Oxygen detection method and equipment based on oxygen production equipment and storage medium

Abstract

The invention discloses an oxygen detection method and equipment based on oxygen production equipment and a storage medium, and relates to the technical field of oxygen detection. According to the method, the real-time torsion angle of the dumbbell ball is calculated and compared with the threshold value, when the torsion angle exceeds the threshold value, the swing amplitude is estimated through the Kalman filtering algorithm in combination with airflow disturbance data, then the support vector machine model is input to classify the airflow disturbance influence degree, the judgment rule is generated, and the measurement precision problem caused by gas flow is solved; meanwhile, a temperature and pressure sensor is used for collecting data and correcting swing amplitude data in combination with a compensation algorithm, so that the problem of temperature and pressure change interference is solved; besides, whether the linear response range of the photoelectric sensor is exceeded or not is judged according to the corrected torsion angle data, when the linear response range is exceeded, sensor array segmentation fitting is adopted, the oxygen concentration is calculated in combination with a fitting algorithm, the problem of sensor response limitation is solved, the accuracy and reliability of oxygen concentration measurement are effectively improved, and the measurement range is effectively widened.

Description

Technical Field

[0001] The present invention relates to the technical field of oxygen detection, and specifically to an oxygen detection method, device, and storage medium based on an oxygen generation device. Background Art

[0002] In the medical field, the accurate measurement of oxygen concentration is crucial. Currently, the commonly used method for measuring oxygen concentration is to utilize the paramagnetic principle of oxygen and determine its concentration by measuring the magnetic field change caused by oxygen. However, in practical applications, due to the influence of gas flow in the detection chamber, the suspended dumbbell-shaped quartz glass balls will generate a certain swing, resulting in a change in the position of the reflected light spot on the photoelectric sensor, thereby affecting the measurement accuracy. Summary of the Invention

[0003] The purpose of the present invention is to provide an oxygen detection method, device, and storage medium based on an oxygen generation device, which realizes high-precision, high-stability, and real-time monitoring of the oxygen concentration in the detection chamber, significantly improves the reliability and safety of oxygen detection in the oxygen generation device, and provides a strong guarantee for the accurate measurement of oxygen concentration in medical and other fields.

[0004] The purpose of the present invention can be achieved through the following technical solutions: The present application provides an oxygen detection method based on an oxygen generation device, including the following steps: S1. Obtain the magnetic field change data caused by the oxygen concentration in the detection chamber, reduce external interference through a high-stability magnetic field source to obtain a magnetic field change signal, and then calculate the torsion angle of the dumbbell ball. When the torsion angle exceeds a preset threshold, judge the influence of air flow disturbance on the swing of the dumbbell ball to obtain the swing amplitude data; S2. According to the swing amplitude data, use a temperature sensor and a pressure sensor to collect the temperature and pressure fluctuation data in the detection chamber in real time to obtain temperature and pressure compensation parameters, and combine a preset compensation algorithm to correct the swing amplitude data to obtain the corrected torsion angle data of the dumbbell ball; S3. According to the corrected torsion angle data of the dumbbell ball, judge whether it exceeds the linear response range of the photoelectric sensor. When it exceeds the range, use multiple photoelectric sensors to form an array to obtain segmented linear fitting data; S4. According to the segmented linear fitting data, combine a preset fitting algorithm to calculate the corresponding relationship between the oxygen concentration and the torsion angle to obtain the oxygen concentration measurement result, and judge whether there is non-linear distortion. When there is distortion, recalculate through the segmented fitting data of the sensor array to obtain the corrected oxygen concentration value.

[0005] Further, obtaining the magnetic field change data caused by the oxygen concentration in the detection chamber and reducing external interference through a high-stability magnetic field source to obtain a magnetic field change signal specifically includes: Adopt a high-stability magnetic field source to generate a uniform and stable magnetic field through the magnetic field source, reduce the influence of external environmental magnetic field interference on the indoor magnetic field change, arrange multiple magnetic field sensors indoors to obtain magnetic field change data at different positions, and improve the spatial resolution and coverage of the magnetic field change data through multi-point sampling; Preprocess the obtained magnetic field change data, and according to the characteristics of the magnetic field change data, adopt the wavelet transform signal processing method to extract the characteristic information in the magnetic field change signal as an indicator of the oxygen concentration change; Then establish a mapping relationship model between the magnetic field change signal and the oxygen concentration change, and train the mapping model through machine learning algorithms such as support vector machines or neural networks; In the mapping relationship model, by designing a real-time monitoring algorithm for the oxygen concentration change, continuously collect and analyze the magnetic field change signal, and judge whether the current oxygen concentration is within the normal range. When it is lower than the preset threshold, give a timely warning.

