Non-contact high-precision liquid volume measuring method for liquid nitrogen generating device

Abstract

The invention discloses a non-contact high-precision liquid nitrogen generation device liquid level detection device measurement method, and belongs to the technical field of low-temperature medium liquid level parameter detection. The device comprises a man-machine interaction module, a liquid nitrogen storage tank, a multi-sensor platform, a damping support frame and an embedded module, the multi-sensor platform is integrated with a distributed weighing sensor and a temperature and vibration auxiliary sensing unit, and a multi-layer graded damping assembly is arranged in the damping support frame; according to the method, benchmark and compensation coefficients are established through multi-dimensional automatic calibration, multi-source data are collected and preprocessed in real time, the liquid nitrogen liquid level height is obtained through multi-dimensional interference correction, data cross validation and fusion calculation, precision is optimized through regular dynamic calibration, and redundancy switching and linkage control are supported. The device is suitable for real-time monitoring of the liquid amount of industrial-grade and laboratory-grade liquid nitrogen generation devices, the detection precision is high, the anti-interference capacity is high, installation is convenient, a tank does not need to be refitted, and continuous and reliable detection and safe operation are guaranteed.

Description

Technical Field

[0001] This invention belongs to the field of cryogenic medium liquid level parameter detection technology, specifically involving non-contact liquid volume detection technology for liquid nitrogen generation devices. Background Technology

[0002] With the development of cryogenic refrigeration technology, liquid nitrogen generation devices have been widely used in industrial production, laboratory research, and other scenarios. Liquid nitrogen volume and level detection during storage are crucial for ensuring the stable operation of liquid nitrogen generation devices and achieving a dynamic balance between liquid nitrogen supply and demand. Currently, the main technical solutions for liquid nitrogen volume detection are divided into two categories: gravimetric methods and contact-based level detection.

[0003] Traditional single-sensor weighing methods measure the total weight of the liquid nitrogen storage tank using a single weighing sensor and calculate the volume of liquid nitrogen within the tank based on a fixed liquid nitrogen density. This approach lacks an auxiliary sensing unit, making the acquired weight signal susceptible to interference from external factors such as electromagnetic fields during transmission. Furthermore, it fails to address the vibrations generated by the air compressor operating with the liquid nitrogen generation device, which directly amplifies the weight measurement error. Additionally, this method requires manual calibration of the support surface using a level, a cumbersome process whose accuracy is easily affected by human error, leading to inaccuracies in the volume calculation.

[0004] Contact-type liquid level detection solutions typically use transparent level indicator containers paired with color-changing indicator strips or float assemblies to indicate the liquid level. During detection, the detection component needs to be immersed in the liquid nitrogen storage tank or brought into direct contact with the liquid nitrogen. Over long-term use, the detection component is susceptible to corrosion and aging due to the low-temperature environment. Furthermore, immersion installation can damage the original sealing structure of the liquid nitrogen storage tank, posing a safety hazard of liquid nitrogen leakage. These solutions rely on manual visual observation for readings and cannot achieve automatic calibration and real-time monitoring of the liquid nitrogen volume, making them unsuitable for continuous operation scenarios of liquid nitrogen generation devices.

[0005] Besides the two mainstream solutions mentioned above, some existing technologies employ liquid nitrogen volume detection schemes using single pressure sensors or ultrasonic sensors. These schemes generally suffer from limited detection dimensions, failing to comprehensively cover multiple sources of interference during the operation of the liquid nitrogen generator, such as air compressor vibration, ambient temperature drift, and localized sensor errors. This results in insufficient detection accuracy, poor operational stability, and a significant decrease in accuracy over long-term use. Furthermore, existing detection schemes generally lack redundancy design; a single sensor failure directly leads to detection malfunction, failing to meet the long-term, continuous, and highly reliable operation requirements of liquid nitrogen generators, thus hindering their stable application under complex operating conditions. Summary of the Invention

[0006] The purpose of this invention is to provide a non-contact, high-precision liquid nitrogen generation device liquid volume measurement method to solve the problems of low detection accuracy, weak anti-interference ability, reliance on manual calibration, and cumbersome operation of existing liquid nitrogen volume detection schemes, thereby improving the accuracy and operational stability of liquid nitrogen volume detection.

[0007] To achieve the above objectives, the present invention provides the following technical solution: A non-contact, high-precision liquid nitrogen generation device liquid volume measurement method includes the following steps: The detection device uses multiple sets of sensor units to collect weight, temperature and vibration data, and performs multi-dimensional automatic calibration on the collected data to obtain the detection reference value, compensation coefficient and correlation model. The real-time weight data, real-time ambient temperature data, and real-time vibration signal of the liquid nitrogen storage tank during the operation of the liquid nitrogen generation device are collected synchronously by the multiple sets of sensing units. The collected real-time data is preprocessed to obtain preprocessed data. Based on the baseline value, compensation coefficient, and correlation model obtained from the multi-dimensional automatic calibration, a multi-source data fusion calculation operation is performed on the preprocessed data to obtain the liquid nitrogen level height. Dynamic calibration is performed on the compensation coefficient and associated model parameters according to a set cycle. When the data deviation of the sensing unit exceeds a preset threshold, a redundancy switching operation is initiated to ensure continuous operation of the detection process. The liquid nitrogen level and related detection data are displayed. When the detection data exceeds a preset threshold, an alarm is triggered, and a linkage signal is output to the liquid nitrogen generation device control system.

[0008] In one possible implementation, during the execution of the multi-dimensional automatic calibration operation... When the liquid nitrogen storage tank is empty and the liquid nitrogen generating device is not running, the multiple sets of sensing units collect local weight data of the empty tank and temperature data at the corresponding locations. The local weight data of the empty can is corrected using a temperature correction formula to eliminate the interference of ambient temperature on the weight data, and the temperature-corrected weight data of the empty can is obtained. The calculation operation is performed based on the assumption that the sensing unit is uniformly stressed under ideal horizontal conditions to obtain the ideal local weight. The difference between the temperature-corrected empty can weight data and the ideal local weight is used to perform a calculation operation to obtain the support surface levelness deviation coefficient. The levelness deviation is compensated by a weighted algorithm to generate a total reference value for the empty tank and correct the initial weighting coefficients, thus completing the calibration of the empty tank reference and levelness deviation.

