Apparatus, method, and program

Abstract

The present invention provides an apparatus comprising: a measured value acquisition unit that acquires a plurality of measured values with respect to a fluid; a state determination estimation unit that estimates a state of the fluid on the basis of at least one of the plurality of acquired measurement values; a fluid mass estimation unit that estimates at least one of an amount of liquid or dryness with respect to the fluid from at least one of the plurality of acquired measurement values, based on the estimated state of the fluid; and an output unit that outputs at least one of the estimated amount of liquid or the degree of dryness. The apparatus further includes a selection unit that selects a calculation from a plurality of calculations corresponding to the estimated state of the fluid, and the fluid mass estimation unit estimates at least one of the liquid amount or the dryness with respect to the fluid from at least one of the plurality of acquired measurement values by the selected calculation.

Description

Technical Field

[0001] The present invention relates to an apparatus, a method and a program. The contents of the following patent applications are incorporated herein by reference: Patent application No. 2023-052768 filed in Japan on March 29, 2023 Background Art

[0002] Patent Document 1 describes “a method for measuring humidity, wherein the flow rate of wet steam circulating through a pipe is measured by a vortex flowmeter or one of an ultrasonic flowmeter and an orifice flowmeter, and the humidity in the pipe is obtained using the value measured by the flowmeter.” Patent Document 2 describes “a vortex flow measuring device that, when used, can typically measure the flow rate of the first phase of a flowable two- or more-phase medium using the vortex frequency of the Karman vortex recorded by an vortex sensor, and also simultaneously and timely (i.e., online) detect at least a second phase that appears in the form of a distributed particle and / or droplet flow. Patent Document 3 describes “the mass flow rate of the second-phase wall flow can be quantitatively measured in a simple manner based on the standard deviation or kurtosis and vortex frequency determined from the measurement signal during use of the vortex flow measuring device, and with reference to the correlation.” List of citations Patent Literature Patent Document 1: Japanese Patent Application Publication No. 2013-18951 Patent Document 2: Japanese Patent No. 5443514 Patent Document 3: Japanese Patent No. 5355724 Summary of the Invention

[0003] A first aspect of the present invention provides a device comprising: a measurement value acquisition unit that acquires multiple measurement values ​​about a fluid; a state determination estimation unit that estimates the state of the fluid based on at least one of the multiple measurement values ​​acquired; a fluid quality estimation unit that estimates at least one of the liquid amount or dryness of the fluid based on the estimated state of the fluid and at least one of the multiple measurement values ​​acquired; and an output unit that outputs at least one of the estimated liquid amount or dryness.

[0004] The apparatus may further include a selection unit configured to select a calculation corresponding to the estimated state of the fluid from a plurality of calculations, and the fluid quality estimation unit may estimate at least one of the liquid amount or the dryness of the fluid based on at least one of the plurality of acquired measurement values ​​using the selected calculation. The selection unit may select different calculations depending on the flow state of the fluid estimated by the state determination estimation unit.

[0005] In any of the above-described apparatuses, the state determination estimation unit may estimate whether the state of the fluid is a laminar flow, a laminar wave flow, or an annular mist flow based on at least one of the acquired plurality of measurement values.

[0006] In the above-mentioned device, the state determination and estimation unit can determine whether the state of the fluid is wet steam based on at least one of the multiple measurement values ​​obtained. If it is determined that the state of the fluid is wet steam, it can estimate whether the state of the fluid is stratified flow, stratified wave flow or annular mist flow based on at least one of the multiple measurement values ​​obtained.

[0007] In the above-mentioned device, when the measured value of the temperature of the fluid exceeds the boiling point of the fluid based on the measured value of the pressure of the fluid acquired by the measurement value acquisition unit, the state determination estimation unit can determine that the state of the fluid is not wet steam, and if the state determination estimation unit determines that the state of the fluid is not wet steam, the fluid mass estimation unit can estimate the liquid amount of the fluid to be 0 or the dryness of the fluid to be 100%.

[0008] In any of the above-mentioned devices, the state determination estimation unit can estimate the state of the fluid based on at least one of the acquired multiple measurement values ​​by using a first model for estimating the state of the fluid based on at least one of the multiple measurement values ​​about the fluid.

[0009] In any of the above-described apparatuses, the fluid quality estimation unit may estimate the at least one of the liquid amount or the dryness of the fluid based on at least one of the acquired plurality of measurement values ​​by using a second model for estimating at least one of the liquid amount or the dryness of the fluid based on the estimated state of the fluid from at least one of the plurality of measurement values. The selection unit may select one second model corresponding to the estimated state of the fluid from a plurality of second models for estimating at least one of the liquid amount or the dryness of the fluid based on at least one of the plurality of measurement values.

[0010] Any of the above-mentioned devices may further include a learning processing unit that generates, for each state of the fluid, the second model for estimating at least one of the liquid amount or the dryness of the fluid based on at least one of the multiple measurement values ​​of the fluid.

[0011] In any of the above devices, the measurement value acquisition unit may acquire at least one of flow rate, temperature, pressure, viscosity, concentration, density, signal amplitude, or vortex frequency measured for the fluid as the plurality of measurement values ​​regarding the fluid.

[0012] In any of the above devices, the measurement value acquisition unit may acquire the plurality of measurement values ​​regarding the fluid from a vortex flowmeter. The second model may be a regression model for outputting the dryness or the amount of liquid based on an input of at least one of signal strength, density, or vortex frequency measured by the vortex flowmeter.

