Gas density measurement method based on gas densitometer

Abstract

The present invention relates to the field of intelligent control technology for gas sensing and measurement, and specifically discloses a gas density measurement method based on a gas densitometer. During the measurement process, the gas sound velocity data and molecular adsorption dynamic response time data are collected in real time, and the gas compressibility fluctuation characteristic value and the response time adsorption hysteresis characteristic value are extracted respectively, and a comprehensive stability characteristic vector is constructed as a key indicator for evaluating the measurement state. By establishing a gas measurement reliability evaluation model based on a random forest algorithm, a quantitative score of the credibility of each measurement is achieved, and parameters such as the sampling frequency and measurement period are automatically adjusted according to the scoring results to form a closed-loop feedback mechanism. The present invention improves the accuracy and stability of gas density measurement, enhances the adaptability and robustness of the system under complex working conditions, and has the advantages of high intelligence, fast response speed, and wide application range. It is suitable for multi-component gas online monitoring and high-precision density detection scenarios.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of gas sensing and measurement, and in particular to a gas density measurement method based on a gas densitometer. Background Art

[0002] Gas density measurement has important application value in industrial process control, environmental monitoring, energy metering and other fields. Traditional gas density measurement methods mainly include buoyancy method, vibration densitometer, sound velocity method, etc. These methods have their own advantages and disadvantages under different working conditions. Among them, gas density meters based on the sound velocity principle are widely used due to their advantages such as non-contact and fast response. Their basic principle is to infer gas density by measuring the propagation speed of sound waves in the gas. However, in practical applications, gas density measurement is easily affected by environmental factors (such as temperature and pressure changes), gas composition fluctuations, and sensor surface adsorption effects, resulting in deviations in measurement results. Existing technologies generally use single parameter calibration or static compensation methods to improve measurement accuracy. However, in dynamically changing gas environments, these methods are difficult to effectively deal with complex real-time interference, resulting in reduced measurement reliability.

[0003] The existing technology has the following deficiencies:

[0004] In existing gas density measurement technologies, a problem that seriously affects measurement accuracy is the coupling interference between the dynamic adsorption hysteresis effect of gas molecules on the sensor surface and the sound velocity fluctuations. Specifically, when gas molecules adsorb or desorb on the sensor surface, their dynamic response time will produce nonlinear delays due to changes in gas composition or environmental conditions. This process will further affect the stability of sound wave propagation, resulting in complex deviations in the sound velocity measurement value that are difficult to separate by conventional methods. Traditional methods usually treat sound velocity data and adsorption effects as independent factors, ignoring the synergistic effect of the two in the time domain and frequency domain, making it impossible to accurately identify hidden instabilities in the measurement process. In addition, existing technologies lack a means of jointly quantitatively analyzing gas compressibility fluctuations and adsorption hysteresis characteristics, resulting in the system being unable to dynamically evaluate measurement credibility and adaptively adjust parameters. The existence of this problem makes it difficult to ensure the long-term stability and reliability of gas density measurements under complex working conditions. Summary of the Invention

[0005] The object of the present invention is to provide a gas density measurement method based on a gas densitometer to solve the above-mentioned problems.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] The gas density measurement method based on the gas densitometer includes the following steps:

[0008] S1: During the gas density measurement process, real-time acquisition of gas sound velocity data and molecular adsorption dynamic response time data;

[0009] S2: Process the sound velocity data of the gas and calculate the gas compressibility fluctuation characteristic value based on the deviation degree of the sound velocity measurement value to evaluate the deviation degree of the gas during the measurement process;

[0010] S3: Processing the gas molecular adsorption dynamic response time data, and calculating the response time adsorption hysteresis characteristic value according to the delay degree of the molecular adsorption dynamic response time, which is used to evaluate the interaction strength between the gas molecules and the density meter;

[0011] S4: The gas compressibility fluctuation characteristic value and the response time adsorption hysteresis characteristic value are constructed into a comprehensive stability characteristic vector, which is input into the gas measurement reliability evaluation model for analysis;

[0012] S5: Based on the analysis results, the reliability level of the current gas density measurement is output, and the measurement parameters are automatically adjusted according to the reliability level.

