High-precision ultrasonic heat metering algorithm and device with adaptive environmental compensation

Abstract

This invention relates to the field of ultrasonic heat metering technology, and discloses a high-precision ultrasonic heat metering algorithm and device with adaptive environmental compensation. The metering device includes a data acquisition module, a flow field distortion intensity calculation module, a flow field type classification module, a classification compensation module, a flow rate calculation module, a reliability assessment module, a temperature self-calibration module, a heat calculation module, and a parameter update module. The algorithm constructs a flow field distortion intensity index based on propagation time difference, echo amplitude asymmetry, and propagation time fluctuation. It calls the corresponding propagation time correction model for four types of distortion: eccentric, vortex, pulsating, and local interference types. Combined with sound velocity inversion and measurement reliability gating, it achieves online temperature self-calibration, ultimately completing the comprehensive calculation of heat values ​​and adaptive parameter updates.

Description

Technical Field

[0001] The embodiments disclosed herein relate to the field of ultrasonic heat metering technology, and in particular to a high-precision ultrasonic heat metering algorithm and device with adaptive environmental compensation. Background Technology

[0002] Ultrasonic heat meters are widely used in centralized heating, central air conditioning, and industrial heat energy metering due to their non-contact measurement, lack of moving parts, and wide range. However, with the continuous expansion of urban heating networks and the increasing demands for metering accuracy, the stable metering capability of ultrasonic heat meters under complex operating conditions has become a core bottleneck restricting their widespread application.

[0003] In actual operation, upstream disturbance sources such as pipe bends, valves, and pump sets can create various flow field distortions, including flow deviation, vortices, and pulsations, within the measurement section. This causes the ultrasonic wave propagation time difference to deviate from the theoretical value under the ideal laminar flow assumption, thus introducing systematic flow measurement errors. Simultaneously, the supply and return water temperature sensors operate in a high-temperature and high-humidity environment for extended periods, causing their zero-point drift and sensitivity drift to accumulate over time, increasing the temperature difference measurement error. These two types of errors superimpose in heat calculations, significantly impacting the accuracy of the measurement results.

[0004] Existing ultrasonic heat meter compensation schemes typically employ fixed correction coefficients or single filtering strategies to process propagation time, failing to distinguish between different types of flow field distortion, resulting in a mismatch between the compensation model and actual distortion characteristics. Regarding temperature calibration, existing schemes largely rely on periodic manual verification, lacking online self-calibration capabilities based on sound velocity inversion, and are unable to track and compensate for sensor drift in real time. The inventors of this disclosure note that current ultrasonic heat meter metering schemes do not consider the impact of different flow field distortion types on the selection of compensation strategies, nor the gating effect of measurement reliability on the temperature drift estimation update process. This leads to large flow compensation residuals under complex operating conditions and temperature drift estimation is susceptible to interference from abnormal measurement data, resulting in error accumulation. Summary of the Invention

[0005] The purpose of this invention is to provide a high-precision ultrasonic heat meter measurement algorithm and device with adaptive environmental compensation, which solves the technical problems of existing ultrasonic heat meters in complex flow field environments, such as the accumulation of propagation time measurement deviation due to flow field distortion, the drift error generated by long-term operation of temperature sensors, and the decrease in heat metering accuracy caused by the coupling effect of the two.

[0006] This disclosure provides at least one embodiment of an adaptive environmental compensation ultrasonic heat meter measurement method. The method includes: acquiring ultrasonic propagation link parameters, temperature parameters, and auxiliary parameters according to a set sampling period, and performing preprocessing operations on the acquired data to form a stable input sequence. The ultrasonic propagation link parameters include at least upstream propagation time, downstream propagation time, upstream echo amplitude, downstream echo amplitude, upstream propagation time standard deviation, and downstream propagation time standard deviation. The temperature parameters include at least the original measured values ​​of the supply water temperature, the original measured values ​​of the return water temperature, and the ambient temperature. The auxiliary parameters include at least the signal-to-noise ratio, gain, and transducer drive state. The method also includes calculating a flow field distortion intensity index based on the preprocessed propagation time data, echo amplitude data, and propagation time fluctuation data. and the flow field distortion intensity index With distortion determination threshold The comparison is performed, wherein the propagation time data includes upstream propagation time and downstream propagation time, the echo amplitude data includes upstream echo amplitude and downstream echo amplitude, and the propagation time fluctuation data includes upstream propagation time standard deviation and downstream propagation time standard deviation; in response to the flow field distortion intensity index Exceeding the distortion determination threshold A classification feature vector is constructed based on propagation time difference characteristics, echo amplitude characteristics, fluctuation characteristics, and time derivative characteristics, and the distortion type label is determined based on the classification feature vector. According to the distortion type label The corresponding propagation time correction model is invoked to classify and compensate the upstream and downstream propagation times of the current sampling period, resulting in the classified and compensated propagation time. Based on the classified and compensated propagation time, the corrected medium velocity and volumetric flow rate are calculated using the ultrasonic time-of-flight method. The measurement reliability is calculated based on the flow field distortion intensity index, signal-to-noise ratio, and distortion type label. The system controls the temperature drift parameter update strategy based on the comparison result between the measurement confidence level and the confidence threshold; within a high confidence window, it estimates the medium sound velocity based on the propagation time after classification compensation and inverts the local fluid temperature inversion value within the measurement pipe section, and completes online temperature self-calibration by combining the heat meter installation location and the original measured values ​​of the supply and return water temperatures; and calculates and accumulates the heat increment based on the volumetric flow rate and the calibrated supply and return water temperatures, wherein the flow field distortion intensity index... The propagation time difference offset term, echo amplitude asymmetry term, and propagation time normalization fluctuation term are obtained by weighted summation.

[0007] For example, in the method provided in at least one embodiment of this disclosure, the flow field distortion intensity index Obtain it using the following formula: ; in, This represents the propagation time difference between upstream and downstream within the current sampling period. To calibrate the reference propagation time difference under the operating conditions, and These are the upstream and downstream echo amplitudes, respectively. and These represent the standard deviations of propagation time upstream and downstream, respectively. and The upstream and downstream propagation times are respectively. To prevent extremely small positive numbers with a denominator of zero, , , The weighting coefficients are satisfied. .

[0008] For example, in the method provided in at least one embodiment of this disclosure, the classification feature vector It includes five components: the normalized offset of the propagation time difference relative to the reference value, the echo amplitude asymmetry, the normalized fluctuation intensity of the propagation time, the time derivative of the propagation time difference, and the transient deviation of the current propagation time difference relative to the mean of the sliding window; the distortion type label It is classified into one of the following: skew-type distortion, vortex-type distortion, pulsating distortion, and local interference distortion. Skew-type distortion is characterized by the echo amplitude asymmetry continuously deviating from zero and the sign remaining consistent over multiple consecutive sampling periods. Vortex-type distortion is characterized by the propagation time difference and echo amplitude asymmetry exhibiting periodic oscillations within the sliding window. Pulsating distortion is characterized by a significant increase in normalized fluctuation intensity and the absence of a periodic pattern. Local interference distortion is characterized by a sudden change in echo amplitude or propagation time difference within a short period of time and the duration of the change is short.

