RU2660026C1 - Method and device for measurement data receiver calibration - Google Patents

Abstract

FIELD: measuring equipment.

SUBSTANCE: technical solution relates to methods for calibrating the measuring means, and more particularly to methods for calibrating the measuring means in application to non-stationary processes. Proposed method for calibrating a measurement data receiver executed on a computer device having access to a database includes: obtaining data from a measurement data receiver; definition of the data type; obtaining from the database the valid values and the minimum number of values for a particular type of data; verification of data in response to the fact that the data included in the range of allowed values are determined; obtaining from a database of models a certain type of data and a process model; data processing, in particular: building a process model, filtering the input signal based on the obtained process model, constructing a model of a measurement data receiver; determine the calibration parameters for the measurement data receiver; save calibration parameters for the measurement data receiver.

EFFECT: improving the accuracy of measurement data receiver calibration for non-stationary processes.

7 cl, 2 dwg

Description

Technical field

The technical solution relates to methods for calibrating measuring instruments. More specifically, to methods for calibrating measuring instruments as applied to non-stationary processes.

State of the art

Irregular chaotic processes are widespread in nature and in technology. These include the turbulent movement of liquids and gases associated with the processes of respiration of living organisms, boiling of liquids, as well as vibrations of buildings, etc., for example, the behavior of temperature during its measurements in turbulent flows of liquids or gases has an irregular and chaotic model, which complicates its accurate measurement.

An extremely accurate description of chaotic processes is necessary to control technical objects of exceptional importance (nuclear power plants, thermal power plants, building vibrations, the state of gas turbines, etc.), in meteorology, medicine, economics and many other areas. Such tasks, as a rule, require an assessment of the stability of the behavior of systems and processes or, contain unidentified elements and relationships. They can be solved by dynamic modeling methods, however, its application to real systems still presents a certain complexity. The limited characteristics of physical measuring instruments (inertia, low sensitivity, high noise level) often make it difficult to obtain reliable information about the process. Therefore, there is an urgent task of increasing the accuracy and reliability of measurement data receivers, for example, temperature sensors, pressure sensors used to obtain the model and characteristics of a non-stationary chaotic process.

Several methods for measuring chaotic processes using dynamic modeling are known in the art.

For example, US Patent Application No. US2009113049 describes the use of embedded dynamic models of resonators, relays, circuit breakers, power stabilizers provided by manufacturers of these electronic components. US Patent No. US5828567 describes the use of autoregression to compare the received physical sensor signal with the stored one and determine its residual life. US Patent Application No. US2015302313 describes a measurement prediction of a high-speed sensor of a wind power plant based on a self-learning model of autoregressive moving average (ARMA).

In autoregressive time series models, autoregressive models with external inputs are distinguished. Autoregressive models with external inputs are models that work as a device, for example, a physical device that converts an external input signal into a ready-made dynamic model at the output. Autoregressive models with external inputs can be either linear (ARX, ARMAX) or nonlinear (NARX, NARMAX). To obtain a reliable model of a non-stationary chaotic process, autoregression of a moving average with external inputs (ARMAX) is used, which allows you to adaptively adapt the resulting dynamic model to the process under study. For example, US Patent No. US5513649 describes the use of the ARMAX model for filtering an electroencephalogram signal, however, the linear model used in it is poorly suited to describe a nonlinear signal from the brain. The same applies to the Kalman filter (used, for example, in the method described in German patent No. DE60106045 for processing an unsteady signal from a sensor) having a local linearization. The Kalman filter cannot adapt to real fast-changing processes and adequately describe them. Especially if it is necessary to build non-linear dynamic models of processes that are in nature and technology, since real processes in nature and technology are non-linear and dynamic in the vast majority of cases. The Fourier transform, wavelets, spline interpolation, models based on differential equations that require cumbersome calculations are also not suitable for describing nonlinear processes.