[0006] Furthermore, after obtaining the magnetic field change signal, it also includes: obtaining the magnetic field change signal of the dumbbell ball, calculating the real-time torsion angle of the dumbbell ball through the magnetic field change signal, comparing the calculated real-time torsion angle of the dumbbell ball with the preset torsion angle threshold, and judging whether the torsion angle of the dumbbell ball exceeds the preset threshold; When the torsion angle of the dumbbell ball exceeds the preset threshold, obtain the current airflow disturbance data, analyze the influence degree of the airflow disturbance on the swing of the dumbbell ball through the airflow disturbance data, and then according to the airflow disturbance data and the dumbbell ball torsion angle data, adopt the Kalman filter algorithm to evaluate the swing amplitude of the dumbbell ball to obtain an estimated value of the swing amplitude of the dumbbell ball; Compare the estimated value of the swing amplitude of the dumbbell ball with the preset swing amplitude threshold, and judge whether the swing amplitude of the dumbbell ball exceeds the preset threshold; When the swing amplitude of the dumbbell ball exceeds the preset threshold, input the torsion angle data and swing amplitude data of the dumbbell ball into a pre-trained support vector machine model, and classify the influence degree of the airflow disturbance through the support vector machine model to obtain a classification result of the influence degree of the airflow disturbance; According to the classification result of the influence degree of the airflow disturbance, adopt the decision tree algorithm to generate a judgment rule for the influence degree of the airflow disturbance, and apply the judgment rule to the dumbbell ball swing monitoring to perform real-time judgment and warning on the airflow disturbance.

[0007] Furthermore, according to the swing amplitude data, use temperature sensors and pressure sensors to collect and detect indoor temperature and pressure fluctuation data in real time to obtain temperature and pressure compensation parameters, specifically including: Combined with a temperature sensor and a pressure sensor, the indoor temperature and pressure fluctuation data are collected in real time, and then the original temperature and pressure data collected by the sensors are preprocessed to obtain effective temperature and pressure data; According to the historical temperature and pressure data, a multiple regression model of temperature-pressure-swing amplitude is established, and the model parameters are trained by machine learning algorithms. The temperature and pressure data collected in real time are input into the trained regression model to predict the estimated value of the swing amplitude in the current environment; Then calculate the temperature and pressure compensation parameters, and by comparing the actually measured swing amplitude with the model prediction value, obtain the deviation amount caused by temperature and pressure.

[0008] Furthermore, combined with a preset compensation algorithm, the swing amplitude data is corrected to obtain the corrected dumbbell ball torsion angle data, specifically including: The Kalman filter algorithm is used to filter the swing amplitude data, establish the state equation and observation equation of the dumbbell ball movement, and use recursive calculation to obtain the optimal estimated value to obtain the filtered swing amplitude data; According to the temperature and pressure compensation parameters and combined with a preset compensation algorithm, non-linear compensation is performed on the filtered swing amplitude data. By establishing a non-linear mapping relationship between the swing amplitude of the dumbbell ball and the torsion angle, the corresponding compensation value is determined by using the look-up table method or the interpolation method to obtain the compensated torsion angle data; The least squares method is used to perform curve fitting on the compensated torsion angle data. By selecting a suitable fitting function and using the principle of minimizing the sum of squared residuals to determine the coefficients of the fitting function, the fitted torsion angle data curve is obtained; According to the fitted torsion angle data curve, the torsion angle data is smoothed. By setting the smoothing window and the weighting coefficient, the data is smoothed by using methods such as moving average or exponential smoothing to obtain the smoothed torsion angle data; The adaptive threshold method is used to perform threshold processing on the smoothed torsion angle data. By dynamically adjusting the threshold size and using the threshold judgment condition to perform binary processing on the data, the threshold-processed torsion angle data is obtained; According to the threshold-processed torsion angle data, the corrected dumbbell ball torsion angle data is obtained by using the weighted fusion or decision tree method.

[0009] Furthermore, according to the corrected dumbbell ball torsion angle data, it is judged whether it exceeds the linear response range of the photoelectric sensor. When it exceeds the range, an array composed of multiple photoelectric sensors is used to obtain piecewise linear fitting data, specifically including: Obtain the corrected torsional angle data of the dumbbell ball, and input the data into the judgment module. The judgment module determines whether the torsional angle data exceeds the linear response range of a single photoelectric sensor according to the preset linear response range threshold; When the torsional angle data does not exceed the linear response range, a single photoelectric sensor is used to obtain the complete torsional angle data. When the torsional angle data exceeds the linear response range, the sensor array module is triggered, and according to the preset sensor layout scheme, a matching number of photoelectric sensors are selected to form a sensor array; The sensor array module controls each photoelectric sensor to obtain the torsional angle data in different angle ranges respectively, forming segmented data. Then, the obtained segmented torsional angle data is input into the data fitting module, and the segmented linear fitting algorithm is used to fit each segment of data to obtain a complete torsional angle fitting curve.

