A multi-modal hydrogen dispenser dynamic metering calibration system

Abstract

The present application belongs to the field of hydrogenation metering, and relates to a data analysis technique, which is used for solving the problem that the prior art cannot switch the corresponding metering mode in time according to the filling stage, and specifically relates to a multi-mode hydrogenation machine dynamic metering calibration system, which comprises a state recognition module, a metering switching module and a compensation optimization module connected in sequence, and the state recognition module and the compensation optimization module are both communicatively connected with a calibration processing module; the state recognition module is used for performing phased recognition on the hydrogenation process of the hydrogenation machine; in the filling process, real-time recognition parameters are collected, and the filling stage in the hydrogenation process of the hydrogenation machine is recognized according to the recognition parameters; the present application solves the error accumulation problem in the dynamic filling process by constructing a stage conversion recognition mechanism. The prior art does not involve the metering mode switching function, and the present scheme effectively reduces the systematic error by adapting different stage characteristics through the multi-mode metering mode.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of hydrogenation metering, and relates to a data analysis technology, in particular to a multi-modal hydrogenation machine dynamic metering calibration system. BACKGROUND

[0002] The hydrogenation machine metering system is a device for measuring and recording the amount of hydrogen gas in the hydrogenation process, which is mainly used in hydrogen energy automobile hydrogenation stations, and the core functions thereof include metering and pricing to ensure the accuracy and fairness of the hydrogenation process; a hydrogen detection alarm device needs to be arranged in the hydrogenation machine, and the alarm is triggered when the hydrogen concentration reaches 0.4%, and the machine is automatically stopped and the valve is closed when the hydrogen concentration reaches 1.6%.

[0003] The hydrogenation machine calibration method and device disclosed in the patent with the publication number CN118687072B can calibrate the hydrogen gas metering accuracy, test the hydrogen gas filling protocol and test the hydrogen gas filling curve when the hydrogenation machine fills the vehicle with hydrogen gas, thereby ensuring the factory quality of the hydrogenation machine; however, the calibration method cannot identify the filling stage of the hydrogenation machine, so it cannot switch the corresponding metering mode in time according to the filling stage, nor can it analyze the priority of the compensation algorithm according to the stage switching characteristics, resulting in that the metering error of the hydrogenation process cannot be effectively controlled.

[0004] In view of the above technical problems, the present application provides a solution. SUMMARY

[0005] The application aims to provide a multi-modal hydrogenation machine dynamic metering calibration system to solve the problem that the prior art cannot switch the corresponding metering mode in time according to the filling stage;

[0006] The technical problem to be solved by the application is how to provide a metering mode that can be switched in time according to the filling stage.

[0007] The object of the application can be achieved by the following technical solutions:

[0008] A multi-modal hydrogenation machine dynamic metering calibration system, comprising a state recognition module, a metering switching module and a compensation optimization module connected in sequence, and the state recognition module and the compensation optimization module are both communicatively connected with a calibration processing module;

[0009] The state recognition module is used for stage recognition of the hydrogenation process of the hydrogenation machine: real-time acquisition of recognition parameters during the filling process, recognition of the filling stage in the hydrogenation process of the hydrogenation machine according to the recognition parameters, generation of a metering switching signal when the filling stage is switched, and sending of the metering switching signal to the metering switching module and the calibration processing module;

[0010] The metering switching module is used for metering mode switching analysis of the hydrogenation process of the hydrogenation machine: when receiving a metering switching signal, the hydrogenation metering process is performed according to the current filling stage of the hydrogenation machine;

[0011] The calibration processing module is used for compensation calibration analysis when the filling stage of the hydrogenation machine is converted: when receiving a metering switching signal, two filling stages converted by the filling process form a conversion stage group, and a matching algorithm is assigned to the conversion stage group; the matching algorithm is used for dynamic compensation analysis when the stage conversion of the conversion stage group is performed;

[0012] The compensation optimization module is used for optimization analysis of the compensation calibration process of the hydrogenation machine.

[0013] Further, the identification parameters include hydrogen temperature, filling pressure and gas flow, and the filling stage includes a pre-cooling stage, an initial filling stage, a rapid filling stage, a deceleration filling stage and a final filling stage.

[0014] Further, the specific process of the hydrogenation metering process includes: extracting the current filling stage of the hydrogenation machine and marking it as a configuration stage, calling the metering mode corresponding to the configuration stage for hydrogenation metering processing, and the metering mode includes mass metering mode, volume metering mode and flow metering mode.

[0015] Further, the specific process of assigning a matching algorithm to the conversion stage group includes: randomly assigning a dynamic compensation algorithm to the conversion stage group and marking it as the matching algorithm of the conversion stage group, and the dynamic compensation algorithm includes real-time data fusion algorithm, transition zone smoothing algorithm and inertia compensation algorithm.