[0009] In one possible implementation, during the execution of the multi-dimensional automatic calibration operation... A known mass of liquid nitrogen is injected into the liquid nitrogen storage tank, and multiple sets of real-time weight data are collected through the multiple sets of sensing units; A temperature correction operation is performed on the collected real-time weight data to obtain temperature-corrected weight data; Based on the aforementioned levelness deviation coefficient, the weighting coefficients are corrected using the least squares method to minimize data deviation and complete the sensor unit consistency calibration. Start the liquid nitrogen generation device and collect vibration signals within a set time period through the multiple sets of sensing units; The vibration signal is subjected to feature extraction to obtain vibration feature parameters. A grouped vibration baseline database is established based on the vibration feature parameters. The directional vibration compensation coefficient of each group of sensing units is calibrated to complete the vibration baseline calibration of the air compressor. To maintain the normal operation of the liquid nitrogen generation device, vibration signals of the workshop under normal conditions for a set duration are collected through the multiple sets of sensing units. A feature extraction operation is performed on the vibration signal to obtain random vibration feature parameters. Based on the random vibration feature parameters, a random vibration identification threshold is set and a random vibration compensation model is established to complete the random vibration baseline calibration.

[0010] In one possible implementation, during the execution of the multi-dimensional automatic calibration operation... Based on the collected temperature data, the average temperature is calculated, and the correlation between liquid nitrogen density and temperature is modeled using an interpolation algorithm to establish a correlation model between liquid nitrogen density and temperature, thereby completing the dynamic calibration of density. Keep the liquid nitrogen generator in a stopped state, perform a static monitoring operation for a set duration, and record liquid nitrogen volume data and ambient temperature data at different times; A calculation is performed based on the recorded data to obtain the baseline evaporation rate; The correlation between evaporation rate and ambient temperature is modeled using a fitting algorithm to establish a correlation model between evaporation rate and ambient temperature, and to complete the baseline calibration of evaporation rate.

[0011] In one possible implementation, during the preprocessing operation on the acquired real-time data, The real-time data is filtered using an outlier removal algorithm to remove extreme data that exceeds a set range. The extreme data that were removed were replaced using data from adjacent time points; The moving average filtering algorithm is used to average the data over multiple consecutive time periods to smooth out high-frequency noise and obtain preprocessed data.

[0012] In one possible implementation, during the multi-source data fusion calculation operation on the preprocessed data... Using the temperature compensation coefficient, directional vibration compensation coefficient, and random vibration compensation coefficient, a correction operation is performed on the preprocessed real-time weight data to compensate for the interference caused by temperature drift, directional vibration, and random vibration, thereby obtaining an effective local weight. The vibration signal of each group of sensing units is monitored in real time using the sliding window method, and the root mean square value and instantaneous peak value of vibration within the window are calculated. If the vibration parameters exceed the random vibration identification threshold, it is determined that there is irregular random vibration, triggering dynamic compensation operation and updating the random vibration compensation coefficient. If the vibration parameters do not exceed the threshold, the baseline value of the random vibration compensation coefficient remains unchanged.

[0013] In one possible implementation, after obtaining the effective local weight, Perform pairwise absolute value difference calculation on multiple groups of effective local weights, compare the calculation results with a preset consistency threshold, and determine whether there is a slight deviation between the two groups of sensing units. If it is determined that there are two sets of sensor units with slight deviations, the corresponding weight coefficients are corrected by the deviation sharing algorithm, while the weight coefficients of the remaining sensor units remain unchanged. The effective local weights are weighted and fused using the corrected weighting coefficients to obtain the total weight of the liquid nitrogen and the empty tank.

[0014] In one possible implementation, after obtaining the total weight of the liquid nitrogen and the empty tank... Based on the real-time average ambient temperature, the real-time liquid nitrogen density is obtained by performing calculations using the density correlation model. By combining the total weight with the total baseline value of the empty tank, a calculation is performed to obtain the uncorrected volume of liquid nitrogen from evaporation. The real-time evaporation rate is obtained by performing calculations based on the liquid nitrogen volume data from continuous detection cycles and the evaporation rate correlation model. After deducting the liquid volume loss due to evaporation, the corrected liquid nitrogen volume is obtained.

[0015] In one possible implementation, after obtaining the corrected liquid nitrogen volume, Based on the pre-defined volume height correlation function of the liquid nitrogen storage tank, the liquid nitrogen level height is obtained by performing a solution operation through an iterative algorithm. During the iteration process, the difference in liquid level height between two adjacent iterations is continuously compared. When the difference meets the preset termination condition, the iteration stops and the final liquid nitrogen level height is output.

[0016] In one possible implementation, during the execution of the dynamic calibration operation... The calibration process is automatically triggered at set intervals, and the baseline values ​​of weighting coefficient, temperature compensation coefficient, directional vibration compensation coefficient, random vibration compensation coefficient and evaporation rate correlation model parameters are re-corrected based on historical test data. When the data deviation of a certain group of sensor units exceeds the set threshold, the faulty sensor unit is automatically blocked, the weights are redistributed based on the data of the remaining normally functioning sensor units, and the liquid volume calculation continues. When the detected data exceeds the preset threshold, an audible and visual alarm is triggered, and a linkage signal is output to the liquid nitrogen generator control system through the industrial communication interface to automatically adjust the generator's operating power or start / stop status.

[0017] Compared with the prior art, the advantages of this invention are as follows: This invention adopts a structure of multiple sets of distributed weighing sensors with grouped auxiliary sensing units, and with the automatic calibration mechanism of multi-sensor data, it replaces the operation method of relying on manual calibration of the support surface level in the prior art. It eliminates the need for manual calibration using a level, and simplifies the process of device deployment and maintenance.

[0018] This invention employs a completely non-contact structure. During the detection process, there is no need to immerse the detection component inside the liquid nitrogen storage tank, nor is there any direct contact between the detection component and the liquid nitrogen. This avoids the corrosion and aging problems that can occur when the detection component is exposed to a low-temperature environment for extended periods. Furthermore, it does not damage the original sealing structure of the liquid nitrogen storage tank, ensuring the safety of the liquid nitrogen generation device during operation. Existing contact-type liquid level detection solutions not only require installation inside the tank but also rely on manual visual readings, making real-time monitoring impossible. The structure of this invention enables continuous real-time monitoring of the liquid nitrogen volume, making it suitable for applications requiring continuous operation of liquid nitrogen generation devices.