[0013] As a further solution of the present invention: the deviation degree of the evaluation gas during the measurement process specifically includes:

[0014] During the gas density measurement process, the sound velocity data of the gas is obtained and processed. According to the degree of deviation of the sound velocity measurement value, the gas compressibility fluctuation characteristic value is calculated to determine whether the gas compressibility fluctuation characteristic value is greater than or equal to a preset threshold. If so, the deviation of the gas during the measurement process is within the abnormal range. If not, the deviation of the gas during the measurement process is within the normal range.

[0015] As a further solution of the present invention: the process of obtaining the gas compressibility fluctuation characteristic value is:

[0016] During the gas density measurement process, the sound velocity data of the gas is obtained, the sound velocity data is integrated into a gas sound velocity data sequence, the average value of all sound velocity measurements is calculated, the sound velocity measurement value of each measurement point is subtracted from the average value, the difference calculation result is ratio-calculated with the average value of all sound velocity measurements, and the relative deviation value of each measurement point is obtained;

[0017] All relative deviation values ​​are used to form a set of time series data, and the time series data of the relative deviation values ​​are subjected to fast Fourier transform to obtain the complex value form of the frequency domain signal;

[0018] The energy spectral density corresponding to each frequency component in the frequency domain signal is calculated, and the ratio of the mean to the variance of the energy spectral density of all frequency components is calculated to obtain the characteristic value of the gas compressibility fluctuation.

[0019] As a further solution of the present invention: the evaluation of the interaction strength between the gas molecules and the density meter specifically includes:

[0020] During the gas density measurement process, the gas molecular adsorption dynamic response time data is obtained and processed. According to the delay degree of the molecular adsorption dynamic response time, the response time adsorption hysteresis characteristic value is calculated to determine whether the response time adsorption hysteresis characteristic value is greater than or equal to a preset threshold. If so, the interaction between the gas molecules and the density meter is abnormal. If not, the interaction between the gas molecules and the density meter is normal.

[0021] As a further solution of the present invention: the process of obtaining the response time adsorption hysteresis characteristic value is:

[0022] During the gas density measurement process, the dynamic response time data of gas molecular adsorption is obtained;

[0023] The gas molecular adsorption dynamic response time data is normalized to obtain a normalized matrix;

[0024] Based on the standardized data matrix, the covariance matrix is ​​constructed. Perform eigenvalue decomposition to obtain the eigenvalue set and the corresponding unit orthogonal eigenvector set;

[0025] Before selection The eigenvectors corresponding to the largest eigenvalues ​​form the projection moments, and the standardized data matrix is ​​projected into the dimensionality reduction space;

[0026] The standard deviation of the projection values ​​on the first principal component of the principal component score matrix was calculated to obtain the response time adsorption hysteresis eigenvalue.

[0027] As a further solution of the present invention, the gas compressibility fluctuation characteristic value and the response time adsorption hysteresis characteristic value are constructed into a comprehensive stability characteristic vector, and input into the gas measurement reliability evaluation model for analysis, specifically including:

[0028] During the gas density measurement process, the gas compressibility fluctuation characteristic value and the response time adsorption hysteresis characteristic value are obtained, and the gas compressibility fluctuation characteristic value and the response time adsorption hysteresis characteristic value are constructed into a comprehensive stability characteristic vector. The vector serves as the input of the gas measurement reliability assessment model to minimize the error between the predicted credibility score and the actual credibility score. The gas measurement reliability assessment model is trained as the training target. According to the trained gas measurement reliability assessment model, the credibility score of the volume density measurement process is output. The gas measurement reliability assessment model is a random forest model.

[0029] As a further solution of the present invention: the training process of the gas measurement reliability assessment model is:

[0030] During the model training process, the collected multiple measurement samples and their corresponding feature vectors and actual credibility scores are divided into training data sets and test data sets; the training data sets are used to tune the hyperparameters of the number of decision trees, split depth, and minimum number of node samples in the random forest model. The training goal is to minimize the error between the predicted credibility score output by the model and the actual credibility score; the square mean of the difference between the predicted credibility score and the actual credibility score is used as the optimization objective function, and the prediction accuracy and generalization ability of the model are improved through continuous iteration.

[0031] As a further solution of the present invention, the outputting of the credibility level of the current gas density measurement specifically includes:

[0032] During the gas density measurement process, it is determined whether the credibility score of the current gas density measurement is greater than or equal to a preset threshold. If so, the current gas density measurement is credible; otherwise, the current gas density measurement is unreliable.