[0016] Further, the specific process of the compensation optimization module for optimization analysis of the compensation calibration process of the hydrogenation machine includes: generating an optimization period, after completing the stage conversion in the optimization period, obtaining the calibration coefficient of the dynamic compensation analysis process of the matching algorithm, marking the configuration optimization algorithm of the conversion stage group through the calibration coefficient, and directly assigning the configuration optimization algorithm of the conversion stage group when performing compensation calibration analysis in the next optimization period.

[0017] Further, the process of obtaining the calibration coefficient includes: obtaining instantaneous data SS, stable data WD and fluctuation data BD in the dynamic compensation analysis process of the matching algorithm, constructing a compensation data row vector BH of the conversion stage group from the instantaneous data SS, the stable data WD and the fluctuation data BD of all conversion stage groups in the optimization period, BH = [SS, WD, BD], generating a weight column vector QK = [a1, a2, a3], and performing dot product calculation on the compensation data row vector BH and the weight column vector QK to obtain the calibration coefficient of the conversion data group.

[0018] Further, the transient data SS is the switching transient error corresponding to the matching algorithm for dynamic compensation analysis, the stable data WD is the stable time corresponding to the matching algorithm for dynamic compensation analysis, and the fluctuation data BD is the pressure fluctuation amplitude value corresponding to the matching algorithm for dynamic compensation analysis.

[0019] Further, the specific process of marking the configuration optimization algorithm of the conversion stage group includes: summing and averaging the calibration coefficients of the same dynamic compensation algorithm corresponding to the same conversion stage group to obtain the calibration optimization value of the conversion stage group and the dynamic compensation algorithm, and marking the dynamic compensation algorithm with the smallest calibration optimization value as the configuration optimization algorithm of the conversion stage group.

[0020] The present application has the following advantages:

[0021] 1. By constructing a stage conversion recognition mechanism, the error accumulation problem in the dynamic filling process is solved. The prior art does not involve metering mode switching function, and the present scheme effectively reduces systematic error by adapting different stage characteristics through multi-modal metering mode.

[0022] 2. By recognizing the filling stage in real time and triggering mode switching, the metering deviation caused by mode mismatch is eliminated. By dynamically distributing compensation algorithms and continuously optimizing algorithm matching relationship, the pressure fluctuation of different conversion stages is targetedly suppressed; ultimately forming a complete error control chain from stage recognition, mode switching to compensation optimization, significantly improving the metering stability under complex working conditions;

[0023] 3. Through the collaborative analysis of multi-dimensional parameters, the hydrogen temperature change curve in the pre-cooling stage can be accurately captured, the maximum pressure difference interval of the rapid filling stage can be recognized, and the flow decay rate can be tracked in real time in the deceleration filling stage, so as to realize the strict matching of the metering mode switching time and the stage characteristics. BRIEF DESCRIPTION OF DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0025] Figure 1 The system block diagram of the first embodiment of the present application is shown in the figure.

[0026] Figure 2 The method flow chart of the second embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0027] The technical solutions of the present application will be clearly and completely described below in conjunction with the embodiments. 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 belong to the scope of the present application.

[0028] Embodiment one: as shown in a multi-modal hydrogenation machine dynamic metering calibration system, comprising state recognition module, metering switching module and compensation optimization module connected in sequence, the state recognition module and the compensation optimization module are both communicatively connected with the calibration processing module. The state recognition module is used for phase recognition of the hydrogenation process of the hydrogenation machine, the metering switching module is used for metering mode switching analysis, the calibration processing module is used for compensation calibration analysis, and the compensation optimization module is used for optimization of the compensation calibration process. Figure 1

[0029] Among them, the state recognition module refers to a functional unit for real-time acquisition of hydrogenation process physical parameters and recognition of phase conversion nodes, which can be realized by using a temperature sensor array and a multi-channel data acquisition card, and the phase conversion is determined by analyzing the temperature gradient and the flow rate change rate. The metering switching module is a control unit for switching the metering mode according to the phase characteristics, which can be realized by using an embedded processor to load a pre-designed metering mode mapping table, for example, calling a mass metering mode in the pre-cooling stage. The calibration processing module is a compensation algorithm distribution unit for processing phase conversion errors, which can be realized by using an FPGA chip to realize dynamic algorithm library calling, and assigning inertia compensation algorithms for adjacent phase combinations. The compensation optimization module is a self-learning unit for optimizing the matching relationship of algorithms, which can train a neural network model using historical data to establish a mapping relationship between the conversion phase group and the optimal algorithm.