[0010] Furthermore, according to the segmented linear fitting data, combined with the preset fitting algorithm, calculate the corresponding relationship between the oxygen concentration and the torsional angle to obtain the oxygen concentration measurement result, specifically including: Obtain the set of data points of the segmented linear fitting, use the least squares method to perform linear fitting on each data segment to obtain the linear function expression of this segment, and then according to the preset fitting algorithm, determine the position of the segmentation point of the segmented linear function, and connect the linear functions of adjacent segments to obtain a complete segmented linear function; Obtain the currently measured torsional angle value, calculate the theoretical value of the oxygen concentration corresponding to this torsional angle through the segmented linear function, obtain the actual measured value of the oxygen concentration sensor, and compare it with the theoretically calculated value. When the deviation exceeds the preset threshold, it is judged as abnormal and an alarm prompt is given; According to the oxygen concentration values measured multiple times, use the Kalman filter algorithm to filter out noise to obtain a more accurate and stable oxygen concentration measurement result, and store the corresponding relationship between the measured oxygen concentration result and the torsional angle in the database.

[0011] Output the final oxygen concentration measurement result according to the corrected oxygen concentration value to complete the measurement process.

[0012] An oxygen detection device based on an oxygen generation device, including a processor, a memory, and computer program instructions stored in the memory. When the computer program instructions are executed by the processor, the above-mentioned oxygen detection method based on an oxygen generation device is implemented.

[0013] A storage medium, on which computer program instructions are stored. When the computer program instructions are executed by the processor, the above-mentioned oxygen detection method based on an oxygen generation device is implemented.

[0014] The beneficial effects of the present invention are: The present invention calculates the real-time torsion angle by obtaining the magnetic field change signal of the dumbbell ball, and compares it with a preset threshold. When the threshold is exceeded, the air flow disturbance data is obtained, and the Kalman filtering algorithm is used to estimate the swing amplitude. Then, the relevant data is input into the support vector machine model to classify the influence degree of the air flow disturbance, and a judgment rule is generated and applied to the monitoring, thereby effectively solving the problem that the swing of the dumbbell-shaped quartz glass ball caused by the indoor gas flow affects the measurement accuracy, and improving the accuracy of oxygen concentration measurement; A temperature sensor and a pressure sensor are used to collect the temperature and pressure fluctuation data in the detection chamber in real time to obtain compensation parameters. Combining with a preset compensation algorithm, a series of correction processes are performed on the swing amplitude data, including Kalman filtering, nonlinear compensation, curve fitting, smoothing processing, and threshold processing, etc. Finally, the corrected torsion angle data of the dumbbell ball is obtained, thereby solving the interference problem of temperature and pressure changes on the oxygen concentration measurement result, and making the measurement data more accurate and reliable; According to the corrected torsion angle data of the dumbbell ball, it is judged whether it exceeds the linear response range of the photoelectric sensor. When it exceeds, an array composed of multiple photoelectric sensors is used to obtain segmented data and perform segmented linear fitting to obtain a complete torsion angle fitting curve. Then, combining with a preset fitting algorithm to calculate the corresponding relationship between the oxygen concentration and the torsion angle, the oxygen concentration measurement result is obtained, and when there is nonlinear distortion, it is recalculated and corrected, thereby effectively expanding the measurement range and solving the problem of inaccurate measurement caused by the limited linear response range of a single photoelectric sensor, ensuring that the oxygen concentration can be accurately measured at different torsion angles. Description of the Drawings

[0015] For better understanding and implementation, the technical solution of the present application will be described in detail below with reference to the drawings.

[0016] Figure 1 It is a schematic flow chart of an oxygen detection method based on an oxygen generation device provided by the present application; Figure 2 It is a schematic flow chart of obtaining the magnetic field change signal of an oxygen detection method based on an oxygen generation device provided by the present application; Figure 3 It is a schematic flow chart of obtaining the temperature and pressure compensation parameters of an oxygen detection method based on an oxygen generation device provided by the present application. Specific Embodiments

[0017] To further illustrate the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, exemplary embodiments will be described in detail herein, and the examples are shown in the accompanying drawings. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of methods and systems consistent with some aspects of the present application as detailed in the appended claims.

[0018] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms "a", "the", and "said" used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0019] The following will describe in detail the specific embodiments, features, and effects of the present invention in conjunction with the accompanying drawings and preferred embodiments.

[0020] Please refer to Figures 1 - 3 , this embodiment provides an oxygen detection method based on an oxygen generation device, including the following steps: S1. Obtain the magnetic field change data caused by the oxygen concentration in the detection chamber, reduce external interference through a high-stability magnetic field source to obtain a magnetic field change signal, and then calculate the torsion angle of the dumbbell ball. When the torsion angle exceeds the preset threshold, judge the influence of air flow disturbance on the swing of the dumbbell ball to obtain the swing amplitude data; Further, obtaining the magnetic field change data caused by the oxygen concentration in the detection chamber and reducing external interference through a high-stability magnetic field source to obtain a magnetic field change signal specifically includes: S11. Adopt a high-stability magnetic field source to generate a uniform and stable magnetic field through the magnetic field source, reduce the influence of external environmental magnetic field interference on the indoor magnetic field change, improve the accuracy of the magnetic field change data, arrange multiple magnetic field sensors in the room to obtain the magnetic field change data at different positions, and improve the spatial resolution and coverage of the magnetic field change data through multi-point sampling; S12. Preprocess the obtained magnetic field change data, including removing outliers, smoothing noise, etc., to improve the signal-to-noise ratio and quality of the magnetic field change data. According to the characteristics of the magnetic field change data, adopt signal processing methods such as wavelet transform to extract the characteristic information in the magnetic field change signal, such as signal amplitude, frequency, phase, etc., as an indication of the oxygen concentration change; S13. Then, establish a mapping relationship model between the magnetic field change signal and the oxygen concentration change. Through machine learning algorithms such as support vector machines or neural networks, train the mapping model to achieve the function of accurately estimating the indoor oxygen concentration from the magnetic field change signal; S14. In the mapping relationship model, by designing a real-time monitoring algorithm for the oxygen concentration change, continuously collect and analyze the magnetic field change signal, and judge whether the current oxygen concentration is within the normal range. When it is lower than the preset threshold, give a timely warning.

