And thirdly, the automation is lacked, the dependence on manpower is large, the calibration cannot adapt to variable environments, and once the environments change, the recalibration is needed.
And fourthly, systematization is not carried out, valuable data such as sampling data, calibration history and the like are not recorded, tracing is not carried out, and a large amount of meaningful analysis is not carried out.
Disclosure of Invention
The invention aims to provide an automatic calibration method and system of a sensor, which are used for solving the technical problems that the traditional sensor calibration is high in one-sided performance, poor in adaptability, low in automation degree and incapable of recording data.
In order to achieve the above purpose, the invention provides the following technical scheme:
an automatic calibration system of a sensor comprises a sensor terminal, a gateway, a cloud platform, a standard metering module, a calibration module and a big data platform;
the sensor terminal comprises a plurality of sensors, the sensors are target calibration objects, and the sensors are used for acquiring sensor data and uploading the sensor data to the gateway;
the gateway is used for connecting the sensor and the cloud platform through a network and transmitting sensor data uploaded by the sensor to the cloud platform;
the standard metering module, the calibration module and the big data platform are all connected to the cloud platform;
the standard metering module is used for acquiring real value data of sensor data corresponding to each sensor and transmitting the real value data to the cloud platform;
the cloud platform is used for transmitting the obtained sensor data and real value data to the big data platform;
the big data platform is used for storing and transmitting the obtained sensor data, real value data and other condition data which are automatically counted and influence the accuracy of the sensor to the calibration module;
the calibration module is used for analyzing the obtained sensor data, the real value data and other condition data which are automatically counted and influence the accuracy of the sensor to obtain a calibration model corresponding to the sensor for target calibration;
the calibration model is used for calibrating data during sensor calibration and outputting a calibrated result.
Further, in the present invention, a part of the sensors in the sensor terminal constitute a basic environment sensor, and the basic environment sensor is a sensor corresponding to a physical parameter that is preselected to affect the accuracy of other sensors.
Further, in the present invention, the calibration module includes a data obtaining module, a data filtering module and a data modeling module;
the data acquisition module is used for acquiring data from a big data platform;
the data filtering module is used for filtering and screening the data in the acquired data module to remove abnormal data and repeated data;
the data modeling module is used for generating a correlation condition list, a calibration ordered list and a calibration model for a target calibrated sensor;
the correlation condition list consists of sensors or other condition data corresponding to physical parameters which are screened from basic environment sensors and other condition data by a variable correlation calculation method and influence the accuracy of the sensors for target calibration;
the calibration ordered list is the influence degree of each sensor on the accuracy of the sensors in the whole system, which is ordered by referring to the relevance list corresponding to each target calibrated sensor, and is ordered in the order of the influence degree of the accuracy of the sensors in the whole system;
the calibration model is used to calibrate the target calibrated sensor in the order of the calibration ordered list and with reference to the real value data.
Further, in the invention, the system further comprises a machine learning module, wherein the machine learning module is connected to the big data platform and the calibration module;
the machine learning module is used for starting and generating a new calibration model to replace the calibration model generated by the calibration module when the accuracy of the calibration model generated by the calibration module is low;
the machine learning module is used for obtaining a correlation condition list of a target calibrated sensor from the calibration module, obtaining sensor data in the correlation list from a big data platform, and performing characterization processing on the sensor data; meanwhile, the machine learning module is used for obtaining sensor data, real value data and other condition data in the correlation condition list of the target calibration sensor from the big data platform;
and the machine learning module performs machine learning based on the obtained data to obtain a calibration model of the sensor for target calibration.
Further, in the present invention, the machine learning module learns according to the following method:
step 1, referring to a correlation condition list, simulating different environmental conditions for a sensor calibrated by a target;
step 2, recording sensor data of the sensor calibrated by the target under each environmental condition within a period of time, solving characteristic values of the sensor data, and simultaneously obtaining real values under corresponding environmental conditions;
and 3, performing machine learning by referring to the real value and the characteristic value to obtain a calibration model.
Further, in the present invention, the accuracy index of the calibration model generated by the calibration module itself includes one or more of a sum variance, a sum of squared errors, a mean square deviation, a root mean square, a standard deviation, and a determination coefficient between the calibration result and the true value.
Further, in the present invention, the calibration model is stored on the sensor terminal and/or the cloud platform.
The invention also discloses an automatic calibration method of the sensor, which is consistent with the basic idea of the system and is introduced as follows.