[0030] Specifically, the state recognition module captures the pressure inflection point generated when the rapid filling stage is converted to the deceleration filling stage by real-time monitoring of hydrogen temperature and gas flow parameters, and triggers the generation of metering switching signals. After receiving the signal, the metering switching module immediately switches the flow metering mode to the volume metering mode to avoid metering errors caused by sudden changes in flow rate. The calibration processing module marks the current conversion phase combination as a "rapid-deceleration" conversion group, calls a transition zone smoothing algorithm from the algorithm library, and performs real-time compensation on pressure fluctuations. After completing ten filling cycles, the compensation optimization module analyzes the calibration coefficients of different algorithms used in the conversion group, and finally determines the inertia compensation algorithm as the optimal algorithm for the conversion group.

[0031] ​Compared with the prior art, the patent with publication number CN118687072B can only complete static precision verification, while the scheme solves the error accumulation problem in the dynamic filling process by constructing a stage conversion recognition mechanism. The prior art does not involve a metering mode switching function, while the scheme effectively reduces systematic errors by adapting different stage characteristics through multi-modal metering modes. The prior art lacks a compensation algorithm optimization mechanism, while the scheme reduces the pressure fluctuation amplitude by about 40% by optimizing the algorithm distribution strategy through self-learning.

[0032] Through the above technical scheme, the application realizes error closed-loop control of the hydrogen filling machine in the stage conversion process. By identifying the filling stage in real time and triggering mode switching, the measurement deviation caused by mode mismatch is eliminated. By dynamically distributing the compensation algorithm and continuously optimizing the algorithm matching relationship, the pressure fluctuation in different conversion stages is targetedly suppressed. Finally, a complete error control chain from stage recognition, mode switching to compensation optimization is formed, which significantly improves the measurement stability under complex working conditions.

[0033] In the hydrogenation process, hydrogen temperature, filling pressure and gas flow are used as identification parameters to divide the filling process into pre-cooling stage, initial filling stage, rapid filling stage, deceleration filling stage and final filling stage.

[0034] Among them, the hydrogen temperature refers to the real-time temperature value of the hydrogen medium in the hydrogenation process, which can be collected by using a thermocouple or an infrared temperature sensor. This parameter is used to reflect the influence of thermodynamic state change on the volume measurement accuracy in the hydrogenation process. The filling pressure refers to the real-time pressure value of hydrogen in the conveying pipeline, which can be monitored by using a pressure transmitter. This parameter is used to judge the change trend of hydrogenation rate and the critical point of stage switching. The gas flow refers to the volume or mass of hydrogen passing through the hydrogen gun per unit time, which can be measured by using a turbine flowmeter or a Coriolis force mass flowmeter. This parameter is used to characterize the filling intensity and assist in stage division. The pre-cooling stage refers to the initial preparation stage of heat balance adjustment of the hydrogenation system by low-temperature medium. The rapid filling stage refers to the core pressurization stage of filling the vehicle-mounted hydrogen storage container with maximum flow rate. The deceleration filling stage refers to the transition stage of actively reducing the filling rate according to the pressure gradient change.

[0035] Specifically, dynamic monitoring of hydrogen temperature can capture the volume expansion effect caused by phase change or heat exchange during the hydrogenation process, providing a data basis for compensating volume measurement errors. Real-time analysis of filling pressure can identify pressure inflection points, such as when the set pressure threshold is reached during the rapid filling stage to trigger stage switching. Continuous tracking of gas flow can determine filling rate mutations, such as when the flow rate exceeds the preset range to determine the entry into the deceleration filling stage. The pre-cooling stage confirms the system thermal equilibrium state by monitoring the hydrogen temperature reaching the set low temperature threshold, the initial filling stage starts with gas flow from zero value as the stage starting mark, and the final filling stage determines the completion of filling by the difference between the filling pressure and the target pressure being less than the set tolerance.

[0036] Compared with the prior art, the existing hydrogenation machine verification method only judges the filling state by fixed threshold, cannot dynamically adjust the measurement reference according to the thermal equilibrium state of the pre-cooling stage, and cannot identify the pressure change characteristics of the rapid filling stage and the deceleration filling stage. The scheme can accurately capture the hydrogen temperature change curve in the pre-cooling stage, identify the maximum pressure difference interval in the rapid filling stage, and track the flow decay rate in real time in the deceleration filling stage, so as to realize strict matching of the measurement mode switching time and the stage characteristics.

[0037] Through the above technical scheme, the present application solves the problem of measurement mode switching lag caused by inaccurate stage recognition in the hydrogenation process, avoids the temperature drift error caused by using volume measurement mode in the rapid filling stage, and reduces the compensation algorithm mismatch caused by inertia effect in the deceleration filling stage. Based on the dynamic calibration of the temperature data in the pre-cooling stage, the influence of system thermal delay on the initial measurement accuracy can be eliminated; through cross verification of pressure and flow parameters, the stage switching boundary can be accurately determined to ensure that the optimal measurement strategy is used in different filling stages.

[0038] Extract the current filling stage of the hydrogenation machine and mark it as the configuration stage, retrieve the measurement mode corresponding to the configuration stage for hydrogenation measurement processing, and the measurement mode includes mass measurement mode, volume measurement mode and flow measurement mode.