[0013] A second aspect of the present invention provides a method comprising: obtaining, by a computer, a plurality of measurement values ​​about a fluid; estimating, by the computer, a state of the fluid based on at least one of the plurality of measurement values ​​obtained; estimating, by the computer, at least one of the liquid amount or dryness of the fluid based on at least one of the plurality of measurement values ​​obtained, according to the estimated state of the fluid; and outputting, by the computer, at least one of the estimated liquid amount or dryness.

[0014] A third aspect of the present invention provides a program that causes a computer to function as: a measurement value acquisition unit that acquires multiple measurement values ​​about a fluid; a state determination estimation unit that estimates the state of the fluid based on at least one of the multiple measurement values ​​acquired; a fluid quality estimation unit that estimates at least one of the liquid amount or dryness of the fluid from at least one of the multiple measurement values ​​acquired based on the estimated state of the fluid; and an output unit that outputs at least one of the estimated liquid amount or dryness.

[0015] The summary does not necessarily describe all necessary features of an embodiment of the present invention. The present invention may also be a sub-combination of the above features. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is an example of a block diagram showing the system 10 according to the present embodiment. Figure 2 This is an explanatory diagram for explaining the flow state of a fluid in a pipe. Figure 3 : is a flowchart showing an operation example of the processing device 30 according to the present embodiment. Figure 4 It is an explanatory diagram for explaining an example of the first model. Figure 5 It is an explanatory diagram for explaining an example of the second model. Figure 6 An example of a computer 2200 is shown in which aspects of the present invention may be embodied in whole or in part. DETAILED DESCRIPTION

[0017] Hereinafter, the present invention will be described by way of its embodiments, but the following embodiments do not limit the present invention according to the claims. In addition, not all feature combinations described in the embodiments are essential to the solution of the present invention.

[0018] Figure 1 is an example of a block diagram illustrating a system 10 according to this embodiment. Note that these blocks are functionally divided functional blocks and may not necessarily match the actual device configuration. That is, in this diagram, a device represented by a block may not necessarily be constructed by a single device. Furthermore, in this diagram, a device represented by a separate block may not necessarily be constructed by separate devices. The same applies to the subsequent block diagrams.

[0019] System 10 collects data from instruments 20 installed in facilities such as factories. Examples of factories include, in addition to chemical and biological industrial plants, plants for managing and controlling wells such as gas and oil fields and their surroundings, plants for managing and controlling hydroelectric, thermal, and nuclear power generation, plants for managing and controlling environmental power generation such as solar and wind power, plants for managing and controlling water and wastewater treatment, dams, and the like. System 10 includes one or more instruments 20, a processing device 30, and an external device 40.

[0020] One or more instruments 20 are connected to a processing device 30 via a wired or wireless connection. At least one of the one or more instruments 20 may be a field instrument, and as an example, a vortex flowmeter that measures flow velocity by measuring vortices generated in the fluid. Alternatively, the instrument 20 may be, for example, a pressure sensor, a flowmeter, a temperature sensor, a density meter, a viscometer, a concentration meter, or the like. The instrument 20 may transmit measurement values ​​regarding the fluid flowing through the pipeline of the facility to the processing device 30.

[0021] The processing device 30 is connected to an external device 40. The processing device 30 can be a computer such as a personal computer (PC), tablet computer, smartphone, workstation, server computer, or general-purpose computer, or a computer system comprising multiple connected computers. Alternatively, the processing device 30 can be an internal computer of a field instrument such as a vortex flowmeter. Such a computer system is also a computer in a broad sense. Furthermore, the processing device 30 can be implemented as one or more virtual computer environments executable on the computer.

[0022] The processing device 30 estimates at least one of the liquid amount or dryness of the fluid based on the measurement values ​​about the fluid acquired from the one or more instruments 20. The processing device 30 includes a measurement value acquisition unit 100, a state determination estimation unit 110, a selection unit 120, a fluid quality estimation unit 130, a learning processing unit 140, and an output unit 150.

[0023] Here, the fluid in this embodiment may include at least one of a gas (vapor) into which water or the like evaporates, or a liquid into which a gas condenses. Furthermore, for example, the estimated liquid amount represents the mass of the condensed liquid in the vapor. For example, the dryness to be estimated represents the ratio of the gas vapor mass to the vapor mass.

[0024] The measurement value acquisition unit 100 is connected to one or more instruments 20 and the state determination and evaluation unit 110. The measurement value acquisition unit 100 acquires a plurality of measurement values ​​about the fluid from the instruments 20. The measurement value acquisition unit 100 may acquire a plurality of different types of measurement values ​​measured by one instrument 20, or may acquire a plurality of measurement values ​​measured by a plurality of instruments 20, respectively.

[0025] The state determination estimation unit 110 is connected to the learning processing unit 140 and the selection unit 120. The state determination estimation unit 110 estimates the state of the fluid based on at least one of the plurality of measurement values ​​acquired by the measurement value acquisition unit 100. The state determination estimation unit 110 can estimate the flow state of the fluid in the pipe as the state of the fluid based on the measurement value. Note that reference will be made later to Figure 2 The state estimation unit 110 can estimate the state of the fluid based on at least one of the multiple measurement values ​​acquired by the measurement value acquisition unit 100 by using a first model for estimating the state of the fluid from at least one of the multiple measurement values ​​about the fluid.

