An intelligent method for monitoring gas flow

Abstract

The present application provides an intelligent gas flow monitoring method, including: detecting the current temperature of the gas, building a gas temperature prediction model, and obtaining a specific value of the temperature change; using a linear regression algorithm to establish a gas thermal expansion or contraction prediction model, and calculating the degree of gas thermal expansion or contraction based on gas temperature change data; predicting gas volume changes based on real-time temperature and pressure monitoring; building a flow rate and volume relationship model based on gas flow rate and volume change data, and predicting gas flow rate changes based on gas volume changes; calculating the difference between the predicted flow rate and the predicted flow rate through real-time flow measurement data; automatically adjusting and calibrating the flow measurement equipment based on abnormality detection and stability evaluation of the flow data; automatically adjusting the calibration period of the flow measurement equipment based on historical and real-time monitoring data; and using the predicted data and real-time data to calibrate the flow measurement of the equipment.

Description

Technical Field

[0001] The invention relates to the field of information technology, and in particular to an intelligent gas flow monitoring method. Background Art

[0002] In modern industry and scientific research, accurate measurement of gas flow is an important part of many critical processes. It is widely used in many fields such as energy management, chemical industry, environmental monitoring and laboratory research. However, accurate measurement of gas flow faces several technical challenges. The physical properties of gas cause it to expand or contract when the temperature changes, which affects its volume and flow rate. This phenomenon complicates accurate flow measurement in environments with large temperature fluctuations. In addition to temperature, other environmental parameters such as pressure and humidity can also affect the flow characteristics of gas. Maintaining measurement accuracy and consistency under changing environmental conditions is a challenge. Due to changes in environmental conditions and long-term use of equipment, flow measurement equipment may experience performance deviations, so regular calibration is required. However, determining when to calibrate and how to accurately perform calibration operations is a technical challenge. With the complexity of industrial processes and the increasing requirements for data accuracy, relying solely on traditional measurement methods can no longer meet the needs. How to effectively process and analyze measurement data and how to accurately predict flow changes based on environmental changes have become key issues. Under the rapid development of industrial automation and intelligent trends, traditional manual monitoring and calibration methods have gradually become inefficient and backward. How to realize the automation and intelligence of the flow measurement process and improve efficiency and accuracy is a current technical problem. Therefore, the main problem facing the field of gas flow measurement is how to accurately measure and predict gas flow under changing environmental conditions, while improving the automation and intelligence level of measurement equipment. Solving these problems is crucial to ensuring the efficiency, safety and environmental compliance of industrial processes. Summary of the invention

[0003] The present invention provides an intelligent gas flow monitoring method, which mainly includes:

[0004] Detect the current temperature of the gas, build a gas temperature prediction model, and obtain the specific value of the temperature change; use the linear regression algorithm to establish a gas thermal expansion or contraction prediction model, and calculate the degree of gas thermal expansion or contraction based on the gas temperature change data; predict the gas volume change based on real-time temperature and pressure monitoring; build a flow rate and volume relationship model based on the gas flow rate and volume change data, and predict the gas flow rate change based on the gas volume change; calculate the difference between the predicted flow rate and the real-time flow measurement data; automatically adjust and calibrate the flow measurement equipment based on the anomaly detection and stability evaluation of the flow data; automatically adjust the calibration period of the flow measurement equipment based on the historical and real-time monitoring data; use the predicted data and real-time data to calibrate the flow measurement of the equipment.

[0005] Furthermore, the current temperature of the detected gas is used to construct a gas temperature prediction model to obtain a specific value of the temperature change, including:

[0006] The real-time temperature data of the gas is obtained through the temperature sensor, and the data acquisition system is used to continuously record the gas temperature data provided by the sensor; the timestamp of each temperature data point is obtained through the sensor data, and the obtained temperature data is sorted into time series; based on the historical gas temperature data, the linear regression algorithm is used for model training, and a gas temperature prediction model is constructed to predict the trend of the gas temperature data, determine the maximum, minimum and average values ​​in the temperature data, and obtain the specific value of the temperature change; determine whether the temperature change exceeds the preset normal operating range; if the temperature change exceeds the normal range, an abnormal report is obtained, including the temperature change value and timestamp.

[0007] Furthermore, the use of a linear regression algorithm to establish a gas thermal expansion or contraction prediction model and to calculate the degree of gas thermal expansion or contraction based on gas temperature change data includes:

[0008] The real-time temperature data is obtained through the gas temperature sensor; the real-time temperature data is preprocessed, including removing noise and outliers; a gas thermal expansion or contraction prediction model is established using the processed temperature data through a linear regression algorithm; a preliminary prediction result of the gas thermal expansion or contraction is obtained based on the current temperature data; the prediction error of each data point is obtained by calculating the difference between the predicted value and the actual observed value, and the overall error of the entire prediction result is calculated by using the method of calculating the mean square error; the accuracy of the model is determined based on the distribution of the error; if the error value is less than a preset threshold and is evenly distributed, the model is considered to be accurate, and if the error value is greater than a preset threshold or is unevenly distributed, it indicates that the model needs to be further optimized; if the model needs to be adjusted, the parameters are adjusted to improve the gas thermal expansion or contraction prediction model; based on the improved gas thermal expansion or contraction prediction model, an optimized gas thermal expansion or contraction prediction result is obtained.

[0009] Further, predicting the gas volume change based on real-time temperature and pressure monitoring includes:

[0010] Use the gas temperature prediction model to obtain the temperature prediction results of the gas at a certain moment or time period in the future; use the ideal gas law PV=nRT, substitute the predicted temperature value into the ideal gas law, and calculate the expected gas volume change, where P is pressure, V is volume, n is the amount of gas substance, R is the ideal gas constant, and T is temperature; obtain the current pressure data of the gas measured in real time through the pressure sensor; combine the real-time measured pressure data with the predicted temperature data to recalculate the gas volume change; determine the initial prediction value of the gas volume change after combining the temperature and pressure data; obtain historical volume change data of the same type of gas under similar conditions; compare the current prediction results with historical data; adjust and correct the current volume change prediction based on the comparison of historical data; continuously monitor the temperature and pressure of the gas, and update the data in real time, and further refine and update the volume change prediction based on the latest monitoring data to determine the final gas volume change prediction.

[0011] Furthermore, the method of establishing a velocity and volume relationship model based on the gas velocity and volume change data and predicting the gas velocity change based on the gas volume change includes:

[0012] Obtain historical gas flow rate and volume change data, use the Navier-Stokes equation to process the initial volume change data, and establish a flow rate and volume relationship model; set the initial parameters of the relationship model based on historical data to predict the flow rate change of the gas; compare the flow rate prediction results obtained through the flow rate and volume relationship model with the real-time measured volume change data; if the deviation between the predicted result and the actual data is greater than the threshold value, dynamically adjust the parameters in the algorithm; continuously monitor the volume change and flow rate change of the gas, and if the difference between the actual measured volume change or flow rate and the predicted value is greater than the preset threshold, use the preset adjustment The flow rate prediction is corrected by an adjustment strategy; the stability of the flow rate change prediction is evaluated using an autoregressive model based on the gas flow rate data; if the flow rate change prediction is abnormal, a Kalman filter is used to perform real-time dynamic correction on the flow rate prediction; the flow rate prediction data corrected by the Kalman filter is obtained, and real-time monitoring is performed again to verify the effectiveness of the correction; if there is still a deviation in the corrected flow rate change data, the parameters are continuously adjusted using the Navier-Stokes equations until the flow rate change prediction data is stable; it also includes: calculating and predicting the flow rate of the multiphase fluid based on the mixed fluid characteristics and real-time data adjustment.

