A hydrogen storage tank intelligent early warning device and early warning method

Abstract

The application provides a hydrogen storage tank intelligent early warning device and method, which comprises a sensing system, a processing unit and an early warning system, wherein the sensing system performs real-time sensing on multiple parameters of selected points of the hydrogen storage tank; the sensing system is connected with the processing unit to transmit sensed data to the processing unit for analysis and judgment; the early warning system receives early warning instructions sent by the processing unit through wireless technology and sends corresponding alarm signals, the intelligent early warning device and method perform sensing on multiple parameters of the hydrogen storage tank, data coupling on the multiple parameters, mutual verification through physical connection between the multiple parameters, and real-time monitoring on parameter threshold and parameter change rate, the application realizes more accurate monitoring and judgment on parameter abnormalities of the hydrogen storage tank and earlier early warning, facilitates timely adoption of effective safety protection measures on the hydrogen storage tank, ensures safety and reliability of the hydrogen storage tank, and realizes real-time monitoring and early warning functions of multiple parameters of the hydrogen storage tank.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of hydrogen equipment state detection control, and particularly relates to a hydrogen storage tank intelligent early warning device and method. BACKGROUND

[0002] With the development of hydrogen energy industry, hydrogen storage tanks are increasingly applied to common application scenarios such as vehicles. Since hydrogen is extremely active, the ignition volume percentage of hydrogen in air environment is 4% to 75%, and the range is larger in oxygen-rich environment, and hydrogen is extremely easy to ignite. Therefore, the relatively closed space where hydrogen and oxygen are gathered is very dangerous. Since a hydrogen storage tank stores a large amount of hydrogen, the safety factor of the hydrogen storage tank is extremely high, and it is necessary to detect and analyze the internal environment of the hydrogen storage tank in real time and perform intelligent early warning, and it is also necessary to monitor the main structure of the hydrogen storage tank in real time, determine the health status of the main body of the hydrogen storage tank, and predict and warn possible damage and failure of the hydrogen storage tank to ensure the safety of the main body of the hydrogen storage tank.

[0003] In order to improve the safety factor of the hydrogen storage tank, the prior art generally only monitors the structure of the hydrogen storage tank in real time by winding a fiber optic strain sensor on the inner container of the hydrogen storage tank or monitors the pressure in the hydrogen storage tank in real time, but the technology has the problems of single monitoring method and monitoring content, cannot monitor the complex working environment inside the hydrogen storage tank as a whole, and the single detection content cannot completely reflect the health status of the hydrogen storage tank as a whole, and single-factor detection is prone to false detection, missed detection, false alarm and other phenomena. SUMMARY

[0004] In view of the problems in the prior art, the present application provides a hydrogen storage tank intelligent early warning device and method, which can more comprehensively and accurately reflect the health status of the hydrogen storage tank by multi-parameter sensing, and further improve the comprehensiveness and accuracy of detection by multi-parameter coupling. The present application not only compares and analyzes the real-time parameter data and reference parameter data of the hydrogen storage tank, but also creatively analyzes and judges the parameter change rate, further improves the hydrogen storage tank early warning logic, and solves the problem that the health of the hydrogen storage tank is monitored by a single sensor monitoring a single content in the prior art, and the accuracy of health monitoring depends on the detection range and working effect of the single sensor, which is prone to false detection, missed detection, false alarm and other phenomena, thereby causing safety hazards in the use of the hydrogen storage tank.

[0005] The present application achieves the above technical purposes by the following technical means.