The NARMAX nonlinear autoregression method allows high-reliability modeling of nonlinear non-stationary processes. Based on the obtained model, signal filtering or calibration of measuring instruments can be carried out. For example, it is known to use NARMAX models made using the least squares (OLS) algorithm for adaptive nonlinear filtering of GSM signals (Costa JP, Pitarque T., Thierry E. Using orthogonal least squares identification for adaptive nonlinear filtering of GSM signals // Acoustics, Speech, and Signal Processing, 1997. ICASSP-97., 1997 IEEE International Conference on. - IEEE, 1997. - T. 3. - S. 2397-2400). The signal is filtered by selecting the appropriate coefficients in the model.

However, as practice has shown, this solution does not allow to achieve the necessary level of filtering - the characteristic features of the signal are smoothed out, but at the same time, undesirable frequencies may not be detected.

Disclosure

The present invention is to improve the accuracy of measurement in irregular chaotic processes, in particular, by increasing the accuracy of calibration of the receiver of the measurement data.

The first object of the present invention is a method for calibrating a measurement data receiver, executed on a computer device having access to a database, including:

- receiving data from the receiver of measurement data;

- definition of the type of data;

- obtaining from the database of the region of permissible values and the minimum number of values for a certain type of data;

- data verification

in response to the fact that the data included in the range of acceptable values is determined;

- obtaining from the database of models of a certain type of data and process model

- data processing, in particular:

- building a process model;

- filtering the input signal based on the obtained process model;

- building a model of the receiver of measurement data;

- determination of calibration parameters for the receiver of measurement data;

- saving calibration parameters for the receiver of measurement data.

An embodiment of the invention is possible in which, at the step of receiving data from a measurement data receiver, receiving data from at least one additional measurement data receiver is performed.

An embodiment of the invention is possible in which the data from the receiver of the measurement data and the data from at least one additional receiver of the measurement data are one type of data.

An embodiment of the invention is possible in which data from a receiver of measurement data and data from at least one additional receiver of measurement data are data of various kinds.

An embodiment of the invention is possible in which data processing is performed at least partially based on a pre-trained machine learning algorithm.

An embodiment of the invention is possible in which the machine learning algorithm is preliminarily trained in constructing a process model, filtering an input signal, and constructing a model of a receiver of measurement data; moreover, as input parameters of the machine learning algorithm, a set of data received from the receiver of measurement data included in the range of permissible values and the type of data are set, and learning the machine learning algorithm is a training with a teacher based on a training sample.

Another object of the present invention is a computer device for calibrating a measurement data receiver, including a communication interface, a processor operatively coupled to a data type determining module, a data verification module, a data processing module and a communication interface, the processor is configured to calibrate the measurement data receiver .

The application of the method is not limited to the calibration of the measurement receivers, it can also be used, in particular, to determine the health and residual life of the measurement receivers, to determine the quality parameters and / or process properties, as well as in other cases obvious to the specialist.

The technical result is an increase in the accuracy of calibration of the receiver of the measurement data by clarifying the model for describing the chaotic process obtained by filtering the signal from the receiver of the measurement data using the non-linear NARMAX autoregressive model.

In the proposed filtering option, the NARMAX method employs the principle of two sliding windows, which allows for a more accurate adjustment of the model for the described process. At the same time, the characteristic features of the process are retained.

The system includes a database (DB) containing process models for various types of data. In the process of calibrating the data receiver or filtering the signal, the current information from the database is loaded.

The proposed method allows you to calibrate complex measuring instruments, for example, heat flux receivers (PTP). The dynamic characteristics of various types of PTP include the transition, pulse-transition, amplitude-and phase-frequency, transition matrices characterizing the thermal bonds in the PTP, the matrix pulse-transfer transfer function that determines the ratios of the inputs U and outputs Y, as well as the transfer function matrix. Each type of anti-TB drug may have differences in performance that should be considered.