[0013] The method of adjusting, calculating and predicting the flow rate of the multiphase fluid according to the mixed fluid characteristics and real-time data specifically includes:

[0014] According to the influence of the combination of multiphase fluids on the flow rate, a multiphase fluid flow rate prediction model V = k × (w oil ×V oil +w water ×Vwater +w gas ×V gas ), where V represents the total flow rate, w oil is the volume ratio of oil in the mixture, w water is the volume ratio of water in the mixture, w gas is the volume ratio of the gas in the mixture, V oil is the base flow rate of the oil, V water is the basic velocity of water, V gas is the basic flow rate of the gas, and k is an adjustment factor. The basic flow rate V is measured in pure oil, pure water and pure gas environments respectively. oil , V water , V gas , record the flow rate data of each phase fluid at different pressures and temperatures. By mixing different proportions of oil, water, and gas, measure the total flow rate V of the mixed fluid. Use nonlinear regression analysis to determine the coefficient k in the multiphase fluid flow rate prediction model. In actual operation scenarios, use sensors to monitor the ratio w of oil, water, and gas in real time. oil , w water , w gas . According to the real-time monitored fluid ratio, the adjusted multiphase fluid flow rate prediction model is used to predict the flow rate. Based on the real-time and historical data analysis, the value of k is dynamically adjusted to cope with the changes in fluid ratio. Regularly compare the difference between the predicted flow rate and the actual measured flow rate to evaluate the accuracy and reliability of the multiphase fluid flow rate prediction model. Based on the performance evaluation results, further adjust the parameters in the formula to optimize the multiphase fluid flow rate prediction model. Implement a complete monitoring system to continuously obtain flow rate, ratio and environmental condition data of multiphase fluids. Establish a feedback mechanism to continuously optimize the prediction formula based on the comparison of monitoring data and prediction results.

[0015] Furthermore, the calculation of the difference between the real-time flow measurement data and the predicted flow rate includes:

[0016] The sensor is used to record the fluid flow in real time, and the filtering and noise elimination technology is used to process the acquired flow data; according to the gas volume data, the velocity and volume relationship model is used to obtain the predicted value of the gas flow rate; the difference between the real-time monitored flow rate and the flow rate predicted by the velocity and volume relationship model is calculated; it is detected whether the velocity difference exceeds the preset safety threshold, and if the difference exceeds the threshold, the abnormal detection mechanism is triggered; according to the abnormal detection result, based on the real-time flow measurement data and analysis algorithm, the parameters of the velocity and volume relationship model are automatically adjusted; the flow measurement data is updated in real time to keep the data real-time and up-to-date; the velocity difference is continuously monitored and analyzed, and if the difference remains within the preset range, the model parameters do not need to be adjusted; all flow and velocity data are automatically recorded and stored; it also includes: according to the real-time environmental variables and flow velocity data, the measurement and prediction process of the fluid flow rate is continuously optimized.

[0017] The process of continuously optimizing the measurement and prediction of fluid flow rate according to real-time environmental variables and flow rate data specifically includes:

[0018] According to the density, temperature and pressure of the fluid, a flow rate prediction model is established. The flow rate prediction model formula is F = a × D b ×T c ×P d , where F is the flow rate, D is the fluid density, T is the temperature, P is the pressure, and a, b, c, d are parameters determined based on experimental data, indicating the degree of influence of these variables on the flow rate. Obtain the data of D, T, P and the actual flow rate F of the fluid under different environmental conditions. Use regression analysis to determine the parameters a, b, c, d in the formula. Measure D, T, P in real time and apply the formula F = a × D b ×T c ×P d Make flow velocity prediction. Record the predicted flow velocity F pred Compared with the actual measured flow rate F actual According to F pred With F actual Dynamically adjust the values ​​of a, b, c, and d based on the differences between the two. Continuously monitor changes in environmental conditions and flow data. Regularly update the flow rate prediction model parameters with newly acquired data to ensure that the model remains accurate and effective in a changing environment. Regularly evaluate the performance of the prediction model under extreme environmental conditions, such as extreme temperatures, high or low pressure, abnormal humidity levels, chemicals, or pollutants. Adjust model parameters based on performance evaluation results to cope with extreme environmental conditions. Prepare comprehensive performance reports that summarize the performance of the model under different conditions and provide report-based decision support.

[0019] Furthermore, the automatic adjustment and calibration of the flow measurement device based on the abnormality detection and stability assessment of the flow data includes:

[0020] According to the difference between the flow measurement data and the predicted flow rate, the flow data is classified into normal flow data and abnormal flow data. If the difference between the flow measurement data and the predicted flow rate exceeds a predetermined threshold, it is judged as abnormal flow data. If the judgment result is abnormal flow data, it is judged whether the abnormality is caused by the error of the flow measurement device by comparing the data at different time points or comparing with the readings of other sensors. If the abnormality is judged to be caused by the error of the equipment, the status and performance parameters of the equipment are obtained to evaluate its current working condition. The historical parameter data obtained by the flow measurement device, including flow rate, temperature, and pressure, are used to train the multi-layer perceptron model. The flow measurement device is continuously used to The readings obtained at different flow rates, temperatures, and pressures are used to evaluate the stability and accuracy of the flow measurement device output using a pre-trained multi-layer perceptron model; if the multi-layer perceptron analysis shows that the device output is unstable or inaccurate, the device automatically enters calibration and automatically adjusts its internal parameters to correct the output error; after the calibration is completed, the device automatically returns to normal working mode, and the device continues to measure flow and output data; after calibration, the device diagnostic program is used to evaluate the device's output data again to determine whether the device has been successfully calibrated. If the calibration is successful, the calibration process is terminated; if the calibration is unsuccessful, the device is recalibrated; it also includes: implementing equipment performance monitoring and dynamic calibration adjustments based on environmental changes.

[0021] The implementation of equipment performance monitoring and dynamic calibration adjustment according to environmental changes specifically includes:

[0022] Through sensors, flow, temperature, and pressure data are obtained under different temperatures, different fluid pressures, and different fluid types. Based on the data obtained under each environmental condition, a decision tree algorithm is used to train the model to identify data patterns and anomalies unique to the environment and determine the impact of temperature and pressure changes on flow measurement. Using historical and real-time data, a deep learning model based on a multilayer perceptron is established to identify data features under different environmental conditions and adapt to various environmental conditions. Based on the output of the multilayer perceptron model, the calibration strategy is dynamically adjusted to adapt to the current environmental conditions. The calibration settings of the device are automatically adjusted based on the analysis results of the multilayer perceptron model. The calibrated device is tested under different environmental conditions, and the performance data of the device is recorded and compared with the output of the prediction model to verify the calibration effect. The performance of the device under various environmental conditions is continuously monitored, and real-time data is regularly fed back to the multilayer perceptron model for iterative optimization of the model. Based on the test data under various environments, a comprehensive device performance evaluation report is established, which includes the performance of the device under various conditions, the calibration effect, and any areas that need further improvement. Based on the performance evaluation report, regular equipment maintenance and software updates are performed according to the problems and deficiencies pointed out in the report.

[0023] Furthermore, the automatic adjustment of the calibration period of the flow measurement device according to the historical and real-time monitoring data includes:

[0024] Obtain historical equipment calibration records, flow measurement data, and environmental parameters, including temperature and pressure; use autoregressive models to analyze calibration frequency and performance fluctuations in historical data, and identify key factors that lead to changes in calibration requirements, including changes in environmental conditions or increases or decreases in equipment usage frequency; use multi-layer perceptrons to build an equipment calibration prediction model based on historical and real-time data to predict future calibration requirements; input historical calibration data and real-time monitoring parameters to train the equipment calibration prediction model; use the equipment calibration prediction model to determine the current equipment status and possible calibration requirements based on the equipment's flow data and environmental parameters; determine whether the equipment needs to be calibrated and determine the best time for calibration based on the output of the equipment calibration prediction model and real-time data analysis; when the equipment calibration prediction model predicts that the equipment needs to be calibrated, automatically trigger the calibration process and ensure that the calibration process meets the equipment's calibration requirements and operating specifications; after calibration is completed, record equipment performance data and evaluate the calibration effect by comparing the flow measurement accuracy before and after calibration; regularly adjust the parameters of the equipment calibration prediction model based on the calibration results and equipment operating data.

