System and method for realizing hydrogen hydrogenation machine performance test conditions

Abstract

The invention provides a system and a method for realizing performance test conditions of a hydrogenation machine. The system comprises an explosion-proof cabin; and the main control unit controls the refrigeration regulation and control valve to regulate the environment temperature in the explosion-proof cabin body. And the main control unit controls the heating regulation and control valve to regulate the environment temperature in the explosion-proof cabin body. And the main control unit controls the pressure regulation and control valve to change the gas quantity and speed of the inert gas entering the explosion-proof cabin body so as to change the pressure environment in the explosion-proof cabin body. And the main control unit forms a control signal according to monitoring results of the sensors and the environment monitoring unit, and controls the hydrogen concentration, the environment pressure and the environment temperature in the explosion-proof cabin body to be respectively kept in corresponding preset ranges, so as to ensure that the testing environment not only meets the national standard requirement, but also ensures the safety.

Description

Technical Field

[0001] This application relates to the field of hydrogen energy safety engineering testing and evaluation technology, specifically, to a system and method for realizing the performance testing conditions of a hydrogen refueling machine. Background Technology

[0002] Hydrogen refueling machines are key equipment in hydrogen refueling stations, used to dispense high-pressure hydrogen from storage containers into onboard hydrogen storage tanks. Their refueling performance is a crucial indicator for evaluating hydrogen refueling stations. The refueling performance of a hydrogen refueling machine is mainly reflected in three technical indicators: the temperature of the onboard hydrogen storage tank, the pressure within the onboard hydrogen storage tank, and the refueling rate. Key factors affecting the refueling performance of a hydrogen refueling machine include ambient temperature, hydrogen fuel temperature, the volume of the onboard hydrogen storage tank, the initial pressure of the onboard hydrogen storage tank, and the type of onboard hydrogen storage tank. When testing the refueling performance of a hydrogen refueling machine, the other four conditions, except for ambient temperature, are relatively easy to control. According to compliance requirements, the refueling performance of a hydrogen refueling machine should be tested within the upper and lower limits of ambient temperature (-25℃ to 40℃). The ambient temperature parameters required for the performance test must be adjusted within a closed space. Since hydrogen itself is flammable and explosive, the test environment must meet the requirements while ensuring safety during the testing process. Therefore, the following three issues need to be addressed regarding the control of the ambient temperature in the closed environment required for the performance testing of hydrogen refueling machines:

[0003] (1) Thermal insulation, flame retardant and explosion prevention in closed environment: Hydrogen is a flammable and explosive gas with a wide range of combustion concentrations. If it leaks in a closed environment, it can easily cause safety problems. In case of combustion and explosion, reliable flame retardant and explosion prevention measures are needed to avoid the impact on the surrounding environment.

[0004] (2) Simultaneously possessing high and low temperature environment realization functions: Tests such as hydrogenation machine performance testing often need to be carried out at high and low temperatures respectively, requiring no risky substances to be introduced into the working fluid during the cooling and heating process, and a feasible high and low temperature environment realization system with a simple and energy-saving working fluid supply system.

[0005] (3) Safety assurance of closed hydrogen-related environment: Tests such as hydrogen refueling machine performance testing need to be carried out in a closed environment with high or low temperature. Hydrogen is prone to leakage, and dangerous phenomena such as overpressure and hydrogen concentration exceeding the limit are relatively easy to occur, posing safety hazards such as fire and explosion. Summary of the Invention

[0006] The technical problem to be solved by this application is that the existing test conditions for testing the performance of hydrogen refueling machines are difficult to meet both the requirements of airtightness and safety. Therefore, this application provides a system and method to achieve the test conditions for hydrogen refueling machine performance.

[0007] Firstly, the technical solution of this application provides a system for achieving performance testing conditions for a hydrogen refueling machine, comprising:

[0008] An explosion-proof chamber is installed outside the hydrogen refueling machine to be tested. The explosion-proof chamber includes an inner explosion-proof plate, an outer explosion-proof plate, and a flame-retardant interlayer disposed between the inner explosion-proof plate and the outer explosion-proof plate. An explosion-proof valve is installed on the explosion-proof chamber.