The sensor collects sensor data and uploads the sensor data to the gateway;
the gateway carries out network connection on the sensor and the cloud platform and transmits sensor data uploaded by the sensor to the cloud platform;
the standard metering module acquires real value data of sensor data corresponding to each sensor and transmits the real value data to the cloud platform;
the cloud platform transmits the obtained sensor data and the real value data to the big data platform;
the big data platform stores the obtained sensor data, real value data and other condition data which are automatically counted and influence the accuracy of the sensor, and transmits the data to the calibration module;
the calibration module analyzes the obtained sensor data, the real value data and other condition data which are automatically counted and influence the accuracy of the sensor to obtain a calibration model corresponding to the sensor for target calibration;
the calibration model calibrates data during sensor calibration and outputs a calibrated result.
Further, in the above method, the calibration module includes a data acquisition module, a data filtering module, and a data modeling module;
the data acquisition module acquires data from a big data platform;
the data filtering module filters and screens the data in the acquired data module to remove abnormal data and repeated data;
the data modeling module generates a correlation condition list, a calibration ordered list and a calibration model for a target calibrated sensor;
the correlation condition list consists of sensors or other condition data corresponding to physical parameters which are screened from basic environment sensors and other condition data by a variable correlation calculation method and influence the accuracy of the sensors for target calibration; the basic environment sensor is a sensor corresponding to a physical parameter which is pre-selected from a sensor terminal and influences the accuracy of other sensors;
the calibration ordered list is the influence degree of each sensor on the accuracy of the sensors in the whole system, which is ordered by referring to the relevance list corresponding to each target calibrated sensor, and is ordered in the order of the influence degree of the accuracy of the sensors in the whole system;
and the calibration model calibrates the target calibrated sensor according to the sequence of the calibration ordered list and by referring to the real value data.
Further, in the above method, when the accuracy of the calibration model generated by the calibration module itself is low, the machine learning module is enabled to generate a new calibration model to replace the calibration model generated by the calibration module itself;
the machine learning module obtains a correlation condition list of a target calibrated sensor from the calibration module, obtains sensor data in the correlation list from a big data platform, and performs characterization processing on the sensor data; meanwhile, the machine learning module obtains sensor data, real value data and other condition data in the correlation condition list of the sensor of the target calibration from a big data platform;
and the machine learning module performs machine learning based on the obtained data to obtain a calibration model of the sensor for target calibration.
Has the advantages that:
according to the technical scheme, the invention provides a set of sensor detection method and system, and supports calibration of various sensors, and compared with the traditional technical scheme, the technical scheme of the invention has the following advantages:
1. the sensor calibration device can support automatic calibration of various complex and changeable sensors and has universality.
2. The method and the system support batch calibration of the sensors, can calibrate the sensors in large batch at the same time, and save time and labor.
3. The method and the system can generate calibration models for various sensors, various linear and nonlinear regular sensors, various directly integrated sensor components or integrated sensor finished products can support, and the calibration models of the sensors can be automatically adjusted according to the change of the environment.
4. The method is simple to operate, the system automatically collects data, calculates data and generates a calibration model, the model can be directly stored in the sensor terminal, the model can also be stored in the cloud, and offline and online use is supported.
It should be understood that all combinations of the foregoing concepts and additional concepts described in greater detail below can be considered as part of the inventive subject matter of this disclosure unless such concepts are mutually inconsistent.
The foregoing and other aspects, embodiments and features of the present teachings can be more fully understood from the following description taken in conjunction with the accompanying drawings. Additional aspects of the present invention, such as features and/or advantages of exemplary embodiments, will be apparent from the description which follows, or may be learned by practice of specific embodiments in accordance with the teachings of the present invention.
Drawings
The drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component that is illustrated in various figures may be represented by a like numeral. For purposes of clarity, not every component may be labeled in every drawing. Embodiments of various aspects of the present invention will now be described, by way of example, with reference to the accompanying drawings, in which:
FIG. 1 is a schematic diagram of the system of the present invention;
FIG. 2 is a functional block diagram of a calibration module according to the present invention;
FIG. 3 is a functional block diagram of the machine learning module of the present invention;
FIG. 4 is a schematic diagram of a complete flow of the calibration module of the present invention;
FIG. 5 is a schematic diagram of the entire system calibration process according to the present invention.
Detailed Description
In order to better understand the technical content of the present invention, specific embodiments are described below with reference to the accompanying drawings.
In this disclosure, aspects of the present invention are described with reference to the accompanying drawings, in which a number of illustrative embodiments are shown. Embodiments of the present disclosure are not necessarily intended to include all aspects of the invention. It should be appreciated that the various concepts and embodiments described above, as well as those described in greater detail below, may be implemented in any of numerous ways, as the disclosed concepts and embodiments are not limited to any one implementation. In addition, some aspects of the present disclosure may be used alone, or in any suitable combination with other aspects of the present disclosure.