[0019] This invention utilizes a dedicated graded vibration damping support frame, combined with a quadruple vibration-resistant structure incorporating a grouped directional vibration compensation algorithm, a random vibration dynamic compensation mechanism, and multi-sensor fusion, to directionally attenuate the transmission of vibrations generated by the operation of the air compressor accompanying the liquid nitrogen generator to the weighing sensor. Simultaneously, it can effectively identify irregular random vibrations caused by the operation of other equipment and personnel movement within the workshop, adapting to multi-source vibration interference scenarios in complex workshop conditions. This solves the problem of existing technologies lacking vibration compensation mechanisms specifically for the operation of liquid nitrogen generators, leading to excessive detection errors due to vibration interference.

[0020] This invention employs a parallel redundant structure of multiple sets of weighing sensors, coupled with a periodic dynamic calibration process. When a single set of sensors fails, the system can automatically switch to redundant operation mode, disabling the faulty sensor and redistributing weights based on the remaining normal sensors to continue detection calculations. This prevents interruption of the detection function, meeting the long-term, continuous, and highly reliable operation requirements of liquid nitrogen generation devices. Simultaneously, a new multi-sensor data cross-validation mechanism is added. For scenarios where two sets of sensors simultaneously exhibit slight deviations, the weight coefficients are corrected through data consistency verification and deviation amortization algorithms. This avoids the accumulation of slight sensor deviations leading to decreased detection accuracy, further improving the stability of detection accuracy during long-term operation of the device.

[0021] This invention establishes a real-time evaporation rate correction mechanism. It calculates the real-time evaporation rate of liquid nitrogen by combining ambient temperature, tank insulation status, and liquid volume changes within a continuous monitoring period. This rate is then incorporated into the calculation of liquid nitrogen density and remaining volume, effectively offsetting the liquid volume calculation error caused by natural evaporation and improving the detection stability of the device under long-term static conditions or fluctuating ambient temperatures. Simultaneously, the device can be linked with the liquid nitrogen generation device control system, automatically adjusting the operating power or start / stop status of the generation device based on the liquid nitrogen level detection results, achieving dynamic balance control of liquid nitrogen generation and storage. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of the liquid nitrogen generation device in an embodiment of the present invention; Figure 2 This is a side view of the liquid nitrogen generating device in an embodiment of the present invention; Figure 3 This is a flowchart illustrating the measurement method in an embodiment of the present invention.

[0024] In the attached diagram: 1. Multi-sensor platform; 2. Liquid nitrogen storage tank; 3. Vibration damping support frame; 4. Embedded module; 5. Human-machine interaction module; 6. Refrigeration unit; 7. Air inlet; 8. Safety valve; 9. Drain outlet; 10. Tank fixing feet; 11. Differential bus; 12. Temperature sensor; 13. Vibration sensor; 14. Signal processing module; 15. Resistance strain gauge load cell; 16. Rubber shock-absorbing pad; 17. Spring damper. Detailed Implementation

[0025] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present application.

[0026] [[IDID=3]]Example: It should be noted that the terms "including" and "having" in the embodiments of the present invention and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily limit to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0027] Figure 1 and Figure 2 A non-contact high-precision liquid volume detection device for a liquid nitrogen generation device to which the measurement method of this embodiment can be applied is shown, which includes: a shock-absorbing support frame 3, a multi-sensor platform 1, a liquid nitrogen storage tank 2, an embedded module 4, and a human-computer interaction module 5. The multi-sensor platform 1 is fixedly arranged on the upper part of the shock-absorbing support frame 3; the liquid nitrogen storage tank 2 is fixedly arranged on the upper surface of the multi-sensor platform 1. The signal output end of the multi-sensor platform 1 is signal-connected to the signal input end of the embedded module 4, and the signal output end of the embedded module 4 is signal-connected to the signal input end of the human-computer interaction module 5. The embedded module 4 is used to receive the acquisition data of the multi-sensor platform 1 and output the processing result to the human-computer interaction module 5.

[0028] Specifically, the shock-absorbing support frame 3 can be a customized rectangular box structure with hierarchical shock-absorbing ability; the multi-sensor platform 1 can be an integrated sensing platform with a distributed layout for collecting multi-dimensional data of weight, temperature, and vibration; the liquid nitrogen storage tank 2 can be a sealed low-temperature storage container supporting the liquid nitrogen generation device; the embedded module 4 can be an intelligent operation module equipped with a customized real-time operating system; the human-computer interaction module 5 can be an industrial-level interaction terminal with data display and instruction input functions.

[0029] In specific implementation, the detection device is deployed by integrating the shock-absorbing support frame 3, multi-sensor platform 1, liquid nitrogen storage tank 2, embedded module 4, and human-machine interaction module 5, establishing communication connections between the modules. The deployed multi-sensor platform 1 collects multiple sets of weight, temperature, and vibration data. The embedded module 4 performs multi-dimensional automatic calibration on the collected data to obtain the detection benchmark value, compensation coefficient, and correlation model. Simultaneously, the multi-sensor platform 1 collects real-time weight data, real-time ambient temperature data, and real-time vibration signals from the liquid nitrogen storage tank 2 during the operation of the liquid nitrogen generation device, and performs pre-processing on the collected real-time data. The system performs processing operations to obtain preprocessed data; based on the reference value, compensation coefficient, and correlation model obtained by automatic calibration, it performs multi-source data fusion calculation on the preprocessed data to obtain the liquid nitrogen level height; it performs dynamic calibration operations on the compensation coefficient and correlation model parameters according to a set cycle, and initiates a redundancy switching operation when the sensor data deviation exceeds a preset threshold to ensure continuous operation of the detection process; it displays the liquid nitrogen level height and related detection data through the human-machine interaction module 5, triggers an alarm operation when the detection data exceeds a preset threshold, and outputs a linkage signal to the liquid nitrogen generation device control system (which can be an embedded module 4).

[0030] As an optional implementation, in some embodiments, the shock-absorbing support frame 3 has a built-in spring damper 17, a rubber damping pad 16 and a rigid support surface. The spring damper 17 is disposed at the bottom layer of the shock-absorbing support frame 3, the rubber damping pad 16 is disposed above the spring damper 17, the rigid support surface is disposed above the rubber damping pad 16, and the upper surface of the rigid support surface is fixedly connected to the multi-sensor platform 1.