The resulting process description models can be applied to determine equipment quality, predict events, and in many other cases.

In the context of the present description, unless clearly indicated otherwise, "electronic device", "computer device", "client device", "server", "remote server" and "computer system" mean hardware and / or system software suitable to solve the corresponding problem. Thus, some non-limiting examples of hardware and / or software include computers (servers, desktop computers, laptops, netbooks, etc.), smartphones, tablets, network equipment (routers, switches, gateways, etc.) and / or their a combination.

In the context of the present description, unless clearly indicated otherwise, “database” means any structured data set that does not depend on the specific structure, database management software, hardware of the computer on which data is stored, used or otherwise found available for use. The database may reside on the same hardware that runs the process that stores or uses the information stored in the database, or it may reside on separate hardware, such as a dedicated server or multiple servers.

In the context of the present description, unless clearly indicated otherwise, the term “module” means software (corresponding to a specific hardware context) that is necessary and sufficient to perform the specific specified (s) function (s).

In the context of the present description, unless specifically indicated otherwise, the term "receiver of measurement data" means one or a group of devices designed to generate a signal of measurement information in a form convenient for transmission, further conversion, processing and (or) storage. Non-limiting options for the receiver of measurement data are electronic sensors designed to measure (control) and convert one physical quantity or several physical quantities into a digital signal. For example, temperature sensors, pressure sensors, vibration sensors, humidity sensors, concentration sensors for various gases.

In the context of the present description, unless specifically indicated otherwise, the term "machine-readable medium" means a medium of absolutely any type and character, including RAM, ROM, disks (CDs, DVDs, floppy disks, hard drives, etc.) , USB flash drives, solid state drives, tape drives, etc.

In the context of the present description, unless clearly indicated otherwise, the terms "model of the data receiver", "process model", "model of the data type" mean a dynamic model.

Additional and / or alternative characteristics, aspects and advantages of embodiments of the present technology will become apparent from the following description, the attached drawings and the attached claims.

Brief Description of the Drawings

The invention is illustrated by drawings, which depict:

in FIG. 1 is a schematic diagram of a system configured to implement the invention;

in FIG. 2 is a flowchart of a method in accordance with an embodiment of the present invention.

The implementation of the invention

In FIG. 1 is a schematic diagram of a system 100 configured in accordance with non-limiting embodiments of the present invention. It is important to keep in mind that the following description of system 100 is a description of illustrative embodiments of the present technology. Thus, the entire following description is presented only as a description of an illustrative example of the present invention. This description is not intended to determine the scope or scope of the present invention. Some useful examples of modifications to the system 100 may also be covered by the following description. The purpose of this is also solely an aid in understanding, and not defining, the scope and scope of the present invention. These modifications are not an exhaustive list, and it will be understood by those skilled in the art that other modifications are possible. In addition, this should not be interpreted so that where examples of modifications have not been set forth, no modifications are possible, and / or that what is described is the only embodiment of this element of the present invention. As will be clear to a person skilled in the art, this is most likely not the case. In addition, it should be borne in mind that the system 100 is in some specific manifestations a fairly simple embodiment of the present invention, and in such cases is presented here in order to facilitate understanding. As will be understood by a person skilled in the art, many embodiments of the present invention will have much greater complexity.

With reference to FIG. 1, system 100 includes a computer device 102, a measurement data receiver 104, and a database 106. Computer device 102 is configured to receive data from the measurement data receiver 104 and exchange data with the database 106. How exactly the computer device 102 is associated with the receiver measurement data 104 and database 106 are not particularly limited.

An embodiment of the present technical solution is possible in which the system 100 includes at least one additional receiver of measurement data (not shown), and the computer device 102 is configured to receive data from at least one additional data receiver (not shown).

An embodiment of the present technical solution is possible in which the type of measurement data receiver 104 and the type of additional measurement data receiver (not shown) are the same.