[0009] The refrigeration control unit includes a refrigeration medium transmission pipeline, a refrigeration control valve, a refrigeration pipeline status sensor, and a refrigeration execution pipeline, wherein the refrigeration execution pipeline is a refrigeration pipeline installed inside the explosion-proof cabin.

[0010] The heating control unit includes a heating medium transmission pipeline, a heating control valve, a heating pipeline status sensor, and a heating execution pipeline, wherein the heating execution pipeline is a heating pipeline installed inside the explosion-proof cabin.

[0011] The pressure control unit includes an inert gas medium transmission pipeline, an inert gas medium transmission pipeline status sensor, and a pressure control valve. The outlet end of the inert gas transmission pipeline is connected to the interior of the explosion-proof cabin.

[0012] An environmental monitoring unit is used to monitor the hydrogen concentration, ambient pressure, and ambient temperature inside the explosion-proof cabin.

[0013] The main control unit receives monitoring results from the environmental monitoring unit, the refrigeration pipeline status sensor, the heating pipeline status sensor, and the inert gas medium transmission pipeline status sensor. Based on the monitoring results, it controls at least one of the refrigeration control valve, the heating control valve, and the pressure control valve to maintain the hydrogen concentration, environmental pressure, and environmental temperature inside the explosion-proof cabin within their respective preset ranges.

[0014] Preferably, in the system for achieving the performance testing conditions of the hydrogen refueling machine, the refrigeration control unit includes a first refrigeration valve, a second refrigeration valve, a first flow controller, and a third refrigeration valve; the refrigeration pipeline status sensor includes a first pressure sensor and a first temperature sensor, wherein:

[0015] The refrigerant transmission pipeline is connected to a storage tank, which stores the refrigerant, which is a liquid medium of the inert gas.

[0016] The first end of the first refrigeration valve is connected to the liquid outlet of the storage tank;

[0017] The second end of the first refrigeration valve is connected to the vaporizer, and the second end of the vaporizer is connected to the pressure control unit;

[0018] The first end of the second refrigeration valve is connected to the second end of the vaporizer, and the second end of the second refrigeration valve is connected to the first end of the first flow controller;

[0019] The second end of the first flow controller is connected to the first end of the refrigeration execution pipeline, and the second end of the refrigeration execution pipeline is connected to the inert gas recovery device via the third refrigeration valve;

[0020] The first pressure sensor and the first temperature sensor are disposed between the second end of the first flow controller and the first end of the refrigeration actuation pipeline.

[0021] Preferably, the system for achieving the performance testing conditions of the hydrogen refueling machine,

[0022] In the heating control unit, the heating control valve includes a first heating valve, a second flow controller, a second heating valve, and a third flow controller; the heating pipeline status sensor includes a second pressure sensor, a third pressure sensor, a second temperature sensor, and a third temperature sensor, wherein:

[0023] The heating medium transmission pipeline is connected to the heat source box, which is used to store the heating medium.

[0024] The first end of the first heating valve is connected to the liquid outlet of the heat source box, and the second end of the first heating valve is connected to the first end of the heating execution pipeline after passing through the second flow controller;

[0025] The first end of the second heating valve is connected to the second end of the heating execution pipeline, and the second end of the second heating valve is connected to the liquid inlet of the heat source box after passing through the third flow controller;

[0026] The second pressure sensor and the second temperature sensor are disposed between the heat source box and the first end of the heating execution pipeline; the third pressure sensor and the third temperature sensor are disposed between the second end of the second heating valve and the liquid inlet of the heat source box.

[0027] Preferably, the system for achieving the performance testing conditions of the hydrogen refueling machine,

[0028] The heat source box includes a box body, a first speed control valve located at the box body outlet, a heater, and a second speed control valve located at the box body inlet;

[0029] The heating control unit also includes a speed regulating pump, which is disposed between the heater and the first heating valve.

[0030] Preferably, in the system for achieving the performance testing conditions of the hydrogen refueling machine, the pressure control unit includes a pressure regulating valve and a fifth flow controller, and the inert gas medium transmission pipeline status sensor includes a fourth pressure sensor and a fourth temperature sensor, wherein:

[0031] The first end of the pressure regulating valve is connected to the second end of the vaporizer, the second end of the pressure regulating valve is connected to the first end of the fifth flow controller, and the second end of the fifth flow controller is connected to the interior of the explosion-proof chamber via a pipeline;

[0032] The fourth pressure sensor and the fourth temperature sensor are located between the second end of the fifth flow controller and the explosion-proof chamber.