[0031] Specifically, the spring damper 17 can be a vibration damping component adapted to the vibration frequency of the air compressor; the rubber damping pad 16 can be a damping pad layer made of high-damping nitrile rubber; and the rigid support surface can be a flat support surface made of thickened carbon steel plate.

[0032] As an optional implementation, in some embodiments, the multi-sensor platform 1 has multiple sets of weighing sensors and multiple sets of auxiliary sensing units built in. The multiple sets of weighing sensors are evenly distributed inside the multi-sensor platform 1, and each set of auxiliary sensing units is installed on the housing of a set of weighing sensors. The signal output terminal of each set of auxiliary sensing units is connected to the signal acquisition terminal of the multi-sensor platform 1.

[0033] Specifically, the load cell can be an industrial-grade resistance strain gauge load cell 15; multiple load cells can be four load cells, evenly distributed in a rectangle at the four corners of the bottom of the storage tank; the auxiliary sensing unit can be a sensing component that integrates temperature acquisition and vibration acquisition functions, and synchronously acquires environmental and operational data with the corresponding load cell.

[0034] As an optional implementation, in some embodiments, the auxiliary sensing unit includes a temperature sensor 12 and a vibration sensor 13. The temperature sensor 12 is fixed next to the housing of the weighing sensor, and the vibration sensor 13 is fixed to the side of the housing of the weighing sensor. The signal output terminals of the temperature sensor 12 and the vibration sensor 13 are both connected to the signal acquisition terminal of the multi-sensor platform 1 for transmitting temperature data and vibration data of the corresponding location to the multi-sensor platform 1.

[0035] Specifically, the temperature sensor 12 can be an industrial-grade AA-grade PT1000 platinum resistance temperature sensor 12; the vibration sensor 13 can be a piezoelectric accelerometer, with the sensor's sensitive axis set perpendicular to the multi-sensor platform 1.

[0036] As an optional implementation, in some embodiments, the multi-sensor platform 1 also incorporates a signal processing module 14. The signal processing module 14 incorporates a multi-channel signal conditioning circuit. The signal input terminal of the signal processing module 14 is connected to the signal output terminal of each group of weighing sensors, temperature sensors 12, and vibration sensors 13. The signal output terminal of the signal processing module 14 is connected to the signal input terminal of the embedded module 4 via a differential bus 11, for transmitting the pre-processed acquired data to the embedded module 4.

[0037] Specifically, the multi-channel signal conditioning circuit can be a circuit that synchronously amplifies and filters multiple signals such as weight, temperature, and vibration; the differential bus 11 can be a differential data transmission bus with CRC check.

[0038] As an optional implementation, in some embodiments, the embedded module 4 adopts a dual-core industrial-grade MCU plus FPGA architecture. The signal input terminal of the embedded module 4 is connected to the signal output terminal of the multi-sensor platform 1, and the control output terminal of the embedded module 4 is connected to the control input terminal of the human-machine interaction module 5, for transmitting detection results and control commands to the human-machine interaction module 5.

[0039] Specifically, the dual-core industrial-grade MCU can be an industrial-grade microcontroller with an ARM Cortex-M7 core; the FPGA can be a field-programmable gate array used for parallel processing of multiple sensor data; and the customized real-time operating system can be an RTOS (Real-Time Operating System) installed inside the embedded module 4.

[0040] As an optional implementation, in some embodiments, the human-machine interaction module 5 includes an industrial-grade touch display terminal. The signal input terminal of the touch display terminal is connected to the signal output terminal of the embedded module 4, and the instruction output terminal of the touch display terminal is connected to the instruction input terminal of the embedded module 4, for displaying detection data and transmitting operation instructions to the embedded module 4.

[0041] Specifically, an industrial-grade touch display terminal can be an industrial-grade capacitive touch display screen.

[0042] As an optional implementation, in some embodiments, the upper part of the liquid nitrogen storage tank 2 is provided with a refrigeration unit 6 installation port, an air inlet 7, a safety valve 8 interface and a drain port 9. The bottom of the liquid nitrogen storage tank 2 is fixedly connected to the upper surface of the multi-sensor platform 1 through a tank fixing bracket 10. The tank body of the liquid nitrogen storage tank 2 is connected to the corresponding pipeline of the liquid nitrogen generating device.

[0043] Specifically, the refrigeration unit 6 installation port can be a standard interface on the upper part of the liquid nitrogen storage tank 2 for installing the refrigeration unit 6; the air inlet 7 can be a pipeline interface for inputting protective gas; the safety valve 8 interface can be a standard interface for installing a safety relief valve; the drain port 9 can be a pipeline interface for outputting liquid nitrogen; and the tank fixing bracket 10 can be four metal fixing brackets symmetrically distributed at the bottom of the tank.

[0044] As an optional implementation, in some embodiments, the human-computer interaction module 5 further includes an audible and visual alarm unit. The control input terminal of the audible and visual alarm unit is connected to the control output terminal of the touch display terminal, and is used to receive alarm commands from the touch display terminal and issue audible and visual alarm signals.

[0045] Specifically, the audible and visual alarm unit can be an alarm component that integrates a buzzer and a three-color warning light. The buzzer can output buzzing signals of different frequencies, and the three-color warning light can switch light colors according to different alarm levels.

[0046] See Figure 3 The non-contact, high-precision liquid nitrogen generation device liquid volume measurement method of this embodiment may include the following steps: Step 100: Collect weight, temperature and vibration data using multiple sets of sensing units of the detection device, and perform multi-dimensional automatic calibration on the collected data to obtain the detection reference value, compensation coefficient and correlation model.

[0047] Specifically, the multiple sensing units can be a combination of four industrial-grade resistance strain gauge load cells, combined with grouped temperature sensors and vibration sensors; the multi-dimensional automatic calibration operation can be a combination of empty tank baseline calibration, sensing unit consistency calibration, vibration baseline establishment, density model calibration, and evaporation rate baseline calibration; the reference value can be the total reference value of the empty tank or the ideal local weight; the compensation coefficient can be the temperature compensation coefficient, the levelness deviation coefficient, the directional vibration compensation coefficient, or the random vibration compensation coefficient; the correlation model can be the correlation model between liquid nitrogen density and temperature, or the correlation model between evaporation rate and ambient temperature.