An embodiment of the present technical solution is possible in which the type of the receiver of the measurement data 104 and the type of the additional receiver of the measurement data (not shown) are different. For example, the measurement data receiver 104 is a temperature sensor measuring a temperature of a steam turbine, and the additional measurement data receiver is a pressure sensor of a steam turbine.

 By way of example, but not limitation, computer device 102 may be coupled to a measurement data receiver 104 and / or database 106 over a wired data network (Ethernet based connection). The computer device 102 may be coupled to a measurement data receiver 104 and / or a database via a wireless communication network (for example, inter alia, a 3G network connection, a 4G network connection, Wireless Fidelity or briefly WiFi®, Bluetooth® and the like).

Alternatively, the computer device 102 may be coupled to a measurement data receiver 104 and receive data from it via a computer-readable medium (not shown) onto which data from the measurement data receiver 104 is previously stored (written) and then data is read on the computer device 102.

An embodiment of the system 100 is possible in which the computer device 102 and the measurement data source 104 are a single device having access to the database 106 and configured to carry out the steps of the proposed method.

The computer device 102 includes a processor 108, at least one communication interface (not shown) configured to receive data from a measurement data receiver 104 and also exchange data with other devices, for example, with at least one additional measurement data receiver, a base data 106, etc.. The processor 108 is operatively coupled to a data type determination unit 110, a data verification unit 112, and a data processing unit 114.

According to a non-limiting embodiment of the system 100 of FIG. 1, modules 110, 112, 114 are part of a computer device 102. However, embodiments of a system 100 are possible in which at least one of the modules 110, 112, 114 is not part of a computer device 102, but, for example, is located on a separate server ( not shown) and connected to the computer device 102 via a data network (not shown). Thus, exactly how the processor 108 of the computer device 102 is functionally connected with the modules 110, 112, 114 is not a limiting condition of the present technical solution. When using modules 110, 112, 114, the processor 108 is configured to perform various operations described below.

Computer device 102 may be a conventional computer server, such as a Dell ™ PowerEdge ™ server that uses the Microsoft ™ Windows Server ™ operating system. It is important to keep in mind that computer device 102 can be implemented on any suitable hardware and / or application software and / or system software, or a combination thereof. In the presented embodiment of the present technical solution, not limiting its scope, the computer device 102 is a single server. In other embodiments of the present technical solution, not limiting its scope, the functionality of the computer device 102 can be divided, and can be performed using several computer devices.

Modifications to the communication interface (not shown) are well known in the art. How exactly the computer device 102 is connected to the measurement data receiver 104 and the database 106 is not particularly limited. By way of example, but not limitation, the computer device 102 may be coupled to a measurement data receiver 104 via a wired data network, a wireless communication network, or a combination thereof (for example, inter alia, an Ethernet-based connection, a 3G network connection, a 4G network connection , Wireless Fidelity or shortly Wi-Fi®, Bluetooth® and the like). The computer device 102 may be connected to a database 106 located on a separate server via a wireless communication network, a wired communication network, or a combination thereof.

With reference to FIG. 1, one or more of the modules 110, 112, 114 of the computer device 102 may, in this context, be a physical device, apparatus, or a plurality of modules made using hardware, for example, using a special purpose integrated circuit (ASIC) or programmable logic integrated circuit (FPGA), or a combination of hardware and software, for example, using a microprocessor system and a set of instructions that implement the functionality of the module, which (when executed) transform the microprocessor ssornuyu system unit for special purposes. One or more of the modules 110, 112, 114 may also be implemented as a combination of hardware and software, with some specific functions being implemented through hardware, and other functions being implemented by a combination of hardware and software.

In other embodiments of the present technology, one or more of the modules 110, 112, 114 may also be implemented as software for executing various types of processors. One or more of the modules 110, 112, 114 may, for example, include one or more physical or logical blocks of computer instructions that can be organized, for example, as an object, procedure, or function. However, these executable modules do not have to be physically located together, and they can include various instructions stored in different places, which, when logically connected to each other, include this module and perform the stated task for this module.