[0039] The configuration stage refers to the current filling process state of the hydrogen filling machine, and can be determined by real-time collection of hydrogen temperature, filling pressure and gas flow parameters by a sensor, and is used to provide a stage identifier for subsequent metering mode selection. The quality metering mode refers to a metering method based on the principle of conservation of mass to calculate the mass of hydrogen, and can be implemented by using a mass flow meter combined with a temperature and pressure compensation algorithm, and is suitable for filling stages with significant phase change or significant density change. The volume metering mode refers to a metering method by measuring the volume of hydrogen, and can be implemented by using a turbine flow meter combined with a pressure sensor, and is suitable for filling stages with stable flow. The flow metering mode refers to a metering method based on instantaneous flow integration to calculate the total amount, and can be implemented by using an ultrasonic flow meter or a differential pressure flow meter, and is suitable for dynamic filling stages with rapid flow fluctuations.

[0040] Specifically, in the pre-cooling stage, the temperature of hydrogen decreases significantly, and at this time, the mass metering mode is used to compensate for the density change caused by the phase change process. After entering the rapid filling stage, the gas flow reaches the peak value and the pressure fluctuates frequently, and the flow metering mode is used to capture the instantaneous flow change to realize dynamic integration. When the filling enters the deceleration stage, the gas flow tends to be flat and the pressure is stable, and the volume metering mode is switched to for steady-state metering. Thus, the differences in physical characteristics of different filling stages are treated by the adaptive metering mode, avoiding the accumulation of errors caused by the mismatch between the single metering method and the stage characteristics.

[0041] Compared with the prior art, the traditional hydrogen filling machine calibration method cannot identify the filling stage and only uses a fixed metering mode, for example, using volume metering in the phase change stage results in the lack of density compensation, or using mass metering in the flow mutation stage causes instantaneous errors. The present scheme triggers the dynamic switching of the metering mode through stage recognition, so that the most suitable metering method is used in each stage, solving the system error caused by the mismatch between the mode and the stage in the prior art;

[0042] Through the above technical scheme, the present application realizes automatic switching of the optimal metering mode of the hydrogen filling machine in different filling stages, effectively reduces the metering deviation caused by the density change in the phase change process, the instantaneous error of flow mutation and the compensation lag in the steady-state process, and significantly improves the overall accuracy of the whole cycle hydrogen filling metering.

[0043] To convert the phase group to assign a matching algorithm, including randomly assigning a dynamic compensation algorithm to the transition phase group and marking it as the matching algorithm of the transition phase group. The dynamic compensation algorithm includes real-time data fusion algorithm, transition zone smoothing algorithm and inertia compensation algorithm.

[0044] The conversion stage group refers to the stage conversion process formed by two adjacent filling stages during the filling process, which can be divided by collecting the time stamp of the stage switching trigger signal to determine the range of the compensation algorithm. The dynamic compensation algorithm refers to a calculation method designed for the error characteristics generated when the hydrogen filling machine switches between different filling stages. Specifically, a multi-sensor data fusion method based on Kalman filtering can be used to implement a real-time data fusion algorithm to eliminate instantaneous measurement errors. A sliding window mean filtering method can be used to implement a transition zone smoothing algorithm to alleviate the measurement fluctuations caused by pressure sudden changes. A delay compensation model based on historical data prediction can be used to implement an inertia compensation algorithm to correct the measurement deviation caused by mechanical system response delay.

[0045] Specifically, when the measurement switching signal is received, the conversion stage group is formed by the current filling stage and the next filling stage. The system randomly assigns a dynamic compensation algorithm to each conversion stage group as the initial matching algorithm, for example, when switching from the pre-cooling stage to the fast filling stage, the real-time data fusion algorithm can be randomly selected. This algorithm runs continuously during stage switching, collects multi-modal sensor data in real time, and eliminates instantaneous errors caused by temperature gradients or flow rate changes through data fusion processing. When the system detects that the stage switching is complete, the matching algorithm stops running and hands over control to the measurement mode of the next stage. In the subsequent optimization period, the system evaluates the performance of different algorithms in the same conversion stage group based on the calibration coefficients, and gradually establishes the algorithm priority relationship.

[0046] Compared with the prior art, the patent with publication number CN118687072B can detect the hydrogen measurement accuracy, but does not consider the dynamic compensation requirement of measurement error during stage switching. The present scheme randomly assigns dynamic compensation algorithms matched with stage characteristics, and for the first time actively corrects errors during stage switching in the hydrogen filling machine control system. Compared with the adaptation deviation caused by the fixed compensation mode in the prior art, the present scheme uses a multi-algorithm parallel trial-and-error mechanism to cover error sources in different physical scenarios, such as real-time data fusion algorithms for sensor instantaneous errors and inertia compensation algorithms for mechanical delay problems, forming a multi-dimensional compensation capability.