[0033] As a further solution of the present invention, the automatic adjustment of measurement parameters according to the credibility level specifically includes:

[0034] If the current gas density measurement is unreliable, the current measurement state is determined to be unstable, and the parameter adaptive adjustment mechanism is triggered. When the gas compressibility fluctuation characteristic value is greater than or equal to the preset threshold, the system automatically increases the sampling frequency; when the response time adsorption hysteresis characteristic value is greater than or equal to the preset threshold, the system extends the measurement period and enhances data collection of the adsorption response process.

[0035] Beneficial effects of the present invention:

[0036] (1) The present invention introduces a multi-physical field collaborative sensing mechanism to construct a gas density measurement method based on the combination of acoustic characteristics and molecular adsorption dynamics, which significantly improves the accuracy, stability and environmental adaptability of the measurement system. Specifically, during the measurement process, the system synchronously collects the sound velocity data and molecular adsorption dynamic response time data of the gas, and extracts the characteristic values ​​that can reflect the gas compressibility fluctuation characteristics and the hysteresis characteristic values ​​that reflect the interaction strength between the gas molecules and the sensor interface. Among them, the gas compressibility fluctuation characteristic value is obtained by performing frequency domain energy analysis on the sound velocity time series, which can effectively characterize the nonlinear compression behavior and transient fluctuation characteristics of the gas under different pressure or composition conditions; and the response time adsorption hysteresis characteristic value is based on the principal component analysis technology to quantitatively characterize the time delay of the adsorption process, thereby revealing the dynamic hysteresis effect of gas molecules in the adsorption / desorption process on the sensor surface and its influence on the measurement stability.

[0037] These two eigenvalues, derived from both macroscopic acoustic response and microscopic molecular behavior, form a comprehensive stability assessment system with clear physical meaning and rigorous mathematical expression, enabling dynamic quality monitoring of the entire gas density measurement process. This method can identify abnormal fluctuations caused by changes in gas composition, environmental disturbances, or sensor adsorption effects in real time, effectively improving the accuracy and repeatability of gas density measurements and avoiding data errors caused by measurement instability.

[0038] (2) Based on the extracted gas compressibility fluctuation characteristic values ​​and response time adsorption hysteresis characteristic values, the present invention constructs a comprehensive stability characteristic vector with physical interpretability and statistical robustness, and uses this as the input feature to train and optimize the gas measurement reliability assessment model using the random forest algorithm. The model establishes a nonlinear mapping relationship between the measurement stability characteristics and the credibility score through multi-dimensional feature learning, and can output a highly interpretable quantitative credibility score for each gas density measurement process. In practical applications, the system dynamically adjusts key measurement parameters based on the scoring results. For example, when high fluctuations or strong adsorption hysteresis are detected, the sampling frequency is automatically increased, the measurement cycle is extended, or the data filtering strength is enhanced, thereby achieving real-time optimization of the measurement strategy. This closed-loop feedback mechanism not only gives the measurement system the ability of intelligent diagnosis and adaptive adjustment, but also significantly enhances its operational robustness and environmental adaptability under complex and variable working conditions. It breaks through the technical limitations of traditional gas density measurement methods that rely on static calibration and lack state perception capabilities, and provides a new technical path and system solution for high-precision, continuous and stable online monitoring of gas density. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The present invention will be further described below with reference to the accompanying drawings.

[0040] Figure 1 It is a flow chart of the gas density measurement method based on the gas densitometer of the present invention. DETAILED DESCRIPTION

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0042] See also Figure 1 As shown, the present invention is a gas density measurement method based on a gas densitometer, comprising the following steps:

[0043] S1: During the gas density measurement process, real-time acquisition of gas sound velocity data and molecular adsorption dynamic response time data;

[0044] S2: Process the sound velocity data of the gas and calculate the gas compressibility fluctuation characteristic value based on the deviation degree of the sound velocity measurement value to evaluate the deviation degree of the gas during the measurement process;

[0045] S3: Processing the gas molecular adsorption dynamic response time data, and calculating the response time adsorption hysteresis characteristic value according to the delay degree of the molecular adsorption dynamic response time, which is used to evaluate the interaction strength between the gas molecules and the density meter;

[0046] S4: The gas compressibility fluctuation characteristic value and the response time adsorption hysteresis characteristic value are constructed into a comprehensive stability characteristic vector, which is input into the gas measurement reliability evaluation model for analysis;

[0047] S5: Based on the analysis results, the reliability level of the current gas density measurement is output, and the measurement parameters are automatically adjusted according to the reliability level.