[0026] The selection unit 120 is connected to the learning processing unit 140 and the fluid quality estimation unit 130. The selection unit 120 selects a calculation corresponding to the estimated state of the fluid from a plurality of calculations. The selection unit 120 can select different calculations depending on the flow state of the fluid. The selection unit 120 can select a second model corresponding to the state of the fluid from a plurality of second models for estimating at least one of the liquid content or dryness of the fluid based on at least one of a plurality of measured values ​​of the fluid.

[0027] The fluid quality estimation unit 130 is connected to the learning processing unit 140 and the output unit 150. Based on the state of the fluid estimated by the state determination estimation unit 110, the fluid quality estimation unit 130 estimates at least one of the liquid amount or dryness of the fluid based on at least one of the multiple measurement values ​​acquired by the measurement value acquisition unit 100. The fluid quality estimation unit 130 can estimate at least one of the liquid amount or dryness of the fluid based on at least one of the multiple measurement values ​​acquired by the measurement value acquisition unit 100 by using a calculation selected by the selection unit 120. Based on the state of the fluid estimated by the state determination estimation unit 110, the fluid quality estimation unit 130 can estimate at least one of the liquid amount or dryness of the fluid based on at least one of the multiple measurement values ​​acquired by the measurement value acquisition unit 100 using a second model for estimating at least one of the liquid amount or dryness of the fluid based on at least one of the multiple measurement values. The fluid quality estimation unit 130 can calculate one of the dryness and liquid amount based on the estimated result of the other by using the vapor flow rate of the fluid.

[0028] The learning processing unit 140 performs a learning process on the plurality of models used in the processing device 30. The learning processing unit 140 generates, for each state of the fluid, a second model for estimating at least one of the amount of liquid or the dryness of the fluid based on at least one of the plurality of measured values ​​of the fluid. The learning processing unit 140 may generate a first model for estimating the state of the fluid based on at least one of the plurality of measured values ​​of the fluid.

[0029] The output unit 150 is connected to the external device 40. The output unit 150 outputs at least one of the liquid amount or the dryness estimated by the fluid quality estimation unit 130 to the external device 40.

[0030] The external device 40 may be a display device that displays the estimation result of the processing device 30 , a PC that stores the estimation result of the processing device 30 , or the like.

[0031] Figure 2 This is an explanatory diagram for explaining the flow state of a fluid in a pipe. Figure 2 The XZ cross section and YZ cross section of the pipe are shown for the laminar flow, laminar wave flow and annular mist flow as flow states. Figure 2 In the figure, the fluid flows in the pipe along the Y-axis.

[0032] For example, in a wet steam state, where liquid water mixes with gaseous vapor, the fluid containing evaporated water has three flow patterns (flow states). In stratified flow, the liquid flows slowly at the bottom of the pipe. In stratified wavy flow, the liquid flows while generating waves at the bottom of the pipe. In annular mist flow, the liquid flows in contact with the entire circumference of the pipe inner wall, and the number of droplets mixed with the vapor increases. These flow patterns vary depending on the gas flow rate and the humidity of the fluid. These three flow patterns differ in the Karman vortex state and the sensor signal used to detect the vortex in vortex flowmeter measurement, and their impact on fluid measurements is also different. This applies to fluids other than steam water.

[0033] Therefore, in this embodiment, the liquid amount or dryness of the fluid as an indicator of steam quality is estimated by calculation corresponding to the state of the fluid.

[0034] Figure 3 : is a flowchart showing an example of the operation of the processing device 30 according to this embodiment. Figure 3 In the embodiment, the instrument 20 is, for example, a vortex flowmeter.

[0035] A vortex flowmeter has a detection unit through which the fluid to be measured flows. When the fluid reaches a certain flow rate or higher, vortices called Karman vortices are generated downstream of the detection unit's vortex generator. Karman vortices are generated alternately on the left and right sides (or top and bottom) of the vortex generator's axis, causing the pressure in the fluid to fluctuate. Two piezoelectric elements are arranged within the vortex generator. These elements detect the pressure generated by the vortexes and also counteract external vibrations transmitted through pipes, etc., through which the fluid flows. The vortex flowmeter measures the period of the Karman vortices generated by the two piezoelectric elements as pressure fluctuations to determine the fluid's volumetric flow rate. Furthermore, a temperature sensor such as a resistance temperature detector and an internal sensor such as a pressure sensor are arranged within the detection unit to measure the fluid's temperature and pressure. The vortex flowmeter also measures the fluid's temperature from the internal sensors and the fluid's pressure from the pressure sensor. Based on the fluid's temperature and pressure, the density is determined and the mass flow rate (volume flow rate × density) is calculated. In addition, the vortex flowmeter can also detect the effect of vortex flow on the speed of sound by using an ultrasonic sensor or an optical sensor. The vortex flowmeter can transmit the multiple measurement values ​​obtained as described above to the processing device 30.