The automatic calibration system of the sensor shown in fig. 1 includes a sensor terminal, a gateway, a cloud platform, a standard metering module, a calibration module, a big data platform, and a management control end.
The sensor terminal comprises a plurality of sensors, and the types of the sensors can be various, such as temperature and humidity, light brightness, color, formaldehyde, CO2, CO, PM2.5, PM10, TVOC, current and the like. The sensor terminal can be an integrated sensor component and can also be an integrated sensor finished product. The difference lies in that the components are integrated, the MCU of the terminal needs to collect signals, the signal values are converted into measurement values, and the integrated finished product can directly take the measurement values.
The sensor is a target calibration object and is used for acquiring sensor data and uploading the sensor data to the gateway.
The gateway is used for connecting the sensor and the cloud platform through a network and transmitting sensor data uploaded by the sensor to the cloud platform.
Of course, the sensor terminal can be provided with a gateway function according to the requirement, and then the sensor data can be directly transmitted to the cloud platform.
And the standard metering module, the calibration module, the big data platform and the management control end are all connected to the cloud platform.
The standard metering module is integrated with a peripheral standard metering instrument and used for acquiring real value data of sensor data corresponding to each sensor and transmitting the real value data to the cloud platform. The standard metering module is only used during calibration, and the sensor is not required to be actually used.
And the cloud platform is used for transmitting the obtained sensor data and the real value data to the big data platform.
The big data platform is used for transmitting the obtained sensor data, real value data and other condition data (including but not limited to sensor use duration, running duration and the like) which are automatically counted and influence the accuracy of the sensor to the calibration module, and meanwhile, the data are also stored on the big data platform.
The calibration module is a core part of the specific embodiment of the present invention, and is configured to analyze the obtained sensor data, real value data, and other condition data that are automatically counted and affect the accuracy of the sensor, and obtain a calibration model corresponding to the sensor to be calibrated.
The calibration model is used for calibrating data during sensor calibration and outputting a calibrated result.
The management control end is used for providing the user to check each sensor and providing convenience for the calibration of each sensor.
Because the number of types of sensors in the whole system is large, and the factors influencing the accuracy of the sensors are large, the calibration of the sensors can be called as moving the whole body in a single movement. Therefore, how to reasonably and effectively calibrate the sensors in the whole system is a problem to be practically solved by the invention.
In view of the above problems, the applicant tried to find the influence weights of different sensors on the system based on the influence factors of the sensors, thereby adjusting the sensors in order of the magnitude of the weights, and ensuring efficient calibration of the whole system.
Therefore, the embodiment of the present invention proposes the concept of "basic environment sensor", that is, a part of the sensors in the sensor terminal constitute the basic environment sensor, and the basic environment sensor is a sensor corresponding to a preselected physical parameter which affects the accuracy of other sensors.
For example, temperature has an effect on electronic components, and therefore temperature necessarily has an effect on the accuracy of many sensors, so that a sensor related to the detected temperature, i.e., a temperature sensor, can be classified as a basic environmental sensor.
The specific basic environment sensors can be flexibly selected and combined according to the requirements of the sensors. To put it back, if the choice of the underlying environmental sensors cannot be determined, then all sensors with possible correlation can be included, with the algorithms and processes unchanged except for the increase in data samples that need to be collected and processed.
The combination of the basic environmental sensors described above is to select a corresponding list of correlation conditions for each particular sensor in turn.
This requires the use of a calibration module, which is described in detail below.
As shown in fig. 2, 3 and 4, the calibration module includes an acquisition data module, a data filtering module and a data modeling module.
The data acquisition module is used for acquiring data from a big data platform, namely sensor data, real value data and other condition data which are automatically counted and influence the accuracy of the sensor.
The data filtering module is used for filtering and screening the data in the data acquisition module to remove abnormal data and repeated data and obtain an effective data set.
The data modeling module is used for generating a correlation condition list, a calibration ordered list and a calibration model for the target calibrated sensor.
The list of correlation conditions is obtained by a correlation condition selection module in the data modeling module, which is composed of sensor or other condition data (including but not limited to sensor usage duration, operation duration, etc.) corresponding to physical parameters selected from basic environmental sensors, other condition data by a variable correlation calculation method (including but not limited to pearson correlation algorithm) that affect the accuracy of the sensor calibrated to the target.