The processor 108, by means of a data type determination module 110, that is, data analysis, is configured to, based on a comparison of the data received by the computer device 102 from the measurement data receiver 104 with various types of data that have been previously stored in the database 106, determine the type of data. Alternatively or additionally, the data type determination module 110 is configured to extract data type information from a presentation method.

The processor 108 through the module 112 data verification is configured to check the data received from the receiver of the measurement data 104 for entry into the range of valid values. The range of acceptable values is the first sliding window that allows primary filtering of the received data. The second sliding window is the amount of data of a certain type needed to process the data and build a dynamic model of the data receiver.

The processor 108 through the data processing module 114 is configured to build a process model; filtering the input signal based on the obtained process model, building a model of the receiver of measurement data, determining calibration parameters for the receiver of measurement data.

The processor 108 is configured to determine and store calibration parameters for the measurement data receiver 104 in the database 106, which can be retrieved by an operator (not shown) to manually calibrate the data receiver, based on the received model of the receiver of measurement data and received data from the receiver of measurement data 104 measurement 104. An embodiment of the present technical solution is possible in which the processor 108 is configured to automatically calibrate the receiver of measurement data 104 associated with the computer device 102 over a data network.

An embodiment of the present technical solution is possible in which the data processing module 114 is made at least partially based on a pre-trained machine learning algorithm.

In some embodiments of the present technical solution, the machine learning algorithm is preliminarily trained in constructing a process model, filtering an input signal, and constructing a model of a receiver of measurement data; moreover, as input parameters of the machine learning algorithm, a set of data received from the receiver of measurement data included in the range of permissible values, data type and model of the data process is set, and learning the machine learning algorithm is a training with a teacher based on a training sample or training based on autoregressive models. As additional input parameters of the machine learning algorithm, one or more of: the type of receiver of measurement data can be specified; calibration parameters for a given type of receiver of measurement data obtained previously.

In FIG. 2 is a flowchart of a method 200 running within the system 100 of FIG. 1, and performed in accordance with an embodiment of the present technical solution.

Step 202 - Receiving Data from a Measurement Data Receiver

The method 200 begins at step 202, in which the computer device 102 receives data from the receiver of the measurement data 104.

The data includes at least measurements taken at regular intervals, information about the frequency of measurements, the values of the measured values at the time of each measurement.

For example, for the case where the measurement data receiver 104 is a temperature sensor mounted on a steam turbine of a nuclear power plant, the computer device 102 receives temperature data for a predetermined measurement time period. For example, daily temperature measurements of the blades of a steam turbine with a measurement frequency of 15 minutes. An embodiment of the present technology is possible in which computer device 102 receives data from at least one additional measurement data receiver (not shown). For example, for the case where the additional measurement data receiver is a pressure sensor mounted on a steam turbine of a nuclear power plant, the computer device 102 receives pressure data for a predetermined measurement time period. As an example, but not limitation, suppose that the period and frequency of measurements of temperature and pressure of a steam turbine coincide.

The data can be obtained by the computer device 102 both directly from the receiver (s) of the measurement data and through a computer-readable medium onto which said data is previously stored. An embodiment is also possible in which data from one receiver of measurement data is received directly over a data network, and data from another receiver of measurement data is obtained via a computer-readable medium. kind of data. The data format is determined by the model and receiver type of the measurement data. The data format may be further converted by the computer device 102 before performing the subsequent steps of the method 200.

The method then proceeds to step 204.

Step 204 - determining the type of data, the type of process, the polling frequency of the measured quantity; obtaining from the database the area of valid values and the minimum number of values for a certain type of data

At step 204, the processor 108 of the computer device 102, by the data type determination unit 110, determines the type of data received from the measurement data receiver 104.