[0025] Furthermore, the use of the predicted data and the real-time data to calibrate the flow measurement of the device includes:

[0026] Obtain real-time data from flow measurement equipment, including flow velocity, fluid density, and temperature; if the predicted data is within the preset error range, use the recurrent neural network algorithm to train the model based on the real-time data, build a prediction model for equipment performance changes, detect changes in equipment performance or identify possible anomalies; obtain a preliminary data quality report based on the analysis results of the recurrent neural network algorithm, which includes an assessment of data consistency, any identified abnormal patterns, and an overall assessment of equipment performance; based on the predicted flow velocity, fluid density, and temperature data of the equipment in a certain period of time in the future and the current real-time flow velocity, fluid density, and temperature data obtained from the flow measurement equipment, use the long short-term memory network algorithm to compare the differences between the predicted data and the real-time data at different time points, and determine the trend of the difference over time; obtain the predicted The analysis results of the difference between the predicted data and the real-time data, including the size, direction and possible causes of the difference; determine whether the equipment needs to be calibrated based on the analysis results, and if the difference exceeds a predetermined threshold, determine that calibration is required; if it is determined that calibration is required, call the equipment's calibration program, which automatically adjusts the equipment settings based on the analysis results to reduce the difference between the predicted and real-time data; apply the determined calibration parameters to the flow measurement device; after the calibration is completed, obtain the equipment's real-time flow measurement data again, compare the data before and after calibration, and evaluate the calibration effect; use the equipment performance change prediction model to evaluate the performance of the equipment after calibration based on the calibrated real-time data to ensure that all adjustments achieve the expected results; continuously monitor the performance of the flow measurement equipment, and regularly use the equipment performance change prediction model to analyze and optimize the data.

[0027] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0028] The present invention provides an intelligent gas flow monitoring method. The accuracy and efficiency of gas flow measurement in industrial and scientific fields are significantly improved. By integrating advanced temperature prediction models and flow calculation methods, it overcomes the measurement challenges caused by thermal expansion or contraction of gas under temperature changes, ensuring the accuracy and consistency of measurements under fluctuating environmental conditions. Through real-time monitoring and data analysis, the method can predict the volume change of gas, and accurately predict the flow rate change of gas based on this, thereby providing reliable flow data under changing environmental conditions. In addition, the present invention significantly reduces the need and frequency of equipment maintenance and improves operational efficiency through automated anomaly detection and calibration mechanisms. The intelligent calibration cycle adjustment function further optimizes the long-term performance of the equipment and reduces the need for manual intervention. This method not only improves the safety and efficiency of industrial processes, but also helps to achieve higher environmental compliance, which is of great significance to modern industrial automation and intelligent transformation. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1The present invention is a flow chart of an intelligent gas flow monitoring method.

[0030] Figure 2 It is a schematic diagram of an intelligent gas flow monitoring method of the present invention.

[0031] Figure 3 It is another schematic diagram of an intelligent gas flow monitoring method of the present invention. DETAILED DESCRIPTION

[0032] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0033] In this embodiment, an intelligent gas flow monitoring method may specifically include:

[0034] S101, detecting the current temperature of the gas, constructing a gas temperature prediction model, and obtaining a specific value of the temperature change.

[0035] The real-time temperature data of the gas is obtained through the temperature sensor, and the data acquisition system is used to continuously record the gas temperature data provided by the sensor. The timestamp of each temperature data point is obtained through the sensor data, and the obtained temperature data is sorted in time series. According to the historical gas temperature data, the linear regression algorithm is used for model training, and a gas temperature prediction model is constructed to predict the trend of the gas temperature data, determine the maximum, minimum and average values ​​in the temperature data, and obtain the specific value of the temperature change. Determine whether the temperature change exceeds the preset normal operating range. If the temperature change exceeds the normal range, an abnormal report is obtained, including the temperature change value and timestamp.

[0036] For example, a temperature sensor measures the gas temperature once a minute and records the data continuously. In one hour, the sensor records a temperature change from 120°C to 150°C. Each temperature data point has a timestamp, such as 120°C at 12:00PM and 150°C at 12:30PM. Using the gas temperature data from the past week to train a linear regression model, the model reveals that the temperature usually rises at noon, such as from 120°C to 150°C. The model predicts the temperature trend in the next few hours and finds that the temperature is expected to reach 160°C. If the safe operating temperature range of the reactor is 100°C to 150°C, the maximum temperature predicted by the model is 160°C, which exceeds this range. Since the predicted temperature exceeds the normal operating range, the system automatically generates an abnormality report. The report indicates that the temperature is expected to reach 160°C at about 1:00PM, which is beyond the safe range.

[0037] S102. Using a linear regression algorithm, a gas thermal expansion or contraction prediction model is established, and the degree of gas thermal expansion or contraction is calculated based on the gas temperature change data.

[0038] Obtain real-time temperature data through the gas temperature sensor. Perform data preprocessing on the real-time temperature data, including removing noise and outliers. Use the processed temperature data through a linear regression algorithm to establish a gas thermal expansion or contraction prediction model. According to the current temperature data, obtain the preliminary prediction results of gas thermal expansion or contraction. By calculating the difference between the predicted value and the actual observed value, obtain the prediction error of each data point, and use the method of calculating the mean square error to calculate the overall error of the entire prediction result; determine the accuracy of the model based on the distribution of the error; if the error value is less than the preset threshold and the distribution is uniform, the model is considered to be accurate; if the error value is greater than the preset threshold or the distribution is uneven, it indicates that the model needs further optimization. If the model needs to be adjusted, adjust the parameters to improve the gas thermal expansion or contraction prediction model. According to the improved gas thermal expansion or contraction prediction model, obtain the optimized gas thermal expansion or contraction prediction results;

[0039] For example, use a gas temperature sensor to measure the real-time temperature of the gas and obtain real-time temperature data

[0040] [25,28,1,32,29]; First, remove the noise and outliers of the real-time temperature data. The temperature data of the third data point is detected to be 1, which is an outlier and is removed. The processed temperature data is [25,28,32,29]; Next, a linear regression algorithm is used to establish a gas thermal expansion or contraction prediction model. According to the processed temperature data, the following linear regression model y=43x+15 is established, where y represents the thermal expansion or contraction degree of the gas, and x represents the temperature. According to the current temperature data [25,28,32,29], the corresponding gas thermal expansion or contraction prediction results [1090,1219,1391,1262] are obtained; Next, the difference between the predicted value and the actual observed value is calculated to obtain the prediction error of each data point; Then, the method of calculating the mean square error is used to calculate the overall error of the entire prediction result. According to the distribution of the error, it can be judged whether the gas thermal expansion or contraction prediction model overestimates or underestimates the expansion or contraction degree of the gas. If the error value is less than the preset threshold and is evenly distributed, such as the mean square error is less than 2 and the error distribution is within ±5, the model is considered accurate. If the error value is greater than the preset threshold or is unevenly distributed, it indicates that the model needs to be further optimized; if the mean square error is less than 5 and the error distribution is within ±5, it indicates that the model is accurate. If it is found that the model needs to be adjusted, parameter adjustment or algorithm adjustment can be performed to improve the gas thermal expansion or contraction prediction model. Finally, based on the improved gas thermal expansion or contraction prediction model, the optimized gas thermal expansion or contraction prediction results can be obtained.

[0041] S103. Predict gas volume changes based on real-time temperature and pressure monitoring.

[0042] Use the gas temperature prediction model to obtain the temperature prediction results of the gas at a certain moment or time period in the future. Use the ideal gas law PV=nRT, substitute the predicted temperature value into the ideal gas law, and calculate the expected gas volume change, where P is pressure, V is volume, n is the amount of gas substance, R is the ideal gas constant, and T is temperature. Obtain the current pressure data of the gas measured in real time through the pressure sensor. Combine the real-time measured pressure data with the predicted temperature data to recalculate the gas volume change; determine the initial predicted value of the gas volume change after combining the temperature and pressure data. Obtain historical volume change data of the same type of gas under similar conditions. Compare the current prediction results with historical data. Adjust and correct the current volume change prediction based on the comparison of historical data. Continuously monitor the temperature and pressure of the gas, and update the data in real time. According to the latest monitoring data, further refine and update the volume change prediction to determine the final gas volume change prediction.