[0036] In step S300, the measurement value acquisition unit 100 acquires at least one of the flow velocity, temperature, pressure, viscosity, concentration, density, signal amplitude, or vortex frequency of the fluid measured by the instrument 20 as a plurality of measurement values ​​regarding the fluid. The measurement value acquisition unit 100 may acquire a plurality of measurement values ​​measured or calculated by a vortex flowmeter serving as the instrument 20. Furthermore, the measurement value acquisition unit 100 may acquire measurement values ​​measured by the vortex flowmeter and calculate measurement values ​​such as the volume flow rate or mass flow rate of the fluid. For example, the measurement value acquisition unit 100 may acquire at least one of the flow velocity derived from the vortex frequency, the signal amplitude detected by two piezoelectric elements of a vortex generator, or the frequency components from the vortex flowmeter as a measurement value, and the processing device 30 may use the measurement values ​​to perform state estimation and parameter estimation in subsequent steps. By using these measurement values, the accuracy of the state estimation regarding the fluid and the accuracy of the estimation of the liquid amount or dryness can be improved.

[0037] The measurement value acquisition unit 100 may output a plurality of measurement values ​​associated with each other that are acquired simultaneously or in the same cycle to the state determination and estimation unit 110 and the fluid quality estimation unit 130. In subsequent steps, the state determination and estimation unit 110 and the fluid quality estimation unit 130 may perform estimation for each group of the plurality of measurement values ​​associated with each other that are acquired simultaneously or in the same cycle by the measurement value acquisition unit 100.

[0038] In step S305, the state determination and estimation unit 110 determines whether the state of the fluid is superheated steam based on the measured values ​​of the fluid acquired by the measurement value acquisition unit 100. The state determination and estimation unit 110 can determine whether the state of the fluid is superheated steam by comparing the measured values ​​of the fluid acquired by the measurement value acquisition unit 100 with a threshold value. When the measured value of the fluid temperature, based on the measured value of the fluid pressure acquired by the measurement value acquisition unit 100, exceeds the boiling point of the fluid (the boiling point of the liquid contained in the fluid), the state determination and estimation unit 110 can determine that the state of the fluid is superheated steam (i.e., not wet steam). For example, the state determination and estimation unit 110 can obtain the boiling point corresponding to the measured value of the fluid pressure based on a vapor curve of the liquid in the fluid, and determine that the fluid is superheated steam when the measured value of the fluid temperature exceeds the boiling point. If the measured value of the fluid temperature exceeds the boiling point, the processing device 30 can proceed to step S330. If the measured value of the fluid temperature is equal to or less than the boiling point, the processing device 30 can proceed to step S310.

[0039] In step S310, the state determination and estimation unit 110 determines whether the state of the fluid is wet steam based on the measured value of the fluid acquired by the measured value acquisition unit 100. The state determination and estimation unit 110 can determine whether the fluid is wet steam based on at least one of the multiple measured values ​​by using a wet steam determination model that estimates whether the fluid is wet steam based on at least one of the multiple measured values. The wet steam determination model can be a model for outputting a determination result, such as logistic regression, a neural network, a support vector machine, a classification tree, change point detection, a k-nearest neighbor algorithm, or a k-means algorithm.

[0040] In addition, when the temperature measurement value of the fluid acquired by the measurement value acquisition unit 100 is less than the boiling point of the fluid, the state determination and estimation unit 110 may determine that the state of the fluid is wet steam (i.e., not superheated steam). In this case, the state determination and estimation unit 110 may determine the state of the fluid by using a steam curve.

[0041] In steps S305 and S310, the state determination estimation unit 110 determines whether the state of the fluid is wet steam based on at least one of the plurality of measurement values ​​acquired by the measurement value acquisition unit 100. If the state determination estimation unit 110 determines in step S305 that the fluid is not superheated steam and in step S310 that the fluid is wet steam, the process may proceed to step S320, and if the state determination estimation unit determines that the fluid is not wet steam, the process may proceed to step S330.

[0042] In step S320, if the state of the fluid is determined to be wet steam in steps S305 and S310, the state determination and estimation unit 110 estimates whether the state of the fluid is stratified flow, stratified wave flow, or annular mist flow based on at least one of the multiple measurement values ​​acquired by the measurement value acquisition unit 100. The state determination and estimation unit 110 can estimate the flow state of the fluid by using a first model for estimating the state of the fluid based on at least one of the multiple measurement values ​​of the fluid. The first model can be a classification model and can be a model such as logistic regression, a neural network, a support vector machine, a classification tree, a change point detection, a k-nearest neighbor algorithm, or a k-means algorithm.

[0043] In step S325, the selection unit 120 selects a calculation corresponding to the fluid state estimated by the state determination and estimation unit 110 from a plurality of calculations for estimating at least one of the liquid amount or the dryness. The selection unit 120 may select a second model corresponding to the fluid state from a plurality of second models for estimating at least one of the liquid amount or the dryness. The selection unit 120 may select a second model corresponding to the flow state of the fluid estimated by the state determination and estimation unit 110 from a second model corresponding to stratified flow, a second model corresponding to stratified wave flow, or a second model corresponding to annular mist flow. The plurality of second models may be generated through a learning process for each corresponding flow state and may be different from each other. The selection unit 120 may store the plurality of second models pre-generated by the learning processing unit 140.

