Method, device and equipment for correcting SF6 gas flow measurement value and storage medium
By correcting the SF6 gas flow rate measurement value using a neural network model, the measurement error problem caused by the influence of temperature and gas pressure was solved, and high-precision gas flow rate measurement was achieved.
Patent Information
- Application Number
- CN202111521340.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-12-13
- Publication Date
- 2025-10-24
- Estimated Expiration
- 2041-12-13
AI Technical Summary
Existing gas flow meters have measurement errors when measuring SF6 gas flow due to the influence of temperature and gas pressure.
A neural network model is used to correct the SF6 gas flow rate measurement value. By obtaining the temperature and pressure values, the model outputs the corrected gas flow rate value using the trained correction value prediction model, and then superimposes it with the gas flow rate measurement value to obtain the corrected gas flow rate value.
It achieves compensation for the measurement error of the SF6 gas flow metering device and obtains a more accurate gas flow value.
Smart Images
Figure CN114298091B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of gas flow measurement, in particular to a SF6 gas flow measurement value correction method, device, equipment and storage medium. BACKGROUND
[0002] Sulfur hexafluoride (SF6) has good electrical insulation performance and excellent arc extinguishing performance. Its dielectric strength is 2.5 times that of nitrogen under the same pressure, the breakdown voltage is 2.5 times that of air, and the arc extinguishing capacity is 100 times that of air. It is a new generation of ultra-high voltage insulation medium material superior to air and oil. Sulfur hexafluoride is widely used in electronic devices such as circuit breakers, high-voltage transformers, gas-enclosed combination capacitors, high-voltage transmission lines, and transformers due to its good insulation performance and arc extinguishing performance. The electrical industry uses it as an insulating material for high-voltage switches, large-capacity transformers, high-voltage cables, and gases due to its high dielectric strength and good arc extinguishing performance.
[0003] SF6 gas needs to be measured when it is applied. However, when using traditional gas flow meters to measure the flow of SF6 gas, errors exist in the measurement value due to the influence of temperature and pressure. Therefore, it is necessary to provide a SF6 gas flow measurement value correction method. SUMMARY
[0004] The purpose of the embodiments of the present application is to provide a SF6 gas flow measurement value correction method, device and storage medium, which aims to solve the problem of errors in the measurement of SF6 gas flow by existing gas flow meters.
[0005] To achieve the above purpose, the present application provides the following technical solutions:
[0006] In a first aspect, the embodiments of the present application provide a SF6 gas flow measurement value correction method, comprising the following steps:
[0007] Obtain the SF6 gas flow measurement value and the corresponding temperature value and pressure value;
[0008] Input the above temperature value and pressure value into the trained correction value prediction model, wherein the correction value prediction model comprises a plurality of neural network models corresponding to different temperature segment and pressure segment combinations, and the correction value prediction model uses the corresponding neural network model to predict and output the corresponding gas flow correction value according to the temperature segment and pressure segment to which the input temperature value and pressure value belong;
[0009] Superimpose the SF6 gas flow measurement value and the gas flow correction value to obtain the corrected gas flow value.
[0010] Further, the method for obtaining the trained correction value prediction model is as follows:
[0011] a plurality of characteristic sample data, each of the characteristic sample data comprising a temperature value, a gas pressure value, and a corresponding SF6 gas flow measurement error value, and dividing the plurality of characteristic sample data into a plurality of groups according to different combinations of temperature segments and gas pressure segments;
[0012] establishing a neural network model with two inputs and one output;
[0013] training the neural network model established above by using each group of characteristic sample data, taking the temperature value and the gas pressure value as inputs, and taking the SF6 gas flow measurement error value as output, to obtain a plurality of neural network models corresponding to different combinations of temperature segments and gas pressure segments, i.e., the trained correction value prediction model.