[0043] For example, the gas temperature prediction model predicts that the temperature of the gas at this time tomorrow will rise from the current 27°C to 77°C. Apply the ideal gas law to calculate the volume change: Use the ideal gas law PV=nRTP. In order to apply this law, we need to convert the temperature from Celsius to K, because the temperature unit in the ideal gas law is K. The conversion formula is T(K)=T(°C)+273.15T(K)=T(°C)+273.15. The initial temperature T1=27°C=300.15K, and the predicted temperature T2=77°C=350.15K. Under the initial conditions, the pressure P=1atm, the amount of substance n=1mol, and the gas constant R=0.0821L\cdotpatm / (K\cdotpmol); the initial volume V1 can be calculated by V1=nRT1P

[0044] V1 = 1 × 0.0821 × 300.151 ≈ 24.65 L. The predicted volume V2 can be calculated by V2 = nRT2P

[0045] V2 = 1 × 0.0821 × 350.151 ≈ 28.75 L. The prediction results show that due to the increase in temperature, the volume of the gas will increase from about 24.65 L to about 28.75 L. This indicates that the gas will expand and the volume will increase.

[0046] S104: Establish a velocity and volume relationship model based on the gas velocity and volume change data, and predict the gas velocity change based on the gas volume change.

[0047] Obtain historical gas flow rate and volume change data, use the Navier-Stokes equation to process the initial volume change data, and establish a flow rate and volume relationship model. Set the initial parameters of the relationship model based on historical data to predict the flow rate change of the gas. Compare the flow rate prediction results obtained by the flow rate and volume relationship model with the real-time measured volume change data. If the deviation between the predicted result and the actual data is greater than the threshold, dynamically adjust the parameters in the algorithm. Continuously monitor the volume change and flow rate change of the gas. If the difference between the actual measured volume change or flow rate and the predicted value is greater than the preset threshold, use the preset adjustment strategy to correct the flow rate prediction. According to the gas flow rate data, use the autoregressive model to evaluate the stability of the flow rate change prediction. If the flow rate change prediction is abnormal, use the Kalman filter to perform real-time dynamic correction on the flow rate prediction. Obtain the flow rate prediction data corrected by the Kalman filter and perform real-time monitoring again to verify the effectiveness of the correction. If there is still a deviation in the corrected flow rate change data, continue to use the Navier-Stokes equation to dynamically adjust the parameters, and continue to adjust until the flow rate change prediction data is stable.

[0048] For example, to predict the change of gas flow rate in a gas pipeline. According to the analysis of historical data, the initial parameters are obtained. The average historical gas flow rate is 10m / s, and the average historical volume change is 5m^3 / s. First, the Navier-Stokes equation is used to establish a flow rate and volume relationship model to predict the flow rate in the next time period, and the predicted value is 12m / s. Then, the volume change of the gas pipeline is measured in real time, and the actual value is 6m^3 / s. According to the preset threshold, if the deviation between the predicted result and the actual data is greater than the threshold of 1m / s, the parameters in the algorithm need to be adjusted dynamically. Next, the volume change and flow rate change of the gas are continuously monitored. In the next time period, the actual measured volume change is 7m^3 / s, which is significantly different from the predicted value of 6m^3 / s. According to the preset adjustment strategy, the flow rate prediction is corrected, and the corrected predicted flow rate is 15m / s. Then, the stability of the flow rate change prediction is evaluated using the autoregressive model based on the gas flow rate data. The evaluation results show that the flow rate prediction is continuously stable, and more refined flow rate change data is obtained. However, at a certain time period, the flow velocity prediction suddenly became abnormal and could not be predicted stably. At this time, the Kalman filter was used to perform real-time dynamic correction on the flow velocity prediction, and the corrected flow velocity prediction data was 18m / s. Then, real-time monitoring was performed again to verify the effectiveness of the correction. The deviation between the corrected flow velocity change data and the actual measured data was small. If there is still a deviation in the corrected flow velocity change data, continue to use the Navier-Stokes equation to dynamically adjust the parameters until the flow velocity change prediction data is stable.

[0049] Calculate and predict multiphase flow rates based on mixed fluid properties and real-time data adjustments.

[0050] According to the influence of the combination of multiphase fluids on the flow rate, a multiphase fluid flow rate prediction model is established

[0051] V=k×(w oil ×V oil +w water ×V water +w gas ×V gas ), where V represents the total flow rate, w oil is the volume ratio of oil in the mixture, w water is the volume ratio of water in the mixture, w gas is the volume ratio of the gas in the mixture, V oil is the base flow rate of the oil, V water is the basic velocity of water, V gas is the basic flow rate of the gas, and k is an adjustment factor. The basic flow rate V is measured in pure oil, pure water and pure gas environments respectively. oil , V water , V gas, record the flow rate data of each phase fluid at different pressures and temperatures. By mixing different proportions of oil, water, and gas, measure the total flow rate V of the mixed fluid. Use nonlinear regression analysis to determine the coefficient k in the multiphase fluid flow rate prediction model. In actual operation scenarios, use sensors to monitor the ratio w of oil, water, and gas in real time. oil , w water , w gas . According to the real-time monitored fluid ratio, the adjusted multiphase fluid flow rate prediction model is used to predict the flow rate. Based on the real-time and historical data analysis, the value of k is dynamically adjusted to cope with the changes in fluid ratio. Regularly compare the difference between the predicted flow rate and the actual measured flow rate to evaluate the accuracy and reliability of the multiphase fluid flow rate prediction model. Based on the performance evaluation results, further adjust the parameters in the formula to optimize the multiphase fluid flow rate prediction model. Implement a complete monitoring system to continuously obtain flow rate, ratio and environmental condition data of multiphase fluids. Establish a feedback mechanism to continuously optimize the prediction formula based on the comparison of monitoring data and prediction results.

[0052] For example, in a pure water environment, the basic flow rate V is measured water In a pure oil environment, the basic flow velocity V oil In a pure gas environment, the basic flow rate V gas According to the total flow rate V measured by mixing different proportions of oil, water and gas in the laboratory setting, when w oil =5,w water =3,w gas =2, the total flow velocity V is measured to be 12m / s. oil =4,w water =4,w gas =2, the total flow velocity V is measured to be 15m / s. oil =2,w water =3,w gas =5, the total flow velocity V is measured to be 18m / s. Based on the above data, nonlinear regression analysis is performed to obtain the formula V=k×(w oil × Voil +w water ×V water +w gas ×V gas ) is 9. In actual operation scenarios, sensors are used to monitor the ratio of oil, water, and gas in real time. oil ,w water ,w gas If the real-time monitored ratio of oil, water and gas is w oil =3,w wate r=4,w gas=3. Flow rate prediction is performed based on the adjusted formula:

[0053] V = 9 × (3 × 8 + 4 × 10 + 3 × 6) = 9 × 82 = 738 m / s; Regularly compare the difference between the predicted flow rate and the actual measured flow rate to evaluate the accuracy and reliability of the multiphase fluid flow rate prediction model. Based on the performance evaluation results, further adjust the parameters in the formula to optimize the multiphase fluid flow rate prediction model. Implement a comprehensive monitoring system to continuously obtain multiphase fluid flow rate, ratio and environmental condition data. Establish a feedback mechanism to continuously optimize the prediction formula based on the comparison of monitoring data and prediction results.

[0054] S105. Calculate the difference between the real-time flow rate measurement data and the predicted flow rate.

[0055] Use sensors to record fluid flow in real time, and use filtering and noise elimination techniques to process the acquired flow data. According to the gas volume data, use the velocity and volume relationship model to obtain the predicted value of the gas flow rate. Calculate the difference between the real-time monitored flow rate and the flow rate predicted by the velocity and volume relationship model. Detect whether the velocity difference exceeds the preset safety threshold. If the difference exceeds the threshold, trigger the anomaly detection mechanism. According to the anomaly detection results, based on the real-time flow measurement data and analysis algorithm, automatically adjust the parameters of the velocity and volume relationship model. Update the flow measurement data in real time to keep the data real-time and up-to-date. Continuously monitor and analyze the velocity difference. If the difference remains within the preset range, there is no need to adjust the model parameters. Automatically record and store all flow and velocity data.