[0006] A hydrogen storage tank intelligent early warning device, comprising:

[0007] A perception system is used for real-time multi-parameter perception of selected H points of a hydrogen storage tank, 1<H<N; wherein N is the total number of grid division of the tank body surface of the hydrogen storage tank;

[0008] A processing unit is connected with the perception system, and is used for receiving real-time parameter data transmitted by the perception system;

[0009] A warning system is connected with the processing unit through a wireless module, and is used for receiving a warning signal sent by the processing unit;

[0010] wherein,

[0011] The processing unit selects a main parameter from the multi-parameter of the H points, and selects reference values of the multi-parameter of all points in the identification library according to the main parameter values of the H points and the state of the hydrogen storage tank; the processing unit reversely calculates the multi-parameter data of all points according to the physical model based on the multi-parameter of the H points, and obtains multi-parameter derivative data of all points through data coupling, and verifies the multi-parameter derivative data of all points with the reference data of the multi-parameter of all points selected from the identification library, and sends a warning instruction to the warning system if the difference between the two exceeds a preset threshold; the processing unit calculates and analyzes the change rate of the derivative data of the multi-parameter of all points, and sends a warning instruction to the warning system if the change rate of a certain parameter exceeds a preset threshold under the state of the hydrogen storage tank.

[0012] Further, the perception system comprises a pressure sensor, a strain sensor, a temperature sensor and a hydrogen sensor; the pressure sensor is used for detecting the pressure of the tank body of the hydrogen storage tank, the temperature sensor is used for detecting the temperature of the tank body of the hydrogen storage tank, and the hydrogen sensor is used for detecting whether there is leaked hydrogen gas near the outside of the hydrogen storage tank; a plurality of strain sensors are installed on the hydrogen storage tank, and are used for detecting the strain of different point positions of the hydrogen storage tank; the pressure sensor, the strain sensor, the temperature sensor and the hydrogen sensor are connected with the processing unit respectively.

[0013] Further, the processing unit comprises an identification library, a reference data selection module, a data coupling module, a parameter change rate generation module, a logic judgment module and a wireless module;

[0014] The identification library contains multi-parameter data reference values of all points of the tank body under all states of the normal operation of the hydrogen storage tank;

[0015] The reference data selection module selects reference data in the identification library according to the selected main parameter and the basic state of the hydrogen storage tank;

[0016] The data coupling module reverses the received H-point multi-parameter data according to a physical model to obtain multi-parameter reverse data of all points of the tank and perform data fusion to obtain derivative data of the multi-parameters of all points of the tank; the parameter change rate generation module is configured to calculate the real-time change rate of the derivative data of the multi-parameters of all points; the logic judgment module compares the reference data of the multi-parameters of all points with the derivative data of the multi-parameters of all points, logically judges the real-time change rate of the derivative data of the multi-parameters of any one point, and sends a warning instruction to the warning system through the wireless module according to the judgment result.

[0017] Further, the warning system comprises an audible and light module and a wireless receiving and processing module; the wireless receiving and processing module is configured to receive and process the warning instruction sent by the processing unit, and the audible and light module is configured to execute a corresponding warning alarm according to the warning instruction.

[0018] A warning method of an intelligent warning device for a hydrogen storage tank, comprising the following steps:

[0019] The surface of the hydrogen storage tank is divided into N grids based on grid division of a simulation model, and each grid center represents a point of the hydrogen storage tank; H monitoring points are selected at random, and at least one sensor is installed at each monitoring point to detect the pressure, temperature, strain and presence of leaked hydrogen of the hydrogen storage tank;

[0020] Standard data is collected under normal operation of the hydrogen storage tank, M working states of the hydrogen storage tank are set, M×N×4-dimensional standard data sets and M×N×4-dimensional parameter change rate thresholds are collected, and a recognition library is generated;

[0021] M×N×4-dimensional abnormal data is collected under abnormal operation of the hydrogen storage tank, and the abnormal data and the standard data are trained by a neural network model to obtain a tank data fusion model based on a BP neural network; the tank data fusion model based on the BP neural network is embedded into a processing unit to form a data coupling module together with a physical model;

[0022] The multi-parameter data of H points of the hydrogen storage tank are monitored in real time by a perception system; a main parameter is selected from the received multi-parameter data of the H points, and reference values of the multi-parameters of all points are selected from the recognition library according to the main parameter values of the H points and the state of the hydrogen storage tank;

[0023] The multi-parameters of the H points are reversely obtained according to a physical model to obtain reverse multi-parameter data of all points, and the reverse multi-parameter data is coupled to obtain derivative data of the multi-parameters of all points; the real-time change rate of the derivative data of the multi-parameters of all points is calculated;

[0024] When the difference between the derivative data of the multi-parameter of all points and the reference data of the multi-parameter of all points selected from the identification library exceeds the preset threshold, or if the change rate of a certain parameter exceeds the preset threshold of the hydrogen storage tank in this state, a warning is given.