[0048] The process of performing the multi-dimensional automatic calibration includes: collecting local weight data of the empty liquid nitrogen storage tank through the multiple sets of sensing units when the liquid nitrogen storage tank is empty and the liquid nitrogen generation device is not running. and temperature data at the corresponding locations The temperature correction formula is used to correct the local weight data of the empty can, eliminating the interference of ambient temperature on the weight data, and obtaining the temperature-corrected weight data of the empty can. For example... In the formula, For the first Initial temperature compensation coefficient for the sensor group. For the first Temperature data at corresponding locations collected by the group of sensing units. For standard reference temperature, For the first The empty can weight data after temperature correction for the group of sensing units; calculations are performed based on the assumption of uniform force on the sensing units under ideal horizontal conditions to obtain the ideal local weight. In the formula, This is the total reference value for the calibrated empty tank. The ideal local weight of a single sensing unit under ideal horizontal conditions is given. A calculation is performed using the difference between the temperature-corrected empty can weight data and the ideal local weight to obtain the support surface levelness deviation coefficient, for example... In the formula, For the first The levelness deviation coefficient of the sensor group positions; compensation operation is performed on the levelness deviation using a weighted algorithm, for example... Generate the total baseline value for empty tanks and correct the initial weighting coefficients. Complete the calibration of the empty tank's reference and levelness deviation, where, For the first The group of sensing units includes initial weighting coefficients for levelness compensation. For the first The levelness deviation coefficient of the support surface at the position of the group of sensing units.

[0049] Specifically, the temperature correction formula can be a weight correction calculation formula based on the temperature compensation coefficient; the standard reference temperature can be 25℃; the assumption of uniform force on the sensing unit under ideal horizontal conditions can be the calculation premise that the weight borne by multiple sensing units is completely consistent under ideal horizontal installation conditions; the ideal local weight can be the weight value that a single sensing unit should bear under ideal horizontal conditions; the horizontality deviation coefficient can be the quantified value of the horizontal deviation of the support surface of a single sensing unit position; and the weighting algorithm can be a calculation method based on the horizontality deviation coefficient to correct the initial weight coefficient of the sensing unit.

[0050] Furthermore, the process of performing the multi-dimensional automatic calibration operation includes: A known mass of liquid nitrogen is injected into the liquid nitrogen storage tank, and multiple sets of real-time weight data are collected through the multiple sets of sensing units; a temperature correction operation is performed on the collected real-time weight data to obtain temperature-corrected weight data. Combined with the aforementioned levelness deviation coefficient The weighting coefficients are corrected using the least squares method, for example... To minimize data deviation and complete sensor unit consistency calibration, in the formula, For the first The weighting coefficients of the group sensing units after correction For the first The liquid ammonia weight data after temperature correction for the group of sensing units. This represents the standard local weight of liquid nitrogen that a single sensing unit should withstand under ideal conditions. The standard mass of liquid nitrogen to be injected into the liquid nitrogen storage tank.

[0051] Specifically, the known mass of liquid nitrogen can be standard mass of liquid nitrogen; the least squares method can be a numerical fitting method that minimizes the deviation of multiple sets of data.

[0052] The liquid nitrogen generation device is started, and vibration signals within a set time period are collected through the multiple sets of sensing units; feature extraction is performed on the vibration signals to obtain vibration feature parameters; a grouped vibration baseline database is established based on the vibration feature parameters; the directional vibration compensation coefficient of each group of sensing units is calibrated to complete the air compressor vibration baseline calibration.

[0053] Specifically, the vibration characteristic parameters can be vibration frequency, amplitude, and root mean square value; the set duration can be 30 seconds; the grouped vibration baseline database can be a database that stores the vibration characteristic parameters and compensation coefficients corresponding to each group of sensing units; the directional vibration compensation coefficient can be a correction coefficient for the directional vibration of the air compressor of the liquid nitrogen generation device.

[0054] To maintain the normal operation of the liquid nitrogen generation device, vibration signals from the workshop under normal conditions for a set duration are collected through the multiple sets of sensing units. Feature extraction is performed on the vibration signals to obtain random vibration characteristic parameters. Based on these random vibration characteristic parameters, a random vibration identification threshold is set, and a random vibration compensation model is established. Complete the random vibration baseline calibration, where, This is the baseline value for the random vibration compensation coefficient. The root mean square value, This represents the root mean square value of the directional vibration baseline.

[0055] Specifically, the random vibration characteristic parameters can be the instantaneous peak value, root mean square value, and frequency bandwidth; the normal workshop setting duration can be 30 minutes; the random vibration identification threshold can be an abnormal vibration judgment value set based on the normal random vibration characteristics; and the random vibration compensation model can be a dynamic correction calculation model for irregular random vibrations in the workshop.

[0056] Furthermore, the process of performing the multi-dimensional automatic calibration operation includes: Based on the collected temperature data, the average temperature is calculated. An interpolation algorithm is then used to model the relationship between liquid nitrogen density and temperature, establishing a correlation model between the two. To complete the density dynamic calibration, in the formula, This represents the dynamic density of liquid nitrogen at the real-time average ambient temperature. Standard temperature The density of liquid nitrogen below This is a primary temperature coefficient. It is a second-order temperature coefficient. This represents the real-time average temperature.

[0057] Specifically, the average temperature can be the arithmetic mean of the ambient temperature data collected by the four sets of sensing units; the interpolation algorithm can be a quadratic interpolation algorithm; the correlation model between liquid nitrogen density and temperature can be a dynamic calculation model of liquid nitrogen density based on temperature correction; and the liquid nitrogen density at standard temperature can be 808.5 g / cm³.

[0058] Keep the liquid nitrogen generator shut down and perform a static monitoring operation for a set period of time to record liquid nitrogen volume data and ambient temperature data at different times; based on the recorded data, perform calculations to obtain the baseline evaporation rate. In the formula, For temperature The baseline evaporation rate of liquid nitrogen; This is the liquid nitrogen volume data at the initial moment of settling. To let stand Liquid nitrogen volume data at any given time. The static monitoring period; a modeling operation is performed on the relationship between evaporation rate and ambient temperature using a fitting algorithm to establish a correlation model between evaporation rate and ambient temperature, for example... In the formula, The evaporation rate of liquid nitrogen is the baseline at ambient temperature T. This is the temperature coefficient of the evaporation rate. Standard temperature The baseline evaporation rate is determined by setting the base evaporation rate, where T is the ambient temperature, and the correction factor for the tank insulation condition is calibrated to complete the baseline calibration of the evaporation rate.