[0056] For example, there is a sensor that records the gas flow in an industrial pipeline in real time. The sensor records data once a minute. The raw flow data it measures may contain noise. Filtering techniques, such as the moving average method, are used to process this data to eliminate the noise. If the flow values ​​measured for five consecutive times are 100, 105, 98, 102, and 99 L / min, the filtered value is the average of these five numbers, that is, 100.8 L / min. Next, based on the volume change data of the gas, a pre-established flow rate and volume relationship model is used to predict the flow rate. If the model predicts that the flow rate at a specific volume should be 105 L / min. Then, the actual measured flow rate of 100.8 L / min is compared with the predicted flow rate of 105 L / min, and the difference is found to be 4.2 L / min. If the set safety threshold is 5 L / min, then this difference of 4.2 L / min does not exceed the threshold, so it will not trigger the anomaly detection mechanism. The sensor continues to record data, and the system updates the flow data in real time. If the difference measured in the future continues to remain within the threshold range, the model parameters do not need to be adjusted. All flow and velocity data are automatically recorded and stored for future analysis and reference.

[0057] Continuously optimize the process of measuring and predicting fluid flow rates based on real-time environmental variables and flow rate data.

[0058] According to the density, temperature and pressure of the fluid, a flow rate prediction model is established. The flow rate prediction model formula is F = a × D b ×T c ×P d , where F is the flow rate, D is the fluid density, T is the temperature, P is the pressure, and a, b, c, d are parameters determined based on experimental data, indicating the degree of influence of these variables on the flow rate. Obtain the data of D, T, P and the actual flow rate F of the fluid under different environmental conditions. Use regression analysis to determine the parameters a, b, c, d in the formula. Measure D, T, P in real time and apply the formula F = a × D b ×T c ×P d Make flow velocity prediction. Record the predicted flow velocity F pred Compared with the actual measured flow rate F actual According to F pred With F actual Dynamically adjust the values ​​of a, b, c, and d based on the differences between the two. Continuously monitor changes in environmental conditions and flow data. Regularly update the flow rate prediction model parameters with newly acquired data to ensure that the model remains accurate and effective in a changing environment. Regularly evaluate the performance of the prediction model under extreme environmental conditions, such as extreme temperatures, high or low pressure, abnormal humidity levels, chemicals, or pollutants. Adjust model parameters based on performance evaluation results to cope with extreme environmental conditions. Prepare comprehensive performance reports that summarize the performance of the model under different conditions and provide report-based decision support.

[0059] For example, in a chemical plant process, the flow rate of gas J is monitored and it is hoped that the change in the gas flow rate under different temperature and pressure conditions can be predicted. Under standard laboratory conditions, the flow rate F of gas J is measured to be 15m / s and the density D is 0.09kg / m3. Using regression analysis to determine the parameters, the result is

[0060] a=1.2,b=1,c=-1,d=1a=1.2,b=1,c=-1,d=1. If the laboratory conditions change to 320K and 1.1atm, measure D,T,P in real time, and apply the formula to predict the flow rate, the prediction result F is obtained pred The actual measured flow velocity F is 16m / s. actualis 15.8m / s. Record this difference and make minor adjustments to the parameters, such as adjusting a to 1.25. Continue to monitor environmental changes, such as temperature and pressure, and record the corresponding flow rates. Re-evaluate model performance under extreme conditions, such as when the temperature rises to 40°C and the pressure rises to 1.5atm. If the performance does not meet expectations, further adjust the model parameters. Prepare a report summarizing the performance of the flow rate prediction model under different temperature and pressure conditions.

[0061] S106. Automatically adjust and calibrate the flow measurement equipment based on the anomaly detection and stability evaluation of the flow data.

[0062] According to the difference between the flow measurement data and the predicted flow rate, the flow data is classified into normal flow data and abnormal flow data. If the difference between the flow measurement data and the predicted flow rate exceeds a predetermined threshold, it is judged as abnormal flow data. If the judgment result is abnormal flow data, it is judged whether the abnormality is caused by the error of the flow measurement device by comparing the data at different time points or comparing with the readings of other sensors. If the abnormality is judged to be caused by the error of the device, the state and performance parameters of the device are obtained to evaluate its current working condition. The historical parameter data obtained by the flow measurement device, including flow rate, temperature, and pressure, are used to train the multi-layer perceptron model; the readings continuously obtained by the flow measurement device at different flow rates, temperatures, and pressures are used to evaluate the stability and accuracy of the output of the flow measurement device using the pre-trained multi-layer perceptron model. If the multi-layer perceptron analysis shows that the output of the device is unstable or inaccurate, the device automatically enters calibration and automatically adjusts its internal parameters to correct the output error. After the calibration is completed, the device automatically returns to the normal working mode, and the device continues to measure the flow and output data. After calibration, use the device diagnostic program to evaluate the output data of the device again to determine whether the device has been successfully calibrated. If the calibration is successful, end the calibration process. If the calibration is unsuccessful, recalibrate.

[0063] For example, the predicted flow rate of the flow measurement device is 100m / s, and the actual measured flow data is 105m / s. According to the set threshold of 5m / s, the difference exceeds the threshold, indicating that there is data anomaly, and the data is divided into normal and abnormal categories. By comparing the data at different time points, it is found that under the same operating conditions, the flow data at other time points are all 100m / s, indicating that the anomaly may be caused by equipment errors. The multi-layer perceptron model is trained by the historical parameter data obtained by the flow measurement device, including flow rate, temperature, and pressure; for the status and performance parameters of the device, its current working condition can be evaluated. If the temperature parameter of the device is 35℃ and the pressure parameter is 2MPa, according to the historical data analysis, these parameters are within the normal working range. The pre-trained multi-layer perceptron model is used to evaluate the stability and accuracy of the flow measurement device. According to the data continuously obtained by the device, the multi-layer perceptron analysis shows that the stability of the device output is 98% and the accuracy is 96%. According to the analysis results of the multi-layer perceptron, there is a certain instability and inaccuracy in the device output. The device automatically enters the calibration process and automatically adjusts the internal parameters to correct the output error. After the calibration is completed, the output data of the device is evaluated again using the device diagnostic program. After calibration, the stability of the data output by the device has increased to 99% and the accuracy has increased to 98%. According to the evaluation results of the device diagnostic program, the device has been successfully calibrated and the calibration process is completed. The device returns to normal operating mode and continues to measure flow and output data. If the evaluation results show that the device has not been successfully calibrated, the calibration process is repeated.

[0064] Implement equipment performance monitoring and dynamic calibration adjustments based on environmental changes.

[0065] Through sensors, flow, temperature, and pressure data are obtained under different temperatures, different fluid pressures, and different fluid types. Based on the data obtained under each environmental condition, a decision tree algorithm is used to train the model to identify data patterns and anomalies unique to the environment and determine the impact of temperature and pressure changes on flow measurement. Using historical and real-time data, a deep learning model based on a multilayer perceptron is established to identify data features under different environmental conditions and adapt to various environmental conditions. Based on the output of the multilayer perceptron model, the calibration strategy is dynamically adjusted to adapt to the current environmental conditions. The calibration settings of the device are automatically adjusted based on the analysis results of the multilayer perceptron model. The calibrated device is tested under different environmental conditions, and the performance data of the device is recorded and compared with the output of the prediction model to verify the calibration effect. The performance of the device under various environmental conditions is continuously monitored, and real-time data is regularly fed back to the multilayer perceptron model for iterative optimization of the model. Based on the test data under various environments, a comprehensive device performance evaluation report is established, which includes the performance of the device under various conditions, the calibration effect, and any areas that need further improvement. Based on the performance evaluation report, regular equipment maintenance and software updates are performed according to the problems and deficiencies pointed out in the report.