[0009] For example, in the method provided in at least one embodiment of this disclosure, the propagation time after classification compensation is obtained according to the following formula: , ,in, and These are the upstream and downstream reference propagation times under the calibration conditions. and These are the average upstream and downstream propagation times within the sliding window, respectively. , , To match the distortion type label The corresponding weight coefficients and satisfy .

[0010] For example, in the method provided in at least one embodiment of this disclosure, the weighting coefficient , , Obtain it using the following formula: ; ; ; in, To match the distortion type label The relevant attenuation coefficient, and To match the distortion type label The relevant basic weight parameters, The value is the flow field distortion intensity index.

[0011] For example, in the method provided in at least one embodiment of this disclosure, the measurement confidence level Obtain it using the following formula: ; in, The signal-to-noise ratio within the current sampling period. For reference signal-to-noise ratio, To match the distortion type label The corresponding type credibility factor and satisfy When the flow field distortion intensity index The distortion threshold was not exceeded. hour, The value is 1.

[0012] For example, in the method provided in at least one embodiment of this disclosure, the online temperature self-calibration includes: calculating the medium sound velocity based on the propagation time after classification compensation. The local fluid temperature in the measuring pipe section was obtained by inverting the relationship between sound velocity and temperature. Based on the measurement reliability Through the credibility gating function Control the update step size of the temperature drift estimate, where the supply water temperature drift estimate is... and return water temperature drift estimate Update according to the recursive formula, where the update driving term in the recursive formula is the local fluid temperature inversion value. The deviation between the original temperature measurement value and the temperature measured on the installation side; and the deviation through the temperature calibration objective function. Regarding the calibrated water supply temperature and calibrated return water temperature The calibrated supply and return water temperatures are obtained by finding the minimum value of the temperature calibration objective function. This includes drift compensation constraints, installation-side sound velocity inversion consistency constraints, and reference temperature difference constraints.

[0013] For example, in the method provided in at least one embodiment of this disclosure, the confidence gating function Defined as: ; in, The gating attenuation coefficient is... As a confidence threshold; the estimated water supply temperature drift. The recursive formula is: ; in, This is the current sampling sequence number. The inertial factor is the temperature drift factor. To update the gain for temperature drift, and These are the original measured values ​​of the supply water temperature and the return water temperature, respectively.

[0014] For example, the method provided in at least one embodiment of this disclosure further includes: adaptively updating the parameters involved in the classification compensation process based on long-term operating data and calibration residuals, wherein the updated parameters include the distortion determination threshold. The basic weight parameters and attenuation coefficients for each distortion type, and the type credibility factor. Credibility threshold Temperature drift inertia factor and temperature drift update gain.

[0015] At least one embodiment of this disclosure also provides an adaptive environmental compensation ultrasonic heat meter metering device, the device including a data acquisition module, a flow field distortion intensity calculation module, a flow field type classification module, a classification compensation module, a flow rate calculation module, a reliability assessment module, a temperature self-calibration module, and a heat calculation module; The data acquisition module is configured to acquire ultrasonic propagation link parameters, temperature parameters, and auxiliary parameters according to a set sampling period, and perform preprocessing operations on the acquired data to output a stable input sequence. The ultrasonic propagation link parameters include at least upstream propagation time, downstream propagation time, upstream echo amplitude, downstream echo amplitude, upstream propagation time standard deviation, and downstream propagation time standard deviation. The temperature parameters include at least the original measured values ​​of the supply water temperature, the original measured values ​​of the return water temperature, and the ambient temperature. The auxiliary parameters include at least the signal-to-noise ratio, gain, transducer drive status, and heat meter installation location parameters. The flow field distortion intensity calculation module is configured to calculate the flow field distortion intensity index based on the propagation time difference, echo amplitude asymmetry, and normalized propagation time fluctuation intensity in the stable input sequence. The flow field distortion intensity index is compared with the distortion judgment threshold to output the distortion judgment result; the flow field type classification module is configured to construct a classification feature vector and output a distortion type label based on propagation time difference features, echo amplitude features, fluctuation features, and time derivative features when the flow field distortion intensity index exceeds the distortion judgment threshold; the classification compensation module is configured to call the corresponding propagation time correction model according to the distortion type label and output the propagation time after classification compensation; the flow rate calculation module is configured to calculate the corrected medium velocity and volumetric flow rate based on the propagation time after classification compensation using the ultrasonic time difference method; the... The reliability assessment module is configured to calculate the measurement reliability based on the flow field distortion intensity index, signal-to-noise ratio, and distortion type label, and output a high-reliability window decision or a low-reliability window decision based on the comparison result of the measurement reliability and the reliability threshold; the temperature self-calibration module is configured to estimate the medium sound velocity based on the propagation time after classification compensation within the high-reliability window and invert the local fluid temperature inversion value in the measurement pipe section, and complete the online temperature self-calibration by combining the heat meter installation location parameters and the original measured values ​​of supply and return water temperatures; the heat calculation module is configured to calculate the heat increment based on the volumetric flow rate and the calibrated supply and return water temperatures and accumulate it.

[0016] The adaptive environmental compensation high-precision ultrasonic heat meter metering algorithm and device provided in this disclosure have the following beneficial effects.

[0017] Firstly, by constructing a flow field distortion intensity index formed by a weighted sum of a propagation time difference offset term, an echo amplitude asymmetry term, and a propagation time normalization fluctuation term, and triggering a classification compensation process when the flow field distortion intensity index exceeds the distortion judgment threshold, additional compensation calculations are only introduced when there is significant distortion in the flow field. Under normal operating conditions, the calculation path remains simple, which helps to reduce unnecessary computational overhead.

[0018] Secondly, by classifying flow field distortion into four types—deviation type, vortex type, pulsation type, and local disturbance type—and configuring a propagation time correction model weight coefficient combination that matches the physical cause of each type of distortion, a clear correspondence exists between the compensation strategy and the actual distortion characteristics, which helps to reduce the compensation residual under different distortion types.

[0019] Third, by introducing a measurement confidence level constructed from the product of flow field distortion intensity index, signal-to-noise ratio, and type confidence factor, and modulating the update step size of the temperature drift estimate with a confidence level gating function, the update of the temperature drift estimate is compressed or frozen during sampling periods with low measurement confidence, which helps to avoid abnormal measurement data from causing continuous interference to the drift estimation results.