[0014] Further, the dividing the plurality of characteristic sample data into a plurality of groups according to different combinations of temperature segments and gas pressure segments specifically comprises:
[0015] segmenting a certain range of temperature and gas pressure respectively, and arranging the segmented intervals of the two, each combination comprising a temperature interval and a gas pressure interval, and dividing the plurality of characteristic sample data into a plurality of groups according to the above combinations.
[0016] Further, the establishing a neural network model with two inputs and one output specifically comprises:
[0017] firstly, building a traditional BP neural network framework and a PSO algorithm framework, initializing the velocity and position vector of the PSO algorithm particles, setting the error interval threshold to be corrected, and calculating the mean square error function at the same time;
[0018] then, updating the individual and global extreme values of the BP neural network, and updating the velocity and position vector of the PSO algorithm particles;
[0019] subsequently, judging whether the accuracy requirement is met according to the calculated mean square error function value, if the accuracy requirement is met, directly assigning the particle velocity and position information to the BP neural network to complete the neural network optimization, if the accuracy requirement is not met, increasing the iteration number until the mean square error function value meets the accuracy requirement, and completing the establishment of the neural network model with two inputs and one output.
[0020] Further, the mean square error function is expressed as follows:
[0021]
[0022] wherein n is the number of the neural network training samples after optimization, Y p (i) is the expected output value of the neural network, is the actual output value of the output layer of the neural network.
[0023] Further, the method further comprises:
[0024] The gas flow correction value output by the trained correction value prediction model and the corresponding temperature value and gas pressure value are used as feature samples to further train the model.
[0025] In a second aspect, the embodiments of the present application provide an SF6 gas flow measurement value correction device, comprising:
[0026] A data acquisition module is configured to acquire an SF6 gas flow measurement value and corresponding temperature and gas pressure values.
[0027] A correction value acquisition module is configured to input the temperature and gas pressure values into a trained correction value prediction model, wherein the correction value prediction model comprises a plurality of neural network models corresponding to different temperature segment and gas pressure segment combinations, and the correction value prediction model uses the corresponding neural network model to predict and output a corresponding gas flow correction value according to the temperature segment and gas pressure segment to which the input temperature and gas pressure values belong.
[0028] An error correction module is configured to superimpose the SF6 gas flow measurement value and the gas flow correction value to obtain a corrected gas flow value.
[0029] In a third aspect, the embodiments of the present application provide an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method described above when executing the computer program.
[0030] In a fourth aspect, the embodiments of the present application provide a computer readable storage medium, wherein the computer readable storage medium stores a data processing program, and the data processing program is executable by a processor to implement the steps of the method described above.
[0031] Compared with the prior art, the present application has the following advantages:
[0032] The SF6 gas flow measurement value correction method, device, equipment and storage medium provided by the present application first acquire an SF6 gas flow measurement value and corresponding temperature and gas pressure values, then input the temperature and gas pressure values into a trained correction value prediction model corresponding to the temperature segment and gas pressure segment to which the temperature and gas pressure values belong, the model outputs a gas flow correction value corresponding to the temperature and gas pressure values, and finally superimposes the SF6 gas flow measurement value and the gas flow correction value to obtain a corrected gas flow value; the temperature and gas pressure are segmented to obtain neural network models corresponding to different temperature segment and gas pressure segment combinations, the model outputs a gas flow correction value with higher precision, and ultimately compensates for the measurement error of the SF6 gas flow measurement device. BRIEF DESCRIPTION OF DRAWINGS
[0033] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments of the present application. It should be understood that the following drawings only show some of the embodiments of the present application, and therefore should not be regarded as a limitation to the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.
[0034] Figure 1 A flow chart of a SF6 gas flow measurement value correction method provided by an embodiment of the present application;
[0035] Figure 2 A structural block diagram of a SF6 gas flow measurement value correction device provided by an embodiment of the present application. DETAILED DESCRIPTION
[0036] The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application. It should be noted that similar reference numerals and letters in the following drawings represent similar items, and therefore, once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings.