[0066] For example, through the sensor, the flow, temperature, and pressure data under different temperatures, different fluid pressures, and different fluid types are obtained. When the temperature is 25°C, the fluid pressure is 2bar, and the fluid type is water, the flow measured by the sensor is 10L / min, the temperature is 27°C, and the pressure is 2bar. According to the data obtained under each environmental condition, the decision tree algorithm is used for model training to identify the data patterns and anomalies unique to the environment, and determine the impact of temperature and pressure changes on flow measurement. It can be found that under high temperature and high pressure, the flow measurement value will be too large. Using historical and real-time data, a deep learning model based on multilayer perceptron is established to identify data features under different environmental conditions and adapt to various environmental conditions. The complex relationship between flow, temperature, and pressure under different temperatures, pressures, and fluid types can be learned. According to the output of the multilayer perceptron model, the calibration strategy is dynamically adjusted to adapt to the current environmental conditions. When the flow measurement value output by the multilayer perceptron model deviates greatly from the actual measurement value, the calibration settings of the sensor can be automatically adjusted. When the multilayer perceptron model analyzes that the sensor measurement value is too large under high temperature and high pressure environment, the calibration settings can be automatically adjusted to correct the deviation. Test the calibrated equipment under different environmental conditions, record the performance data of the equipment, and compare it with the output of the prediction model to verify the calibration effect. For example, place the calibrated equipment in an environment with a temperature of 30°C, a fluid pressure of 3 bar, and a fluid type of oil for testing, and compare it with the flow value output by the prediction model to verify the calibration effect. Continuously monitor the performance of the equipment under various environmental conditions, and regularly feed back real-time data to the multi-layer perceptron model for iterative optimization of the model. Collect the performance data of the equipment under different temperatures, pressures, and fluid types every day, and use these data to update the multi-layer perceptron model to improve the accuracy of performance prediction. Based on the test data under various environments, establish a comprehensive equipment performance evaluation report. The report includes the performance of the equipment under different temperatures, pressures, and fluid types, the calibration effect, and any areas that need further improvement. The report points out that in high temperature environments, the flow measurement results of the equipment have large deviations, and the calibration strategy of the sensor needs to be further improved. Based on the performance evaluation report, according to the problems and deficiencies pointed out in the report, regularly perform equipment maintenance and software updates. According to the deviation problems in high temperature environments pointed out in the report, calibrate or replace the sensor, and update the relevant parameters in the multi-layer perceptron model to improve the accuracy of performance prediction.

[0067] S107. Automatically adjust the calibration period of the flow measurement device according to historical and real-time monitoring data.

[0068] Obtain the historical calibration records, flow measurement data, and environmental parameters, including temperature and pressure; analyze the calibration frequency and performance fluctuations in the historical data through the autoregressive model, and identify the key factors that lead to changes in calibration requirements, including changes in environmental conditions or increases or decreases in the frequency of equipment use. Based on historical and real-time data, use a multi-layer perceptron to build a device calibration prediction model to predict future calibration needs. Input historical calibration data and real-time monitoring parameters to train the device calibration prediction model. Use the device calibration prediction model to determine the current device status and possible calibration requirements based on the flow data and environmental parameters of the device. Based on the output of the device calibration prediction model and real-time data analysis, determine whether the device needs to be calibrated and determine the best time for calibration. When the device calibration prediction model predicts that the device needs to be calibrated, it automatically triggers the calibration process and ensures that the calibration process meets the calibration requirements and operating specifications of the device. After the calibration is completed, record the device performance data and evaluate the calibration effect by comparing the flow measurement accuracy before and after calibration. Regularly adjust the parameters of the device calibration prediction model based on the calibration results and the operating data of the device.

[0069] For example, there is a flow meter device that needs to be calibrated to ensure measurement accuracy. First, collect the historical calibration records, flow measurement data, and environmental parameters, including temperature and pressure, of the device. The calibration frequency and performance fluctuations in the historical data can be analyzed by an autoregressive model. It is found that the calibration frequency was performed once every three months on average in the past year, and the performance fluctuations were kept within 5% before and after calibration. Next, it is necessary to identify the key factors that lead to changes in calibration needs. By analyzing historical data, it is found that changes in calibration needs are related to changes in environmental conditions and the increase or decrease in the frequency of equipment use. When the temperature rises in summer, the calibration needs of the equipment may increase. Based on historical and real-time data, a multi-layer perceptron is used to build a device calibration prediction model. Input historical calibration data and real-time monitoring parameters, and train the model to predict future calibration needs. Based on historical data, the model may predict that calibration is required within the next quarter. Continuously monitor the flow data and related environmental parameters of the device, and use the device calibration prediction model to determine the current device status and possible calibration needs. Based on the flow data and temperature and pressure parameters of the device, the model may determine that the current device needs to be calibrated. Based on the output of the prediction model and real-time data analysis, it can be determined whether the device needs to be calibrated and the best time for calibration can be determined. The model predicts that the equipment needs to be calibrated in the next month. The calibration process can be automatically triggered at the optimal time point and ensure that the calibration meets the requirements and operating specifications. After the calibration is completed, the equipment performance data is recorded and the effect of the calibration is evaluated by comparing the flow measurement accuracy before and after the calibration. The measurement accuracy after calibration is improved to 3%. According to the calibration results and the operation data of the equipment, the parameters of the prediction model are adjusted regularly. The parameters can be adjusted according to the data of the most recent year to improve the accuracy of the model. If the average flow measurement data of the equipment in the past week is 1000L / min, and the performance fluctuation range after the last calibration is 50L / min. At the same time, the average temperature in the environmental parameters in the past week is 25℃, and the average pressure in the past week is 2bar. According to the historical data analysis, it is found that when the temperature rises to 30℃, the calibration demand of the equipment will increase by 10%. According to the model prediction, there is a 50% probability that the equipment will need to be calibrated in the next quarter. According to the real-time data analysis, the current temperature is 28℃ and the pressure is 5bar. The model determines that the equipment needs to be calibrated. The calibration process is automatically triggered according to the optimal time point, and the measurement accuracy after calibration is improved to 3%. Based on calibration results and equipment operation data, the parameters of the prediction model are regularly adjusted to improve accuracy.

[0070] S108. Use the predicted data and real-time data to calibrate the flow measurement of the equipment.

[0071] Obtain real-time data of the flow measurement device, including flow velocity, fluid density, and temperature; if the predicted data is within the preset error range, use the recurrent neural network algorithm to train the model based on the real-time data, build a prediction model for device performance changes, detect changes in device performance or identify possible anomalies. Through the analysis results of the recurrent neural network algorithm, a preliminary data quality report is obtained, which includes an assessment of data consistency, any abnormal patterns identified, and an overall assessment of device performance. Based on the predicted flow velocity, fluid density, and temperature data of the device in a certain period of time in the future and the current real-time flow velocity, fluid density, and temperature data obtained from the flow measurement device, the long short-term memory network algorithm is used to compare the differences between the predicted data and the real-time data at different time points to determine the trend of the difference over time. Obtain the analysis results of the difference between the predicted data and the real-time data, including the size, direction, and possible causes of the difference. Determine whether the device needs to be calibrated based on the analysis results. If the difference exceeds a predetermined threshold, it is determined that calibration is required. If it is determined that calibration is required, call the device's calibration program, which automatically adjusts the device settings based on the analysis results to reduce the difference between the predicted and real-time data. Apply the determined calibration parameters to the flow measurement device. After calibration, obtain the real-time flow measurement data of the device again, compare the data before and after calibration, and evaluate the calibration effect. Use the device performance change prediction model to evaluate the performance of the device after calibration based on the real-time data after calibration to ensure that all adjustments achieve the expected results. Continuously monitor the performance of the flow measurement device and regularly use the device performance change prediction model to analyze and optimize the data.

[0072] For example, real-time data from flow measurement equipment is obtained, including flow velocity, fluid density, and temperature; if the predicted data is within the preset error range, a recurrent neural network algorithm is used to train the model based on the real-time data, build a prediction model for equipment performance changes, detect changes in equipment performance, or identify possible anomalies. Based on the analysis results of the recurrent neural network algorithm, a preliminary data quality report is obtained, which includes an assessment of data consistency, any identified abnormal patterns, and an overall assessment of equipment performance. Real-time data was obtained from the flow measurement device, flow velocity: 50m / s, fluid density: 1000kg / m 3 , temperature: 25℃; Next, use the recurrent neural network algorithm to train the model for these data and build a prediction model for equipment performance changes. The preliminary data quality report obtained through the model analysis obtained a data consistency assessment. According to historical data, the real-time data is consistent with the previous data. No abnormal patterns have been identified so far. According to the model analysis, the equipment performance is normal. Next, use the long short-term memory network algorithm to compare the predicted data with the real-time data to determine whether there is a significant difference between the two. If the predicted data is, flow rate: 51m / s, fluid density: 1001kg / m 3, temperature: 26℃; the difference results obtained through analysis may be that the flow velocity increased by 1m / s and the fluid density increased by 1kg / m 3 , the temperature increased by 1°C. Compared with the predicted data, the real-time data shows a slight increase in flow rate, fluid density and temperature. The increase in ambient temperature causes the fluid density and temperature to increase, while the increase in flow rate may be due to other factors. Based on the analysis results, it is determined that the device needs to be calibrated because the difference exceeds the predetermined threshold. Next, the calibration procedure of the device is called to automatically adjust the device settings based on the analysis results to reduce the difference between the predicted and real-time data. After the calibration is completed, the real-time flow measurement data of the device is obtained again, the data before and after calibration are compared, and the calibration effect is evaluated. The real-time data after calibration is: flow rate: 55m / s, fluid density: 1005kg / m 3 , temperature: 25℃; through comparative evaluation, it can be determined that the calibration effect is good, and the difference between the predicted and real-time data is significantly reduced. Finally, the performance of the equipment after calibration is evaluated based on the real-time data after calibration using the equipment performance change prediction model to ensure that all adjustments have achieved the expected results. Continuously monitor the performance of the flow measurement equipment and regularly use the equipment performance change prediction model to analyze and optimize the data.