[0047] Through the above technical solutions, the present application can automatically select the appropriate error compensation algorithm during the filling stage switching of the hydrogen filling machine, effectively reducing the instantaneous measurement deviation caused by pressure sudden changes, mechanical delay or sensor drift. For the different characteristics of hydrogen flow and temperature changes in different conversion stage combinations, accurate error correction is achieved through a dynamic algorithm allocation mechanism, making the measurement data of each stage of the hydrogen filling process smoothly transition, and the overall calibration accuracy is improved.

[0048] The specific process of the compensation optimization module for optimizing and analyzing the compensation calibration process of the hydrogenation machine includes generating an optimization period, obtaining a calibration coefficient of a matching algorithm for a dynamic compensation analysis process after completing a stage conversion in the optimization period, marking a configuration optimization algorithm of a conversion stage group through the calibration coefficient, and directly distributing the configuration optimization algorithm through the conversion stage group when performing compensation calibration analysis in the next optimization period.

[0049] The optimization period refers to a preset time period for periodically evaluating the performance of the compensation algorithm. It can be implemented by using a fixed time length or a stage conversion number threshold, for example, setting the optimization period to trigger evaluation after completing five stage conversions. This provides a quantifiable evaluation window for dynamic optimization. The calibration coefficient refers to a numerical indicator that quantifies the performance of the compensation algorithm. It can be implemented by weighted calculation of instantaneous data, stable data, and fluctuation data, for example, using row vector and weight column vector dot product to obtain the calibration coefficient. This provides an objective performance evaluation benchmark for algorithm optimization. The configuration optimization algorithm refers to the priority compensation algorithm marked by the calibration coefficient. It can be implemented by using numerical comparison to select the algorithm with the smallest calibration optimization value as the optimal choice, for example, determining the mapping relationship between the conversion stage group and the algorithm through historical data statistics. This replaces the random distribution mechanism with a data-driven approach.

[0050] Specifically, after completing the refueling stage conversion in the optimization period, the compensation optimization module extracts the instantaneous data, stable data, and fluctuation data generated by the matching algorithm in the dynamic compensation analysis process from the calibration processing module, and calculates the calibration coefficient through the preset weight. This coefficient is used to mark the adaptability of the current conversion stage group and the dynamic compensation algorithm, for example, recording the algorithm with the lowest calibration coefficient as the configuration optimization algorithm. In the next optimization period, when the same conversion stage group appears again, the system directly calls the marked configuration optimization algorithm for compensation analysis without re-random distribution. This forms a closed-loop mechanism based on iterative optimization of historical data, gradually reducing the uncertainty error of algorithm selection.

[0051] Compared with the prior art, the existing hydrogenation machine verification method can only verify the measurement accuracy statically and cannot dynamically adjust the compensation algorithm according to stage conversion, for example, the patent with publication number CN118687072B does not involve an adaptive optimization mechanism for the compensation algorithm. The present scheme establishes a dynamic correlation between algorithm performance and conversion stages by periodically evaluating the calibration coefficient, applies the historically optimal algorithm configuration to subsequent periods, and effectively solves the error accumulation problem caused by random distribution.

[0052] By the technical solution, the adaptive optimization of the compensation algorithm at the stage conversion is realized, the calibration coefficient marking and the historical optimal algorithm calling mechanism are used, and the accumulation of the measurement error caused by the mismatch between the algorithm and the stage characteristics is significantly reduced, and the dynamic calibration precision of the hydrogenation process is improved.

[0053] The acquisition process of the calibration coefficient includes: acquiring instantaneous data SS, stable data WD and fluctuation data BD in the dynamic compensation analysis process of the matching algorithm, forming a compensation data row vector BH of the conversion stage group by the instantaneous data SS, the stable data WD and the fluctuation data BD of all conversion stage groups in the optimization period, BH = [SS, WD, BD], generating a weight column vector QK = [a1, a2, a3], and performing dot product calculation on the compensation data row vector BH and the weight column vector QK to obtain the calibration coefficient of the conversion data group.

[0054] The instantaneous data SS is the switching instantaneous error of the hydrogen measurement system during the dynamic compensation analysis, which can be calculated by the real-time acquisition data difference of the pressure sensor and the flowmeter at the stage conversion moment, and is used to evaluate the response precision of the compensation algorithm at the stage switching moment.

[0055] The stable data WD is the time required for the hydrogen measurement system to reach a stable state after the compensation analysis is completed, which can be calculated by recording the time point when the pressure fluctuation value enters the preset threshold range by the timer, and is used to measure the optimization efficiency of the compensation algorithm on the system stability.