[0033] Preferably, in the system for achieving the performance testing conditions of the hydrogen refueling machine, the environmental monitoring unit includes a hydrogen concentration sensing probe, a pressure sensor probe, and a temperature sensor probe disposed inside the explosion-proof chamber. The hydrogen concentration sensing probe detects the hydrogen concentration inside the explosion-proof chamber, the pressure sensor probe detects the environmental pressure inside the explosion-proof chamber, and the temperature sensor probe detects the environmental temperature inside the explosion-proof chamber.

[0034] The hydrogen concentration sensing probe, the pressure sensor probe, and the temperature sensor probe are provided with explosion-proof sealing interfaces at the positions where they contact the explosion-proof cabin.

[0035] Preferably, in the system for achieving the performance testing conditions of the hydrogen refueling machine, the explosion-proof chamber is equipped with an explosion-proof door;

[0036] The explosion-proof door includes a door body, and a fixing plate is pressed onto the edge of the door body. The fixing plate is pressed between the outer explosion-proof plate and the inner explosion-proof plate of the explosion-proof compartment.

[0037] Preferably, the system for achieving the performance testing conditions of the hydrogen refueling machine further includes:

[0038] The venting control unit includes a gas venting pipeline and a venting control valve;

[0039] One end of the gas venting pipeline is connected to the interior of the explosion-proof cabin, and the venting control valve is located between the gas venting pipeline and the explosion-proof cabin;

[0040] The main control unit is also used to control the operation of the discharge regulating valve.

[0041] Secondly, this application provides a method for implementing hydrogen refueling machine performance testing conditions, applied in the main control unit of the system for implementing hydrogen refueling machine performance testing conditions as described in any of the technical solutions of the first aspect, including:

[0042] The working status data of the refrigerant transmission pipeline monitored by the refrigeration pipeline status sensor, the working status data of the heating pipeline monitored by the heating pipeline status sensor, and the working status data of the inert gas medium transmission pipeline monitored by the inert gas medium transmission pipeline status sensor are obtained as sample input data within a historical time period.

[0043] The hydrogen concentration, environmental pressure, and environmental temperature inside the explosion-proof cabin, as monitored by the environmental monitoring unit corresponding to the sample data during the historical time period, are obtained as sample output data.

[0044] The sample input data and the sample output data are merged according to the condition that the collection time is the same to obtain the key variable data grouping result of the internal environment control of the explosion-proof cabin. Each group is a time series data.

[0045] The grouping results are divided into a training dataset and a test dataset to obtain the training dataset of the initial learning model. The initial learning model includes an anomaly detection initial model and an anomaly prediction initial model.

[0046] The training dataset is input into the anomaly monitoring initial model to learn and train the monitoring results of the refrigeration pipeline status sensor, the heating pipeline status sensor, the inert gas medium transmission pipeline status sensor, and the environmental monitoring unit to detect anomalies.

[0047] The training dataset is input into the anomaly prediction initial model to learn and train the prediction of anomalies in the monitoring results of the environmental monitoring unit.

[0048] The initial models for anomaly monitoring and anomaly prediction, which have completed the learning and training, will be used as the control models for the hydrogen refueling machine test environment.

[0049] Preferably, in the method for achieving the performance testing conditions of the hydrogen refueling machine:

[0050] The initial model for anomaly monitoring includes:

[0051] The input training dataset is processed through two time block structures in sequence. Based on a predetermined period, the one-dimensional time series data is decomposed into a two-dimensional time series data matrix that includes a trend component and a short-term fluctuation component.

[0052] A convolutional neural network is used to extract the temporal trend features from the two-dimensional time-series data matrix output by two time-block structures.

[0053] The deep learning algorithm structure extracts all time trend features from the output of the convolutional neural network and performs summation to obtain the time series prediction result.