[0073] The description of the above embodiments is only used to help understand the technical solutions and core ideas of the present application. Ordinary technical personnel in this field should understand that they can still modify the technical solutions recorded in the aforementioned embodiments, or replace some of the technical features therein with equivalents. However, these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. An intelligent gas flow monitoring method, characterized in that: The method comprises: Detect the current temperature of the gas, build a gas temperature prediction model, and obtain the specific value of the temperature change; use a linear regression algorithm to establish a gas thermal expansion or contraction prediction model, and calculate the degree of gas thermal expansion or contraction based on the gas temperature change data; predict the gas volume change based on real-time temperature and pressure monitoring; build a flow rate and volume relationship model based on the gas flow rate and volume change data, and predict the gas flow rate change based on the gas volume change; calculate the difference between the predicted flow rate and the predicted flow rate through real-time flow measurement data; automatically adjust and calibrate the flow measurement equipment based on abnormality detection and stability evaluation of the flow data; automatically adjust the calibration period of the flow measurement equipment based on historical and real-time monitoring data; use predicted data and real-time data to calibrate the flow measurement of the equipment; The present temperature of the detected gas is constructed, and a gas temperature prediction model is constructed to obtain a specific value of the temperature change, including: obtaining real-time temperature data of the gas through a temperature sensor, and using a data acquisition system to continuously record the gas temperature data provided by the sensor; obtaining a timestamp of each temperature data point through sensor data, and performing time series sorting on the obtained temperature data; according to historical gas temperature data, a linear regression algorithm is used to perform model training, and a gas temperature prediction model is constructed to predict the trend of the gas temperature data, determine the maximum and minimum values ​​and the average value in the temperature data, and obtain a specific value of the temperature change; The method of using a linear regression algorithm to establish a prediction model for thermal expansion or contraction of gas comprises: obtaining real-time temperature data through a gas temperature sensor; performing data preprocessing on the real-time temperature data, including removing noise and outliers; and using a linear regression algorithm to use the processed temperature data to establish a prediction model for thermal expansion or contraction of gas; The method of establishing a velocity and volume relationship model based on gas velocity and volume change data and predicting gas velocity changes based on gas volume changes includes: obtaining historical gas velocity and volume change data, using the Navier-Stokes equation to process initial volume change data, and establishing a velocity and volume relationship model; setting initial parameters of the relationship model based on historical data to predict gas velocity changes; comparing the velocity prediction result obtained by the velocity and volume relationship model with the real-time measured volume change data; dynamically adjusting the parameters in the algorithm if the deviation between the prediction result and the actual data is greater than a threshold; continuously monitoring the volume change and Flow rate changes: if the difference between the actual measured volume change or flow rate and the predicted value is greater than the preset threshold, the preset adjustment strategy is used to correct the flow rate prediction; based on the gas flow rate data, the autoregressive model is used to evaluate the stability of the flow rate change prediction; if the flow rate change prediction is abnormal, the Kalman filter is used to perform real-time dynamic corrections to the flow rate prediction; the flow rate prediction data corrected by the Kalman filter is obtained, and real-time monitoring is performed again to verify the effectiveness of the correction; if the corrected flow rate change data still has deviations, the Navier-Stokes equations are continued to be used to dynamically adjust the parameters, and the adjustment is continued until the flow rate change prediction data is stable.

2. The method according to claim 1, wherein: The current temperature of the gas is detected, a gas temperature prediction model is constructed, and a specific value of the temperature change is obtained, including: Determine whether the temperature change exceeds the preset normal operating range; if the temperature change exceeds the normal range, obtain an abnormal report including the temperature change value and timestamp.

3. The method according to claim 1, wherein: The calculation of the degree of thermal expansion or contraction of the gas based on the gas temperature change data includes: According to the current temperature data, a preliminary prediction result of the thermal expansion or contraction of the gas is obtained; the prediction error of each data point is obtained by calculating the difference between the predicted value and the actual observed value, and the overall error of the entire prediction result is calculated by calculating the mean square error; the accuracy of the model is determined according to the distribution of the error; if the error value is less than the preset threshold and the distribution is uniform, the model is considered to be accurate, if the error value is greater than the preset threshold or the distribution is uneven, it indicates that the model needs to be further optimized; if the model needs to be adjusted, the parameters are adjusted to improve the gas thermal expansion or contraction prediction model; according to the improved gas thermal expansion or contraction prediction model, the optimized gas thermal expansion or contraction prediction result is obtained.

4. The method according to claim 1, wherein: The method of predicting gas volume changes based on real-time temperature and pressure monitoring includes: Use the gas temperature prediction model to obtain the temperature prediction result of the gas at a certain moment or time period in the future; use the ideal gas law PV=nRT, substitute the predicted temperature value into the ideal gas law, and calculate the expected gas volume change, where P is pressure, V is volume, n is the amount of gas substance, R is the ideal gas constant, and T is temperature; obtain the current pressure data of the gas measured in real time through the pressure sensor; combine the real-time measured pressure data with the predicted temperature data to recalculate the gas volume change; determine the initial predicted value of the gas volume change after combining the temperature and pressure data; Obtain historical volume change data for the same type of gas under similar conditions; compare current prediction results with historical data; adjust and correct current volume change predictions based on historical data comparison; continuously monitor gas temperature and pressure, and update data in real time, further refine and update volume change predictions based on the latest monitoring data, and determine the final gas volume change prediction.

5. The method according to claim 1, wherein: The method of establishing a velocity and volume relationship model based on the gas velocity and volume change data and predicting the gas velocity change based on the gas volume change includes: Calculate and predict the flow rate of multiphase fluids based on mixed fluid characteristics and real-time data adjustments; The method of adjusting, calculating and predicting the flow rate of the multiphase fluid according to the mixed fluid characteristics and real-time data specifically includes: establishing a multiphase fluid flow rate prediction model V=k×(w oil ×V oil +w water ×V water +w gas ×V gas ), where V represents the total flow rate, w oil is the volume ratio of oil in the mixture, w water is the volume ratio of water in the mixture, w gas is the volume ratio of the gas in the mixture, V oil is the base flow rate of the oil, V water is the basic velocity of water, V gas is the basic flow rate of the gas, k is an adjustment factor; the basic flow rate V is measured in pure oil, pure water and pure gas environments respectively oil , V water , V gas , record the flow rate data of each phase fluid at different pressures and temperatures; measure the total flow rate V of the mixed fluid by mixing different proportions of oil, water, and gas; use nonlinear regression analysis to determine the coefficient k in the multiphase fluid flow rate prediction model; in actual operation scenarios, use sensors to monitor the ratio w of oil, water, and gas in real time oil , w water , w gas ; According to the real-time monitored fluid ratio, the adjusted multiphase fluid flow rate prediction model is used to predict the flow rate; based on the real-time and historical data analysis, the value of k is dynamically adjusted to cope with the changes in fluid ratio; the difference between the predicted flow rate and the actual measured flow rate is regularly compared to evaluate the accuracy and reliability of the multiphase fluid flow rate prediction model; based on the performance evaluation results, the parameters in the formula are further adjusted to optimize the multiphase fluid flow rate prediction model; a complete monitoring system is implemented to continuously obtain the flow rate, ratio and environmental condition data of the multiphase fluid; a feedback mechanism is established to continuously optimize the prediction formula based on the comparison of monitoring data and prediction results.