[0048] In S1, during the gas density measurement process, the gas sound velocity data and molecular adsorption dynamic response time data are collected in real time, including:

[0049] During gas density measurement, the specific process for collecting real-time data on gas sound velocity and molecular adsorption dynamic response time is as follows: First, a specially designed sound velocity sensor module and adsorption response detection module are configured within the gas density measurement system. The sound velocity sensor module, consisting of a pair of ultrasonic transmitting and receiving probes, is installed at either end of a sealed measurement chamber. It continuously transmits and receives high-frequency sound wave signals passing through the gas, calculating the time difference between each transmission and reception to determine the gas sound velocity. Simultaneously, the adsorption response detection module utilizes a highly sensitive piezoelectric crystal to monitor the frequency changes caused by the adsorption and desorption of gas molecules on the sensor surface, recording the time required for each adsorption event to reach a stable state as the molecular adsorption dynamic response time.

[0050] The acquired sound velocity data and molecular adsorption dynamic response time data are transmitted in real time to the data processing unit, and each data set is accurately timestamped for subsequent analysis. The sound velocity data is obtained by calculating the difference in acoustic wave propagation time at each sampling point, while the molecular adsorption dynamic response time is determined by monitoring the stabilization time of the frequency change curve during the adsorption event. This process ensures that the gas compressibility fluctuation characteristic values ​​and the response time adsorption hysteresis characteristic values ​​can be accurately extracted, providing solid data support for the subsequent construction of a comprehensive stability characteristic vector and the evaluation of the reliability of gas density measurements.

[0051] In S2, the sound velocity data of the gas is processed. According to the deviation degree of the sound velocity measurement value, the gas compressibility fluctuation characteristic value is calculated to evaluate the deviation degree of the gas during the measurement process. Specifically, it includes:

[0052] During the gas density measurement process, the sound velocity data of the gas is obtained and processed. According to the degree of deviation of the sound velocity measurement value, the gas compressibility fluctuation characteristic value is calculated to determine whether the gas compressibility fluctuation characteristic value is greater than or equal to a preset threshold. If so, the deviation of the gas during the measurement process is within the abnormal range. If not, the deviation of the gas during the measurement process is within the normal range.

[0053] The process of obtaining the gas compressibility fluctuation characteristic value is as follows:

[0054] During the gas density measurement process, the sound velocity data of the gas is obtained, the sound velocity data is integrated into a gas sound velocity data sequence, the average value of all sound velocity measurements is calculated, the sound velocity measurement value of each measurement point is subtracted from the average value, the difference calculation result is ratio-calculated with the average value of all sound velocity measurements, and the relative deviation value of each measurement point is obtained;

[0055] All relative deviation values ​​are used to form a set of time series data, and the time series data of the relative deviation values ​​are subjected to fast Fourier transform to obtain the complex value form of the frequency domain signal;

[0056] Calculate the energy spectral density corresponding to each frequency component in the frequency domain signal. The calculation expression is: ;in, Represents frequency components The energy spectral density, Indicates the frequency components, Represents frequency components The complex value form of is used to calculate the ratio of the mean to the variance of the energy spectral density of all frequency components to obtain the characteristic value of gas compressibility fluctuation.

[0057] It should be noted that: by performing time domain normalization processing on the gas sound velocity measurement data, the relative deviation of each measurement point from the average value is extracted, and the obtained time series data is converted into a frequency domain signal, the spectrum information in complex value form is obtained by fast Fourier transform, and the energy spectrum density of each frequency component is further calculated. Finally, the ratio of the mean to the variance of the energy spectrum density is used as the characteristic value of the gas compressibility fluctuation. In this implementation scheme, the statistical characteristics of frequency domain energy distribution are applied to the analysis of gas compressibility fluctuations for the first time, which can not only effectively reflect the non-stationarity and complexity of the sound velocity fluctuation of the gas during the measurement process, but also realize the quantitative evaluation of abnormal gas compressibility fluctuations through mathematical modeling, thereby improving the sensitivity and accuracy of the gas density measurement system in judging the stability of the measurement state, thereby enhancing the intelligence level and adaptability of the overall measurement method.