[0054] An anomaly detection output structure is generated by obtaining a reasonable threshold based on the test dataset in the training dataset, performing anomaly detection using the reasonable threshold and the time-series prediction result, and outputting the anomaly monitoring result.

[0055] The initial model for anomaly prediction includes:

[0056] The historical time series construction module extracts the historical time series features of the time series prediction results;

[0057] The relevant time series construction module concatenates the historical time features, uses a scaled reset matrix and performs position encoding on a scaled basis, and then passes through a multi-head attention module and a bidirectional cross-attention module to extract inter-time series features and intra-time series features respectively, and determines the degree of influence of each time series feature on the relevant time series features.

[0058] The relevant real-time construction module performs real-time correction on the internal features of each time series based on the introduced real-time relevant sensor data;

[0059] The prediction result output module sums the output results of the historical time series construction module, the relevant time series construction module, and the relevant real-time construction module and outputs the sum. It then determines the prediction result, performs anomaly monitoring on the prediction result, and outputs the anomaly prediction result.

[0060] The technical solution provided in this application has the following technical effects compared with the prior art:

[0061] The system and method provided in this application for achieving performance testing conditions for a hydrogen refueling machine include an explosion-proof chamber located outside the hydrogen refueling machine under test. The explosion-proof chamber comprises an inner explosion-proof plate, an outer explosion-proof plate, and a flame-retardant interlayer between the inner and outer explosion-proof plates, creating a sealed environment with explosion-proof, flame-retardant, and heat-insulating functions. In the refrigeration control unit, the refrigerant flows through a refrigerant transmission pipeline to the refrigeration execution pipeline within the explosion-proof chamber. A refrigeration pipeline status sensor monitors the refrigeration pipeline status and sends the data to the main control unit. The main control unit controls the refrigeration control valve to change parameters such as the flow rate and velocity of the refrigerant entering the refrigeration execution pipeline, thereby regulating the ambient temperature inside the explosion-proof chamber. Similarly, in the heating control unit, the heating medium flows through a heating medium transmission pipeline to the heating execution pipeline inside the explosion-proof chamber. A heating pipeline status sensor monitors the heating pipeline status and sends the data to the main control unit. The main control unit controls the heating control valve to change parameters such as the flow rate and velocity of the heating medium entering the heating execution pipeline, thereby regulating the ambient temperature inside the explosion-proof chamber. In the pressure control unit, an inert gas medium transmission pipeline transmits inert gas into the explosion-proof chamber. An inert gas medium transmission pipeline status sensor sends the operating status of the inert gas medium transmission pipeline to the main control unit. The main control unit controls the pressure control valve to change the amount and velocity of inert gas entering the explosion-proof chamber, thereby altering the pressure environment inside the chamber. This creates a slight positive pressure inside the explosion-proof chamber, preventing air from entering. The environmental monitoring unit monitors the hydrogen concentration, ambient pressure, and ambient temperature inside the explosion-proof chamber and sends the monitoring results to the main control unit. The main control unit generates control signals based on the monitoring results from each sensor and the environmental monitoring unit, thereby controlling the cooling control valve, heating control valve, and pressure control valve. This ensures that the hydrogen concentration, ambient pressure, and ambient temperature inside the explosion-proof chamber are maintained within their respective preset ranges, thus ensuring that the testing environment meets national standards and guarantees safety. Attached Figure Description

[0062] Figure 1 This is a schematic diagram of the system structure for implementing the performance testing conditions of a hydrogen refueling machine according to one embodiment of this application;

[0063] Figure 2 This is a schematic diagram of the system for implementing the performance testing conditions of a hydrogen refueling machine according to another embodiment of this application;

[0064] Figure 3 This is a flowchart illustrating a method for implementing hydrogen refueling machine performance testing conditions according to one embodiment of this application;

[0065] Figure 4 This is a schematic diagram of the internal network structure of the anomaly monitoring initial model according to one embodiment of this application;

[0066] Figure 5This is another schematic diagram of the internal network structure of the anomaly prediction initial model described in one embodiment of this application. Detailed Implementation

[0067] The specific embodiments of this application will be further described below with reference to the accompanying drawings.