[0020] Fourth, by obtaining the inversion value of the local fluid temperature in the measuring pipe section based on the speed of sound, and using it as one of the constraint terms of the objective function of temperature calibration on the installation side, the temperature after calibration is solved together with the original measured values ​​of the supply and return water temperatures. This enables the temperature calibration process to have online self-calibration capability, and can continuously track sensor drift without relying on external manual verification, which helps to reduce the accumulation of temperature difference measurement errors during long-term operation.

[0021] Fifth, by adaptively updating parameters such as distortion judgment threshold, weight parameters, confidence factor and temperature drift update gain based on the statistical characteristics of long-term compensation residuals in step S900, the parameters can be gradually adjusted with changes in operating conditions, which helps maintain the adaptability of the measurement algorithm in different operating stages. Attached Figure Description

[0022] Figure 1 This is a flowchart of the adaptive environmental compensation ultrasonic heat meter metering method provided in an embodiment of the present invention; Figure 2 This is a flowchart of the adaptive environmental compensation ultrasonic heat meter metering method provided in the embodiments of the present invention; Figure 3 This is a flowchart of the distortion classification and compensation switching process of the adaptive environmental compensation ultrasonic heat meter metering method provided in this embodiment of the invention. Figure 4 This is an algorithm state diagram of the adaptive environmental compensation ultrasonic heat meter metering method provided in this embodiment of the invention; Figure 5 This is a line graph showing the changes in flow field distortion intensity and measurement reliability of the adaptive environmental compensation ultrasonic heat meter measurement method as a function of sampling period, provided in this embodiment of the invention. Figure 6 This is a bar chart showing the compensation weights under different distortion types of the adaptive environmental compensation ultrasonic heat meter metering method provided in this embodiment of the invention. Detailed Implementation

[0023] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, some features described in the examples may be combined in other examples.

[0024] Example 1 like Figures 1-6 As shown, at least one embodiment of this disclosure provides an adaptive environmental compensation ultrasonic heat meter measurement method, which includes the following steps S100-S900.

[0025] Step S100: Data acquisition and preprocessing.

[0026] In step S100, the heat meter main control unit collects ultrasonic propagation link parameters, temperature parameters and auxiliary parameters in real time according to the set sampling period, and performs preprocessing operations on the collected data to form a stable input sequence.

[0027] For example, the parameters collected by the heat meter's main control unit in each sampling cycle include at least: upstream propagation time. Downstream transmission time Upstream echo amplitude Downstream echo amplitude Upstream propagation time standard deviation Downstream propagation time standard deviation Original measured value of water supply temperature Original measured value of return water temperature Ambient temperature And auxiliary parameters such as gain, signal-to-noise ratio, and transducer drive state. Among them, This is the time it takes for the ultrasonic wave to propagate along the direction of fluid flow. The time it takes for the ultrasonic wave to propagate against the direction of fluid flow; and These represent the peak amplitudes of the echo signals received by the transducers in the corresponding directions; and The standard deviation of the propagation time in multiple consecutive samplings is used to characterize the degree of fluctuation in propagation time.

[0028] In at least one example of the embodiments of this disclosure, the sampling period is configured to be in the range of 0.5 seconds to 5 seconds, preferably 1 second.

[0029] For example, the preprocessing operations include: sequentially performing despicing filtering, amplitude limiting, moving average smoothing, and outlier removal on the acquired raw propagation time series and echo amplitude series. Despicing filtering eliminates spikes introduced by electromagnetic interference or transducer transient response; amplitude limiting truncates sampled values ​​exceeding physically reasonable ranges to preset upper and lower limits; moving average smoothing calculates the arithmetic mean of continuous sampled values ​​within a set window length to suppress high-frequency noise; and outlier removal removes sampled points that deviate from the moving window mean by more than a preset multiple of the standard deviation from the current input sequence. After the above preprocessing, a stable input sequence is formed for subsequent steps.

[0030] It should be noted that the flow field distortion intensity index in step S200 All components are dimensionless ratios, with the propagation time difference offset term being... The time difference is normalized based on the reference propagation time difference, and the echo amplitude asymmetry term is calculated. The amplitude difference is normalized based on the sum of the amplitudes, and the propagation time is normalized to normalize the fluctuation term. The standard deviation is normalized based on the sum of propagation times; all three terms have been dimensionless and can be directly weighted and summed. The classification feature vector in step S300... Each component is constructed using the same benchmark normalization method, and each component is dimensionless. The measurement reliability in step S600... Signal-to-noise ratio components The signal-to-noise ratio is normalized to between 0 and 1 based on a reference signal-to-noise ratio. and Since they have the same dimensions, they can be multiplied directly.

[0031] Step S200: Construct flow field distortion intensity index.

[0032] In step S200, based on the preprocessed propagation time data, echo amplitude data, and propagation time fluctuation data, a flow field distortion intensity index is constructed to quantify the degree to which the current flow field deviates from the normal state. The system determines whether to proceed with the classification compensation process based on the comparison between the flow field distortion intensity index and the distortion judgment threshold.

[0033] For example, flow field distortion intensity index Obtain it in the following ways: ; in, This represents the time difference between upstream and downstream propagation within the current sampling period; The reference propagation time difference is used to stabilize or calibrate the operating conditions. To prevent extremely small positive numbers with a denominator of zero; , , Let be the weighting coefficients of each component, satisfying... In the above formula, the first term reflects the degree of deviation of the propagation time difference relative to the reference value, the second term reflects the degree of asymmetry in the amplitude of the upstream and downstream echoes, and the third term reflects the ratio of the intensity of the propagation time fluctuation to the propagation time itself. All three terms are dimensionless normalized quantities and can be directly weighted and summed.

[0034] In at least one example of the embodiments of this disclosure, when When the current flow field is determined to be in a normal or slightly disturbed state, among which... The distortion threshold is used for subsequent velocity calculations; at this point, the propagation time is directly used without classification compensation. If a significant distortion is detected in the current flow field, the process proceeds to the classification and compensation step S300. For example, the distortion determination threshold... It is configured to be in the range of 0.02 to 0.10.

[0035] Step S300: Construct classification features for flow field distortion types and determine the distortion type.

[0036] In step S300, the flow field distortion intensity index Exceeding the distortion threshold Under the condition of propagation time difference, echo amplitude characteristics, wave characteristics and their time derivative characteristics, a classification feature vector is constructed, and the current flow field distortion is classified into specific distortion types according to the classification feature vector.

[0037] For example, the feature vector for classifying flow field distortion types. Construct it in the following way: ; in, The time derivative of the propagation time difference is obtained by dividing the difference in propagation time difference between adjacent sampling periods by the sampling period; In the sliding window The mean propagation time difference within the component. It should be noted that the fourth component... The unit is time difference per sampling period, which differs in dimension from the other four dimensionless components; in classifying the flow field distortion type feature vector Before inputting into the classification model, the fourth component is separately subjected to mean normalization based on its range to eliminate the influence of dimensional differences on the classification results. The range used for normalization is determined by the maximum and minimum values ​​of the fourth component in historical operating condition samples. (Flow field distortion type classification feature vector) The five components represent: the normalized offset of the propagation time difference relative to the reference value, the asymmetry of the echo amplitude, the intensity of the normalized fluctuation of the propagation time, the rate of change of the propagation time difference, and the transient deviation of the current propagation time difference from the recent window mean.