[0021] Specifically, through a series of methods such as adopting a high-stability magnetic field source, arranging magnetic field sensors at multiple points, preprocessing the magnetic field change data, and establishing a mapping relationship model, high-precision, high-stability, and real-time monitoring of the indoor oxygen concentration are achieved. Specifically, the high-stability magnetic field source reduces external interference, multi-point sampling improves the spatial resolution and coverage of the magnetic field change data, the preprocessing and signal processing methods improve the data quality and the accuracy of feature extraction, and the mapping model trained by the machine learning algorithm realizes the accurate estimation of the oxygen concentration from the magnetic field change signal. The real-time monitoring algorithm can continuously collect and analyze the magnetic field change signal, timely judge whether the oxygen concentration is normal and give a warning. This effectively solves the problems such as large external interference, low data accuracy, and untimely monitoring that may exist in traditional oxygen detection methods, improves the reliability and safety of oxygen detection in oxygen production equipment, and provides more accurate oxygen concentration monitoring guarantee for relevant application scenarios.

[0022] Further, after obtaining the magnetic field change signal, it also includes: obtaining the magnetic field change signal of the dumbbell ball, calculating the real-time torsion angle of the dumbbell ball through the magnetic field change signal, comparing the calculated real-time torsion angle of the dumbbell ball with a preset torsion angle threshold, and judging whether the torsion angle of the dumbbell ball exceeds the preset threshold; When the torsion angle of the dumbbell ball exceeds the preset threshold, obtain the current air flow disturbance data, analyze the influence degree of the air flow disturbance on the swing of the dumbbell ball through the air flow disturbance data, and then according to the air flow disturbance data and the dumbbell ball torsion angle data, use the Kalman filter algorithm to evaluate the swing amplitude of the dumbbell ball to obtain an estimated value of the swing amplitude of the dumbbell ball; Compare the estimated value of the swing amplitude of the dumbbell ball with a preset swing amplitude threshold, and judge whether the swing amplitude of the dumbbell ball exceeds the preset threshold; When the swing amplitude of the dumbbell ball exceeds the preset threshold, input the torsion angle data and swing amplitude data of the dumbbell ball into a pre-trained support vector machine model, and classify the influence degree of the air flow disturbance through the support vector machine model to obtain a classification result of the influence degree of the air flow disturbance; According to the classification results of the influence degree of air flow disturbance, a decision tree algorithm is used to generate judgment rules for the influence degree of air flow disturbance, and the judgment rules are applied to the subsequent monitoring of the swing of the dumbbell ball to realize real-time judgment and early warning of the influence of air flow disturbance.

[0023] Specifically, by acquiring the magnetic field change signal of the dumbbell ball and calculating its real-time torsion angle, combining the air flow disturbance data, using the Kalman filter algorithm to estimate the swing amplitude of the dumbbell ball, and classifying the influence degree of air flow disturbance with the help of a support vector machine model, and then using the decision tree algorithm to generate judgment rules, the accurate monitoring and real-time early warning of the influence of air flow disturbance are realized, effectively improving the accuracy and reliability of oxygen detection, and solving the problem of detection error caused by air flow disturbance to the swing of the dumbbell ball.

[0024] S2. According to the swing amplitude data, use temperature sensors and pressure sensors to collect the temperature and pressure fluctuation data in the detection room in real time, obtain the temperature and pressure compensation parameters, and combine the preset compensation algorithm to correct the swing amplitude data to obtain the corrected torsion angle data of the dumbbell ball; Furthermore, according to the swing amplitude data, use temperature sensors and pressure sensors to collect the temperature and pressure fluctuation data in the detection room in real time, obtain the temperature and pressure compensation parameters, specifically including: S21. Through the combination of temperature sensors and pressure sensors, collect the temperature and pressure fluctuation data in the room in real time, and then preprocess the original temperature and pressure data collected by the sensors, including data cleaning, outlier filtering, etc., to obtain effective temperature and pressure data; S22. According to the historical temperature and pressure data, establish a multiple regression model of temperature-pressure-swing amplitude, train the model parameters through machine learning algorithms such as support vector machines or random forests, and input the real-time collected temperature and pressure data into the trained regression model to predict the estimated swing amplitude value in the current environment; S23. Then calculate the temperature and pressure compensation parameters, and obtain the deviation amount caused by temperature and pressure by comparing the actually measured swing amplitude with the model prediction value.

