CN114136387B - Multi-channel ultrasonic flowmeter error compensation method based on SVM (support vector machine) algorithm - Google Patents
Multi-channel ultrasonic flowmeter error compensation method based on SVM (support vector machine) algorithm Download PDFInfo
- Publication number
- CN114136387B CN114136387B CN202111438951.3A CN202111438951A CN114136387B CN 114136387 B CN114136387 B CN 114136387B CN 202111438951 A CN202111438951 A CN 202111438951A CN 114136387 B CN114136387 B CN 114136387B
- Authority
- CN
- China
- Prior art keywords
- ultrasonic flowmeter
- support vector
- vector machine
- error
- compensation
- 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
- G01F—MEASURING VOLUME, VOLUME FLOW, MASS FLOW OR LIQUID LEVEL; METERING BY VOLUME
- G01F1/00—Measuring the volume flow or mass flow of fluid or fluent solid material wherein the fluid passes through a meter in a continuous flow
- G01F1/66—Measuring the volume flow or mass flow of fluid or fluent solid material wherein the fluid passes through a meter in a continuous flow by measuring frequency, phase shift or propagation time of electromagnetic or other waves, e.g. using ultrasonic flowmeters
- G01F1/667—Arrangements of transducers for ultrasonic flowmeters; Circuits for operating ultrasonic flowmeters
- G01F1/668—Compensating or correcting for variations in velocity of sound
Landscapes
- Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Fluid Mechanics (AREA)
- General Physics & Mathematics (AREA)
- Details Of Flowmeters (AREA)
- Measuring Volume Flow (AREA)
Abstract
The invention discloses a multichannel ultrasonic flowmeter error compensation method based on SVM algorithm, which comprises the steps of measuring and obtaining a plurality of channel flow data through a multichannel ultrasonic flowmeter, and extracting the ambient temperature and the medium temperature; preprocessing the sound channel flow data to obtain an asymmetric coefficient and an integral speed; inputting relevant parameters into a support vector machine model, setting errors as output, and adding the output errors to the indicating values obtained by measurement to compensate to obtain a compensation result; and comparing the compensated data with the original data, and verifying the method by comparing the relative errors before and after compensation. The invention can obviously improve the measurement precision of the ultrasonic flowmeter and reduce the error generated by influencing factors.
Description
Technical Field
The invention relates to the technical field of ultrasonic flow measurement, in particular to a multichannel ultrasonic flow meter error compensation method based on an SVM algorithm.
Background
The measurement accuracy of the ultrasonic flowmeter is influenced by factors such as temperature, flow fluctuation and flow field, wherein the more significant factor is asymmetry of the flow field. The influence of factors such as environment temperature, medium temperature and asymmetric coefficient on the measurement error of the ultrasonic flowmeter is researched aiming at an unstable flow field caused by a gate valve in a pipeline.
The method is researched by aiming at the error compensation method, an ultrasonic flowmeter error compensation model based on a support vector machine is designed, factors such as the ambient temperature, the medium temperature and the asymmetric coefficient, the speed of eight sound channels after passing through a circular integration method and the like are used as input, and the error is used as output to train the model. After the model training is finished, the prediction error is compensated back to the original measurement value, and the effect of error compensation is achieved.
(1) And (3) ambient temperature compensation: foreign scholars are dedicated to research novel materials, and the temperature adaptability is improved by finding out an energy conversion material with better performance so as to achieve the effect of improving the precision of the ultrasonic transducer.
(2) Medium temperature compensation: the idea of the BP neural network is applied to the error compensation of the ultrasonic flowmeter caused by medium temperature, the temperature data and the flow data are collected by designing and optimizing a hardware circuit, and the higher the temperature is, the larger the deviation between an actual measured value and an ideal value is. Therefore, the temperature data and the flow data are used as the input of the BP network, the agreed magnitude value is used as the output to establish a neural network model for training, the measurement error caused by temperature change is compensated, and the measurement precision is greatly improved.
(3) Flow field error compensation
And the multi-layer perceptron network and the radial basis network are adopted to compensate the flow field error. When the axial linear velocity is measured by ultrasonic waves, the speed of each point in the axial direction is supplemented and corrected by adopting a network interpolation method.
The above-mentioned techniques have the following disadvantages: only aiming at a single factor and having low compensation precision.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide an error compensation method for a multi-channel ultrasonic flowmeter based on an SVM algorithm, which improves the measurement accuracy of the multi-channel ultrasonic flowmeter, compensates errors and comprehensively compensates a plurality of influence factors.