[0025] Further, the establishment of the tank data fusion model based on the BP neural network is specifically:

[0026] Collecting hydrogen storage tank data to obtain a training data set D, the data set D includes a union D1 and a union D2; wherein the union D1 includes multi-parameter data of all points of a standard tank body in M states under normal operation and multi-parameter data of all points of a standard tank body in M states under abnormal operation; the union D2 includes reverse multi-parameter data of all points obtained by reverse calculation of H points in M states under normal operation according to a physical model and reverse multi-parameter data of all points obtained by reverse calculation of H points in M states under abnormal operation according to a physical model;

[0027] Selecting all data D1n1 at point n1 from the union D1 as the target value of the neural network, and selecting all data D2n1 at point n1 from the union D2 as the input data of the neural network to obtain a neural network training data set D n1 ;

[0028] The neural network training data set D n1 Preprocess to obtain a data set F n1 ; divide the data set F n1 into a training set and a test set at random; use the training set data in the data set F n1 , combine the BP neural network algorithm, establish a tank data fusion model based on the BP neural network, and perform model test training.

[0029] Further, the real-time change rate of the derivative data of the multi-parameter of all points is calculated as follows: subtract the single parameter data obtained at the current time from the single parameter data obtained at the previous time to obtain a single parameter change value, and divide the single parameter change value by the time interval to obtain the real-time parameter change rate of the single parameter change value.

[0030] Further, the warning instruction is divided into four levels:

[0031] When the difference between the derivative data of the multi-parameter of all points and the reference data of the multi-parameter of all points selected from the identification library does not exceed 5%, but the change rate of a certain parameter exceeds the preset threshold of the hydrogen storage tank in this state, a first-level warning is given;

[0032] When the difference between the multi-element parameter derivative data of all points and the multi-element parameter reference data of all points selected from the identification library exceeds 5%, but the change rate of a certain parameter does not exceed the preset threshold of the hydrogen storage tank in this state, a secondary early warning is issued;

[0033] When the difference between the multi-element parameter derivative data of all points and the multi-element parameter reference data of all points selected from the identification library exceeds 5%, and the change rate of a certain parameter exceeds the preset threshold of the hydrogen storage tank in this state, a tertiary early warning is issued;

[0034] When the multi-element parameter derivative data of all points is 90% of the safety range limit value of the hydrogen storage tank, and the change rate of a certain parameter exceeds the preset threshold of the hydrogen storage tank in this state, a quaternary early warning is issued.

[0035] The beneficial effects of the present application are:

[0036] The intelligent early warning device and method for the hydrogen storage tank disclosed by the present application can more comprehensively and accurately reflect the health status of the hydrogen storage tank through multi-element parameter sensing, and further improve the comprehensiveness and accuracy of detection through multi-element parameter coupling. In addition, the present application not only compares and analyzes the real-time parameter data and reference parameter data of the hydrogen storage tank, but also creatively analyzes and judges the parameter change rate, further improves the hydrogen storage tank early warning logic, and solves the problem that the existing technology uses a single sensor to monitor a single content to monitor the health of the hydrogen storage tank. The accuracy of health monitoring depends entirely on the detection range and working effect of the single sensor, and it is easy to have false detection, missed detection, false alarm, and thus cause safety hazards in the use of the hydrogen storage tank. BRIEF DESCRIPTION OF DRAWINGS

[0037] 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 the prior art description. The drawings in the following description are 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.

[0038] Figure 1 The intelligent early warning device system framework diagram for the hydrogen storage tank disclosed by the present application.

[0039] Figure 2 The data coupling module flow chart in the processing unit of the embodiment of the present application.

[0040] Figure 3 The logic judgment module flow chart in the processing unit of the embodiment of the present application.

[0041] Figure 4 The structure diagram of the sensing system of the embodiment of the present application.