[0056] The fluctuation data BD is the extreme value of the fluctuation amplitude of the hydrogen pressure during the dynamic compensation process, which can be calculated by the standard deviation or the peak-to-peak value after continuous acquisition of the pressure sensor, and is used to quantify the suppression ability of the compensation algorithm to the pressure disturbance.

[0057] The compensation data row vector BH is a three-dimensional vector composed of instantaneous data, stable data and fluctuation data, which is formed by arranging the three types of data in order to form a row vector, and is used to comprehensively represent the dynamic performance characteristics of the compensation algorithm in different dimensions.

[0058] The weight column vector QK is a column vector composed of weight coefficients corresponding to different data indicators, which can be determined by training the historical data by expert experience method or machine learning algorithm, for example, a1 is 0.5, a2 is 0.3, and a3 is 0.2, and is used to reflect the priority difference of different performance indicators in the comprehensive evaluation.

[0059] The dot product calculation is a matrix multiplication operation of the compensation data row vector and the weight column vector, which realizes the weighted fusion of multi-dimensional data by linear algebra operation, and is used to generate a scalar calibration coefficient representing the comprehensive performance of the compensation algorithm.

[0060] Specifically, during the hydrogen refueling stage transition process, by collecting real-time pressure fluctuation, flow change and temperature fluctuation data, switching error, system stability required time and pressure fluctuation amplitude extreme value representing the instantaneous response ability of the compensation algorithm are calculated respectively. The three types of data are constructed into three-dimensional row vectors, and the preset weight column vector is multiplied to generate a calibration coefficient that can quantify the comprehensive performance of the compensation algorithm. The coefficient reflects the optimization effect of the compensation algorithm on the system dynamic characteristics during the stage transition process, providing an objective evaluation basis for subsequent algorithm optimization selection. By adjusting the weight coefficient, the evaluation emphasis can be optimized for different refueling conditions, for example, focusing on pressure fluctuation suppression ability in fast refueling stage, and giving priority to system stability in deceleration stage.

[0061] Compared with the prior art, the traditional method usually only evaluates the compensation algorithm by a single index, such as only detecting the pressure fluctuation amplitude or flow error, which leads to the inability to comprehensively evaluate the comprehensive influence of the algorithm on the system dynamic characteristics. While the present scheme can objectively quantify the performance of the compensation algorithm in different performance dimensions by constructing a multi-dimensional data model and introducing a weight distribution mechanism, solving the optimization direction deviation problem caused by a single evaluation index.

[0062] In the dynamic compensation analysis process, the switching instantaneous error is taken as the instantaneous data, the stable time length is taken as the stable data, and the pressure fluctuation amplitude value is taken as the fluctuation data.

[0063] Wherein, the switching instantaneous error refers to the instantaneous deviation caused by the switching of the metering mode at the moment of stage transition, which can be measured by the difference between the pressure sensor and the flowmeter at the switching time. This parameter is used to quantify the response speed and accuracy of the algorithm.

[0064] Wherein, the stable time length refers to the time length required for the compensation algorithm to restore the system to a stable state, which can be realized by recording the time span of the pressure fluctuation entering the preset threshold range using a timer. This parameter reflects the suppression efficiency of the algorithm on system oscillation.

[0065] Wherein, the pressure fluctuation amplitude value refers to the maximum deviation of hydrogen pressure during the stage transition process, which can be realized by calculating the absolute difference between the peak value and the valley value monitored by the pressure sensor. This parameter is used to evaluate the influence of the compensation process on safety.

[0066] Specifically, in the stage conversion process of the hydrogen filling machine, the dynamic compensation algorithm needs to meet the requirements of error control, stable time compression and pressure fluctuation suppression. The switching transient error is obtained by real-time acquisition of the flow and pressure deviation at the switching time, which exposes the defects of the algorithm in transient response. The stable time is obtained by recording the time when the pressure curve enters the steady state interval, which evaluates the convergence efficiency of the algorithm. The pressure fluctuation amplitude is obtained by monitoring the pressure extreme difference, which constrains the safety risk in the compensation process. The three parameters construct the evaluation dimension of the compensation algorithm from the error amplitude, time dimension and pressure stability, providing a data basis with clear physical meaning for the weight distribution of the calibration coefficient.

[0067] Compared with the prior art, the existing hydrogen filling machine calibration method only relies on a single parameter to evaluate the accuracy of the measurement, such as only detecting flow error or pressure deviation, which cannot establish multi-dimensional compensation optimization basis in multi-stage conversion scenarios. The present scheme defines the quantitative indicators of transient error, stable time and pressure fluctuation to form a multi-dimensional data matrix, solving the problem of insufficient algorithm optimization basis caused by missing parameter dimension in the prior art.

[0068] The specific process of marking the configuration optimization algorithm of the conversion stage group includes summing and averaging the calibration coefficients of the same dynamic compensation algorithm corresponding to the same conversion stage group to obtain the calibration optimization value of the conversion stage group and the dynamic compensation algorithm, and marking the dynamic compensation algorithm with the smallest calibration optimization value as the configuration optimization algorithm of the conversion stage group.