[0044] The second model can be a regression model and can be a model such as multiple regression analysis, neural network, support vector regression, Gaussian process regression, regression tree, autoregressive model, logistic regression, classification tree, change point detection, k-nearest neighbor algorithm or k-means algorithm. The neural network can include a convolutional neural network, a recursive neural network or a long / short memory neural network. Since the signal intensity, density and vortex frequency of the piezoelectric element measured by the vortex flowmeter are related to the dryness or liquid amount, the second model can be a regression model for outputting the dryness or liquid amount based on at least one of the input of signal intensity, density or vortex frequency. As an example, since there is a correlation between the flow rate of the fluid calculated based on at least one of the signal intensity, density or vortex frequency obtained from the vortex flowmeter and the dryness or liquid amount of the fluid, the second model can output the dryness or liquid amount of the fluid by using at least one of the signal intensity, density or vortex frequency obtained from the vortex flowmeter as input. Note that the second model can be generated by a learning process using multiple measurement values ​​about the fluid (e.g., at least one of signal strength, density, or vortex frequency) as training data and using the dryness or liquid amount corresponding to the multiple measurement values ​​as teacher data.

[0045] In step S330, the fluid quality estimation unit 130 estimates at least one of the liquid amount or dryness of the fluid. The fluid quality estimation unit 130 may use at least one of the liquid amount or dryness output in response to inputting a plurality of measurement values ​​into the second model selected by the selection unit 120 as an estimation result. The fluid quality estimation unit 130 may input all types of measurement values ​​acquired by the measurement value acquisition unit 100 into the second model, or may input a predetermined type of measurement value from the plurality of measurement values ​​acquired by the measurement value acquisition unit 100 into the second model. The fluid quality estimation unit 130 may estimate one of the liquid amount and dryness, and calculate the other of the two based on the estimated one.

[0046] If the state determination and estimation unit 110 determines in step S305 or S310 that the state of the fluid is not wet steam, the fluid quality estimation unit 130 may estimate the liquid amount of the fluid to be 0 or the dryness of the fluid to be 100%. Since the state determination and estimation unit 110 determines that the fluid is superheated steam or is not wet steam, the fluid quality estimation unit 130 may use the liquid amount of 0 or the dryness of 100% as the estimation result. In this case, the fluid quality estimation unit 130 may calculate the enthalpy or heat of the superheated steam as the estimation result for the fluid based on the multiple measurement values.

[0047] In step S335, the output unit 150 outputs the estimation result of the fluid quality estimation unit 130 to the external device 40. The output unit 150 may output display data for displaying the estimation result on the external device 40. Furthermore, the output unit 150 may output data for notifying the user of the estimation result via an alarm or the like on the external device 40. For example, when the liquid amount or dryness as the estimation result exceeds a predetermined threshold or is equal to or less than a predetermined threshold, the output unit 150 may output data for notifying the user via an alarm on the external device 40. The predetermined threshold may be pre-set by the user or the like. Furthermore, the output unit 150 may output control data for controlling the plant corresponding to the estimation result to the external device 40, and the external device 40 may control the plant based on the control data.

[0048] Figure 4 is an explanatory diagram for explaining an example of the first model. Figure 4 In the example, the first model is a neural network model, with each node represented by a circle. In this first model, the measured values ​​(vortex frequency, flow velocity, signal amplitude, pressure, and temperature) of the fluid acquired by the measurement value acquisition unit 100 from the vortex flowmeter and other sensors are input to the nodes of the input layer, and the estimated results of the fluid flow state are output to the nodes of the output layer. The three nodes in the output layer are: a node that outputs the calculation results of stratified flow (e.g., the probability that the fluid is stratified flow), a node that outputs the calculation results of stratified wave flow (e.g., the probability that the liquid is stratified wave flow), and a node that outputs the calculation results of annular mist flow (e.g., the probability that the fluid is annular mist flow).

[0049] Any number of layers and neurons can be set in the intermediate layers of the first model. A general nonlinear function such as a ramp function, a sigmoid function, or a hyperbolic tangent function can be set as the activation function of the first model. From the perspective of the error backpropagation method, a softmax function can be provided for the output layer in multi-value classification.

[0050] In the first model, in response to the measurement values ​​being input to the input layer, the calculation results corresponding to the weights allocated between the nodes in the intermediate layer can be output to the nodes of the output layer, and the state determination estimation unit 110 can output the flow state with the highest probability as the state of the fluid. Note that the state determination estimation unit 110 can normalize each measurement value to a predetermined range and input the normalized value to each node of the input layer. For the measurement values, the learning processing unit 140 can perform a learning process by using a data set of multiple measurement values ​​as training data and using the flow state with respect to the measurement values ​​as teacher data, and optimize the weights to generate the first model. The multiple measurement values ​​in each flow state used by the learning processing unit 140 can be data obtained in advance through experiments, simulations, factory operations, etc.

[0051] The first model may have at least one node of an input layer, and may input all or some of the plurality of measurement values ​​regarding the fluid acquired by the measurement value acquisition unit 100 to the node of the input layer. Furthermore, the wet steam determination model may be a model similar to the first model, and may be subjected to a learning process similar to that of the first model.

[0052] Figure 5 is an explanatory diagram for explaining an example of the second model. Figure 5 In the example, the second model is a neural network model, and each node is represented by a circle. In the second model, the measured values ​​(vortex frequency, flow velocity, signal amplitude, pressure, and temperature) of the fluid acquired by the measurement value acquisition unit 100 from the vortex flowmeter and other sensors are input to each node of the input layer, and the liquid volume or dryness of the fluid is output to the nodes of the output layer. Any number of layers and any number of neurons can be set in the intermediate layer of the second model. A general nonlinear function such as a ramp function, a sigmoid function, or a hyperbolic tangent function can be set as the activation function of the second model. From the perspective of the error backpropagation method, a softmax function can be provided for the output layer in multi-value classification.