[0037] The term "comprising" or "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or device. Without more limitation, the element defined by the statement "including a" does not exclude the presence of another identical element in the process, method, article or device including the element.
[0038] The terms "first", "second" and the like are only used to distinguish one entity or operation from another entity or operation, and cannot be understood as indicating or implying relative importance, and cannot be understood as requiring or implying any such actual relationship or order between the entities or operations.
[0039] As shown in the Figure 1 The present embodiment provides a SF6 gas flow measurement value correction method, which comprises the following steps:
[0040] S101, obtaining a SF6 gas flow measurement value and corresponding temperature value and gas pressure value.
[0041] The SF6 gas flow measurement value is measured by an SF6 gas flow measurement device, such as an SF6 gas flow meter / SF6 gas flow sensor, and the corresponding temperature value and the gas pressure value can be measured by a temperature sensor and a gas pressure sensor, and the SF6 gas flow measurement device and the temperature sensor and the gas pressure sensor are arranged in series on the SF6 gas channel, that is, the SF6 gas sequentially passes through the above measurement devices.
[0042] In S102, the above temperature value and the gas pressure value are input into the trained correction value prediction model, the correction value prediction model includes a plurality of neural network models corresponding to different combinations of temperature segments and gas pressure segments, and the correction value prediction model uses the corresponding neural network model to predict and output the corresponding gas flow correction value according to the temperature segment and the gas pressure segment to which the input temperature value and the input gas pressure value belong.
[0043] In the embodiment of the present application, the temperature and the gas pressure are segmented and the neural network models corresponding to different combinations of temperature segments and gas pressure segments are obtained, and since the influence of the SF6 gas temperature and the gas pressure on the SF6 gas flow measurement is different under different temperature segments and gas pressure segments, compared with only using one neural network model, the model can output a gas flow correction value with higher precision, so as to obtain a gas flow value with higher precision.
[0044] In S103, the SF6 gas flow measurement value is superimposed with the gas flow correction value to obtain a corrected gas flow value.
[0045] Specifically, the gas flow correction value can be calculated in real time and the SF6 gas flow measurement value can be corrected in real time to obtain a high-precision gas flow value for providing a user and realizing compensation for the measurement error of the SF6 gas flow measurement device.
[0046] Further, the method for obtaining the trained correction value prediction model is as follows:
[0047] (1) Obtain a plurality of feature sample data, the feature sample data is standard data, each feature sample data includes a temperature value, a gas pressure value and a corresponding SF6 gas flow measurement error value, and the plurality of feature sample data is divided into a plurality of groups according to different combinations of temperature segments and gas pressure segments.
[0048] (2) Establish a neural network model with double inputs and single output;
[0049] (3) Use each group of feature sample data, use the temperature value and the gas pressure value as input, and use the SF6 gas flow measurement error value as output, train the above established neural network model, and obtain a plurality of neural network models corresponding to different combinations of temperature segments and gas pressure segments, that is, the trained correction value prediction model.
[0050] Specifically, the step of dividing the plurality of feature sample data into a plurality of groups according to different combinations of temperature segments and pressure segments specifically includes:
[0051] The temperature and pressure ranges are segmented respectively, and the segmented temperature and pressure ranges are arranged and combined, each combination containing a temperature range and a pressure range, the plurality of feature sample data is divided into a plurality of groups according to the ranges to which the temperature and pressure of the feature sample data belong according to the above combinations, each group containing a plurality of feature sample data, and the ranges of the segmented temperature and pressure and the segmentation method can be set according to actual needs, which are not limited in the embodiment.