[0025] Specifically, by collecting the temperature and pressure fluctuation data in the detection room in real time, combining the multiple regression model and machine learning algorithms, calculating the temperature and pressure compensation parameters, and then correcting the swing amplitude data, more accurate torsion angle data of the dumbbell ball are obtained, effectively improving the oxygen detection accuracy, and solving the problem of the influence of temperature and pressure changes on the detection results.

[0026] Furthermore, combine the preset compensation algorithm to correct the swing amplitude data to obtain the corrected torsion angle data of the dumbbell ball, specifically including: The Kalman filter algorithm is used to filter the swing amplitude data, establish the state equation and observation equation of the dumbbell ball movement, and obtain the optimal estimated value through recursive calculation to acquire the filtered swing amplitude data; According to the temperature and pressure compensation parameters and the preset compensation algorithm, perform non-linear compensation on the filtered swing amplitude data. By establishing the non-linear mapping relationship between the swing amplitude of the dumbbell ball and the torsion angle, use the look-up table method or interpolation method to determine the corresponding compensation value, and obtain the compensated torsion angle data; Use the least squares method to perform curve fitting on the compensated torsion angle data. By selecting an appropriate fitting function and using the principle of minimizing the sum of squared residuals to determine the coefficients of the fitting function, obtain the curve of the fitted torsion angle data; According to the curve of the fitted torsion angle data, perform smoothing processing on the torsion angle data. By setting the smoothing window and weighting coefficient, use methods such as moving average or exponential smoothing to smooth the data, and obtain the smoothed torsion angle data; Use the adaptive threshold method to perform threshold processing on the smoothed torsion angle data. By dynamically adjusting the threshold size and using the threshold judgment condition to perform binary processing on the data, obtain the torsion angle data after threshold processing; According to the torsion angle data after threshold processing, use methods such as weighted fusion or decision tree to obtain the corrected torsion angle data of the dumbbell ball.

[0027] Specifically, the swing amplitude data is filtered by the Kalman filter algorithm, then non-linearly compensated in combination with the temperature and pressure compensation parameters, then curve fitting is performed using the least squares method, the fitted data is smoothed, then threshold processing is performed using the adaptive threshold method, and finally the corrected torsion angle data of the dumbbell ball is obtained through methods such as weighted fusion or decision tree. This series of operations effectively improves the accuracy and stability of oxygen detection, makes the detection results more accurate and reliable, and can better reflect the actual oxygen concentration situation.

[0028] S3. According to the corrected torsion angle data of the dumbbell ball, judge whether it exceeds the linear response range of the photoelectric sensor. When it exceeds the range, use multiple photoelectric sensors to form an array to obtain piecewise linear fitting data; Furthermore, according to the corrected torsion angle data of the dumbbell ball, judge whether it exceeds the linear response range of the photoelectric sensor. When it exceeds the range, use multiple photoelectric sensors to form an array to obtain piecewise linear fitting data, specifically including: Obtain the corrected torsion angle data of the dumbbell ball and input the data into the judgment module. The judgment module judges whether the torsion angle data exceeds the linear response range of a single photoelectric sensor according to the preset linear response range threshold; When the torsional angle data does not exceed the linear response range, the complete torsional angle data is directly obtained by a single photoelectric sensor. When the torsional angle data exceeds the linear response range, the sensor array module is triggered, and according to the preset sensor layout scheme, the matching number of photoelectric sensors is selected to form a sensor array; The sensor array module controls each photoelectric sensor to obtain the torsional angle data within different angle ranges respectively, forming segmented data. Then, the obtained segmented torsional angle data is input into the data fitting module, and the segmented linear fitting algorithm is used to fit each segment of data to obtain the complete torsional angle fitting curve.

[0029] Among them, according to the torsional angle fitting curve, the angle change data of the dumbbell ball during the whole torsional process is determined as the input for subsequent motion analysis; The rule for determining the matching number is as follows: when the corrected torsional angle data of the dumbbell ball does not exceed the linear response range of a single photoelectric sensor, only 1 photoelectric sensor is needed; when it exceeds this range, it is calculated based on the ratio of the actual torsional angle range of the dumbbell ball to the linear response range of a single photoelectric sensor. Specifically, in the two-way measurement scenario, the matching number is the result of rounding up this ratio and then multiplying by 2 to ensure coverage of the entire range from the maximum negative torsional angle to the maximum positive torsional angle; in the one-way measurement scenario, the matching number is the result of rounding up this ratio, so as to ensure the measurement accuracy in the corresponding direction.

[0030] Specifically, by judging whether the corrected torsional angle data of the dumbbell ball exceeds the linear response range of the photoelectric sensor, and when it exceeds the range, multiple photoelectric sensors are used to form an array for segmented linear fitting, effectively expanding the detection range, improving the detection accuracy, and being able to accurately obtain the angle change data of the dumbbell ball during the whole torsional process, providing more reliable data support for subsequent oxygen concentration analysis.