[0053] In the second model, in response to measurement values ​​being input to the input layer, calculation results (liquid volume or dryness) corresponding to the weights assigned between nodes in the intermediate layer are output to nodes in the output layer, and the fluid mass estimation unit 130 can output the output liquid volume or dryness as the estimated result. Note that the fluid mass estimation unit 130 can normalize each measurement value to a predetermined range and input the normalized value to each node in the input layer. The learning processing unit 140 can perform a learning process using a data set of multiple measurement values ​​obtained for each flow state as training data and optimize the weights to generate the second model. When the fluid is stratified flow, the learning processing unit 140 can generate a second model corresponding to stratified flow using the multiple measurement values ​​as training data. When the fluid is stratified wavy flow, the learning processing unit 140 can generate a second model corresponding to stratified wavy flow using the multiple measurement values ​​as training data. When the fluid is annular mist flow, the learning processing unit 140 can generate a second model corresponding to annular mist flow using the multiple measurement values ​​as training data. The plurality of measured values ​​in each flow state used by the learning processing unit 140 may be data previously obtained through experiments, simulations, plant operations, or the like.

[0054] The second model may have at least one node of an input layer, and all or some of the plurality of measurement values ​​about the fluid acquired by the measurement value acquisition unit 100 may be input to the node of the input layer.

[0055] The processing device 30 according to the present embodiment can estimate the state of the fluid by using a model corresponding to the state of the fluid, and estimate the liquid amount or dryness so that the estimation accuracy is high.

[0056] The fluid quality estimation unit 130 may perform calculations by using a function or a table that defines a relationship between the measurement value and the amount of liquid or dryness.

[0057] Alternatively, the processing device 30 may not include the learning processing unit 140 and may pre-store a first model and a second model, the first model being generated by an external learning device for estimating the state of the fluid based on at least one of a plurality of measured values ​​regarding the fluid, and the second model being generated by the external learning device for estimating at least one of the liquid volume or dryness of the fluid based on the estimated state of the fluid and at least one of the plurality of measured values ​​regarding the fluid. In this case, the state determination estimation unit 110 can estimate the state of the fluid using the stored first model, and the fluid quality estimation unit 130 can estimate at least one of the liquid volume or dryness using the stored second model. The learning device need only include at least the learning processing unit 140 of this embodiment. The learning device may be included in a computer such as a PC or an external device such as a cloud.

[0058] The various embodiments of the present invention may be described with reference to flowcharts and block diagrams, where blocks in the flowcharts and block diagrams may represent (1) steps of a process that performs an operation or (2) portions of an apparatus responsible for performing an operation. Some steps and portions may be implemented by dedicated circuits, programmable circuits provided with computer-readable instructions stored on a computer-readable medium, and / or processors provided with computer-readable instructions stored on a computer-readable medium. Dedicated circuits may include digital and / or analog hardware circuits, or may include integrated circuits (ICs) and / or discrete circuits. Programmable circuits may include reconfigurable hardware circuits having logical AND, OR, XOR, NAND, NOR, and other logical operations, flip-flops, registers, storage elements, and the like, such as field programmable gate arrays (FPGAs), programmable logic arrays (PLAs), and the like.

[0059] A computer-readable medium may include any tangible device capable of storing instructions for execution by an appropriate device, such that a computer-readable medium storing instructions includes an article of manufacture containing instructions that can be executed to create a method for performing the operations specified in the flowchart or block diagram. Examples of computer-readable media may include electronic storage media, magnetic storage media, optical storage media, electromagnetic storage media, semiconductor storage media, and the like. More specific examples of computer-readable media may include a floppy disk (registered trademark), a magnetic disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), electrically erasable programmable read-only memory (EEPROM), static random access memory (SRAM), compact disc read-only memory (CD-ROM), digital versatile disk (DVD), Blu-ray (registered trademark) disk, memory stick, integrated circuit card, and the like.

[0060] Computer-readable instructions may include: assembly instructions; instruction set architecture (ISA) instructions; machine instructions; machine-dependent instructions; microcode; firmware instructions; state-setting data; or any of source code or object code written in any combination of one or more programming languages ​​including object-oriented programming languages ​​such as Smalltalk (registered trademark), JAVA (registered trademark), and C++, and traditional procedural programming languages ​​such as the "C" programming language or similar programming languages.

[0061] Computer-readable instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable processing device, or a programmable circuit, locally or via a local area network (LAN) or a wide area network (WAN) (such as the Internet), and the computer-readable instructions can be executed to create a method for performing the operations specified in the flowchart or block diagram. Examples of processors include computer processors, processing units, microprocessors, digital signal processors, controllers, microcontrollers, etc.

[0062] Figure 6 An example of a computer 2200 is shown through which various aspects of the present invention can be implemented in whole or in part. The program installed in the computer 2200 can cause the computer 2200 to function as or perform operations associated with the apparatus of the embodiments of the present invention or one or more parts thereof, or perform the operations or one or more parts thereof, and / or cause the computer 2200 to perform a process according to the embodiments of the present invention or steps of the process. Such a program can be executed by the CPU 2212 to cause the computer 2200 to perform some operations associated with some or all of the blocks in the flowcharts and block diagrams described herein.