The calibration sorting list needs to perform calibration sorting on different types of sensors in the system, that is, the above sorting is performed according to the influence of the sensors on the system. Namely, the influence degree of each sensor on the accuracy of the sensors in the whole system is sorted by referring to the relevance list corresponding to the sensors calibrated by each target, and the influence degrees are sorted in the order of the accuracy influence magnitude of the sensors in the whole system.
Specifically, for sensor a, there are N sensors in the correlation list for which parameters affecting the accuracy of sensor a correspond, and the effects of these sensors on sensor a have respective weights. Then for the entire system, the weights of the weighted sensors can be accumulated, the degree of influence of the various sensors on the accuracy of the sensors in the entire system is obtained, and the sensors are sorted in order of magnitude of influence.
Of course, the above-mentioned method of accumulating weights includes, but is not limited to, summing the weights of the corresponding sensors distributed in each relevant list, and obtaining the weight of the sensor of this type for the whole system.
The various sensors are sorted in the order of the above-mentioned weight magnitude with respect to the entire system, and the calibration order is performed in this order.
During calibration, fitting is carried out firstly according to the true value data, a straight line or a polynomial curve can be fitted, a calibration model can be directly obtained for a sensor with a good fitting effect, and a machine learning module needs to be called for a sensor with a poor fitting effect to generate the calibration model.
The accuracy index of the calibration model generated by the calibration module comprises one or more of sum variance, error sum of squares, mean square deviation, root mean square, standard deviation and determination coefficient between the calibration result and the real value.
The following is directed to the machine learning module.
The machine learning module is connected to the big data platform and the calibration module.
The machine learning module is used for starting and generating a new calibration model to replace the calibration model generated by the calibration module when the accuracy of the calibration model generated by the calibration module is low.
The machine learning module is used for obtaining a correlation condition list of a target calibrated sensor from the calibration module, obtaining sensor data in the correlation list from a big data platform, and performing characterization processing on the sensor data; and the machine learning module is used for obtaining sensor data, real value data and other condition data in the correlation condition list of the target calibration sensor from the big data platform.
And the machine learning module performs machine learning based on the obtained data to obtain a calibration model of the sensor for target calibration.
The machine learning module learns according to the following method:
step 1, referring to the correlation condition list, simulating different environmental conditions for the sensor calibrated by the target.
And 2, recording the sensor data of the sensor calibrated by the target under each environmental condition within a period of time, solving the characteristic value of the sensor data, and simultaneously obtaining the true value under the corresponding environmental condition.
And 3, performing machine learning by referring to the real value and the characteristic value to obtain a calibration model.
Whether a calibration module or a machine-learned calibration model, the calibration model is stored on the sensor terminal and/or the cloud platform.
During the working of the sensor, the data which is closer to the true value after calibration can be obtained only by transmitting the sensor data acquired by the sensor, the sensor data which affects the sensor and other influencing factor data into the calibration model.
The automatic calibration method of the sensor corresponding to the system of the embodiment of the invention is as follows:
the sensor collects sensor data in a calibration stage and an actual use stage and uploads the sensor data to the gateway.
The gateway is used for carrying out network connection on the sensor and the cloud platform and transmitting the sensor data uploaded by the sensor to the cloud platform.
And the standard metering module acquires real value data of the sensor data corresponding to each sensor in a calibration stage and transmits the real value data to the cloud platform.
And the cloud platform transmits the obtained sensor data and the real value data to the big data platform.
The big data platform passes acquired sensor data, real-world data, and other condition data that is automatically statistical and affects sensor accuracy to the calibration module.
The calibration module analyzes the obtained sensor data, the real value data and other condition data which are automatically counted and influence the accuracy of the sensor, and obtains a calibration model corresponding to the sensor calibrated by the target.
The calibration model calibrates data during sensor calibration and outputs a calibrated result.
Further, in the above method, the calibration module includes a data acquisition module, a data filtering module, and a data modeling module.
And the data acquisition module acquires data from the big data platform.
And the data filtering module filters and screens the data in the acquired data module to remove abnormal data and repeated data.
The data modeling module generates a correlation condition list, a calibration ordered list, and a calibration model for a target calibrated sensor.
The correlation condition list consists of sensors or other condition data corresponding to physical parameters which are screened from basic environment sensors and other condition data by a variable correlation calculation method and influence the accuracy of the sensors for target calibration; the basic environment sensor is a sensor corresponding to a physical parameter which is pre-selected from the sensor terminal and influences the accuracy of other sensors.
The calibration sorting list is the influence degree of each sensor on the accuracy of the sensors in the whole system, which is sorted by referring to the relevance list corresponding to each target calibrated sensor, and the influence degree is sorted in the order of the accuracy influence of the sensors in the whole system.