Determining the type of measurement data obtained allows you to select the most appropriate process model. In addition, the separation of the obtained data by species allows them to be accumulated and used to refine existing models.

At step 204, the processor 108, through the data type determination module 110, extracts from the data received from the measurement data receiver 104 at least a portion of the information providing the determination of the data type.

Determining the type of data also contributes to a more accurate description of the process being measured. So, for example, each type of thermistor has its own temperature dependence of resistance. Using the wrong calibration curve gives incorrect results.

Then, the processor 108 obtains a valid value region for a certain type of data. For a certain type of data, the region of permissible values is obtained from the database 106, as well as the minimum number of values for a certain type of data,

For example, according to a non-limiting embodiment of the present technology, the range of permissible temperature values of a steam turbine of a nuclear power plant is from 100C to 600C, and the minimum number of values is 200. The range of permissible pressure values of a steam turbine of a nuclear power plant is, for example, from 0.1 to 20 MPa, and the minimum the number of values is 200.

The range of permissible values allows for additional filtering, and the minimum number of values determines the threshold value of the amount of data necessary to build a process model and a receiver model of measurement data for calibrating it with a given accuracy.

The method then proceeds to step 206.

Step 206 - Data Validation

At step 206, the processor 108, through the data verification module 112, verifies that the received data is in compliance with the valid value area, as well as the adequacy of the data for processing.

In response to the fact that the received data from the receiver, the measurement data 104 included in the range of acceptable values is not enough, the method ends.

In response to the fact that the received data from the receiver of the measurement data 104 included in the range of acceptable values is sufficient, the method proceeds to step 208.

It should be noted that when measuring the same duration, the number of useful values that can be used in the calibration can be different. The sufficiency of the obtained data depends, inter alia, on measurement errors, measurement recording errors, and sensor polling frequencies.

An embodiment of the method is possible in which calibration is carried out with varying degrees of accuracy depending on the number of values for a particular type of data. So, for example, the minimum number of values for temperature data is 200, if there are more values, for example, 500, the calibration accuracy will increase.

Step 208 - obtaining from the database of models for a certain type of data and a process model, in response to the fact that the data included in the range of valid values

At step 208, in response to the fact that the data included in the region of allowable values is determined, the processor 108 obtains from the database 106 models the data type that was determined at step 204.

So for the considered non-limiting example of measuring temperature and pressure of a steam turbine of a nuclear power plant, a temperature and pressure data model is obtained from the database 106. The mentioned models for each type of data for each type of process are stored previously in the database 106.

Then, the method proceeds to step 210.

Step 210 - data processing, in particular:

- filtering the input signal based on the obtained process model;

- building a model of the receiver of measurement data

At step 210, the processor 108, through the data processing module 114, constructs a process model based on the data stream, that is, the recorded series of observations, at the same time filtering the input signal and constructing the model of the measurement data receiver.

On the basis of the received data included in the range of permissible values, the obtained process model, the received data is filtered. Perform smoothing of unwanted frequencies while maintaining the characteristic features of the process. In the non-limiting example of the method under consideration, taking into account the model of the process of temperature and pressure changes in the steam turbine of nuclear power plants, the input signal is filtered and the model of the receiver of measurement data 104 is constructed.

 An embodiment of the present technical solution is possible in which the data processing module 114 is made at least partially based on a pre-trained machine learning algorithm.

In some embodiments of the present technical solution, the machine learning algorithm is preliminarily trained in constructing a process model, filtering an input signal, and constructing a model of a receiver of measurement data; moreover, as input parameters of the machine learning algorithm, a set of data received from the receiver of measurement data included in the range of permissible values, data type and model of the data process is set, and learning the machine learning algorithm is a training with a teacher based on a training sample or training based on autoregressive models. As additional input parameters of the machine learning algorithm, one or more of: the type of receiver of measurement data can be specified; calibration parameters for a given type of receiver of measurement data obtained previously.