6. The method according to claim 1, wherein: The difference between the real-time flow measurement data and the predicted flow rate is calculated, including: Use sensors to record fluid flow in real time, and use filtering and noise elimination technology to process the acquired flow data; use the flow rate and volume relationship model to obtain the predicted value of gas flow rate based on gas volume data; calculate the difference between the flow rate monitored in real time and the flow rate predicted by the flow rate and volume relationship model; detect whether the flow rate difference exceeds the preset safety threshold, and if the difference exceeds the threshold, trigger the abnormal detection mechanism; according to the abnormal detection results, based on the real-time flow measurement data and analysis algorithm, automatically adjust the parameters of the flow rate and volume relationship model; update the flow measurement data in real time to keep the data real-time and up-to-date; continuously monitor and analyze the flow rate difference, and if the difference remains within the preset range, there is no need to adjust the model parameters; automatically record and store all flow and flow rate data; also includes: continuously optimize the measurement and prediction process of fluid flow rate based on real-time environmental variables and flow rate data; The process of continuously optimizing the measurement and prediction of fluid flow rate according to real-time environmental variables and flow rate data specifically includes: establishing a flow rate prediction model according to the density, temperature and pressure of the fluid, and the flow rate prediction model formula is F=a×D b ×T c ×P d , where F represents flow rate, D represents fluid density, T represents temperature, P represents pressure, and a, b, c, d are parameters determined based on experimental data, indicating the degree of influence of these variables on flow rate; obtain the data of D, T, P and actual flow rate F of the fluid under different environmental conditions; use regression analysis to determine the parameters a, b, c, d in the formula; measure D, T, P in real time, and apply the formula F = a × D b ×T c ×P d Carry out flow velocity prediction; record the predicted flow velocity F pred Compared with the actual measured flow rate F actual The difference between pred With F actual Dynamically adjust the values ​​of a, b, c, and d based on the differences between them; continuously monitor changes in environmental conditions and flow data; regularly update the flow rate prediction model parameters with newly acquired data to ensure that the model remains accurate and effective in a changing environment; regularly evaluate the performance of the prediction model under extreme environmental conditions, including extreme temperatures, high or low pressure, abnormal humidity levels, chemicals or pollutants; adjust model parameters based on performance evaluation results to cope with extreme environmental conditions; prepare a comprehensive performance report that summarizes the performance of the model under different conditions and provides report-based decision support.

7. The method according to claim 1, wherein: The method of automatically adjusting and calibrating a flow measurement device based on anomaly detection and stability assessment of flow data includes: According to the difference between the flow measurement data and the predicted flow rate, the flow data is classified into normal flow data and abnormal flow data. If the difference between the flow measurement data and the predicted flow rate exceeds a predetermined threshold, it is judged as abnormal flow data. If the judgment result is abnormal flow data, it is judged whether the abnormality is caused by the error of the flow measurement device by comparing the data at different time points or comparing with the readings of other sensors. If the abnormality is judged to be caused by the error of the equipment, the status and performance parameters of the equipment are obtained to evaluate its current working condition. The historical parameter data obtained by the flow measurement device, including flow rate, temperature and pressure, is used to train the multi-layer perceptron. Model; the readings continuously obtained by the flow measurement device at different flow rates, temperatures and pressures are used to evaluate the stability and accuracy of the flow measurement device output using a pre-trained multi-layer perceptron model; if the multi-layer perceptron analysis shows that the device output is unstable or inaccurate, the device automatically enters calibration and automatically adjusts its internal parameters to correct the output error; after the calibration is completed, the device automatically returns to the normal working mode, and the device continues to measure flow and output data; after calibration, the device diagnostic program is used to evaluate the device output data again to determine whether the device has been successfully calibrated. If the calibration is successful, the calibration process is terminated. If the calibration is unsuccessful, the calibration is repeated; It also includes: implementing equipment performance monitoring and dynamic calibration adjustments based on environmental changes; The implementation of equipment performance monitoring and dynamic calibration adjustment according to environmental changes specifically includes: obtaining flow, temperature, and pressure data under different temperatures, different fluid pressures, and different fluid types through sensors; using a decision tree algorithm to train a model based on the data obtained under each environmental condition, identifying data patterns and anomalies unique to the environment, and determining the impact of temperature and pressure changes on flow measurement; using historical and real-time data to establish a deep learning model based on a multilayer perceptron to identify data features under different environmental conditions and adapt to various environmental conditions; dynamically adjusting the calibration strategy based on the output of the multilayer perceptron model to adapt to current environmental conditions; automatically adjusting the calibration settings of the equipment based on the analysis results of the multilayer perceptron model; testing the calibrated equipment under different environmental conditions, recording the performance data of the equipment, and comparing it with the output of the prediction model to verify the calibration effect; continuously monitoring the performance of the equipment under various environmental conditions, regularly feeding back real-time data to the multilayer perceptron model, and iteratively optimizing the model; establishing a comprehensive equipment performance evaluation report based on the test data under various environments, the equipment performance evaluation report including the performance of the equipment under various conditions and the calibration effect; based on the performance evaluation report, regularly performing equipment maintenance and software updates based on the problems and deficiencies pointed out in the report.

8. The method according to claim 1, wherein: The method of automatically adjusting the calibration period of the flow measurement device according to historical and real-time monitoring data comprises: Obtain historical equipment calibration records and flow measurement data, as well as environmental parameters, including temperature and pressure; use autoregressive models to analyze calibration frequency and performance fluctuations in historical data, and identify key factors that lead to changes in calibration requirements, including changes in environmental conditions or increases or decreases in equipment usage frequency; use multi-layer perceptrons to build equipment calibration prediction models based on historical and real-time data to predict future calibration requirements; input historical calibration data and real-time monitoring parameters to train equipment calibration prediction models; use equipment calibration prediction models to determine current equipment status and possible calibration requirements based on equipment flow data and environmental parameters; determine whether equipment needs calibration based on the output of the equipment calibration prediction model and real-time data analysis, and determine the best time for calibration; when the equipment calibration prediction model predicts that equipment needs calibration, automatically trigger the calibration process and ensure that the calibration process meets the equipment's calibration requirements and operating specifications; After the calibration is completed, the equipment performance data is recorded, and the effect of the calibration is evaluated by comparing the flow measurement accuracy before and after calibration; according to the calibration results and the equipment operation data, the parameters of the equipment calibration prediction model are regularly adjusted.

9. The method according to claim 1, wherein: The method of using the predicted data and the real-time data to calibrate the flow measurement of the equipment includes: Obtain real-time data from flow measurement equipment, including flow velocity, fluid density, and temperature; if the predicted data is within the preset error range, use the recurrent neural network algorithm to train the model based on the real-time data, build a prediction model for equipment performance changes, detect changes in equipment performance, or identify possible anomalies; Through the analysis results of the recurrent neural network algorithm, a preliminary data quality report is obtained, which includes an assessment of data consistency, any abnormal patterns identified, and an overall assessment of equipment performance; based on the predicted flow rate, fluid density, and temperature data of the equipment in a certain period of time in the future and the current real-time flow rate, fluid density, and temperature data obtained from the flow measurement device, the long short-term memory network algorithm is used to compare the differences between the predicted data and the real-time data at different time points to determine the trend of the differences over time; the analysis results of the differences between the predicted data and the real-time data are obtained, including the size, direction, and possible causes of the differences; Determine whether the device needs to be calibrated based on the analysis results, and if the difference exceeds a predetermined threshold, determine that calibration is required; If it is determined that calibration is required, call the device's calibration program, which automatically adjusts the device settings based on the analysis results to reduce the difference between the predicted and real-time data; apply the determined calibration parameters to the flow measurement device; after the calibration is completed, obtain the device's real-time flow measurement data again, compare the data before and after calibration, and evaluate the calibration effect; use the device performance change prediction model to evaluate the performance of the device after calibration based on the real-time data after calibration to ensure that all adjustments achieve the expected results; continuously monitor the performance of the flow measurement device, and regularly use the device performance change prediction model to analyze and optimize the data.

Images