And the calibration model calibrates the target calibrated sensor according to the sequence of the calibration ordered list and by referring to the real value data.
Preferably, when the accuracy of the calibration model generated by the calibration module is low, the machine learning module is started to generate a new calibration model to replace the calibration model generated by the calibration module.
The machine learning module obtains a correlation condition list of a target calibrated sensor from the calibration module, obtains sensor data in the correlation list from a big data platform, and performs characterization processing on the sensor data; and simultaneously, the machine learning module obtains sensor data, real value data and other condition data in the correlation condition list of the target calibrated sensor from a big data platform.
And the machine learning module performs machine learning based on the obtained data to obtain a calibration model of the sensor for target calibration.
Naturally, in practical applications, the method described above needs to be adapted, and refer to fig. 5.
The user starts a calibration mode, the instruction is issued to the sensor terminal through the cloud platform, the sensor receives the instruction, enters the calibration mode, starts to collect data rapidly, sends the data to the cloud platform, and after the cloud platform receives the data, the cloud platform processes the data, adds current condition data (temperature, humidity and device use duration), and stores the data in the big data platform. Meanwhile, a standard measuring instrument needs to be started to obtain the true value of the sensor at the same moment. For meters that are complex and cannot be integrated into our system, manual intervention is required, such as the true value of formaldehyde. And then, automatically labeling the data by the system according to the time for acquiring the data and the real data by the sensor, starting model training, obtaining a calibration model after the training is finished, storing the model into the cloud platform corresponding to the sensor, wherein the sensor has a unique ID in the system, and the calibration model of the sensor can be searched according to the unique ID, so that the online calibration at the cloud end can be realized. Of course, the calibration model may also be issued to the sensor terminal through the cloud platform, and the sensor itself records the calibration. And when the calibration process is finished, the system issues an instruction to enable the sensor to enter a working mode, and after the sensor acquires data, the calibration model is used for calibrating the data, and a calibrated result is output.
The system and method described above is applied to, for example, a school classroom room environment system, which is calibrated.
The system comprises an air quality detection sensor, a temperature and humidity sensor and a light brightness sensor, and a user can see the current environment state through a management control end. As the sensors are all directly purchased from the existing products, the accuracy of the sensors has certain errors, and the sensors in the whole environment need to be calibrated. The method comprises the steps of integrating a sensor into an automatic calibration system, starting calibration by a user through an interface, starting data acquisition, simultaneously inputting real data by a standard meter, acquiring sensor data at different temperatures and different air quality environments due to the fact that an air quality sensor is sensitive to temperature and the accuracy of the sensor is affected by temperature change, automatically creating a model for the sensor by the trained system after a large amount of data are acquired, and completing automatic calibration. Thus, the sensor can obtain more accurate results under different temperatures. The effect of temperature on the sensor is eliminated.
For another example, a large number of batches of sensors of the same type need to be factory calibrated, and if the sensors are good in consistency, only one sensor needs to be taken and integrated into an automatic calibration system to simulate different environments, collect data and calibrate. If the consistency of the sensors among various batches is not good, but the consistency of the sensors among various batches is good, one sensor is taken from each batch, the sensor is integrated into an automatic calibration system, different environments are simulated, data are collected, and calibration is carried out. If the consistency of the sensors among each batch is not good, all the sensors are taken and integrated into the system, and the sensors are calibrated one by one.
Although the present invention has been described with reference to the preferred embodiments, it is not intended to be limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention. Therefore, the protection scope of the present invention should be determined by the appended claims.