[0058] In S3, the gas molecular adsorption dynamic response time data is processed. According to the delay degree of the molecular adsorption dynamic response time, the response time adsorption hysteresis characteristic value is calculated to evaluate the interaction strength between the gas molecules and the density meter. Specifically, it includes:

[0059] During the gas density measurement process, the gas molecular adsorption dynamic response time data is obtained and processed. According to the delay degree of the molecular adsorption dynamic response time, the response time adsorption hysteresis characteristic value is calculated to determine whether the response time adsorption hysteresis characteristic value is greater than or equal to a preset threshold. If so, the interaction between the gas molecules and the density meter is abnormal. If not, the interaction between the gas molecules and the density meter is normal.

[0060] The process of obtaining the response time adsorption hysteresis characteristic value is as follows:

[0061] During the gas density measurement process, the dynamic response time data of gas molecular adsorption is obtained;

[0062] The molecular adsorption dynamic response time data of the gas is normalized to obtain a normalized matrix, where the calculation expression for the normalization is: ;in, represents the normalized adsorption response time data, Indicates the The sample in The original adsorption response time at each time point, Indicates the number of sample data collected, Indicates the number of acquisition time points, Indicates the The average adsorption response time of all samples at each time point is Indicates the The standard deviation of all samples at each time point;

[0063] Based on the standardized data matrix, the covariance matrix is ​​constructed and the calculation expression is: ;in, represents the covariance matrix, represents the normalized matrix, represents the transpose operation, Indicates the total number of sample data collected, the covariance matrix Perform eigenvalue decomposition to obtain the eigenvalue set and the corresponding unit orthogonal eigenvector set, and satisfy: ,in, represents the unit orthogonal eigenvector, represents the eigenvalue;

[0064] Before selection The eigenvectors corresponding to the largest eigenvalues ​​form the projection moment, and the standardized data matrix is ​​projected into the dimensionality reduction space to obtain the principal component score matrix. The calculation expression is: ;in, represents the principal component score matrix, Indicates that the eigenvectors form the projection matrix;

[0065] The standard deviation of the projection values ​​on the first principal component of the principal component score matrix was calculated to obtain the response time adsorption hysteresis eigenvalue.

[0066] It should be noted that the technical solution for calculating the response time adsorption hysteresis characteristic value through the dynamic response time data of molecular adsorption has the following technical effects and innovations: This method standardizes the dynamic response time data of gas molecules in the adsorption process on the sensor surface, and extracts the projection standard deviation in the direction of the first principal component based on principal component analysis as the response time adsorption hysteresis characteristic value, thereby realizing a quantitative evaluation of the interaction strength between the gas molecules and the density meter; Compared with the traditional judgment method that only relies on a single response time mean or maximum value, this solution fully explores the potential change trends in multidimensional time series data and improves the sensitivity to the hysteresis and nonlinear behavior of the adsorption process under different measurement conditions; This method not only effectively removes the influence of data dimension and fluctuation amplitude under different measurement conditions, but also enhances the interpretability and robustness of data features through dimensionality reduction, providing the gas density measurement system with high-precision identification capabilities for the stability state of gas adsorption behavior, further improving the reliability of measurement results and the level of equipment self-diagnosis.

[0067] In S4, the gas compressibility fluctuation characteristic value and the response time adsorption hysteresis characteristic value are constructed into a comprehensive stability characteristic vector and input into the gas measurement reliability evaluation model for analysis, including:

[0068] During the gas density measurement process, the gas compressibility fluctuation characteristic value and the response time adsorption hysteresis characteristic value are obtained, and the gas compressibility fluctuation characteristic value and the response time adsorption hysteresis characteristic value are constructed into a comprehensive stability characteristic vector. The vector serves as the input of the gas measurement reliability assessment model to minimize the error between the predicted credibility score and the actual credibility score. The gas measurement reliability assessment model is trained as the training target. According to the trained gas measurement reliability assessment model, the credibility score of the volume density measurement process is output. The gas measurement reliability assessment model is a random forest model.

[0069] During model training, the collected measurement samples, their corresponding feature vectors, and actual credibility scores are divided into training and test datasets. The training dataset is used to tune the hyperparameters of the random forest model, including the number of decision trees, split depth, and minimum number of node samples. The training goal is to minimize the error between the predicted credibility score output by the model and the actual credibility score. Specifically, the mean square of the difference between the predicted score and the actual score is used as the optimization objective function, and the model's prediction accuracy and generalization ability are improved through continuous iteration.