[0031] S4. According to the segmented linear fitting data, combined with the preset fitting algorithm, calculate the corresponding relationship between the oxygen concentration and the torsional angle, obtain the oxygen concentration measurement result, and judge whether there is non-linear distortion. When there is distortion, recalculate through the segmented fitting data of the sensor array to obtain the corrected oxygen concentration value; Furthermore, according to the segmented linear fitting data, combined with the preset fitting algorithm, calculate the corresponding relationship between the oxygen concentration and the torsional angle, obtain the oxygen concentration measurement result, specifically including: Obtain the set of data points of the segmented linear fitting, use the least squares method to perform linear fitting on each data segment to obtain the linear function expression of this segment, and then according to the preset fitting algorithm, determine the position of the segmentation point of the segmented linear function, and connect the linear functions of adjacent segments to obtain the complete segmented linear function; Obtain the currently measured torsional angle value, calculate the theoretical value of the oxygen concentration corresponding to the torsional angle through a piecewise linear function, obtain the actual measured value of the oxygen concentration sensor, compare it with the theoretically calculated value, and when the deviation exceeds the preset threshold, it is judged as abnormal and an alarm prompt is given; According to the oxygen concentration values measured multiple times, use the Kalman filter algorithm to filter out noise, obtain a more accurate and stable oxygen concentration measurement result, and store the corresponding relationship between the measured oxygen concentration result and the torsional angle in the database for subsequent data analysis and model optimization.

[0032] Specifically, calculate the corresponding relationship between the oxygen concentration and the torsional angle by piecewise linear fitting of data combined with a preset fitting algorithm to obtain the oxygen concentration measurement result, and be able to judge whether there is nonlinear distortion. When there is distortion, use the piecewise fitting data of the sensor array to recalculate the corrected oxygen concentration value, thereby improving the accuracy and reliability of the oxygen concentration measurement, providing more accurate data for the oxygen detection of the oxygen production equipment, and ensuring the stability and effectiveness of the oxygen concentration measurement.

[0033] S5. Output the final oxygen concentration measurement result according to the corrected oxygen concentration value to complete the measurement process.

[0034] An oxygen detection device based on an oxygen production device includes a processor, a memory, and computer program instructions stored in the memory. When the computer program instructions are executed by the processor, the above-mentioned oxygen detection method based on an oxygen production device is implemented.

[0035] A storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the above-mentioned oxygen detection method based on an oxygen production device is implemented.

[0036] The oxygen detection method based on an oxygen production device in this embodiment first generates a uniform and stable magnetic field through a highly stable magnetic field source to reduce external interference, and arranges multiple magnetic field sensors indoors to obtain magnetic field change data at different positions, improving the spatial resolution and coverage of the magnetic field change data. Then, preprocess the magnetic field change data, including removing outliers, smoothing noise, etc. Next, use signal processing methods such as wavelet transform to extract the characteristic information in the magnetic field change signal, establish a mapping relationship model between the magnetic field change signal and the oxygen concentration change, train the model through machine learning algorithms such as support vector machines or neural networks, realize the function of accurately estimating the indoor oxygen concentration from the magnetic field change signal, and design a real-time monitoring algorithm to continuously collect and analyze the magnetic field change signal, judge whether the oxygen concentration is normal and give an early warning in time; After obtaining the magnetic field change signal, further obtain the magnetic field change signal of the dumbbell ball to calculate the real-time torsion angle, compare it with the preset threshold. When the threshold is exceeded, obtain the air flow disturbance data, use the Kalman filter algorithm to estimate the swing amplitude of the dumbbell ball, then input the relevant data into the support vector machine model to classify the influence degree of the air flow disturbance, and generate judgment rules for application in monitoring, so as to realize the real-time judgment and early warning of the influence of the air flow disturbance; At the same time, use temperature sensors and pressure sensors to collect the indoor temperature and pressure fluctuation data in real time, obtain the compensation parameters, and combine with the preset compensation algorithm to correct the swing amplitude data, including Kalman filtering, non-linear compensation, curve fitting, smoothing processing and threshold processing, etc., to obtain the corrected torsion angle data of the dumbbell ball and improve the oxygen detection accuracy; Then, according to the corrected torsion angle data of the dumbbell ball, judge whether it exceeds the linear response range of the photoelectric sensor. When it exceeds the range, use multiple photoelectric sensors to form an array for piecewise linear fitting to obtain a complete torsion angle fitting curve, and determine the angle change data of the dumbbell ball during the entire torsion process, providing reliable support for subsequent analysis; Finally, according to the piecewise linear fitting data, combine with the preset fitting algorithm to calculate the corresponding relationship between the oxygen concentration and the torsion angle, obtain the oxygen concentration measurement result, and judge whether there is non-linear distortion. When there is distortion, use the piecewise fitting data of the sensor array to recalculate the corrected oxygen concentration value to improve the measurement accuracy and reliability, and finally output the corrected oxygen concentration measurement result to complete the measurement process.