[0042] Fig.:

[0043] 1 - pressure sensor; 2 - strain sensor; 3 - temperature sensor; 4 - hydrogen sensor. DETAILED DESCRIPTION

[0044] The application will be further described below in conjunction with the drawings and specific examples, but the scope of protection of the application is not limited thereto.

[0045] Embodiments of the application are described in detail below with reference to examples shown in the drawings, wherein the same or similar reference signs represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by reference to the drawings are exemplary and are intended to explain the application, and cannot be understood as a limitation of the application.

[0046] In the description of the application, it should be understood that the orientation or positional relationship indicated by the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "axial", "radial", "vertical", "horizontal", "inner", "outer" and the like are based on the orientation or positional relationship shown in the drawings, and are only for the purpose of facilitating the description of the application and simplifying the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation of the application. In addition, the terms "first", "second" are only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first", "second" can be explicitly or implicitly included one or more of the features. In the description of the application, the meaning of "a plurality of" is two or more, unless otherwise explicitly specified and limited.

[0047] In the present application, unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connection", "fixing" and the like should be broadly understood, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0048] As Figure 1 shown, the intelligent early warning device for hydrogen storage tank according to the present application comprises a sensing system, a processing unit and an early warning system; the sensing system is used for real-time sensing of multi-element parameters of selected H point positions of the hydrogen storage tank, 1<H<N; wherein N is the total number of grid division through the hydrogen storage tank body surface; as Figure 4As shown, the sensing system comprises a pressure sensor 1, a strain sensor 2, a temperature sensor 3 and a hydrogen sensor 4; the pressure sensor 1 is used to detect the pressure of the hydrogen storage tank body, the temperature sensor 3 is used to detect the temperature of the hydrogen storage tank body, and the hydrogen sensor 4 is used to detect whether there is leaked hydrogen gas near the outside of the hydrogen storage tank; a plurality of strain sensors 2 are installed on the hydrogen storage tank, which are used to detect the strain at different positions of the hydrogen storage tank; the pressure sensor 1, the strain sensor 2, the temperature sensor 3 and the hydrogen sensor 4 are respectively connected with the processing unit. In the embodiment, the sensing system is used for real-time sensing of four parameters of selected H points of the hydrogen storage tank.

[0049] The processing unit is connected with the sensing system, and is used to receive real-time parameter data transmitted by the sensing system; the early warning system is connected with the processing unit through a wireless module, and is used to receive early warning signals sent by the processing unit; the processing unit selects a main parameter from the multi-parameter of the H points, and selects reference values of the multi-parameters of all points in the identification library according to the main parameter values of the H points and the state of the hydrogen storage tank; the processing unit obtains the backstepping multi-parameter data of all points by backstepping the multi-parameters of the H points according to the physical model, and obtains the derivative data of the multi-parameters of all points by data coupling, and verifies the derivative data with the reference data of the multi-parameters of all points selected from the identification library; if the difference between the two exceeds a preset threshold, the processing unit sends an early warning instruction to the early warning system; the processing unit calculates and analyzes the change rate of the derivative data of the multi-parameters of all points, and sends an early warning instruction to the early warning system if the change rate of a certain parameter exceeds a preset threshold under the state of the hydrogen storage tank.

[0050] The processing unit comprises an identification library, a reference data selection module, a data coupling module, a parameter change rate generation module, a logic judgment module and a wireless module; the identification library contains reference values of multi-parameter data of all points of the tank body under all states of the normal operation of the hydrogen storage tank; the reference data selection module selects reference data in the identification library according to the selected main parameter and the basic state of the hydrogen storage tank; the data coupling module backsteps the multi-parameter data of the H points received according to the physical model to obtain backstepping data of the multi-parameters of all points of the tank body and performs data fusion to obtain derivative data of the multi-parameters of all points of the tank body; the parameter change rate generation module is used to calculate the real-time change rate of the derivative data of the multi-parameters of all points; the logic judgment module compares the reference data of the multi-parameters of all points with the derivative data of the multi-parameters of all points, logically judges the real-time change rate of the derivative data of the multi-parameters of any point, and sends an early warning instruction to the early warning system through the wireless module according to the judgment result.