[0038] In at least one example of the embodiments of this disclosure, feature vectors are classified according to the type of flow field distortion. The values ​​of each component are used to classify the flow field distortion into the following four types, and they are labeled with distortion type. It means that among them : This indicates skewed distortion, characterized by: echo amplitude asymmetry ( The second component of the propagation time difference (the second component) continuously deviates from zero and the sign remains consistent over multiple consecutive sampling periods. The propagation time difference offset direction is stable and the absolute value of the time derivative of the propagation time difference is small, indicating that the upstream and downstream acoustic paths are affected by a stable asymmetric velocity distribution caused by pipe bends or installation eccentricity. This indicates vortex-type distortion, characterized by: the asymmetry of propagation time difference and echo amplitude exhibiting periodic oscillations within the sliding window; the time derivative of propagation time difference alternating between positive and negative signs; and transient deviation components ( The fifth component oscillates around zero, indicating the presence of vortex structures in the fluid excited by upstream disturbance sources; , indicating pulsating distortion, characterized by: normalized fluctuation intensity ( The third component of the propagation time difference (the absolute value of the time derivative) increases significantly, and the absolute value of the propagation time difference time derivative shows multiple peaks in a short period of time, but it does not have the periodicity of vortex-type distortion, indicating that the flow rate has random short-term fluctuations caused by pumping pulsation or valve regulation. This indicates a localized interference-type distortion, characterized by a sudden change in echo amplitude or propagation time difference within a short period of time (the absolute value of the transient deviation component exceeds the preset mutation threshold). However, the mutation is short-lived and does not form a stable trend. After the mutation ends, each component returns to a near-normal range, indicating the presence of transient interference caused by bubbles, particles, or external vibrations.

[0039] It should be noted that the above classification method can be implemented using rule-based determination based on the threshold combination of each component, table lookup determination, or decision tree classification model. When rule-based determination is used, the determination criteria for each distortion type are based on the flow field distortion type classification feature vector. The magnitude relationship between each component and its corresponding threshold, along with the duration condition, are jointly determined. When using a decision tree classification model, the normalized flow field distortion type classification feature vector is used. As input, with distortion type labels As output, labeled historical working condition samples are used as training data. Gini impurity is used as the node splitting criterion, and the tree structure is trained through recursive splitting. The number of the above four distortion types can be expanded or merged according to the actual application scenario.

[0040] Step S400: Switch the propagation time compensation model according to the distortion type.

[0041] In step S400, based on the distortion type label determined in step S300 Calling and distortion type tags The corresponding propagation time correction model classifies and compensates for the current propagation time difference to obtain the propagation time after classification compensation. and .

[0042] For example, the propagation time after classification compensation is obtained in the following way: ; ; in, and These are the upstream and downstream reference propagation times under stable or calibrated operating conditions, respectively. and Sliding windows The average upstream and downstream propagation time within the region; , , For distortion type tags The corresponding weighting coefficients satisfy In the above formula , , The three quantities have the same dimensions (all are time units), and the weighted sum is consistent in terms of dimensions.

[0043] In at least one example of the embodiments of this disclosure, the weighting coefficients corresponding to different distortion types are configured in the following manner: when (Deviation-type distortion) Compared to and The reference propagation time is increased, giving it a larger weight in the compensation result, which is used to suppress the systematic propagation time shift caused by the steady-state bias. when (During vortex distortion) Compared to and The value is increased, giving the sliding window mean a larger weight in the compensation result, which is used to suppress the periodic fluctuations in propagation time caused by the vortex structure. when (Pulsating distortion) Improved and The time filtering effect is enhanced by reducing the time filter, thus suppressing short-term fluctuations caused by pulsation; when (Local interference type distortion) The weight of the current abnormal sample is significantly reduced and compressed. When the current sample value deviates from the mean of the sliding window by more than a preset deviation threshold, the most recent reliable sample value is used to replace the current sample value in the compensation calculation.

[0044] It should be noted that one preferred way to define the weights is as follows: ; ; ; in, For distortion type tags The relevant attenuation coefficient, and For distortion type tags The relevant basic weight parameters. The above definition method makes the flow field distortion intensity index... When the value increases, the weight of the current measured value Adaptive reduction, reference value weight and window mean weight The corresponding increase allows for adaptive adjustment of the compensation intensity according to the degree of distortion. , , The range of values ​​is updated in step S900.

[0045] Step S500: Correct the flow rate and volumetric flow rate calculation.

[0046] In step S500, the propagation time after classification compensation is obtained in step S400. and The corrected medium velocity and volumetric flow rate are calculated based on the principle of ultrasonic time difference method.

[0047] For example, the corrected medium flow rate Obtain it in the following ways: ; in, Equivalent ultrasonic path length, which is the path length of ultrasonic waves in a fluid from the transmitting transducer to the receiving transducer. This is the projected length of the ultrasonic propagation path along the pipe axis, i.e. , The angle between the ultrasonic propagation path and the pipeline axis.

[0048] Volumetric flow rate Obtain it in the following ways: ; in, This refers to the effective cross-sectional area of ​​the pipe. The calibration correction factor is determined during factory calibration or field calibration and is used to compensate for the difference between the flow velocity distribution and the average flow velocity in the pipeline cross section.

[0049] Step S600: Build measurement confidence and control temperature drift parameter updates.

[0050] In step S600, a measurement reliability index is constructed based on the flow field distortion intensity index, signal quality parameters, and distortion type labels, and an update strategy for the temperature drift parameters is determined based on the comparison results between the measurement reliability index and the reliability threshold.

[0051] For example, measuring credibility Obtain it in the following ways: ; in, The signal-to-noise ratio within the current sampling period; The reference signal-to-noise ratio is used to normalize the signal-to-noise ratio components to a range between 0 and 1. To match the current distortion type label The relevant type credibility factor satisfies The type reliability factor reflects the reliability of measurement results under different distortion types, with the skew distortion type corresponding to... Taking values ​​close to 1, the vortex distortion corresponds to Take the average value, corresponding to pulsating distortion Corresponding to local interference-type distortion Take the smaller value. In the above formula , and All three terms are dimensionless normalized quantities and can be directly multiplied.

[0052] In at least one example of the embodiments of this disclosure, when When the current sampling period is determined to be within a high-confidence window, then... As a confidence threshold, the temperature drift parameter is allowed to be updated with normal gain; when When the current sampling period is determined to be within a low-confidence window, the temperature drift estimate is frozen during updates or updated at a smaller gain to avoid the accumulation of temperature drift estimation errors introduced by unreliable measurement data under significantly distorted operating conditions. For example, a confidence threshold... It is configured in the range of 0.5 to 0.8.