Claims (7)
1. An automatic calibration system for a sensor, comprising: the system comprises a sensor terminal, a gateway, a cloud platform, a standard metering module, a calibration module and a big data platform;
the sensor terminal comprises a plurality of sensors, the sensors are target calibration objects, and the sensors are used for acquiring sensor data and uploading the sensor data to the gateway;
a part of sensors in the sensor terminal form a basic environment sensor, and the basic environment sensor is a sensor corresponding to a preselected physical parameter which influences the accuracy of other sensors;
the gateway is used for connecting the sensor and the cloud platform through a network and transmitting sensor data uploaded by the sensor to the cloud platform;
the standard metering module, the calibration module and the big data platform are all connected to the cloud platform;
the standard metering module is used for acquiring real value data of sensor data corresponding to each sensor and transmitting the real value data to the cloud platform;
the cloud platform is used for transmitting the obtained sensor data and real value data to the big data platform;
the big data platform is used for storing and transmitting the obtained sensor data, real value data and other condition data which are automatically counted and influence the accuracy of the sensor to the calibration module;
the calibration module is used for analyzing the obtained sensor data, the real value data and other condition data which are automatically counted and influence the accuracy of the sensor to obtain a calibration model corresponding to the sensor for target calibration;
the calibration module comprises a data acquisition module, a data filtering module and a data modeling module;
the data acquisition module is used for acquiring data from a big data platform;
the data filtering module is used for filtering and screening the data in the acquired data module to remove abnormal data and repeated data;
the data modeling module is used for generating a correlation condition list, a calibration ordered list and a calibration model for a target calibrated sensor;
the correlation condition list consists of sensors or other condition data corresponding to physical parameters which are screened from basic environment sensors and other condition data by a variable correlation calculation method and influence the accuracy of the sensors for target calibration;
the calibration ordered list is the influence degree of each sensor on the accuracy of the sensors in the whole system, which is ordered by referring to the relevance list corresponding to each target calibrated sensor, and is ordered in the order of the influence degree of the accuracy of the sensors in the whole system;
the calibration model is used for calibrating the sensor calibrated by the target according to the sequence of the calibration ordered list and by referring to the real value data during sensor calibration and outputting a calibrated result.
2. The system for automatic calibration of sensors according to claim 1, characterized in that: the system also comprises a machine learning module, wherein the machine learning module is connected to the big data platform and the calibration module;
the machine learning module is used for starting and generating a new calibration model to replace the calibration model generated by the calibration module when the accuracy of the calibration model generated by the calibration module is low;
the machine learning module is used for obtaining a correlation condition list of a target calibrated sensor from the calibration module, obtaining sensor data in the correlation list from a big data platform, and performing characterization processing on the sensor data; meanwhile, the machine learning module is used for obtaining sensor data, real value data and other condition data in the correlation condition list of the target calibration sensor from the big data platform;
and the machine learning module performs machine learning based on the obtained data to obtain a calibration model of the sensor for target calibration.
3. The system for automatic calibration of sensors according to claim 2, characterized in that: the machine learning module learns according to the following method:
step 1, referring to a correlation condition list, simulating different environmental conditions for a sensor calibrated by a target;
step 2, recording sensor data of the sensor calibrated by the target under each environmental condition within a period of time, solving characteristic values of the sensor data, and simultaneously obtaining real values under corresponding environmental conditions;
and 3, performing machine learning by referring to the real value and the characteristic value to obtain a calibration model.
4. The system for automatic calibration of sensors according to claim 2, characterized in that: the accuracy index of the calibration model generated by the calibration module comprises one or more of sum variance, error sum of squares, mean square deviation, root mean square, standard deviation and determination coefficient between the calibration result and the real value.
5. The system for automatic calibration of a sensor according to any one of claims 1-4, wherein: the calibration model is stored on the sensor terminal and/or the cloud platform.
6. An automatic calibration method for a sensor, characterized by: a sensor in the sensor terminal acquires sensor data and uploads the sensor data to a gateway;
the gateway carries out network connection on the sensor and the cloud platform and transmits sensor data uploaded by the sensor to the cloud platform;
the standard metering module acquires real value data of sensor data corresponding to each sensor and transmits the real value data to the cloud platform;
the cloud platform transmits the obtained sensor data and the real value data to the big data platform;
the big data platform stores the obtained sensor data, real value data and other condition data which are automatically counted and influence the accuracy of the sensor, and transmits the data to the calibration module;
the calibration module analyzes the obtained sensor data, the real value data and other condition data which are automatically counted and influence the accuracy of the sensor to obtain a calibration model corresponding to the sensor for target calibration;
the calibration module comprises a data acquisition module, a data filtering module and a data modeling module;
the data acquisition module acquires data from a big data platform;
the data filtering module filters and screens the data in the acquired data module to remove abnormal data and repeated data;
the data modeling module generates a correlation condition list, a calibration ordered list and a calibration model for a target calibrated sensor;
the correlation condition list consists of sensors or other condition data corresponding to physical parameters which are screened from basic environment sensors and other condition data by a variable correlation calculation method and influence the accuracy of the sensors for target calibration; the basic environment sensor is a sensor corresponding to a physical parameter which is pre-selected from a sensor terminal and influences the accuracy of other sensors;
the calibration ordered list is the influence degree of each sensor on the accuracy of the sensors in the whole system, which is ordered by referring to the relevance list corresponding to each target calibrated sensor, and is ordered in the order of the influence degree of the accuracy of the sensors in the whole system;
and the calibration model calibrates the sensor calibrated by the target according to the sequence of the calibration ordered list and by referring to the real value data during sensor calibration and outputs a calibrated result.