[0037] The whole method effectively solves problems such as external interference, air flow disturbance, temperature and pressure changes, and sensor response limitations through a series of fine steps and advanced algorithms, realizes high-precision, high-stability and real-time monitoring of the oxygen concentration in the detection chamber, provides more accurate and reliable data for the oxygen detection of the oxygen production equipment, ensures the stability and effectiveness of the oxygen concentration measurement, improves the reliability and safety of the oxygen detection of the oxygen production equipment, and provides a strong guarantee for related application scenarios.

[0038] The above is only a preferred embodiment of the present invention, and does not impose any form of limitation on the present invention. Although the present invention has been disclosed above with a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to the above-disclosed technical content to be equivalent embodiments within the scope of the technical solution of the present invention. However, as long as it does not depart from the content of the technical solution of the present invention, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. An oxygen detection method based on an oxygen generation device, characterized in that: It includes the following steps: Obtain the magnetic field change data caused by the oxygen concentration in the detection chamber, reduce external interference through a high-stability magnetic field source to obtain a magnetic field change signal, then calculate the torsion angle of the dumbbell ball. When the torsion angle exceeds the preset threshold, judge the influence of air flow disturbance on the swing of the dumbbell ball to obtain the swing amplitude data; According to the swing amplitude data, use a temperature sensor and a pressure sensor to collect the temperature and pressure fluctuation data in the detection chamber in real time to obtain temperature and pressure compensation parameters, and combine with a preset compensation algorithm to correct the swing amplitude data to obtain the corrected torsion angle data of the dumbbell ball; According to the corrected torsion angle data of the dumbbell ball, judge whether it exceeds the linear response range of the photoelectric sensor. When it exceeds the range, use multiple photoelectric sensors to form an array to obtain piecewise linear fitting data; According to the piecewise linear fitting data, combine with a preset fitting algorithm to calculate the corresponding relationship between the oxygen concentration and the torsion angle to obtain the oxygen concentration measurement result, and judge whether there is non-linear distortion. When there is distortion, recalculate through the piecewise fitting data of the sensor array to obtain the corrected oxygen concentration value.

2. The oxygen detection method based on an oxygen generation device according to claim 1, characterized in that: Obtain the magnetic field change data caused by the oxygen concentration in the detection chamber, reduce external interference through a high-stability magnetic field source to obtain a magnetic field change signal, specifically including: Use a high-stability magnetic field source to generate a uniform and stable magnetic field through the magnetic field source, reduce the influence of external environmental magnetic field interference on the indoor magnetic field change, arrange multiple magnetic field sensors in the room to obtain magnetic field change data at different positions, and improve the spatial resolution and coverage of the magnetic field change data through multi-point sampling; Preprocess the obtained magnetic field change data, and according to the characteristics of the magnetic field change data, use the wavelet transform signal processing method to extract the characteristic information in the magnetic field change signal as an indicator of the oxygen concentration change; Then establish a mapping relationship model between the magnetic field change signal and the oxygen concentration change, and train the mapping model through machine learning algorithms such as support vector machines or neural networks; In the mapping relationship model, by designing a real-time monitoring algorithm for the oxygen concentration change, continuously collect and analyze the magnetic field change signal to judge whether the current oxygen concentration is within the normal range. When it is lower than the preset threshold, give a timely warning.

3. The oxygen detection method based on an oxygen generation device according to claim 2, characterized in that: After obtaining the magnetic field change signal, it also includes: obtaining the magnetic field change signal of the dumbbell ball, calculating the real-time torsion angle of the dumbbell ball through the magnetic field change signal, comparing the calculated real-time torsion angle of the dumbbell ball with the preset torsion angle threshold, and judging whether the torsion angle of the dumbbell ball exceeds the preset threshold; When the torsion angle of the dumbbell ball exceeds the preset threshold, obtain the current air flow disturbance data, analyze the influence degree of the air flow disturbance on the swing of the dumbbell ball through the air flow disturbance data, and then according to the air flow disturbance data and the dumbbell ball torsion angle data, use the Kalman filter algorithm to evaluate the swing amplitude of the dumbbell ball to obtain an estimated value of the swing amplitude of the dumbbell ball; Compare the estimated value of the swing amplitude of the dumbbell ball with the preset swing amplitude threshold to judge whether the swing amplitude of the dumbbell ball exceeds the preset threshold; When the swing amplitude of the dumbbell ball exceeds the preset threshold, the torsion angle data and swing amplitude data of the dumbbell ball are input into a pre-trained support vector machine model. The support vector machine model classifies the influence degree of the airflow disturbance to obtain the classification result of the influence degree of the airflow disturbance. According to the classification result of the influence degree of the airflow disturbance, a decision tree algorithm is used to generate a judgment rule for the influence degree of the airflow disturbance. The judgment rule is applied to the monitoring of the dumbbell ball swing to perform real-time judgment and early warning of the airflow disturbance.