[0053] It should be noted that when the flow field distortion intensity index Exceeding the distortion threshold And when the flow field is in a distorted state, the distortion type label output in step S300 Pick A certain value in, Take the corresponding type credibility factor; when the flow field distortion intensity index At this time, the flow field is in a normal state. Set the value to 1 to measure the confidence level. Degenerate into It can still be used to control the update strategy of temperature drift parameters.

[0054] Step S700: Online self-calibration of temperature based on sound velocity inversion.

[0055] In step S700, within the high confidence window, the medium sound velocity is estimated based on the propagation time after classification compensation, and the local fluid temperature value in the measurement pipe section obtained by sound velocity inversion is used as the installation-side calibration constraint, which together with the original measured values ​​of supply and return water temperatures completes the online temperature self-calibration.

[0056] For example, the speed of sound in a medium Obtain it in the following ways: ; in, This is the equivalent ultrasonic path length. and The upstream and downstream propagation times after classification and compensation.

[0057] Based on the correlation between the sound velocity and temperature of the medium, the inverted value of the local fluid temperature in the measuring pipe section is obtained. This correspondence is determined by the sound velocity-temperature characteristic curve of the medium (e.g., water) within the operating temperature range. Specifically, based on the measured sound velocity values ​​at several known temperature points within the operating temperature range, a polynomial fitting method is used to establish the sound velocity... Regarding temperature Fitting polynomial ,in For the temperature of the medium, To fit a polynomial function; the currently estimated medium sound velocity Substituting the inverse function relation of the fitted polynomial, that is, letting The local fluid temperature inversion value is obtained through numerical solution or table lookup interpolation. .

[0058] Furthermore, depending on whether the heat meter is installed on the supply or return side, the initial temperature on the installation side is defined. for: ; Furthermore, temperature drift updates are performed as follows: Supply water temperature drift estimate. and return water temperature drift estimate Update according to the following recursive formulas: ; ; in, This is the current sampling sequence number; and The temperature drift inertia factor, with a value ranging from 0.90 to 0.99, is used to control the decay rate of the historical drift estimate; and The gain is updated for temperature drift, with a value ranging from 0.001 to 0.05, and is used to control the step size of each update. The credibility gating function is defined as follows: ; in, The threshold attenuation coefficient is preferably 5 or another small constant. This is the credibility threshold. When... When the confidence level is 1, the output of the confidence level gate function is 1, and the temperature drift parameter is updated according to the normal gain; when At this point, the output value of the confidence gating function decays exponentially as confidence decreases, and the update step size of the temperature drift parameter is correspondingly compressed. In the above recursive formula, , and amendments All of these are in units of temperature, and all dimensions are consistent, so they can be added directly.

[0059] It should be noted that the estimated water supply temperature drift in the above recursive formula... Estimated return water temperature drift Using the same installation-side sound velocity inversion temperature constraint term Driven, water supply temperature drift estimation Estimated return water temperature drift Gain is updated through independent temperature drift. and and temperature drift inertia factor and Differentiated drift estimation rates are achieved to track the drift characteristics of the supply and return water sensors separately.

[0060] In at least one example of the embodiments of this disclosure, a temperature calibration objective function is further constructed. To obtain the calibrated supply and return water temperatures: ;

[0061] in, and These are the calibrated supply water temperature and the calibrated return water temperature, respectively, which need to be solved. and These are the supply water temperature drift estimates and return water temperature drift estimates for the current sampling period; The reference temperature difference is obtained based on the system's physical constraints or historical stable operating conditions. , , , These are the weighting coefficients for each item. Temperature calibration objective function. The deviation between the temperature values ​​after the first and second constraint calibrations and the original measured values ​​after drift compensation; the inverse value of the mounting-side temperature and the local fluid temperature after the third constraint calibration. The consistency between them, the fourth constraint being the deviation between the calibrated supply and return water temperature difference and the reference temperature difference. The quantities within parentheses in each term of the temperature calibration objective function are all temperature differences, with consistent dimensions. After squaring, the dimensions of each term are the same, and they can be directly summed under the weighting coefficients. Through the temperature calibration objective function... about and Find the minimum value to obtain the water supply temperature after online calibration. and return water temperature .

[0062] It should be noted that the temperature calibration objective function For about and The quadratic function, the temperature calibration objective function The minimum value can be obtained by adjusting the value of the minimum ... and The analytical solution is obtained by taking the partial derivatives and setting them to zero, without the need for iterative optimization.

[0063] Step S800: Comprehensive calculation of calorie value.

[0064] In step S800, based on the volumetric flow rate after classification compensation obtained in step S500 and the calibrated supply and return water temperatures obtained in step S700 by combining the installation-side sound velocity inversion constraint and the reference temperature difference constraint, the heat increment of the current sampling period is calculated and accumulated.

[0065] For example, the heat increment in the current sampling period Obtain it in the following ways: ; in, The volumetric flow rate after classification compensation; The density of the medium obtained or calculated at the supply and return water temperatures after calibration; The specific heat capacity of the medium obtained or calculated at the supply and return water temperatures after calibration; The calibrated supply and return water temperature difference; This is the duration of the sampling period.

[0066] Cumulative heat Obtain it in the following ways: ; in, This represents the cumulative number of sampling periods up to the current moment. For the first The heat increment per sampling period This is the sampling period number.

[0067] Step S900: Adaptive update of classification model parameters.

[0068] In step S900, the parameters involved in the classification compensation process are adaptively updated based on long-term operating data and calibration residuals.

[0069] For example, the updated parameters include at least the distortion determination threshold. The basic weight parameters corresponding to each distortion type , and attenuation coefficient Type confidence factor corresponding to each distortion type Credibility threshold Temperature drift inertia factor and Temperature drift update gain and And the fitting parameters for the physical properties of the medium.

[0070] In at least one example of the embodiments of this disclosure, the triggering conditions for parameter updates include: the system's cumulative runtime reaching a preset update cycle, or new calibration residual data being acquired during the most recent verification and calibration process. The parameter update method involves comparing the compensated residual sequences for each distortion type accumulated during long-term operation with the calibration reference values, and adjusting the weight parameters and threshold parameters for the corresponding distortion types based on the residual statistical characteristics, so that the deviation between the compensated propagation time and the calibration reference value is further reduced in subsequent operations.

[0071] The adaptive environmental compensation ultrasonic heat meter metering device provided in at least one embodiment of this disclosure includes a data acquisition module, a flow field distortion intensity calculation module, a flow field type classification module, a classification compensation module, a flow rate calculation module, a reliability assessment module, a temperature self-calibration module, a heat calculation module, a parameter update module, and a storage and communication module.