[0070] Finally, when the model training reaches the set convergence condition or the maximum number of training rounds, the model version with the best performance is saved and deployed to the gas density measurement system; in the actual application stage, the comprehensive stability feature vector extracted in the new measurement process is input into the trained gas measurement reliability assessment model. The model automatically outputs the credibility score of this measurement as a quantitative basis for measuring the reliability of the current gas density measurement results, thereby realizing quality control and dynamic evaluation of the entire gas density measurement process.

[0071] During the gas density measurement process, it is determined whether the credibility score of the current gas density measurement is greater than or equal to a preset threshold. If so, the current gas density measurement is credible; otherwise, the current gas density measurement is unreliable.

[0072] In S5, based on the analysis results, the reliability level of the current gas density measurement is output, and the measurement parameters are automatically adjusted according to the reliability level, including:

[0073] If the current gas density measurement is unreliable, the current measurement state is determined to be unstable, and the parameter adaptive adjustment mechanism is triggered. When the gas compressibility fluctuation characteristic value is greater than or equal to the preset threshold, the system automatically increases the sampling frequency and data processing accuracy to capture the rapidly changing fluctuation characteristics; when the response time adsorption hysteresis characteristic value is greater than or equal to the preset threshold, the system extends the measurement cycle and enhances data collection of the adsorption response process to improve the adaptability and stability of the measurement system.

[0074] The working principle of the present invention is as follows: by collecting the sound velocity data of the gas during the measurement process and the molecular adsorption dynamic response time data, the gas compressibility fluctuation characteristic value and the response time adsorption hysteresis characteristic value are calculated respectively, thereby constructing a comprehensive characteristic vector reflecting the measurement stability. The vector is then input into the gas measurement reliability assessment model trained based on the random forest algorithm to achieve a quantitative assessment of the current measurement credibility. The method first uses an ultrasonic sensor module to accurately obtain the gas sound velocity data, combines it with the fast Fourier transform to extract the frequency domain energy distribution characteristics, and calculates the gas compressibility fluctuation characteristic value by the ratio of the energy spectrum density mean to the variance. This can effectively identify the non-stationary fluctuation behavior of the gas during the measurement process. At the same time, a high-sensitivity piezoelectric crystal detection module is used to collect the time series data of the gas molecule adsorption process, and the principal component analysis technology is introduced to extract the first principal component projection standard deviation as the response time adsorption hysteresis characteristic value to quantitatively evaluate the interaction strength between the gas molecules and the sensor and the adsorption stability. Ultimately, the two eigenvalues ​​are combined into a comprehensive stability eigenvector to construct an intelligent evaluation model. Model training is completed by minimizing the error between the predicted and actual scores. The reliability of the measurement state is automatically determined based on the output credibility score. When abnormal fluctuations or adsorption instability are detected, the system adaptively adjusts key parameters such as the sampling frequency and measurement cycle, thereby enabling dynamic quality monitoring and intelligent control of the entire gas density measurement process. This solution significantly improves the robustness, accuracy, and self-regulation capabilities of the gas density measurement system, making it suitable for real-time density detection of multi-component gases under complex working conditions.