[0063] The computer 2200 according to this embodiment includes a CPU 2212, a RAM 2214, a graphics controller 2216, and a display device 2218, which are connected to each other via a main controller 2210. The computer 2200 also includes input / output units such as a communication interface 2222, a hard disk drive 2224, a DVD-ROM drive 2226, and an IC card drive, which are connected to the main controller 2210 via an input / output controller 2220. The computer also includes conventional input / output units such as a ROM 2230 and a keyboard 2242, which are connected to the input / output controller 2220 via an input / output chip 2240.

[0064] The CPU 2212 operates according to the programs stored in the ROM 2230 and the RAM 2214, thereby controlling each unit. The graphics controller 2216 obtains image data generated by the CPU 2212 from a frame buffer or the like in the RAM 2214 or in the graphics controller itself, and displays the image data on the display device 2218.

[0065] The communication interface 2222 communicates with other electronic devices via a network. The hard disk drive 2224 stores programs and data used by the CPU 2212 within the computer 2200. The DVD-ROM drive 2226 reads programs or data from the DVD-ROM 2201 and provides the programs or data to the hard disk drive 2224 via the RAM 2214. The IC card drive reads programs and data from an IC card and / or writes programs and data to an IC card.

[0066] The ROM 2230 stores a boot program executed when the computer 2200 is started and / or programs according to the hardware of the computer 2200. The input / output chip 2240 can also connect various input / output units to the input / output controller 2220 via a parallel port, a serial port, a keyboard port, a mouse port, etc.

[0067] The program can be provided via a computer-readable medium such as a DVD-ROM 2201 or an IC card. The program is read from the computer-readable medium, installed in the hard disk drive 2224, RAM 2214, or ROM 2230, which are also examples of computer-readable media, and executed by the CPU 2212. The information processing described in these programs is read by the computer 2200, and cooperation is provided between the programs and the various hardware resources described above. By implementing information operations or processing based on the use of the computer 2200, a device or method can be constructed.

[0068] For example, when the computer 2200 communicates with an external device, the CPU 2212 executes a communication program loaded in the RAM 2214 and instructs the communication interface 2222 to communicate based on the processing written in the communication program. Under the control of the CPU 2212, the communication interface 2222 reads transmission data stored in a transmission buffer processing area provided in a recording medium such as the RAM 2214, the hard disk drive 2224, the DVD-ROM 2201, or an IC card, and transmits the read transmission data to the network or writes received data received from the network to a reception buffer processing area provided on the recording medium.

[0069] Furthermore, the CPU 2212 can cause all or necessary parts of a file or database that has been stored in an external recording medium such as the hard disk drive 2224, the DVD-ROM drive 2226 (DVD-ROM 2201), an IC card, etc. to be read into the RAM 2214, and perform various types of processing on the data on the RAM 2214. Next, the CPU 2212 rewrites the processed data into the external recording medium.

[0070] Various information, such as various programs, data, tables, and databases, can be stored in the recording medium for information processing. The CPU 2212 can perform various processing on the data read from the RAM 2214 and rewrite the results to the RAM 2214. The processing described throughout this disclosure is specified by the instruction sequence of the program and includes various types of operations, information processing, conditional judgment, conditional branching, unconditional branching, information search / replacement, and the like. Furthermore, the CPU 2212 can search for information in files, databases, and the like in the recording medium. For example, when a recording medium stores multiple entries (each having an attribute value of a first attribute related to an attribute value of a second attribute), the CPU 2212 can search the multiple entries for an entry that matches the condition specified for the attribute value of the first attribute, and read the attribute value of the second attribute stored in the entry, thereby obtaining an attribute value of the second attribute related to the first attribute that satisfies the predetermined condition.

[0071] The above-described program or software module can be stored in a computer-readable medium on or near the computer 2200. In addition, a recording medium such as a hard disk or RAM provided in a server system connected to a dedicated communication network or the Internet can be used as a computer-readable medium, thereby providing the program to the computer 2200 via the network.

[0072] Although the present invention has been described with reference to the embodiments, the technical scope of the present invention is not limited to the above-described embodiments. It is obvious that those skilled in the art can make various changes or improvements to the above-described embodiments. It is also obvious from the scope of the claims that embodiments with such changes or improvements are included in the technical scope of the present invention.

[0073] The operations, procedures, steps, and stages of each process performed by the apparatus, system, program, and method described in the claims, embodiments, or drawings may be performed in any order as long as the order is not indicated by "prior to," "before," or the like, and as long as the output of the previous process is not used for the subsequent process. Even if the process flow is described using words such as "first" or "then" in the claims, embodiments, or drawings, this does not necessarily mean that the processes must be performed in that order. Reference Signs List

[0074] 10: system; 20: Instruments; 30: processing device; 40: external device; 100: measurement value acquisition unit; 110: state determination and estimation unit; 120: selection unit; 130: fluid quality estimation unit; 140: learning processing unit; 150: output unit; 2200: Computer; 2201: DVD-ROM; 2210: Main controller; 2212: CPU; 2214: RAM; 2216: Image controller; 2218: Display device; 2220: Input / output controller; 2222: Communication interface; 2224: Hard disk drive; 2226: DVD-ROM drive; 2230: ROM; 2240: Input / output chip; and 2242: Keyboard.