[0072] The data acquisition module is configured to acquire ultrasonic propagation time, echo amplitude, propagation time fluctuation, supply and return water temperature, ambient temperature, and auxiliary parameters such as gain, signal-to-noise ratio, and transducer drive status, and to perform preprocessing operations on the acquired data to output a stable input sequence.

[0073] The flow field distortion intensity calculation module is configured to receive a stable input sequence from the data acquisition module and calculate the flow field distortion intensity index based on the propagation time difference, echo amplitude asymmetry, and propagation time normalized fluctuation intensity. and the flow field distortion intensity index With distortion determination threshold The comparison is performed to output the distortion determination result.

[0074] The flow field type classification module is configured to classify the flow field distortion intensity index. Exceeding the distortion threshold At that time, a stable input sequence is received and a flow field distortion type classification feature vector is constructed. Classification feature vectors based on flow field distortion type The current flow field distortion is classified into eccentric distortion, vortex distortion, pulsating distortion, or local disturbance distortion, and distortion type labels are output. .

[0075] The classification compensation module is configured to receive distortion type tags. And a stable input sequence, based on the distortion type label Call the corresponding propagation time correction model and output the propagation time after classification compensation. and .

[0076] The traffic calculation module is configured to receive the propagation time after classification compensation. and The corrected medium flow velocity was calculated based on the principle of ultrasonic time-of-flight method. and volumetric flow rate .

[0077] The credibility assessment module is configured to receive flow field distortion intensity indicators. Signal-to-noise ratio and distortion type labels Calculate the reliability of the measurement And based on the measurement reliability With credibility threshold The comparison results output either a high-confidence window decision or a low-confidence window decision.

[0078] The temperature self-calibration module is configured to receive the propagation time after classification compensation under confidence gating. and and the original measured values ​​of supply and return water temperatures and Update the water supply temperature drift estimate based on the sound velocity inversion results. and return water temperature drift estimate And by minimizing the temperature calibration objective function Output calibrated water supply temperature and return water temperature .

[0079] The heat calculation module is configured to receive the volumetric flow rate after classification compensation. and calibrated water supply temperature and return water temperature Calculate the heat increment for the current sampling period. And accumulate the cumulative heat .

[0080] The parameter update module is configured to adjust the distortion threshold based on long-term operating data and calibration residuals. Basic weight parameters and attenuation coefficient Type credibility factor Credibility threshold Temperature drift inertia factor and Temperature drift update gain and And adaptive updates are made to the fitting parameters of the medium properties.

[0081] The storage and communication module is configured to store historical operating data, parameter tables used by each module, and heat calculation results, and output the cumulative heat value to the external system through a preset communication interface. Current operating status and parameter update records.

[0082] In at least one example of the embodiments of this disclosure, the ultrasonic heat meter measurement method with adaptive environmental compensation described above is applied, and its application background is as follows: In branch heat metering scenarios within secondary heating networks, ultrasonic heat meters are often installed near valves, elbows, or reducers. If the upstream straight pipe section is insufficient in length, the fluid is prone to flow deviation when entering the measuring pipe section, causing the velocity distribution along the upstream and downstream ultrasonic propagation paths to become asymmetrical, thus deviating the ideal relationship under calibration conditions from the propagation time difference.

[0083] This example assumes an ultrasonic heat meter is installed at the outlet of a branch road in a heat exchange station, close to a 90° bend upstream, where slight flow distortion has already occurred. The system sampling period is 1 second. This calculation focuses on one sampling period to illustrate the complete processing chain of the algorithm in an engineering scenario.

[0084] The operating conditions in this example are as follows: the medium is heating circulating water; the installation location is near an elbow, and there is a slight stable flow deviation; the current system is in normal heating operation; the sampling period is 1 second; the signal quality is good in the current period, but the flow field distortion has reached the level that needs to be compensated.

[0085] The specific measurement steps in this example are as follows: Step S100: Data acquisition and preprocessing; The purpose of this step is to determine the stable input quantities that can be used for calculation in this cycle. By performing de-glitching, amplitude limiting, and moving average processing on the propagation time, echo amplitude, and temperature data, we can avoid instantaneous outliers directly affecting subsequent distortion determination and heat calculation.

[0086] Assuming that the propagation time, echo amplitude, temperature and signal-to-noise ratio in the current cycle have been preprocessed as described above, the data in Table 1 can be used directly for calculation.

[0087] Table 1 Sensors and Operating Parameters

[0088] At this point, the stable input sequence formed by the system includes at least: Propagation time series: , ; Echo amplitude sequence: , ; Volatility sequence: , ; Temperature sequence: , .

[0089] Step S200: Calculate the flow field distortion intensity index ; The purpose of this step is to combine information from three aspects—propagation time difference offset, echo amplitude asymmetry, and propagation time fluctuation—into a single indicator in order to quickly determine whether the current operating condition has deviated from the normal measurement state.

[0090] First, calculate the propagation time difference: ; According to the formula: ; Calculate each item separately: ; ; ; Substitute the weights: ; ; Judgment result: ; Therefore, the system determines that there is a significant distortion in the current operating condition, and it needs to enter the classification and compensation process.

[0091] Step S300: Determine the distortion type; After confirming the existence of significant distortion, this step further determines the type of distortion. Only after type identification is completed can subsequent steps invoke the corresponding propagation time compensation model, rather than using the same correction logic for all abnormal operating conditions.

[0092] Based on the collected features, we can know that: The upstream and downstream echo amplitudes have long been asymmetrical. > ; The propagation time difference offset direction is stable; Fluctuations exist but do not exhibit obvious periodic oscillations; The derivative does not change drastically.

[0093] Therefore, it is determined to be a skewed distortion: ; Step S400: Calculate the bias compensation weight and correct the propagation time; The purpose of this step is to reallocate the weights of the current sample, the reference sample, and the window mean based on the characteristics of the eccentric flow distortion, so as to obtain a propagation time result that is more consistent with the real flow field, thereby providing a more accurate input for flow velocity calculation.

[0094] The parameters for the eccentric flow pattern are: ; First, calculate the exponent term: ; Calculate the denominator: ; We obtain three types of weights: ; ; ; check: ; Recalculate the compensated propagation time: ; ; Substitute the values:

[0095]

[0096] ;

[0097]

[0098] ; The compensated propagation time difference is: ; Compared to the original value of 0.0672s, the compensated result is closer to the reference value of 0.0600s.

[0099] Step S500: Calculate the corrected flow rate and volumetric flow rate; This step converts the compensated propagation time back into flow velocity and volumetric flow rate. Since heat calculations directly depend on flow rate values, this step is crucial for converting the acoustic compensation results into heat calculation inputs.