7. The method for automatic calibration of a sensor according to claim 6, characterized in that: starting a machine learning module and generating a new calibration model to replace the calibration model generated by the calibration module when the accuracy of the calibration model generated by the calibration module is low;
the machine learning module obtains a correlation condition list of a target calibrated sensor from the calibration module, obtains sensor data in the correlation list from a big data platform, and performs characterization processing on the sensor data; meanwhile, the machine learning module obtains sensor data, real value data and other condition data in the correlation condition list of the sensor of the target calibration from a big data platform;
and the machine learning module performs machine learning based on the obtained data to obtain a calibration model of the sensor for target calibration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811459123.6A CN109631973B (en) | 2018-11-30 | 2018-11-30 | Automatic calibration method and system for sensor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811459123.6A CN109631973B (en) | 2018-11-30 | 2018-11-30 | Automatic calibration method and system for sensor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109631973A CN109631973A (en) | 2019-04-16 |
| CN109631973B true CN109631973B (en) | 2021-04-16 |
Family
ID=66070373
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811459123.6A Active CN109631973B (en) | 2018-11-30 | 2018-11-30 | Automatic calibration method and system for sensor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109631973B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110084379A (en) * | 2019-05-08 | 2019-08-02 | 东莞德福得精密五金制品有限公司 | The method that calibration instrument is calibrated is treated using artificial intelligence cloud computing |
| CN110514785B (en) * | 2019-07-31 | 2022-02-11 | 北京佳华智联科技有限公司 | Sensor data online calibration method and device, electronic equipment and storage medium |
| CN110672143A (en) * | 2019-11-07 | 2020-01-10 | 徐州陀微传感科技有限公司 | Sensor calibration method |
| CN110782640A (en) * | 2019-11-07 | 2020-02-11 | 成都千嘉科技有限公司 | Online self-calibration gas alarm and calibration method thereof |
| CN111307195A (en) * | 2020-03-11 | 2020-06-19 | 深圳市创维电器科技有限公司 | Universal sensor calibration method, device, equipment and computer readable storage medium |
| CN111596006A (en) * | 2020-04-29 | 2020-08-28 | 北京雪迪龙科技股份有限公司 | Calibration method of atmosphere online monitor and monitor |
| CN111651441B (en) * | 2020-05-11 | 2023-05-09 | 北京小米移动软件有限公司 | Data processing method and device and computer storage medium |
| CN111812181A (en) * | 2020-07-21 | 2020-10-23 | 艾感科技(广东)有限公司 | Remote calibration method and system for gas sensor |
| CN113095389B (en) * | 2021-04-02 | 2023-04-07 | 河北先河环保科技股份有限公司 | Method and system for calibrating monitoring network of gridding sensor |
| CN113256959A (en) * | 2021-06-02 | 2021-08-13 | 常州市工业互联网研究院有限公司 | Industrial data acquisition terminal test system |
| CN113391040B (en) * | 2021-07-12 | 2023-09-15 | 北京清环宜境技术有限公司 | Automatic calibration method for data artificial intelligence of atmospheric micro station |
| CN113945684A (en) * | 2021-10-14 | 2022-01-18 | 中国计量科学研究院 | Big data-based micro air station self-calibration method |
| CN114323104A (en) * | 2021-12-23 | 2022-04-12 | 沈阳环境科学研究院 | Online calibration system of vehicle environment detection sensor |
| CN116718220B (en) * | 2023-08-11 | 2023-10-31 | 南京中旭电子科技有限公司 | Hall sensor calibration method and system based on multidimensional data |
| CN117053838A (en) * | 2023-10-11 | 2023-11-14 | 深圳市潼芯传感科技有限公司 | Ranging result calibration method for ranging sensor |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104483042A (en) * | 2014-12-31 | 2015-04-01 | 四川金网通电子科技有限公司 | Cloud-based sensor numerical value demarcation system and method |
| CN106405007A (en) * | 2016-08-30 | 2017-02-15 | 河北先河环保科技股份有限公司 | Novel calibration method for gas sensors and particulate matter sensors |
| CN106441402A (en) * | 2016-08-31 | 2017-02-22 | 北京众清科技有限公司 | Sensor assembly calibration method, device and system |
| CN106908095A (en) * | 2017-01-09 | 2017-06-30 | 浙江大学 | A kind of extraction of sensing data alignment features and appraisal procedure |
| CN106909972A (en) * | 2017-01-09 | 2017-06-30 | 浙江大学 | A kind of learning method of sensing data calibrating patterns |
| JP2017167999A (en) * | 2016-03-18 | 2017-09-21 | 善郎 水野 | Sensor management system |
| CN107621279A (en) * | 2017-09-12 | 2018-01-23 | 歌尔股份有限公司 | Data processing method, sensing data calibration method and device |