[0051] As shown in the figure, Figure 2As shown in the flowchart of the data coupling module in the processing unit of this embodiment of the invention, the processing unit receives the pressure data, temperature data, strain data and hydrogen data of the hydrogen storage tank and inputs them to the multi-element data coupling module. The module filters out obviously abnormal data according to the physical model between each physical quantity and alerts the corresponding monitoring device to report an error. Then, it reverse-engineers the normal data to obtain the data of the entire hydrogen storage tank.

[0052] The early warning system includes an audio-visual module and a wireless receiving and processing module; the wireless receiving and processing module is used to receive early warning commands sent by the processing unit, and the audio-visual module executes the corresponding early warning alarm according to the early warning command.

[0053] like Figure 3 As shown, the method of the intelligent early warning device for hydrogen storage tanks according to the present invention includes the following steps:

[0054] S01: Based on the simulation model mesh generation, the surface of the hydrogen storage tank is divided into N grids, with the center of each grid representing a hydrogen storage tank location; H monitoring points are arbitrarily selected, and pressure sensor 1, strain sensor 2, temperature sensor 3 and hydrogen sensor 4 are installed at each monitoring point to detect the pressure, temperature, strain and whether there is a hydrogen leak in the hydrogen storage tank.

[0055] S02: Standard data collection under normal operation of the hydrogen storage tank. Assuming M operating states of the hydrogen storage tank, and since there are four physical quantities, collect M×N×4 dimensional standard datasets and M×N×4 dimensional parameter change rate thresholds to generate an identification library, and store the identification library in the processing unit; collect M×N×4 dimensional abnormal data of the hydrogen storage tank under abnormal operation. The abnormal data and standard data are trained through a neural network model to obtain a tank data fusion model based on a BP neural network; embed the tank data fusion model based on the BP neural network into the processing unit, forming a data coupling module together with the physical model;

[0056] Data from hydrogen storage tanks is collected to obtain a training dataset D, which includes set D1 and set D2. Set D1 includes multivariate parameter data of all points of the standard tank under M states of normal operation and multivariate parameter data of all points of the standard tank under M states of abnormal operation. Set D2 includes multivariate data of all points obtained by back-deriving the multivariate parameters of H points under M states of normal operation from the physical model and multivariate data of all points obtained by back-deriving the multivariate parameters of H points under M states of abnormal operation from the physical model.

[0057] Select all data points D1n1 at position n1 from set D1 as the target value of the neural network, and select all data points D2n1 at position n1 from set D2 as the input data of the neural network, thus obtaining the neural network training dataset D. n1 ;

[0058] Neural network training dataset D n1 Preprocessing dataset F n1 ; dataset F n1 Randomly divided into training set and test set; using dataset F n1 The training set data in dataset F

[0059] S03: Real-time monitoring of multi-element parameter data of H points of hydrogen storage tank by sensing system; selecting a main parameter in the received multi-element parameters of H points, and selecting reference values of multi-element parameters of all points in the identification library according to the main parameter values of H points and the state of hydrogen storage tank;

[0060] S04: The multi-element data of all points is obtained by backstepping the multi-element parameters of H points according to the physical model, and the multi-element data of all points is obtained by data coupling; calculate the real-time change rate of the derivative data of the multi-element parameters of all points; the specific method for calculating the real-time change rate of the derivative data of the multi-element parameters of all points is: subtracting the single parameter data obtained at the current time from the single parameter data obtained at the last time to obtain the single parameter change value, and dividing the single parameter change value by the time interval to obtain the real-time parameter change rate of the single parameter change value.