Claims

1. A device comprising: a measurement value acquisition unit that acquires a plurality of measurement values ​​regarding the fluid; a state determination estimation unit that estimates a state of the fluid based on at least one of the plurality of measurement values ​​acquired by the measurement value acquisition unit; a fluid quality estimation unit that estimates at least one of a liquid amount or a dryness with respect to the fluid from at least one of the plurality of measurement values ​​acquired by the measurement value acquisition unit, according to the state of the fluid estimated by the state determination estimation unit; and An output unit that outputs at least one of the liquid amount or the dryness estimated by the fluid mass estimation unit.

2. The device according to claim 1, further comprising a selection unit that selects a calculation corresponding to the state of the fluid estimated by the state determination estimation unit from a plurality of calculations, wherein The fluid quality estimation unit estimates at least one of the liquid amount or the dryness regarding the fluid from at least one of the plurality of measurement values ​​acquired by the measurement value acquisition unit through the calculation selected by the selection unit.

3. The device according to claim 1, wherein The state determination estimation unit estimates whether the state of the fluid is a laminar flow, a laminar wave flow, or an annular mist flow based on at least one of the plurality of measurement values ​​acquired by the measurement value acquisition unit.

4. The device according to claim 3, wherein The state determination estimation unit determining whether the state of the fluid is wet steam based on at least one of the plurality of measurement values ​​acquired by the measurement value acquisition unit, and If it is determined that the state of the fluid is wet steam, it is estimated whether the state of the fluid is a laminar flow, a laminar wave flow, or annular mist flow based on at least one of the plurality of measurement values ​​acquired by the measurement value acquisition unit.

5. The device according to claim 4, wherein The state determination estimation unit determines that the state of the fluid is not wet steam when the measured value of the temperature of the fluid exceeds the boiling point of the fluid based on the measured value of the pressure of the fluid acquired by the measured value acquisition unit, and If the state determination estimation unit determines that the state of the fluid is not wet steam, the fluid quality estimation unit estimates the amount of liquid with respect to the fluid as 0 or the dryness with respect to the fluid as 100%.

6. The device according to claim 1, wherein The state determination estimation unit estimates the state of the fluid based on at least one of the plurality of measurement values ​​acquired by the measurement value acquisition unit by using a first model for estimating the state of the fluid from at least one of the plurality of measurement values ​​regarding the fluid.

7. The device according to claim 2, wherein The selection unit selects, based on the state of the fluid estimated by the state determination and estimation unit, a second model corresponding to the state of the fluid estimated by the state determination and estimation unit from a plurality of second models for estimating at least one of the liquid amount or the dryness of the fluid based on at least one of the plurality of measurement values ​​regarding the fluid, and The fluid quality estimation unit estimates at least one of the liquid amount or the dryness regarding the fluid from at least one of the plurality of measurement values ​​acquired by the measurement value acquisition unit by using the second model selected by the selection unit.

8. The device according to claim 7, further comprising A learning processing unit generates, for each state of the fluid, the second model for estimating at least one of the liquid amount or the dryness of the fluid based on at least one of the plurality of measurement values ​​of the fluid.

9. The device according to claim 1, wherein The device stores a first model generated by an external learning device for estimating a state of the fluid based on at least one of the plurality of measurement values ​​of the fluid, and a second model generated by the external learning device for estimating at least one of the fluid amount or the dryness of the fluid based on at least one of the plurality of measurement values ​​of the fluid according to the state of the fluid estimated by the state determination estimation unit, The state determination estimation unit estimates the state of the fluid based on at least one of the plurality of measurement values ​​acquired by the measurement value acquisition unit by using the first model, and The fluid quality estimation unit estimates at least one of the liquid amount or the dryness regarding the fluid from at least one of the plurality of measurement values ​​acquired by the measurement value acquisition unit by using the second model.

10. The device according to claim 1, wherein The measurement value acquisition unit acquires at least one of flow velocity, temperature, pressure, viscosity, concentration, density, signal amplitude, or eddy current frequency measured for the fluid as the plurality of measurement values ​​regarding the fluid.

11. The device according to claim 1, wherein The measurement value acquisition unit acquires the plurality of measurement values ​​regarding the fluid from a vortex flowmeter.

12. A learning device comprising: A learning processing unit generates a second model for each state of the fluid, the second model being used to estimate at least one of a liquid amount or a dryness of the fluid based on at least one of a plurality of measurement values ​​of the fluid.

13. A method comprising: obtaining, by a computer, a plurality of measurements about the fluid; estimating, by the computer, a state of the fluid based on at least one of the plurality of measurements acquired by the computer; estimating, by the computer, at least one of a liquid amount or a dryness of the fluid based on at least one of the plurality of measurement values ​​acquired by the computer based on the state of the fluid estimated by the computer; as well as At least one of the amount of liquid or the dryness estimated by the computer is output by the computer.

14. A program for causing a computer to: a measurement value acquisition unit that acquires a plurality of measurement values ​​regarding the fluid; a state determination estimation unit that estimates a state of the fluid based on at least one of the plurality of measurement values ​​acquired by the measurement value acquisition unit; a fluid quality estimation unit that estimates at least one of a liquid amount or a dryness of the fluid based on at least one of the plurality of measurement values ​​acquired by the measurement value acquisition unit, based on the state of the fluid estimated by the state determination estimation unit; as well as An output unit that outputs at least one of the liquid amount or the dryness estimated by the fluid mass estimation unit.

15. A learning program that enables a computer to be used as A learning processing unit generates, for each state of a fluid, a second model for estimating at least one of a liquid amount or a dryness of the fluid based on at least one of a plurality of measurement values ​​of the fluid.

Images