[0052] Since the initial weights and thresholds of the BP neural network are randomly generated, if the initial weights and thresholds are not properly selected, problems such as slow network convergence speed and falling into local minimum value are prone to occur. The PSO algorithm is used to optimize the BP neural network in the application, the influence of the initial value on the prediction result of the BP neural network is reduced, and the convergence speed and prediction accuracy of the network are improved. Preferably, the step of establishing the neural network model with double inputs and single output specifically includes:
[0053] First, the framework of the traditional BP neural network and the framework of the PSO algorithm are built, the velocity and position vector of the PSO algorithm particle are initialized, the error interval threshold to be corrected is set, and the mean square error function is calculated; then the individual and global extreme value of the BP neural network are updated, and the velocity and position vector of the PSO algorithm particle are updated; subsequently, whether the accuracy requirement is met is judged according to the calculated mean square error function value, if the accuracy requirement is met, the particle velocity and position information are directly assigned to the BP neural network to complete the neural network optimization, if the accuracy requirement is not met, the iteration number is increased until the mean square error function value meets the accuracy requirement, and the establishment of the neural network model with double inputs and single output is completed.
[0054] The mean square error function is expressed as follows:
[0055]
[0056] Wherein n is the number of the optimized neural network training samples, Yp(i) is the expected output value of the neural network, Yp(i) is the actual output value of the neural network output layer.
[0057] Preferably, the method further includes: taking the gas flow correction value output by the trained correction value prediction model and the corresponding temperature value and pressure value as feature samples to further train the model, so as to further improve the accuracy of the model.
[0058] As shown in Figure 2 The embodiment of the application also provides an SF6 gas flow measurement value correction device, which comprises:
[0059] The data acquisition module 201 is configured to acquire the SF6 gas flow measurement value, and corresponding temperature value and air pressure value;
[0060] The correction value acquisition module 202 is configured to input the temperature value and air pressure value into a trained correction value prediction model, wherein the correction value prediction model comprises a plurality of neural network models corresponding to different combinations of temperature segments and air pressure segments, and the correction value prediction model uses the corresponding neural network model to predict and output the corresponding gas flow correction value according to the temperature segment and air pressure segment to which the input temperature value and air pressure value belong.
[0061] The error correction module 203 is configured to superimpose the SF6 gas flow measurement value and the gas flow correction value to obtain a corrected gas flow value.
[0062] The embodiment of the present application further provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the method according to any one of the above when executing the computer program.
[0063] The embodiment of the present application further provides a computer readable storage medium, and the computer readable storage medium stores a data processing program, and the data processing program is executable on a processor to implement the steps of the method according to any one of the above.
[0064] Since the principle of solving problems of the above-mentioned device, electronic device and computer readable storage medium embodiments is similar to that of the above-mentioned method embodiments, the implementation thereof can be referred to the above-mentioned method embodiments, and the repeated parts will not be described herein.
[0065] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt a computer program product in the form of one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program codes.
[0066] The present application is described with reference to flowcharts and / or block diagrams according to the method, device (system), and computer program product of the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the computer or other programmable data processing devices produce a device that implements the functions described in the flowcharts and / or block diagrams. Figure 1one or more processes and / or blocks Figure 1 an apparatus with the functionality to achieve the specified processes or functions in the block or blocks.
[0067] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a Figure 1 one or more processes and / or blocks Figure 1 an apparatus with the functionality to achieve the specified processes or functions in the block or blocks.
[0068] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the processes specified in the flowchart Figure 1 one or more processes and / or blocks Figure 1 an apparatus with the functionality to achieve the specified processes or functions in the block or blocks.
[0069] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.
[0070] The memory can include non-persistent memory and / or volatile memory, such as random access memory (RAM) about which the computer stores information such as computer program instructions. RAM allows information to be loaded quickly into the computer. Non-volatile memory can also be used such as read only memory (ROM), electrically programmable read only memory (EPROM), electrically erasable programmable read only memory (EEPROM), flash memory, or other memory technologies. The memory is an example of computer readable media.
[0071] Computer readable media includes permanent and non-permanent, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), electrically programmable read only memory (EEPROM), flash memory or other memory technologies, compact disc read only memory (CD-ROM), digital versatile disc (DVD), or other optical storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer readable media does not include transitory media such as modulated data signals and carrier waves.