[0100] First, calculate the projected length: ; According to the flow velocity formula: ; Previous coefficient: ; Time ratio item:

[0101] ; therefore: ; Volumetric flow rate:

[0102] ; Unit conversion: ; Step S600: Calculate measurement reliability ; The purpose of this step is to quantify the reliability of the measurement results for this period by comprehensively considering the degree of distortion, signal quality, and distortion type. Reliability not only reflects the reliability of the current measurement results but also determines whether the temperature drift parameter can be updated normally in the next step.

[0103] According to the formula:

[0104] In this example:

[0105] Substitute the values:

[0106]

[0107] determination:

[0108] Therefore, the current sampling period falls within a high-confidence window, and the temperature drift parameter can be updated according to the normal gain.

[0109] Step S700: Sound velocity inversion and online temperature self-calibration; The purpose of this step is to estimate the medium's sound velocity using the propagation time after classification compensation, and to correct the supply and return water temperatures online by using sound velocity inversion temperature constraints, thereby reducing the heat error caused by long-term drift of the temperature sensor.

[0110] 1. Calculate the velocity of sound in the medium ; Substitute: ; Since the heat meter is installed at the branch outlet, the acoustic measurement section corresponds to the local fluid temperature on the water supply side. Based on the water medium sound velocity-temperature relationship, the following can be obtained by referring to a table or interpolation: ; 2. Calculate the installation-side temperature drift drive term. ; ; 3. Update temperature drift estimate Let the drift estimate of the previous period be: ; Since the current window is of high confidence, and in this example the sound velocity inversion temperature directly corresponds to the water supply side, the water supply side drift estimate is updated according to the driving term, while the return water side drift estimate retains its inertia. ; therefore: ; ; ; ; 4. Establish and solve the temperature calibration objective function. First, calculate the temperature reference after drift compensation: ; ; Since the sound velocity-inverted temperature directly constrains the water supply side, the temperature calibration objective function is written as:

[0111] right Taking the partial derivatives of each equation and setting them to zero, we obtain the system of linear equations: ; ; The solution yields: ; ; Temperature difference after calibration: ; Step S800: Calculate the heat increment; The purpose of this step is to substitute the previously obtained corrected flow rate, calibration temperature, and medium property parameters into the heat formula, output the heat increment for this sampling period, and further convert it into success rate and cumulative heat.

[0112] Take the medium property parameters corresponding to the calibration temperature: ; Sampling period: ; Formula for heat increment: ; First, calculate the mass flow rate: ; Calculate the heat increment again: ; Right now: ; Conversion success rate: ; If this operating condition is maintained for 10 minutes, the cumulative heat will be: ; Step S900: Parameter Update Instructions The purpose of this step is to save the key results generated in this cycle as a basis for long-term adaptive updates. It does not change the heat values ​​already output in this cycle, but rather provides historical support for optimizing thresholds, compensation weights, and reliability parameters in subsequent cycles.

[0113] For this sampling period, the system will record the following intermediate results as a basis for long-term parameter updates: ; Distortion type ; Current credibility ; This period-compensated propagation time and volumetric flow rate; Updated results for temperature drift estimation; This period's heat increase.

[0114] If multiple consecutive cycles are identified as skewed flow distortion, the system can fine-tune the skewed flow parameters in subsequent updates. To further reduce long-term residuals.

[0115] In summary, this example demonstrates that the original propagation time difference has deviated from the reference value due to the biased flow condition. Without compensation, the flow velocity and heat will exhibit systematic deviations. After classification compensation, the propagation time difference was reduced from 0.0672μs to 0.0638μs, which is closer to the reference value; If the confidence threshold is met, the system allows the temperature drift parameter to continue to be updated, so that the temperature value is consistent with the sound speed inversion result; By combining flow compensation and temperature calibration, a more stable heat calculation result is finally obtained.

[0116] The embodiments of the present invention have been described above. However, the embodiments are not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make more equivalent embodiments under the guidance of the present embodiments, and all of them are within the protection scope of the present embodiments.

Claims

1. A method for measuring ultrasonic heat meters with adaptive environmental compensation, characterized in that, Includes the following steps: Ultrasonic propagation link parameters, temperature parameters, and auxiliary parameters are collected according to a set sampling period, and the collected data are preprocessed to form a stable input sequence. The ultrasonic propagation link parameters include upstream propagation time, downstream propagation time, upstream echo amplitude, downstream echo amplitude, upstream propagation time standard deviation, and downstream propagation time standard deviation. The temperature parameters include the original measured values ​​of the supply water temperature and the return water temperature. The auxiliary parameters include the signal-to-noise ratio. Based on the preprocessed ultrasonic propagation link parameters and temperature parameters, the flow field distortion intensity index is calculated, and the flow field distortion intensity index is compared with the distortion judgment threshold. The flow field distortion intensity index is obtained by weighted summation of the propagation time difference offset term, the echo amplitude asymmetry term, and the propagation time normalized fluctuation term. In response to the flow field distortion intensity index exceeding the distortion judgment threshold, a classification feature vector is constructed based on the propagation time difference feature, echo amplitude feature, fluctuation feature and propagation time difference time derivative feature, and the distortion type label corresponding to the current flow field is determined according to the classification feature vector; Based on the distortion type label, the propagation time correction model corresponding to the distortion type label is invoked to classify and compensate the upstream propagation time and downstream propagation time of the current sampling period, so as to obtain the classified and compensated upstream propagation time and the classified and compensated downstream propagation time. Based on the upstream propagation time and the downstream propagation time after classification compensation, the corrected medium velocity and volumetric flow rate are calculated according to the ultrasonic time difference method. The measurement confidence level is calculated based on the flow field distortion intensity index, the signal-to-noise ratio, and the distortion type label, and the measurement confidence level is compared with the confidence threshold. When the measurement confidence level is greater than or equal to the confidence level threshold, the current sampling period is determined to be in a high confidence window. Within the high confidence window, the medium sound velocity is estimated based on the upstream propagation time after classification compensation and the downstream propagation time after classification compensation. The local fluid temperature inversion value in the measurement pipe section is obtained based on the medium sound velocity. The local fluid temperature inversion value is associated with the corresponding installation side temperature according to the installation position of the heat meter on the water supply side or the water return side. Then, the temperature drift parameter is updated based on the local fluid temperature inversion value, the original measurement value of the water supply temperature and the original measurement value of the water return temperature, and the online self-calibration of the water supply and return temperatures is completed. The heat increment is calculated and accumulated based on the volumetric flow rate and the calibrated supply and return water temperatures. The upstream propagation time after classification compensation and the downstream propagation time after classification compensation are used as common inputs for both flow calculation and online temperature self-calibration.

2. The adaptive environmental compensation ultrasonic heat meter metering method according to claim 1, characterized in that, The flow field distortion intensity index Obtain it using the following formula: ； in, This represents the propagation time difference between upstream and downstream within the current sampling period. To calibrate the reference propagation time difference under the operating conditions, and These are the upstream and downstream echo amplitudes, respectively. and These represent the standard deviations of propagation time upstream and downstream, respectively. and The upstream and downstream propagation times are respectively. To prevent extremely small positive numbers with a denominator of zero, , , The weighting coefficients are satisfied. .