| CN108469273A (en) * | 2018-02-27 | 2018-08-31 | 济宁中科云天环保科技有限公司 | High in the clouds data joint debugging calibration method based on machine learning algorithm |
| 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 |
-
2018
- 2018-11-30 CN CN201811459123.6A patent/CN109631973B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104483042A (en) * | 2014-12-31 | 2015-04-01 | 四川金网通电子科技有限公司 | Cloud-based sensor numerical value demarcation system and method |
| JP2017167999A (en) * | 2016-03-18 | 2017-09-21 | 善郎 水野 | Sensor management system |
| CN106405007A (en) * | 2016-08-30 | 2017-02-15 | 河北先河环保科技股份有限公司 | Novel calibration method for gas sensors and particulate matter sensors |
| CN106441402A (en) * | 2016-08-31 | 2017-02-22 | 北京众清科技有限公司 | Sensor assembly calibration method, device and system |
| CN106908095A (en) * | 2017-01-09 | 2017-06-30 | 浙江大学 | A kind of extraction of sensing data alignment features and appraisal procedure |
| CN106909972A (en) * | 2017-01-09 | 2017-06-30 | 浙江大学 | A kind of learning method of sensing data calibrating patterns |
| CN107621279A (en) * | 2017-09-12 | 2018-01-23 | 歌尔股份有限公司 | Data processing method, sensing data calibration method and device |
| CN108469273A (en) * | 2018-02-27 | 2018-08-31 | 济宁中科云天环保科技有限公司 | High in the clouds data joint debugging calibration method based on machine learning algorithm |
| 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109631973A (en) | 2019-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109631973B (en) | Automatic calibration method and system for sensor | |
| CN103605095B (en) | A kind of method making the adaptive all vector network analyzers of Electronic Calibration part | |
| CN104111436A (en) | Automatic detection system and method for digital multimeter | |
| CN109492709A (en) | Data predication method and device based on mixed model | |
| CN110389290A (en) | The system and method and device of a kind of electronic component noise testing and life appraisal | |
| CN102095831A (en) | Moisture meter management method and system | |
| CN109297534B (en) | Environmental parameter weight determination method and system for evaluating indoor environmental quality | |
| CN114994261A (en) | Remote calibration method for sensor for ambient air monitoring | |
| CN206696358U (en) | A kind of communication cable performance test apparatus | |
| CN105424074A (en) | Sensor calibration system, control method and control device thereof | |
| CN101393240A (en) | Electric network harmonic on-line supervisory circuit based on virtual instrument and working method thereof | |
| CN117332205B (en) | High-precision automatic optimization method and device for temperature compensation resistance of piezoresistor | |
| WO2020107445A1 (en) | Automatic calibration method and system for sensor | |
| CN105842566B (en) | The system and its mapping method of automation mapping digital filter amplitude-versus-frequency curve | |
| CN116678489A (en) | Quality control method, system, equipment and medium for force sensor | |
| CN111561968A (en) | Sensor-based environmental parameter detection method and device and data processing equipment | |
| CN107765202A (en) | The on-line calibration system and method for AC measurment unit in integrated circuit test system | |
| CN107907764B (en) | Detection method and system suitable for intelligent characteristic verification of intelligent instrument | |
| CN104280098B (en) | Ship liquid level sensor method of testing based on Labview | |
| CN113778042B (en) | Industrial control equipment instruction simulation operation communication system based on informatization modeling | |
| US8103476B2 (en) | Abnormal simulation signal analysis methods and abnormal signal simulation analysis module for 4˜20mA instrumental system | |
| CN203149105U (en) | Digital universal meter automatic calibrating device | |
| CN114487976A (en) | Method and system for evaluating traceability uncertainty of MCM electronic transformer calibrator | |
| CN116879121B (en) | Air particulate matter concentration real-time monitoring system based on optical fiber sensing technology | |
| CN206146591U (en) | Based on pressure transmitter capability test 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: An automatic calibration method and system for sensors Effective date of registration: 20220602 Granted publication date: 20210416 Pledgee: Zhejiang Tailong Commercial Bank Co.,Ltd. Suzhou Likou sub branch Pledgor: DATATELLIT (SUZHOU) Inc. Registration number: Y2022320010187 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right |