[0059] Specifically, the set duration can be 2 hours; the baseline evaporation rate can be the natural evaporation volume of liquid nitrogen per unit time at the corresponding ambient temperature; the fitting algorithm can be a linear fitting algorithm; and the correlation model between the evaporation rate and the ambient temperature can be a dynamic calculation model of the evaporation rate based on ambient temperature correction.

[0060] Step 200: The real-time weight data, real-time ambient temperature data and real-time vibration signal of the liquid nitrogen storage tank during the operation of the liquid nitrogen generation device are collected synchronously through the multiple sets of sensing units. The collected real-time data is preprocessed to obtain preprocessed data.

[0061] Specifically, the preprocessing operation can be a combination of outlier removal and moving average filtering.

[0062] The preprocessing operation performed on the collected real-time data includes: The real-time data is filtered using an outlier removal algorithm to remove extreme data that exceeds a set range; the removed extreme data is replaced using data from adjacent time points; and the data from multiple consecutive time points is averaged using a moving average filtering algorithm to smooth high-frequency noise, resulting in preprocessed data.

[0063] Specifically, the outlier removal algorithm can be a 3σ outlier removal algorithm; the set range can be a range of values ​​that is 3 times the standard deviation of the data mean; extreme data can be abnormal collected values ​​that exceed the set range; adjacent time data can be normal collected data from consecutive time points before and after the extreme data; the moving average filtering algorithm can be a smoothing filtering algorithm that takes the arithmetic mean of the effective data from multiple consecutive time points; high-frequency noise can be high-frequency data fluctuations caused by electromagnetic interference or slight vibrations; multiple consecutive time points can be 5 to 10 consecutive collection time points.

[0064] Step 300: Based on the benchmark value, compensation coefficient and correlation model obtained by the multi-dimensional automatic calibration, perform multi-source data fusion calculation on the preprocessed data to obtain the liquid nitrogen level height.

[0065] Specifically, multi-source data fusion calculation operations can be a combination of grouped weight correction, multi-sensor data cross-validation, weighted fusion, real-time evaporation rate calculation, dynamic density and volume calculation, and liquid level iterative solution.

[0066] The process of performing multi-source data fusion calculation on the preprocessed data includes: using the temperature compensation coefficient, directional vibration compensation coefficient, and random vibration compensation coefficient to perform a correction operation on the preprocessed real-time weight data to compensate for interference caused by temperature drift, directional vibration, and random vibration, thereby obtaining an effective local weight, for example... In the formula, This is the real-time weight data after preprocessing for the i-th group of sensing units. The temperature compensation coefficient for the i-th group of sensing units; The real-time ambient temperature data is preprocessed by the i-th group of sensing units; Let be the directional vibration compensation coefficient of the i-th group of sensing units. The root mean square value of the directional vibration after preprocessing of the i-th group of sensing units. Let be the root mean square value of the directional vibration baseline of the i-th group of sensing units. The directional vibration amplitude after preprocessing of the i-th group of sensing units. Let be the baseline amplitude of the directional vibration of the i-th group of sensing units. is the root mean square value of random vibration after preprocessing of the i-th group of sensing units; The instantaneous peak value of random vibration after preprocessing of the i-th group of sensing units. The instantaneous peak value of the random vibration baseline. Let be the random vibration compensation coefficient of the i-th group of sensing units. The effective local weight of the i-th group of sensing units after multi-dimensional correction is defined. A sliding window method is used to perform real-time monitoring of the vibration signal of each group of sensing units, calculating the root mean square value and instantaneous peak value of the vibration within the window. If the vibration parameters exceed the random vibration identification threshold, it is determined that there is irregular random vibration, triggering a dynamic compensation operation and updating the random vibration compensation coefficients. For example... In the formula, The reference value for the random vibration compensation coefficient of the i-th group of sensing units is given. The threshold value is the root mean square value of random vibration; if the vibration parameter does not exceed the threshold value, the baseline value of the random vibration compensation coefficient remains unchanged.

[0067] Specifically, the sliding window method can be a data processing method with a fixed-length time window; the root mean square value of vibration can be the square root of the average of the squares of the instantaneous values ​​of the vibration signal within the window, reflecting the vibration intensity; the instantaneous peak value can be the maximum instantaneous amplitude of the vibration signal within the window; the random vibration identification threshold can be the vibration root mean square value threshold set during the calibration phase; the dynamic compensation operation can be the calculation process of correcting the random vibration compensation coefficient based on real-time vibration parameters; and the reference value of the random vibration compensation coefficient can be the initial compensation coefficient calibrated during the calibration phase.

[0068] Further, after obtaining the effective local weight, the process includes: performing pairwise absolute difference calculations on multiple sets of the effective local weights; comparing the calculation results with a preset consistency threshold to determine if there are slight deviations between two sets of sensing units; if slight deviations are found between two sets of sensing units, correcting the corresponding weight coefficients using a deviation amortization algorithm, while keeping the weight coefficients of the remaining sensing units unchanged; and performing a weighted fusion operation on the effective local weights using the corrected weight coefficients to obtain the total weight of the liquid nitrogen and the empty tank, for example... In the formula, The total weight after weighted merging. The weight coefficients of the i-th group of sensing units are corrected by the bias allocation algorithm. Let be the effective local weight of the i-th group of sensing units.

[0069] Specifically, the pairwise absolute difference calculation operation can be to combine all groups of effective local weights in pairs and calculate the absolute difference of each group; the preset consistency threshold can be the critical value for determining the consistency of sensor unit data; slight deviation can be the state where the absolute difference between two groups of sensor unit data exceeds the consistency threshold but does not reach the fault threshold; the deviation amortization algorithm can be a correction algorithm that proportionally distributes the deviation of two groups of data to the corresponding weight coefficients; the weighted fusion operation can be a calculation process of weighted summation of effective local weights based on the weight coefficients of each group.

[0070] Further, after obtaining the total weight of the liquid nitrogen and the empty tank, the process includes: performing a calculation operation based on the real-time average ambient temperature using the density correlation model to obtain the real-time liquid nitrogen density; performing a calculation operation combining the total weight and the total baseline value of the empty tank to obtain the uncorrected volume of liquid nitrogen due to evaporation; performing a calculation operation based on the liquid nitrogen volume data of the continuous detection period and the evaporation rate correlation model to obtain the real-time evaporation rate; and deducting the liquid loss caused by evaporation to obtain the corrected volume of liquid nitrogen.