[0061] S05: When the difference between the derivative data of the multi-element parameters of all points and the reference data of the multi-element parameters of all points selected in the identification library exceeds the preset threshold, or if the change rate of a certain parameter exceeds the preset threshold of the hydrogen storage tank in this state, an early warning instruction is issued, and the early warning instruction is divided into four levels: when the difference between the derivative data of the multi-element parameters of all points and the reference data of the multi-element parameters of all points selected in the identification library does not exceed 5%, but the change rate of a certain parameter exceeds the preset threshold of the hydrogen storage tank in this state, a first-level early warning is issued;

[0062] When the difference between the derivative data of the multi-element parameters of all points and the reference data of the multi-element parameters of all points selected in the identification library exceeds 5%, but the change rate of a certain parameter does not exceed the preset threshold of the hydrogen storage tank in this state, a second-level early warning is issued;

[0063] When the difference between the derivative data of the multi-element parameters of all points and the reference data of the multi-element parameters of all points selected in the identification library exceeds 5%, and the change rate of a certain parameter exceeds the preset threshold of the hydrogen storage tank in this state, a third-level early warning is issued;

[0064] When the derivative data of the multi-element parameters of all points is 90% of the safety range limit value of the hydrogen storage tank, and the change rate of a certain parameter exceeds the preset threshold of the hydrogen storage tank in this state, a fourth-level early warning is issued.

[0065] It should be understood that although the present specification is described in terms of various embodiments, each of which describes only one implementation, the specification is intended to cover all possible combinations for each independent hardware or software feature and its alternatives. A person skilled in the art should consider the specification as a whole and the technical solutions in each embodiment can be combined with each other to form other embodiments which can be understood by a person skilled in the art.

[0066] The above detailed description of a series of specific embodiments is only for the feasibility of the present application, and is not intended to limit the protection scope of the present application. Any equivalent embodiments or changes made without departing from the spirit of the present application shall be included in the protection scope of the present application.

Claims

1. An intelligent early warning device for a hydrogen storage tank, characterized in that, Including: A sensing system for real-time multi-parameter sensing of H selected points on the hydrogen storage tank, where 1 < H < N; N is the total number of grid divisions on the surface of the hydrogen storage tank body; A processing unit connected to the sensing system for receiving real-time parameter data transmitted by the sensing system; An early warning system connected to the processing unit through a wireless module for receiving early warning signals sent by the processing unit; Wherein, The processing unit selects a main parameter from the multi-parameters of the H points received. The processing unit selects the reference values of the multi-parameters of all points in the recognition library according to the main parameter values of the H points and the state of the hydrogen storage tank. The processing unit inversely deduces the inverse multi-data of all points based on the multi-parameters of the H points according to the physical model. The processing unit performs data coupling on the inverse multi-data to obtain the derivative multi-parameter data of all points, and mutually verifies it with the reference data of the multi-parameters of all points selected from the recognition library. If the difference between the two exceeds the preset threshold, the processing unit sends an early warning instruction to the early warning system. The processing unit calculates and analyzes the change rate of the derivative data of the multi-parameters of all points. If the change rate of a certain parameter exceeds the preset threshold under this state of the hydrogen storage tank, the processing unit sends an early warning instruction to the early warning system.

2. The intelligent early warning device for hydrogen storage tanks according to claim 1, characterized in that, The sensing system includes a pressure sensor (1), a strain sensor (2), a temperature sensor (3), and a hydrogen sensor (4); the pressure sensor (1) is used to detect the pressure of the hydrogen storage tank body, the temperature sensor (3) is used to detect the temperature of the hydrogen storage tank body, the hydrogen sensor (4) is used to detect whether there is leaked hydrogen near the outside of the hydrogen storage tank; several strain sensors (2) are installed on the hydrogen storage tank for detecting the strain at different point positions of the hydrogen storage tank; the pressure sensor (1), the strain sensor (2), the temperature sensor (3), and the hydrogen sensor (4) are respectively connected to the processing unit.

3. The intelligent early warning device for hydrogen storage tanks according to claim 1, characterized in that, The processing unit includes a recognition library, a reference data selection module, a data coupling module, a parameter change rate generation module, a logic judgment module, and a wireless module; The recognition library contains the reference values of the multi-parameter data of all points on the tank body in all states during the normal operation of the hydrogen storage tank; The reference data selection module selects reference data from the recognition library according to the selected main parameter and the basic state of the hydrogen storage tank; The data coupling module inversely deduces the inverse multi-parameter data of all points on the tank body based on the multi-parameter data of the H points received according to the physical model and performs data fusion to obtain the derivative multi-parameter data of all points on the tank body; the parameter change rate generation module is used to calculate the real-time change rate of the derivative data of the multi-parameters of all points; the logic judgment module makes a comparison logic judgment between the reference data of the multi-parameters of all points and the derivative data of the multi-parameters of all points, makes a logic judgment on the real-time change rate of the derivative data of the multi-parameters of any one point, and sends an early warning instruction to the early warning system through the wireless module according to the judgment result.