[0072] The above merely provides an example of the present application, and is not used to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.
Claims
1. A method of correcting a measured value of SF6 gas flow, characterized by, The method comprises the following steps: obtaining SF6 gas flow measurement values and corresponding temperature values and pressure values; inputting the temperature values and pressure values into a trained correction value prediction model, wherein the correction value prediction model comprises a plurality of neural network models corresponding to different combinations of temperature segments and pressure segments, and the correction value prediction model uses the corresponding neural network model to predict and output a corresponding gas flow correction value according to the temperature segment and the pressure segment to which the input temperature value and pressure value belong; superimposing the SF6 gas flow measurement values and the gas flow correction value to obtain a corrected gas flow value; The method for obtaining the trained correction value prediction model is as follows: obtaining a plurality of feature sample data, wherein each feature sample data comprises a temperature value, a pressure value and a corresponding SF6 gas flow measurement error value, and the plurality of feature sample data are divided into a plurality of groups according to different combinations of temperature segments and pressure segments; establishing a neural network model with double inputs and single output; training the neural network model established above by using each group of feature sample data, taking the temperature value and the pressure value as inputs and taking the SF6 gas flow measurement error value as output, to obtain a plurality of neural network models corresponding to different combinations of temperature segments and pressure segments, i.e., the trained correction value prediction model; The method for dividing the plurality of feature sample data into a plurality of groups according to different combinations of temperature segments and pressure segments specifically comprises: segmenting a certain range of temperature and pressure respectively, and arranging and combining the segmented intervals of the two, wherein each combination comprises a temperature interval and a pressure interval, and the plurality of feature sample data are divided into a plurality of groups according to the above combinations.
2. The SF6 gas flow meter value correction method according to claim 1, characterized by, The method for establishing a neural network model with double inputs and single output specifically comprises: firstly, building a traditional BP neural network framework and a PSO algorithm framework, initializing the velocity and position vector of the PSO algorithm particles, setting the error interval threshold to be corrected, and calculating the mean square error function at the same time; then, updating the individual and global extreme values of the BP neural network, and updating the velocity and position vector of the PSO algorithm particles; subsequently, judging whether the accuracy requirement is met according to the calculated mean square error function value, if the accuracy requirement is met, directly assigning the particle velocity and position information to the BP neural network to complete the neural network optimization, if the accuracy requirement is not met, increasing the iteration number until the mean square error function value meets the accuracy requirement, and completing the establishment of the neural network model with double inputs and single output.
3. The SF6 gas flow meter value correction method according to claim 2, characterized by, The mean square error function is expressed as follows: ， where n is the number of training samples of the optimized neural network, Y p (i) is the desired output value of the neural network, is the actual output value of the output layer of the neural network.
4. The SF6 gas flow meter value correction method according to claim 1, characterized by, The method further comprises: further training the model by taking the gas flow correction value output by the trained correction value prediction model and the corresponding temperature value and pressure value as feature samples.
5. A SF6 gas flow meter value correction device for implementing the method according to any one of claims 1 to 4, characterized in that, It comprises: a data acquisition module for obtaining SF6 gas flow measurement values and corresponding temperature values and pressure values; a correction value acquisition module for inputting the temperature values and pressure values into a trained correction value prediction model, wherein the correction value prediction model comprises a plurality of neural network models corresponding to different combinations of temperature segments and pressure segments, and the correction value prediction model uses the corresponding neural network model to predict and output a corresponding gas flow correction value according to the temperature segment and the pressure segment to which the input temperature value and pressure value belong; An error correction module is configured to superimpose the SF6 gas flow measurement value and the gas flow correction value to obtain a corrected gas flow value.
6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor implements the steps of the method as claimed in any one of claims 1-4 when executing the computer program.
7. A computer readable storage medium characterized in that, The computer readable storage medium stores a data processing program, and the data processing program, when executed by the processor, implements the steps of the method as claimed in any one of claims 1-4.