4. The oxygen detection method based on an oxygen generation device according to claim 1, characterized in that: According to the swing amplitude data, a temperature sensor and a pressure sensor are used to collect the indoor temperature and pressure fluctuation data in real time to obtain the temperature and pressure compensation parameters, specifically including: Through the combination of a temperature sensor and a pressure sensor, the indoor temperature and pressure fluctuation data are collected in real time, and then the original temperature and pressure data collected by the sensor are preprocessed to obtain effective temperature and pressure data. According to the historical temperature and pressure data, a multiple regression model of temperature-pressure-swing amplitude is established. The model parameters are trained by a machine learning algorithm, and the temperature and pressure data collected in real time are input into the trained regression model to predict the estimated value of the swing amplitude in the current environment. Then, the temperature and pressure compensation parameters are calculated. By comparing the actually measured swing amplitude with the model prediction value, the deviation amount caused by temperature and pressure is obtained.

5. The oxygen detection method based on an oxygen generation device according to claim 1, characterized in that: Combined with the preset compensation algorithm, the swing amplitude data is corrected to obtain the corrected torsion angle data of the dumbbell ball, specifically including: The Kalman filtering algorithm is used to filter the swing amplitude data, establish the state equation and observation equation of the dumbbell ball movement, and use recursive calculation to obtain the optimal estimated value to obtain the filtered swing amplitude data. According to the temperature and pressure compensation parameters and the preset compensation algorithm, the filtered swing amplitude data is nonlinearly compensated. By establishing a nonlinear mapping relationship between the swing amplitude and torsion angle of the dumbbell ball, the corresponding compensation value is determined by using the look-up table method or interpolation method to obtain the compensated torsion angle data. The least squares method is used to perform curve fitting on the compensated torsion angle data. By selecting a fitting function and using the principle of minimizing the sum of squared residuals to determine the coefficients of the fitting function, the fitted torsion angle data curve is obtained. According to the fitted torsion angle data curve, the torsion angle data is smoothed. By setting a smoothing window and a weighting coefficient, the data is smoothed by using the moving average or exponential smoothing method to obtain the smoothed torsion angle data. The adaptive threshold method is used to perform threshold processing on the smoothed torsion angle data. By dynamically adjusting the threshold size, the data is binarized by using the threshold judgment condition to obtain the threshold-processed torsion angle data. According to the threshold-processed torsion angle data, the corrected torsion angle data of the dumbbell ball is obtained by using the weighted fusion or decision tree method.

6. The oxygen detection method based on an oxygen generation device according to claim 1, characterized in that: According to the corrected torsion angle data of the dumbbell ball, it is judged whether it exceeds the linear response range of the photoelectric sensor. When it exceeds the range, an array composed of multiple photoelectric sensors is used to obtain the piecewise linear fitting data, specifically including: Obtain the corrected dumbbell ball torsion angle data, and input the data into the judgment module. The judgment module determines whether the torsion angle data exceeds the linear response range of a single photoelectric sensor according to the preset linear response range threshold; When the torsion angle data does not exceed the linear response range, a single photoelectric sensor is used to obtain the complete torsion angle data. When the torsion angle data exceeds the linear response range, the sensor array module is triggered, and according to the preset sensor layout scheme, a matching number of photoelectric sensors are selected to form a sensor array; The sensor array module controls each photoelectric sensor to obtain the torsion angle data in different angle ranges respectively, forming segmented data, and then inputs the obtained segmented torsion angle data into the data fitting module. The segmented linear fitting algorithm is used to fit each segment of data to obtain a complete torsion angle fitting curve.

7. A method for detecting oxygen based on an oxygen generation device according to claim 1, wherein: According to the segmented linear fitting data, combined with the preset fitting algorithm, calculate the corresponding relationship between the oxygen concentration and the torsion angle to obtain the oxygen concentration measurement result, specifically including: Obtain the set of data points of the segmented linear fitting, use the least squares method to perform linear fitting on each data segment to obtain the linear function expression of this segment, and then according to the preset fitting algorithm, determine the position of the segmentation point of the segmented linear function, and connect the linear functions of adjacent segments to obtain a complete segmented linear function; Obtain the currently measured torsion angle value, calculate the theoretical value of the oxygen concentration corresponding to this torsion angle through the segmented linear function, obtain the actual measured value of the oxygen concentration sensor, and compare it with the theoretically calculated value. When the deviation exceeds the preset threshold, it is judged as abnormal and an alarm prompt is given; According to the oxygen concentration values measured multiple times, use the Kalman filter algorithm to filter out noise to obtain a more accurate and stable oxygen concentration measurement result, and store the corresponding relationship between the measured oxygen concentration result and the torsion angle in the database.

8. A method for detecting oxygen based on an oxygen generation device according to claim 1, wherein: It also includes: Output the final oxygen concentration measurement result according to the corrected oxygen concentration value to complete the measurement process.

9. An oxygen detection device based on an oxygen generation device, wherein: It includes a processor, a memory, and computer program instructions stored in the memory. When the computer program instructions are executed by the processor, it implements an oxygen detection method based on an oxygen generation device described in any one of claims 1-8 above.

10. A storage medium having computer program instructions stored thereon, wherein, When the computer program instructions are executed by the processor, it implements an oxygen detection method based on an oxygen generation device described in any one of claims 1-8 above.

Images