[0069] Wherein, the conversion stage group refers to a combination of two filling stages converted by the filling process, which can be realized by the combination of pre-cooling stage and initial filling stage, the combination of rapid filling stage and deceleration filling stage, etc., and is used to define the stage conversion scenario that needs to be compensated and calibrated.

[0070] Wherein, the dynamic compensation algorithm refers to an operation method for error compensation in the stage conversion process, which can be realized by real-time data fusion algorithm, transition zone smoothing algorithm or inertia compensation algorithm, and different algorithms correspond to different compensation logic.

[0071] Wherein, the calibration coefficient refers to a quantitative evaluation index calculated by the dot product of the compensation data row vector and the weight column vector, which can be realized by the weighted calculation result of the instantaneous error, the stable time and the pressure fluctuation amplitude, and is used to reflect the comprehensive performance of the dynamic compensation algorithm in a specific conversion stage group.

[0072] Wherein, the calibration optimization value refers to the result of mean calculation of multiple calibration coefficients of the same dynamic compensation algorithm under the same conversion stage group, which can be realized by historical data statistical method, and is used to eliminate accidental error interference in single compensation process.

[0073] The configuration optimization algorithm refers to the optimal dynamic compensation algorithm screened for a specific conversion stage group, which can be specifically realized by comparing the calibration optimization values of different algorithms and selecting the algorithm corresponding to the minimum value, and is used to ensure the stability of the subsequent compensation process.

[0074] Specifically, in the conversion process of the refueling stage of the hydrogenation machine, a conversion stage group is first formed by the pre-cooling stage and the initial refueling stage. For this conversion stage group, the system randomly allocates different dynamic compensation algorithms for compensation operation in multiple optimization periods, for example, the real-time data fusion algorithm is allocated first, and the inertia compensation algorithm is allocated second. After each compensation is completed, the system calculates the calibration coefficient according to the instantaneous error, the stable time length and the pressure fluctuation amplitude. When the optimization period ends, the system calculates the mean value of all calibration coefficients of the same algorithm under the same conversion stage group, to obtain the calibration optimization value corresponding to the algorithm. For example, the calibration optimization value of the real-time data fusion algorithm in the conversion stage group is 0.85, and the calibration optimization value of the inertia compensation algorithm is 0.72, so the inertia compensation algorithm is marked as the configuration optimization algorithm of the conversion stage group. In the subsequent stage conversion, the system directly calls the configuration optimization algorithm for compensation, avoiding error accumulation caused by repeated trial and error.

[0075] Compared with the prior art, the existing hydrogenation machine verification method adopts a random allocation of dynamic compensation algorithms, which cannot optimize algorithm selection according to historical performance data of stage conversion scenarios, resulting in large fluctuations in compensation effect. The present scheme establishes a mapping relationship between the conversion stage group and the optimal compensation algorithm through statistical calculation of the calibration optimization value, so that the algorithm selection process changes from random trial and error to data-driven optimization.

[0076] Embodiment two: as shown in the figure, a multi-modal hydrogenation machine dynamic metering calibration method, comprising the following steps: Figure 2

[0077] Step one: phase identification of the hydrogenation process of the hydrogenation machine: real-time acquisition of identification parameters in the refueling process, identification of the refueling stage in the hydrogenation process of the hydrogenation machine according to the identification parameters;

[0078] Step two: analysis of metering mode switching of the hydrogenation process of the hydrogenation machine, when the metering switching signal is received, the current refueling stage of the hydrogenation machine is extracted and marked as the configuration stage, the metering mode corresponding to the configuration stage is called to perform hydrogenation metering processing;

[0079] Step three: compensation and calibration analysis of the hydrogenation machine in the refueling stage conversion: when the metering switching signal is received, the two refueling stages converted from the refueling process constitute a conversion stage group, a dynamic compensation algorithm is randomly allocated for the conversion stage group and marked as the matching algorithm of the conversion stage group;

[0080] ​Step four: optimization analysis is carried out on the compensation calibration process of the hydrogenation machine: an optimization cycle is generated, after the completion of the stage conversion in the optimization cycle, the calibration coefficient of the matching algorithm is obtained to carry out the dynamic compensation analysis process, the calibration coefficients of the same dynamic compensation algorithm corresponding to the same conversion stage group are summed and averaged to obtain the calibration optimization value of the conversion stage group and the dynamic compensation algorithm, and the configuration optimization algorithm of the conversion stage group is marked through the calibration optimization value.

[0081] A multi-modal hydrogenation machine dynamic metering calibration system, when working,

[0082] The above is only an example and description of the structure of the present application, and those skilled in the art can make various modifications or supplements to the described specific embodiments or use similar ways to replace, as long as the modifications or supplements do not deviate from the structure of the application or exceed the scope defined by the claims, and should belong to the protection scope of the present application.