The purpose of the invention is realized by the following technical scheme:
a multichannel ultrasonic flowmeter error compensation method based on SVM algorithm includes the following steps:
step A, measuring and acquiring flow data of a plurality of sound channels through a multi-channel ultrasonic flowmeter, and extracting the ambient temperature and the medium temperature;
b, preprocessing the sound channel flow data to obtain an asymmetric coefficient and an integral speed;
step C, inputting relevant parameters into the support vector machine model, setting errors as output, and adding the output errors to the measured indication value for compensation to obtain a compensation result;
and step D, comparing the compensated data with the original data, and verifying the method by comparing the relative errors before and after compensation.
One or more embodiments of the present invention may have the following advantages over the prior art:
the invention can obviously improve the measurement precision of the ultrasonic flowmeter and reduce the error generated by influencing factors.
Drawings
FIG. 1 is a flow chart of a method for error compensation of a multi-channel ultrasonic flowmeter based on an SVM algorithm;
FIG. 2 is a diagram of a multi-channel ultrasonic flowmeter error compensation model based on an SVM algorithm.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail with reference to the following embodiments and accompanying drawings.
As shown in fig. 1, a flow of an error compensation method for a multi-channel ultrasonic flowmeter based on an SVM algorithm includes the following steps:
The step 30 specifically includes: the ambient temperature, the medium temperature, the asymmetry coefficient and the integration speed are input into a support vector machine model, an error, namely a difference value between a measured flow rate and a default quantity value is set as an output, and the output error is added to a measured indicating value for compensation (as shown in figure 2).
In the embodiment, flow data of eight sound channels are obtained through measurement, in the established error compensation model, flow field influence factors such as ambient temperature and medium temperature are used as input of the compensation model, and errors, namely, a difference value between a measured flow rate and an agreed magnitude value is used as output of the compensation model of the support vector machine for training. And after the training is finished, adding the predicted error value and the original measurement value to obtain the compensated flow rate. In error compensation, the selection of a compensation model is particularly important, and from the perspective of practical application, a support vector machine algorithm is selected as a core of the compensation model to train the error compensation model.
The environment temperature and the medium temperature are respectively the environment temperature of the ultrasonic flowmeter, the temperature of a pipeline through which water flows, and the like; the asymmetry coefficient is a quantitative evaluation index calculated by water flow through eight sound channels under the influence of an asymmetric flow field; the integral speed is the speed of the eight sound channels of the ultrasonic flowmeter after the speed is measured by a circular integral method; the ambient temperature, the medium temperature and the asymmetry coefficient, and the circular integration velocity of the eight channels are taken as inputs, and the error is taken as an output. And adding the error trained by the support vector machine model to the measured flow indication value to obtain a compensation result.
The above embodiment may replace the support vector machine algorithm with the convolutional neural network algorithm, and the model input and output are not changed, but only the model constructed by the compensation algorithm is changed.
Although the embodiments of the present invention have been described above, the above descriptions are only for the convenience of understanding the present invention, and are not intended to limit the present invention. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (5)
1. The method for compensating the error of the multi-channel ultrasonic flowmeter based on the SVM algorithm of the support vector machine is characterized by comprising the following steps of:
step A, measuring and acquiring flow data of a plurality of sound channels through a multi-channel ultrasonic flowmeter, and extracting the ambient temperature and the medium temperature;
b, preprocessing the sound channel flow data to obtain an asymmetric coefficient and an integral speed;
step C, inputting relevant parameters into the support vector machine model, setting errors as output, and adding the output errors to the measured indication value for compensation to obtain a compensation result;
step D, comparing the compensated data with the original data, and performing method verification by comparing relative errors before and after compensation;
the step C specifically comprises the following steps: inputting the environment temperature, the medium temperature, the asymmetric coefficient and the integral speed into a support vector machine model, setting an error, namely a difference value between a measured flow rate and an appointed quantity value as an output, and adding the output error to an indicating value obtained by measurement to compensate;
in the established error compensation model, taking environmental temperature and medium temperature flow field influence factors as the input of the compensation model, and taking the error, namely the difference value between the measured flow rate and the appointed quantity value as the output of the compensation model of the support vector machine for training; and after the training is finished, adding the predicted error value and the original measurement value to obtain the compensated flow rate.
2. The multi-channel ultrasonic flowmeter error compensation method based on the SVM algorithm based on a support vector machine model (SVM) algorithm as claimed in claim 1, wherein the multi-channel flow data in step A has certain errors due to the influence of temperature, flow field and true value.
3. The multi-channel ultrasonic flowmeter error compensation method based on the SVM algorithm of the support vector machine model as claimed in claim 1, wherein the environment temperature and the medium temperature are respectively an environment temperature at which the ultrasonic flowmeter is located and a pipeline temperature through which water flows.
4. The multi-channel ultrasonic flowmeter error compensation method based on the SVM algorithm of the support vector machine model as claimed in claim 1, wherein the asymmetric coefficient is a quantitative evaluation index calculated by a multi-channel flow velocity of water flow under the influence of an asymmetric flow field.