At the output of the machine learning algorithm, a receiver model of measurement data 104 is obtained, the input and output characteristics of the values of chaotic invariants for the type of process. As the input and output characteristics determine the correlation dimension D2in. D2out. and correlation entropy K2in. and K2out. Moreover, the input characteristics of the values of chaotic invariants are determined on the basis of the obtained (raw) measurement data, the type of process, and the polling frequency of the measured quantity. The output characteristics of the values of chaotic invariants are determined on the basis of the processed measurement data, the constructed model of the receiver of measurement data 104, the type of process and the polling frequency of the measured quantity.

The method then proceeds to step 212.

Step 212 — Determining Calibration Parameters for the Measurement Data Receiver

At step 212, the input and output characteristics of the values of the chaotic invariants are compared.

The processor 108 of the computer device 102 through the data processing module 114 compares the input and output characteristics of the values of chaotic invariants.

An embodiment is possible in which a threshold value is preliminarily set for calibrating a receiver of measurement data. For example, if the input and output characteristics of the values of chaotic invariants differ by 1% or more, calibrate, and if the difference is less than 1%, do not calibrate the receiver of the measurement data. An embodiment is possible in which different threshold values can be set for different types of data. For example, for a temperature sensor of a steam turbine of a nuclear power plant, the threshold value can be 5%, and for a pressure sensor, 2%.

Calibration parameters for the receiver of measurement data are determined based on the autoregressive method using the orthogonal least squares estimation.

The method then proceeds to step 214.

Step 214 — Saving Calibration Parameters for the Measurement Data Receiver

Calibration parameters can be stored on a computer-readable medium and / or in a database, including, for example, base 106, and / or on the measurement data receiver itself directly.

At step 214, the processor 108 of the computer device 102 stores the calibration parameters for the type of data receiver in the database 106.

An embodiment of the present technical solution is possible, in which the processor 108 stores the calibration parameters in the memory of the corresponding receiver 104 of the measurement data.

At step 214, method 200 may terminate.

Modifications and improvements to the above described embodiments of the present technical solution will be apparent to those skilled in the art. The preceding description is provided as an example only and is not subject to any restrictions. Thus, the scope of the present technical solution is limited only by the scope of the attached claims.

Claims (19)

1. A method of calibrating a receiver of measurement data executed on a computer device having access to a database, including: - receiving data from the receiver of measurement data; - definition of the type of data; - obtaining from the database of the region of permissible values and the minimum number of values for a certain type of data; - data verification in response to the fact that the data included in the range of acceptable values is determined; - obtaining from the database of models of a certain type of data and process model; - data processing, in particular: - building a process model; - filtering the input signal based on the obtained process model; - building a model of the receiver of measurement data; - determination of calibration parameters for the receiver of measurement data; - saving calibration parameters for the receiver of measurement data. 2. The method according to p. 1, in which at the stage of receiving data from the receiver of measurement data, receive data from at least one additional receiver of measurement data. 3. The method according to claim 2, in which the data from the receiver of the measurement data and data from at least one additional receiver of the measurement data are data of the same type. 4. The method according to claim 2, in which the data from the receiver of the measurement data and the data from at least one additional receiver of the measurement data are data of various kinds. 5. The method of claim 1, wherein the data processing is performed at least partially based on a pre-trained machine learning algorithm. 6. The method according to p. 5, in which the machine learning algorithm is pre-trained to build a process model, filter the input signal and build a model of the receiver of measurement data; moreover, as input parameters of the machine learning algorithm, a set of data received from the receiver of measurement data included in the range of permissible values and the type of data are set, and learning the machine learning algorithm is a training with a teacher based on a training sample. 7. A computer device for calibrating a measurement data receiver, including a communication interface, a processor operatively coupled to a data type determining module, a data verification module, a data processing module and a communication interface, the processor is configured to implement the method according to any one of claims. 1-6.

Images