[0083] In the description of the present application, the description of the terms "one embodiment", "example", "specific example" and the like means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are contained in at least one embodiment or example of the present application. In the present application, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0084] The preferred embodiments of the application disclosed above are only used to help explain the application. The preferred embodiments do not describe all the details and limit the application to the specific embodiments. Obviously, many modifications and changes can be made according to the content of the present application. The present application selects and describes these embodiments in order to better explain the principles and practical application of the present application, so that those skilled in the art can well understand and utilize the present application. The present application is limited by the claims and their full scope and equivalents.

Claims

1. A multi-modal hydro-mechanical dynamic metering calibration system, characterized by, The state recognition module, the compensation optimization module and the calibration processing module are sequentially connected, and the state recognition module and the compensation optimization module are in communication connection with the calibration processing module; The state recognition module is configured to recognize the hydrogenation process of the hydrogenation machine in stages: real-time collection of recognition parameters during the refueling process, recognition of the refueling stage in the hydrogenation process of the hydrogenation machine according to the recognition parameters, generation of a metering switching signal when the refueling stage is recognized, and sending of the metering switching signal to the metering switching module and the calibration processing module; The metering switching module is configured to analyze the metering mode switching of the hydrogenation process of the hydrogenation machine: when the metering switching signal is received, hydrogenation metering processing is performed according to the current refueling stage of the hydrogenation machine; The calibration processing module is configured to analyze the compensation calibration of the hydrogenation machine during the conversion of the refueling stage: when the metering switching signal is received, two refueling stages converted by the refueling process form a conversion stage group, and a matching algorithm is assigned to the conversion stage group. The matching algorithm is used to perform dynamic compensation analysis during the stage conversion of the conversion stage group. The compensation optimization module is configured to optimize the compensation calibration process of the hydrogenation machine. The process of obtaining the calibration coefficient includes: obtaining instantaneous data SS, stable data WD and fluctuation data BD during the dynamic compensation analysis of the matching algorithm, forming a compensation data row vector BH of the conversion stage group from the instantaneous data SS, the stable data WD and the fluctuation data BD of all conversion stage groups in the optimization period, BH = [SS, WD, BD], generating a weight column vector QK = [a1, a2, a3], and performing dot product calculation on the compensation data row vector BH and the weight column vector QK to obtain the calibration coefficient of the conversion stage group. The instantaneous data SS is the switching instantaneous error corresponding to the dynamic compensation analysis of the matching algorithm, the stable data WD is the stable time corresponding to the dynamic compensation analysis of the matching algorithm, and the fluctuation data BD is the pressure fluctuation amplitude value corresponding to the dynamic compensation analysis of the matching algorithm. The specific process of marking the configuration optimization algorithm of the conversion stage group includes: summing and averaging the calibration coefficients of the same dynamic compensation algorithm corresponding to the same conversion stage group to obtain the calibration optimization value of the conversion stage group and the dynamic compensation algorithm, and marking the dynamic compensation algorithm with the smallest calibration optimization value as the configuration optimization algorithm of the conversion stage group.

2. A multi-modal hydro-mechanical dynamic metering calibration system according to claim 1, characterized in that, The recognition parameters include hydrogen temperature, refueling pressure and gas flow, and the refueling stage includes a precooling stage, an initial refueling stage, a rapid refueling stage, a deceleration refueling stage and a final refueling stage.

3. A multi-modal hydrodynamic motor dynamic metering calibration system according to claim 1, wherein, The specific process of hydrogenation metering processing includes: extracting the current refueling stage of the hydrogenation machine and marking it as a configuration stage, calling the corresponding metering mode of the configuration stage for hydrogenation metering processing, and the metering mode includes mass metering mode, volume metering mode and flow metering mode.

4. The multi-modal hydrodynamic motor dynamic metering calibration system of claim 1, wherein, The specific process of assigning a matching algorithm to the conversion stage group includes: randomly assigning a dynamic compensation algorithm to the conversion stage group and marking it as the matching algorithm of the conversion stage group, and the dynamic compensation algorithm includes a real-time data fusion algorithm, a transition zone smoothing processing algorithm and an inertia compensation algorithm.

5. A multi-modal hydrodynamic motor dynamic metering calibration system according to claim 1, wherein, The specific process of the compensation optimization module for optimizing and analyzing the compensation calibration process of the hydrogenation machine includes: generating an optimization period, obtaining calibration coefficients of a matching algorithm for a dynamic compensation analysis process after completing stage conversion in the optimization period, marking a configuration optimization algorithm of a conversion stage group through the calibration coefficients, and directly distributing the configuration optimization algorithm through the conversion stage group when performing compensation calibration analysis in the next optimization period.

Images