5. The multi-channel ultrasonic flowmeter error compensation method based on the SVM algorithm of the support vector machine model of claim 1, wherein the integrated velocity is a velocity of the measured velocities of the plurality of channels of the ultrasonic flowmeter by a circular integration method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111438951.3A CN114136387B (en) | 2021-11-25 | 2021-11-25 | Multi-channel ultrasonic flowmeter error compensation method based on SVM (support vector machine) algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111438951.3A CN114136387B (en) | 2021-11-25 | 2021-11-25 | Multi-channel ultrasonic flowmeter error compensation method based on SVM (support vector machine) algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114136387A CN114136387A (en) | 2022-03-04 |
| CN114136387B true CN114136387B (en) | 2022-12-20 |
Family
ID=80389642
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111438951.3A Active CN114136387B (en) | 2021-11-25 | 2021-11-25 | Multi-channel ultrasonic flowmeter error compensation method based on SVM (support vector machine) algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114136387B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118329138A (en) * | 2024-06-14 | 2024-07-12 | 江苏德高物联技术有限公司 | High-precision electromagnetic flowmeter assembly and flowmeter control system |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110060221A1 (en) * | 2009-09-04 | 2011-03-10 | Siemens Medical Solutions Usa, Inc. | Temperature prediction using medical diagnostic ultrasound |
| CN102538912B (en) * | 2011-11-17 | 2014-04-16 | 中国计量科学研究院 | Method for analyzing additive errors of flow field of ultrasonic flowmeter |
| CN106643987B (en) * | 2016-11-10 | 2019-02-22 | 浙江大学 | A kind of multichannel ultrasonic gas flowmeter fault diagnosis and compensation method |
| CN106895890B (en) * | 2017-04-25 | 2019-04-16 | 浙江大学 | A kind of multichannel ultrasonic gas flowmeter sound channel weight coefficient calculation method |
| CN110906993B (en) * | 2019-12-12 | 2021-07-13 | 金卡水务科技有限公司 | Flow meter metering temperature compensation method and ultrasonic flow meter |
| US20210252698A1 (en) * | 2020-02-14 | 2021-08-19 | Nvidia Corporation | Robotic control using deep learning |
-
2021
- 2021-11-25 CN CN202111438951.3A patent/CN114136387B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN114136387A (en) | 2022-03-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102087130B (en) | Computational fluid dynamics (CFD) technology-based method for optimizing acoustic path of multi-path ultrasonic flowmeter assembled in elbow pipe | |
| CN104390680B (en) | Transit time ultrasonic ripple heat quantity flow error correcting method | |
| CN105841762B (en) | The flow metering method and system of ultrasonic water meter | |
| CN106096223A (en) | A kind of five-hole probe data processing method | |
| CN114136387B (en) | Multi-channel ultrasonic flowmeter error compensation method based on SVM (support vector machine) algorithm | |
| CN114235111B (en) | Ultrasonic water meter flow calibration method based on model optimization | |
| CN113483863B (en) | Method and system for determining fine adjustment coefficient of ultrasonic gas flowmeter | |
| CN110346005B (en) | Coriolis mass flowmeter digital signal processing method based on deep learning | |
| CN104537256A (en) | Numerical control low-speed wire cutting machine tool thermal error modeling method based on RBF neural network M-RAN algorithm | |
| CN114091368A (en) | Method for identifying cavitation state of axial flow turbine | |
| CN117109709B (en) | Ultrasonic water meter calibration method | |
| CN108267743B (en) | Fast iterative underwater positioning method based on fitting interpolation | |
| CN117309076A (en) | Gain intelligent adjustment control method for ultrasonic gas meter measurement data | |
| CN114812707B (en) | Metering method of ultrasonic meter | |
| CN113884164B (en) | Self-adaptive calibration method of ultrasonic gas meter | |
| CN116147724A (en) | Metering method suitable for ultrasonic water meter | |
| CN117234245B (en) | High-temperature-resistant high-pressure flow control method and system | |
| Catak et al. | Self-Calibration of Ultrasonic Water Flow Meter | |
| CN118392280B (en) | Operation error calibration method based on intelligent water meter of Internet of things | |
| JP2007107948A (en) | Method and apparatus for measuring quantity of flow and hydraulic machine provided with the same | |
| CN118533246A (en) | Temperature metering method suitable for ultrasonic water meter | |
| RU2682540C9 (en) | The method of setting the flow measurement channel with a restriction device | |
| CN117091677A (en) | Calibration method, system, server and storage medium of ultrasonic water meter | |
| CN118362170A (en) | Temperature metering method suitable for ultrasonic water meter | |
| CN117870786A (en) | Signal acquisition method, system and equipment of ultrasonic flowmeter and readable medium |
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 |