[0071] Specifically, the real-time average ambient temperature can be the arithmetic mean of the data collected by the four temperature sensors within the current detection period. The density correlation model can be a functional model of liquid nitrogen density versus temperature established during the calibration phase, for example... In the formula, Real-time liquid nitrogen density with evaporation rate correction. The real-time evaporation rate of liquid nitrogen; the uncorrected volume of evaporated liquid nitrogen can be the volume of liquid nitrogen calculated solely based on the weight difference and real-time density, for example... In the formula, This represents the remaining volume of liquid nitrogen after evaporation correction. The total baseline value for the empty tank; the continuous detection cycle can be the time interval between two consecutive complete detections; the real-time evaporation rate can be the amount of liquid nitrogen evaporating per unit time calculated based on continuous volume data, for example... In the formula, This represents the real-time evaporation rate of liquid nitrogen. This is a correction factor for the insulation condition of the tank. The baseline evaporation rate is calculated based on the real-time average ambient temperature. This is the liquid nitrogen volume data from the previous testing cycle. This is the uncorrected volume data of evaporated liquid nitrogen for the current detection cycle. The detection cycle duration; the corrected liquid nitrogen volume can be the actual liquid nitrogen volume after deducting evaporation losses.

[0072] Furthermore, after obtaining the corrected liquid nitrogen volume, the liquid nitrogen level height is obtained by performing a solution operation through an iterative algorithm based on the preset volume height correlation function of the liquid nitrogen storage tank. During the iteration process, the difference in liquid level height between two adjacent iterations is continuously compared. When the difference meets the preset termination condition, the iteration stops and the final liquid nitrogen level height is output.

[0073] Specifically, the volume-to-height correlation function can be a function characterizing the relationship between the internal volume and liquid level of the liquid nitrogen storage tank, which is preset by the tank's structural parameters, for example... In the formula, Here are the coefficients of the correlation function polynomial; H is the liquid nitrogen level height. To determine the storage tank volume at a given height, the iterative algorithm could be Newton's method, for example... Numerical solutions for nonlinear volume height correlation functions, such as... The preset termination condition can be a threshold value representing the difference in liquid level height between two adjacent iterations, for example... The final liquid nitrogen level can be the value obtained after iterative convergence.

[0074] Step 400: Perform dynamic calibration on the compensation coefficient and associated model parameters according to the set cycle. When the data deviation of the sensing unit exceeds the preset threshold, start the redundancy switching operation to ensure continuous operation of the detection process.

[0075] Specifically, dynamic calibration can be a parameter correction operation that is automatically triggered every 8 hours; redundancy switching can be a continuous operation that shields the faulty sensor unit and redistributes weights based on the remaining sensor units when a single sensor unit fails.

[0076] The dynamic calibration process includes: automatically triggering the calibration process at set intervals, and re-correcting the reference values ​​of weighting coefficients, temperature compensation coefficients, directional vibration compensation coefficients, random vibration compensation coefficients, and evaporation rate correlation model parameters based on historical detection data; automatically disabling faulty sensing units when the data deviation of a certain group of sensing units exceeds a set threshold, redistributing weights based on the data of the remaining normally functioning sensing units, and continuing to perform liquid volume calculation; triggering an audible and visual alarm when the detection data exceeds a preset threshold, and outputting a linkage signal to the liquid nitrogen generation device control system through the industrial communication interface to automatically adjust the operating power or start / stop status of the generation device.

[0077] Specifically, the set duration can be the time interval for dynamic calibration; historical detection data can be all valid detection data and environmental parameter data stored within the set duration; the set threshold can be the critical deviation value for determining sensor unit failure; the industrial communication interface can be a Modbus RTU communication interface; the linkage signal can be a switch or analog signal used to adjust the operating status of the liquid nitrogen generator; the operating power adjustment can be to increase or decrease the output power of the generator proportionally; and the start / stop status can be to control the generator to turn on or off.

[0078] Step 500: Display the liquid nitrogen level and related detection data. When the detection data exceeds a preset threshold, trigger an alarm and output a linkage signal to the liquid nitrogen generation device control system.

[0079] Specifically, the linkage signal can be an electrical signal output to the control system of the liquid nitrogen generation device to regulate the operating power and control the start and stop of the equipment.

[0080] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0081] The above embodiments are merely illustrative of the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be construed as limiting the scope of protection of the present invention. All equivalent changes or modifications made based on the essence of the content of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A non-contact, high-precision liquid nitrogen generation device liquid volume measurement method, characterized in that, Including the following steps: The detection device uses multiple sets of sensor units to collect weight, temperature and vibration data, and performs multi-dimensional automatic calibration on the collected data to obtain the detection reference value, compensation coefficient and correlation model. The real-time weight data, real-time ambient temperature data, and real-time vibration signal of the liquid nitrogen storage tank during the operation of the liquid nitrogen generation device are collected synchronously by the multiple sets of sensing units. The collected real-time data is preprocessed to obtain preprocessed data. Based on the baseline value, compensation coefficient, and correlation model obtained from the multi-dimensional automatic calibration, a multi-source data fusion calculation operation is performed on the preprocessed data to obtain the liquid nitrogen level height. Dynamic calibration is performed on the compensation coefficient and associated model parameters according to a set cycle. When the data deviation of the sensing unit exceeds a preset threshold, a redundancy switching operation is initiated to ensure continuous operation of the detection process. The liquid nitrogen level and related detection data are displayed. When the detection data exceeds a preset threshold, an alarm is triggered, and a linkage signal is output to the liquid nitrogen generation device control system.

2. The method for measuring the liquid volume of a non-contact high-precision liquid nitrogen generator according to claim 1, characterized in that, During the execution of the aforementioned multi-dimensional automatic calibration operation, When the liquid nitrogen storage tank is empty and the liquid nitrogen generating device is not running, the multiple sets of sensing units collect local weight data of the empty tank and temperature data at the corresponding locations. The local weight data of the empty can is corrected using a temperature correction formula to eliminate the interference of ambient temperature on the weight data, and the temperature-corrected weight data of the empty can is obtained. The calculation operation is performed based on the assumption that the sensing unit is uniformly stressed under ideal horizontal conditions to obtain the ideal local weight. The difference between the temperature-corrected empty can weight data and the ideal local weight is used to perform a calculation operation to obtain the support surface levelness deviation coefficient. The levelness deviation is compensated by a weighted algorithm to generate a total reference value for the empty tank and correct the initial weighting coefficients, thus completing the calibration of the empty tank reference and levelness deviation.