4. The intelligent early warning device for hydrogen storage tanks according to claim 1, characterized in that, The early warning system includes an audio-visual module and a wireless receiving and processing module; the wireless receiving and processing module is used to receive early warning commands sent by the processing unit, and the audio-visual module executes the corresponding early warning alarm according to the early warning command.

5. A warning method for an intelligent early warning device for a hydrogen storage tank according to any one of claims 1-4, characterized in that, Includes the following steps: Based on the simulation model mesh generation, the surface of the hydrogen storage tank is divided into N grids, with the center of each grid representing a hydrogen storage tank location; H monitoring points are arbitrarily selected, and at least one sensor is installed at each monitoring point to detect the pressure, temperature, strain, and presence of hydrogen leakage in the hydrogen storage tank. Standard data collection under normal operation of hydrogen storage tanks: Assuming M operating states of hydrogen storage tanks, collect M×N×4 dimensional standard datasets and M×N×4 dimensional parameter change rate thresholds to generate an identification library; Abnormal M×N×4 dimensional data of hydrogen storage tanks under abnormal operation is collected. Abnormal data and standard data are trained through a neural network model to obtain a tank data fusion model based on a BP neural network. The tank data fusion model based on the BP neural network is embedded into the processing unit and forms a data coupling module together with the physical model. The sensing system monitors the multivariate parameter data of H points on the hydrogen storage tank in real time; a main parameter is selected from the received multivariate parameters of the H points, and reference values ​​of the multivariate parameters of all points are selected from the identification library based on the values ​​of the main parameter of the H points and the status of the hydrogen storage tank. The multivariate parameters of H points are back-inferred from the physical model to obtain the back-inferred multivariate data of all points. The back-inferred multivariate data is then coupled to obtain the multivariate parameter derived data of all points. Calculate the real-time rate of change of the derived data of multivariate parameters for all points; An early warning will be issued when the difference between the multivariate parameter derived data of all points and the multivariate parameter reference data of all points selected in the identification library exceeds a preset threshold, or when the rate of change of a certain parameter exceeds the preset threshold of the hydrogen storage tank under that state.

6. The early warning method of the intelligent early warning device for hydrogen storage tanks according to claim 5, characterized in that, The establishment of the tank data fusion model based on BP neural network is specifically as follows: Data from hydrogen storage tanks is collected to obtain a training dataset D, which includes set D1 and set D2. Set D1 includes multivariate parameter data of all points of the standard tank under M states of normal operation and multivariate parameter data of all points of the standard tank under M states of abnormal operation. Set D2 includes multivariate data of all points obtained by back-deriving the multivariate parameters of H points under M states of normal operation from the physical model and multivariate data of all points obtained by back-deriving the multivariate parameters of H points under M states of abnormal operation from the physical model. Select all data points D1n1 at position n1 from set D1 as the target value of the neural network, and select all data points D2n1 at position n1 from set D2 as the input data of the neural network, thus obtaining the neural network training dataset D. n1 ; Neural network training dataset D n1 Preprocessing yields dataset F n1 ; The dataset F n1 The dataset is randomly divided into training and test sets; using dataset F n1 Using the training set data and the BP neural network algorithm, a tank data fusion model based on the BP neural network was established and tested.

7. The early warning method of the intelligent early warning device for hydrogen storage tanks according to claim 5, characterized in that, The specific method for calculating the real-time rate of change of the multivariate parameters of all points is as follows: subtract the current single parameter data from the single parameter data obtained in the previous time to obtain the single parameter change value, and then divide the single parameter change value by the time interval to obtain the real-time parameter change rate of the single parameter change value.

Images