[0075] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0076] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0077] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0078] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0079] The above is a detailed description of an embodiment of the present invention. However, the content described is only a preferred embodiment of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A gas density measurement method based on a gas densitometer, characterized in that: The following steps are involved: S1: During the gas density measurement process, real-time acquisition of gas sound velocity data and molecular adsorption dynamic response time data; S2: Process the sound velocity data of the gas and calculate the gas compressibility fluctuation characteristic value based on the deviation degree of the sound velocity measurement value to evaluate the deviation degree of the gas during the measurement process; The process of obtaining the gas compressibility fluctuation characteristic value is as follows: During the gas density measurement process, the sound velocity data of the gas is obtained, the sound velocity data is integrated into a gas sound velocity data sequence, the average value of all sound velocity measurements is calculated, the sound velocity measurement value of each measurement point is subtracted from the average value, the difference calculation result is ratio-calculated with the average value of all sound velocity measurements, and the relative deviation value of each measurement point is obtained; All relative deviation values ​​are used to form a set of time series data, and the time series data of the relative deviation values ​​are subjected to fast Fourier transform to obtain the complex value form of the frequency domain signal; Calculate the energy spectrum density corresponding to each frequency component in the frequency domain signal, calculate the ratio of the mean to the variance of the energy spectrum density of all frequency components, and obtain the gas compressibility fluctuation characteristic value; S3: Processing the gas molecular adsorption dynamic response time data, and calculating the response time adsorption hysteresis characteristic value according to the delay degree of the molecular adsorption dynamic response time, which is used to evaluate the interaction strength between the gas molecules and the density meter; The process of obtaining the response time adsorption hysteresis characteristic value is as follows: During the gas density measurement process, the dynamic response time data of gas molecular adsorption is obtained; The gas molecular adsorption dynamic response time data is normalized to obtain a normalized matrix; Based on the standardized data matrix, the covariance matrix is ​​constructed. Perform eigenvalue decomposition to obtain the eigenvalue set and the corresponding unit orthogonal eigenvector set; Before selection The eigenvectors corresponding to the largest eigenvalues ​​form the projection moments, and the standardized data matrix is ​​projected into the dimensionality reduction space; Calculate the standard deviation of the projection value on the first principal component of the principal component score matrix to obtain the response time adsorption hysteresis eigenvalue; S4: The gas compressibility fluctuation characteristic value and the response time adsorption hysteresis characteristic value are constructed into a comprehensive stability characteristic vector, which is input into the gas measurement reliability assessment model for analysis, including: During the gas density measurement process, a gas compressibility fluctuation characteristic value and a response time adsorption hysteresis characteristic value are obtained, and the gas compressibility fluctuation characteristic value and the response time adsorption hysteresis characteristic value are constructed into a comprehensive stability characteristic vector. The vector serves as an input to a gas measurement reliability assessment model to minimize the error between the predicted credibility score and the actual credibility score. The gas measurement reliability assessment model is trained as a training objective. Based on the trained gas measurement reliability assessment model, a credibility score during the volume density measurement process is output. The gas measurement reliability assessment model is a random forest model. S5: Based on the analysis results, the reliability level of the current gas density measurement is output, and the measurement parameters are automatically adjusted according to the reliability level.

2. The gas density measurement method based on a gas densitometer according to claim 1, characterized in that: The deviation degree of the evaluation gas during the measurement process specifically includes: During the gas density measurement process, the sound velocity data of the gas is obtained and processed. According to the degree of deviation of the sound velocity measurement value, the gas compressibility fluctuation characteristic value is calculated to determine whether the gas compressibility fluctuation characteristic value is greater than or equal to a preset threshold. If so, the deviation of the gas during the measurement process is within the abnormal range. If not, the deviation of the gas during the measurement process is within the normal range.

3. The gas density measurement method based on a gas densitometer according to claim 1, characterized in that: The evaluation of the interaction strength between the gas molecules and the density meter specifically includes: During the gas density measurement process, the gas molecular adsorption dynamic response time data is obtained and processed. According to the delay degree of the molecular adsorption dynamic response time, the response time adsorption hysteresis characteristic value is calculated to determine whether the response time adsorption hysteresis characteristic value is greater than or equal to a preset threshold. If so, the interaction between the gas molecules and the density meter is abnormal. If not, the interaction between the gas molecules and the density meter is normal.

4. The gas density measurement method based on a gas densitometer according to claim 1, characterized in that: The training process of the gas measurement reliability assessment model is as follows: During the model training process, the collected multiple measurement samples and their corresponding feature vectors and actual credibility scores are divided into training data sets and test data sets; the training data sets are used to tune the hyperparameters of the number of decision trees, split depth, and minimum number of node samples in the random forest model. The training goal is to minimize the error between the predicted credibility score output by the model and the actual credibility score; the square mean of the difference between the predicted credibility score and the actual credibility score is used as the optimization objective function, and the prediction accuracy and generalization ability of the model are improved through continuous iteration.

5. The gas density measurement method based on a gas densitometer according to claim 1, characterized in that: The output of the reliability level of the current gas density measurement specifically includes: During the gas density measurement process, it is determined whether the credibility score of the current gas density measurement is greater than or equal to a preset threshold. If so, the current gas density measurement is credible; otherwise, the current gas density measurement is unreliable.

6. The gas density measurement method based on a gas densitometer according to claim 1, characterized in that: The automatic adjustment of measurement parameters according to the credibility level specifically includes: If the current gas density measurement is unreliable, the current measurement state is determined to be unstable, and the parameter adaptive adjustment mechanism is triggered. When the gas compressibility fluctuation characteristic value is greater than or equal to the preset threshold, the system automatically increases the sampling frequency; when the response time adsorption hysteresis characteristic value is greater than or equal to the preset threshold, the system extends the measurement period and enhances data collection of the adsorption response process.

Images