3. The method for measuring the liquid volume of a non-contact high-precision liquid nitrogen generator according to claim 2, characterized in that, During the execution of the aforementioned multi-dimensional automatic calibration operation, A known mass of liquid nitrogen is injected into a liquid nitrogen storage tank, and multiple sets of real-time weight data are collected through the multiple sets of sensing units; a temperature correction operation is performed on the collected real-time weight data to obtain temperature-corrected weight data; Based on the aforementioned levelness deviation coefficient, the weighting coefficients are corrected using the least squares method to minimize data deviation and complete the sensor unit consistency calibration. The liquid nitrogen generation device is started, and vibration signals within a set time period are collected through the multiple sets of sensing units; feature extraction is performed on the vibration signals to obtain vibration feature parameters; a grouped vibration baseline database is established based on the vibration feature parameters; the directional vibration compensation coefficient of each group of sensing units is calibrated to complete the air compressor vibration baseline calibration. To maintain the normal operation of the liquid nitrogen generation device, vibration signals of a set duration under normal workshop conditions are collected through the multiple sets of sensing units; feature extraction is performed on the vibration signals to obtain random vibration feature parameters; a random vibration identification threshold is set based on the random vibration feature parameters and a random vibration compensation model is established to complete the random vibration baseline calibration.

4. The method for measuring the liquid volume of a non-contact high-precision liquid nitrogen generator according to claim 1, characterized in that, During the execution of the aforementioned multi-dimensional automatic calibration operation, Based on the collected temperature data, the average temperature is calculated, and the correlation between liquid nitrogen density and temperature is modeled using an interpolation algorithm to establish a correlation model between liquid nitrogen density and temperature, thereby completing the dynamic calibration of density. Keep the liquid nitrogen generating device in a stopped state, perform a static monitoring operation for a set period of time, and record liquid nitrogen volume data and ambient temperature data at different times; perform calculation operations based on the recorded data to obtain the baseline evaporation rate; perform modeling operations on the correlation between evaporation rate and ambient temperature through a fitting algorithm, establish a correlation model between evaporation rate and ambient temperature, and complete the baseline calibration of evaporation rate.

5. The method for measuring the liquid volume of a non-contact high-precision liquid nitrogen generator according to claim 1, characterized in that, During the preprocessing operation on the collected real-time data, The real-time data is filtered using an outlier removal algorithm to remove extreme data that exceeds a set range. The extreme data that were removed were replaced using data from adjacent time points; The moving average filtering algorithm is used to average the data over multiple consecutive time periods to smooth out high-frequency noise and obtain preprocessed data.

6. The method for measuring the liquid volume of a non-contact high-precision liquid nitrogen generator according to claim 1, characterized in that, During the multi-source data fusion calculation operation on the preprocessed data, Using the temperature compensation coefficient, directional vibration compensation coefficient, and random vibration compensation coefficient, a correction operation is performed on the preprocessed real-time weight data to compensate for the interference caused by temperature drift, directional vibration, and random vibration, thereby obtaining an effective local weight. The vibration signal of each group of sensing units is monitored in real time using the sliding window method, and the root mean square value and instantaneous peak value of vibration within the window are calculated. If the vibration parameters exceed the random vibration identification threshold, it is determined that there is irregular random vibration, triggering dynamic compensation operation and updating the random vibration compensation coefficient. If the vibration parameters do not exceed the threshold, the baseline value of the random vibration compensation coefficient remains unchanged.

7. The method for measuring the liquid volume of a non-contact high-precision liquid nitrogen generator according to claim 6, characterized in that, After obtaining the effective local weight, Perform pairwise absolute value difference calculation on multiple groups of effective local weights, compare the calculation results with a preset consistency threshold, and determine whether there is a slight deviation between the two groups of sensing units. If it is determined that there are two sets of sensor units with slight deviations, the corresponding weight coefficients are corrected by the deviation sharing algorithm, while the weight coefficients of the remaining sensor units remain unchanged. The effective local weights are weighted and fused using the corrected weighting coefficients to obtain the total weight of the liquid nitrogen and the empty tank.

8. The method for measuring the liquid volume of a non-contact high-precision liquid nitrogen generator according to claim 7, characterized in that, After obtaining the total weight of the liquid nitrogen and the empty tank, Based on the real-time average ambient temperature, the real-time liquid nitrogen density is obtained by performing calculations using the density correlation model. By combining the total weight with the total baseline value of the empty tank, a calculation is performed to obtain the uncorrected volume of liquid nitrogen from evaporation. The real-time evaporation rate is obtained by performing calculations based on the liquid nitrogen volume data from continuous detection cycles and the evaporation rate correlation model. After deducting the liquid volume loss due to evaporation, the corrected liquid nitrogen volume is obtained.

9. The method for measuring the liquid volume of a non-contact high-precision liquid nitrogen generator according to claim 8, characterized in that, After obtaining the corrected liquid nitrogen volume, Based on the pre-defined volume height correlation function of the liquid nitrogen storage tank, the liquid nitrogen level height is obtained by performing a solution operation through an iterative algorithm. During the iteration process, the difference in liquid level height between two adjacent iterations is continuously compared. When the difference meets the preset termination condition, the iteration stops and the final liquid nitrogen level height is output.

10. The method for measuring the liquid volume of a non-contact high-precision liquid nitrogen generator according to claim 1, characterized in that, During the execution of the dynamic calibration operation, The calibration process is automatically triggered at set intervals, and the baseline values ​​of weighting coefficient, temperature compensation coefficient, directional vibration compensation coefficient, random vibration compensation coefficient and evaporation rate correlation model parameters are re-corrected based on historical test data. When the data deviation of a certain group of sensor units exceeds the set threshold, the faulty sensor unit is automatically blocked, the weights are redistributed based on the data of the remaining normally functioning sensor units, and the liquid volume calculation continues. When the detected data exceeds the preset threshold, an audible and visual alarm is triggered, and a linkage signal is output to the liquid nitrogen generator control system through the industrial communication interface to automatically adjust the generator's operating power or start / stop status.

Images