3. The adaptive environmental compensation ultrasonic heat meter metering method according to claim 1, characterized in that, The classification feature vector It includes five components: normalized offset of propagation time difference relative to reference value, echo amplitude asymmetry, normalized propagation time fluctuation intensity, time derivative of propagation time difference, and transient deviation of current propagation time difference relative to sliding window mean. The distortion type label is determined based on the classification feature vector. Includes: labeling the distortion type It is classified into one of the following: eccentric distortion, vortex distortion, pulsating distortion, and local disturbance distortion; The characteristics of the skew-type distortion are that the echo amplitude asymmetry continuously deviates from zero and the sign remains consistent over multiple consecutive sampling periods; the characteristics of the vortex-type distortion are that the propagation time difference and echo amplitude asymmetry exhibit periodic oscillations within the sliding window; the characteristics of the pulsating distortion are that the normalized fluctuation intensity is significantly increased and does not have a periodic pattern; and the characteristics of the local interference-type distortion are that the echo amplitude or propagation time difference undergoes abrupt changes in a short period of time and the duration of the abrupt changes is short.

4. The adaptive environmental compensation ultrasonic heat meter metering method according to claim 1, characterized in that, The label based on the distortion type Invoke the corresponding propagation time correction model to perform classification compensation for the current propagation time, including: The propagation time after classification compensation is obtained according to the following formula: ； ； in, and These are the upstream and downstream reference propagation times under the calibration conditions. and These are the average upstream and downstream propagation times within the sliding window, respectively. , , To match the distortion type label The corresponding weight coefficients and satisfy .

5. The adaptive environmental compensation ultrasonic heat meter metering method according to claim 4, characterized in that, The weighting coefficient , , Obtain it using the following formula: ； ； ； in, To match the distortion type label The relevant attenuation coefficient, and To match the distortion type label The relevant basic weight parameters, The value is the flow field distortion intensity index.

6. The adaptive environmental compensation ultrasonic heat meter metering method according to claim 1, characterized in that, The measurement reliability Obtain it using the following formula: ； in, The signal-to-noise ratio within the current sampling period. For reference signal-to-noise ratio, To match the distortion type label The corresponding type credibility factor, and satisfying ; When the flow field distortion intensity index The distortion threshold was not exceeded. hour, The value is 1.

7. The adaptive environmental compensation ultrasonic heat meter metering method according to claim 1, characterized in that, The online temperature self-calibration includes: When the measurement reliability Greater than or equal to the credibility threshold At that time, based on the upstream propagation time after the classification compensation and the downstream propagation time after the classification compensation Calculate the speed of sound in the medium The local fluid temperature in the measuring pipe section was obtained by inverting the relationship between sound velocity and temperature. ; Based on the local fluid temperature inversion value The deviation between the original temperature measurement value and the original temperature measurement value corresponding to the installation side, combined with the original water supply temperature measurement value. and the original measured value of return water temperature Through the credibility gating function Controlled water supply temperature drift estimation and return water temperature drift estimate The update step size is used to update the temperature drift parameters; Based on the updated supply water temperature drift estimate Updated return water temperature drift estimate The local fluid temperature inversion value And construct the temperature calibration objective function with reference temperature difference constraints. And by applying the temperature calibration objective function Regarding the calibrated water supply temperature and calibrated return water temperature Find the minimum value to obtain the calibrated supply water temperature and calibrated return water temperature; Wherein, the temperature calibration objective function It includes at least drift compensation constraints, sound speed inversion consistency constraints, and reference temperature difference constraints.

8. The adaptive environmental compensation ultrasonic heat meter metering method according to claim 7, characterized in that, The credibility gate function Defined as: ； in, The gating attenuation coefficient is... As a credibility threshold; The sound velocity of the medium Obtain it using the following formula: ； in, This is the equivalent ultrasonic path length. For the upstream propagation time after classification compensation, The downstream propagation time after classification compensation; The estimated water supply temperature drift The recursive formula is: ； ； The estimated return water temperature drift The recursive formula is: ； The temperature calibration objective function Construct it in the following way: ； in, This is the current sampling sequence number. This is the temperature drift inertia factor corresponding to the estimated water supply temperature drift. Update the gain for the temperature drift estimate. This is the temperature drift inertia factor corresponding to the estimated return water temperature drift. Update the gain for the temperature drift corresponding to the return water temperature drift estimate. and These are the original measured values ​​of the supply water temperature and the return water temperature, respectively. , , , The weighting coefficients for the temperature calibration objective function. For reference temperature difference.

9. The adaptive environmental compensation ultrasonic heat meter metering method according to claim 1, characterized in that, Also includes: The parameters involved in the classification compensation process are adaptively updated based on long-term operating data and calibration residuals. The updated parameters include the distortion determination threshold. The basic weight parameters and attenuation coefficients for each distortion type, and the type credibility factor. Credibility threshold Temperature drift inertia factor and temperature drift update gain.

10. An adaptive environmental compensation ultrasonic heat meter metering device, used to perform the adaptive environmental compensation ultrasonic heat meter metering method according to any one of claims 1 to 9, characterized in that, include: The data acquisition module is configured to acquire ultrasonic propagation link parameters, temperature parameters, and auxiliary parameters according to a set sampling period, and to perform preprocessing operations on the acquired data to output a stable input sequence; The flow field distortion intensity calculation module is configured to calculate the flow field distortion intensity index based on the propagation time difference, echo amplitude asymmetry and propagation time normalized fluctuation intensity in the stable input sequence, and compare the flow field distortion intensity index with the distortion judgment threshold to output the distortion judgment result. The flow field type classification module is configured to construct a classification feature vector and output a distortion type label based on propagation time difference features, echo amplitude features, fluctuation features and time derivative features when the flow field distortion intensity index exceeds the distortion judgment threshold. The classification compensation module is configured to call the corresponding propagation time correction model based on the distortion type label and output the propagation time after classification compensation. The flow calculation module is configured to calculate the corrected medium velocity and volumetric flow rate based on the propagation time after the classification compensation, according to the ultrasonic time difference method. The credibility assessment module is configured to calculate the measurement credibility based on the flow field distortion intensity index, signal-to-noise ratio, and distortion type label, and output a credibility window judgment based on the comparison result between the measurement credibility and the credibility threshold. The temperature self-calibration module is configured to estimate the medium sound velocity based on the propagation time after classification compensation within a high confidence window and invert the local fluid temperature inversion value in the measurement pipe section, and complete the online temperature self-calibration by combining the heat meter installation location and the original measured values ​​of supply and return water temperatures. And a heat calculation module, configured to calculate and accumulate the heat increment based on the volumetric flow rate and the calibrated supply and return water temperatures.

Images