[0068] It is readily understood that, based on the technical solution of this application, various structural and implementation methods can be interchanged by those skilled in the art without altering the essential spirit of this application. Therefore, the following detailed embodiments and accompanying drawings are merely illustrative examples of the technical solution of this application and should not be considered as the entirety of this application or as limitations or restrictions on the technical solution of the application.

[0069] This embodiment provides a system for achieving the performance testing conditions of a hydrogen refueling machine, such as... Figure 1 As shown, it includes:

[0070] An explosion-proof chamber 100 is installed outside the hydrogen refueling machine to be tested. The explosion-proof chamber 100 includes an inner explosion-proof plate 101, an outer explosion-proof plate 102, and a flame-retardant interlayer 103 disposed between the inner explosion-proof plate 101 and the outer explosion-proof plate 102. An explosion-proof door 104 is provided on the explosion-proof chamber 100.

[0071] The refrigeration control unit 200 includes a refrigeration medium transmission pipeline 201, a refrigeration control valve 202, a refrigeration pipeline status sensor 203, and a refrigeration execution pipeline 204, wherein the refrigeration execution pipeline 204 is a refrigeration pipeline installed inside the explosion-proof cabin 100.

[0072] The heating control unit 300 includes a heating medium transmission pipeline 301, a heating control valve 302, a heating pipeline status sensor 303, and a heating execution pipeline 304, wherein the heating execution pipeline is a heating pipeline installed inside the explosion-proof cabin.

[0073] The pressure control unit 400 includes an inert gas medium transmission pipeline 401, a pressure control valve 402, and an inert gas medium transmission pipeline status sensor 403. The outlet end of the inert gas transmission pipeline is connected to the interior of the explosion-proof cabin 100.

[0074] An environmental monitoring unit 500 is used to monitor the hydrogen concentration, ambient pressure, and ambient temperature inside the explosion-proof chamber 100. The interior of the explosion-proof chamber 100 serves as the testing environment.

[0075] The main control unit 600 receives monitoring results from the environmental monitoring unit 500, the refrigeration pipeline status sensor 203, the heating pipeline status sensor 303, and the inert gas medium transmission pipeline status sensor 403. Based on the monitoring results, it controls at least one of the refrigeration control valve 202, the heating control valve 302, and the pressure control valve 402 to maintain the hydrogen concentration, environmental pressure, and environmental temperature inside the explosion-proof cabin 100 within their respective preset ranges.

[0076] The preset range is determined based on the performance test conditions of the hydrogen refueling machine. Specifically, the lower limit of the ambient temperature is -25℃ and the upper limit is 40℃. Other conditions need to be determined according to the requirements of GB / T31138 2022 "Hydrogen Refueling Machine" standard.

[0077] The above-described solution provided in this embodiment includes an explosion-proof chamber 100, which is located outside the hydrogen refueling machine under test. Since the explosion-proof chamber 100 includes an inner explosion-proof plate 101, an outer explosion-proof plate 102, and a flame-retardant interlayer 103 disposed between the inner and outer explosion-proof plates, it can achieve a sealed environment with explosion-proof, flame-retardant, and heat-insulating functions. In the refrigeration control unit 200, the refrigerant can flow through the refrigerant transmission pipeline 201 to the refrigeration execution pipeline 204 inside the explosion-proof chamber 100. The refrigeration pipeline status sensor 203 can monitor the status of the refrigeration pipeline and send it to the main control unit 600. The main control unit 600 can control the refrigeration control valve 202 to change parameters such as the flow rate and velocity of the refrigerant entering the refrigeration execution pipeline 204, thereby regulating the ambient temperature inside the explosion-proof chamber 100. Similarly, in the heating control unit 300, the heating medium can flow through the heating medium transmission pipeline 301 to the heating execution pipeline 304 inside the explosion-proof cabin 100. The heating pipeline status sensor 303 can monitor the status of the heating pipeline and send it to the main control unit 600. The main control unit 600 can control the heating control valve 302 to change the flow rate, flow velocity and other parameters of the heating medium entering the heating execution pipeline 304, thereby regulating the ambient temperature inside the explosion-proof cabin 100. In the pressure control unit 400, the inert gas medium transmission pipeline 401 can transmit inert gas to the explosion-proof chamber 100. The inert gas medium transmission pipeline status sensor 403 can send the working status of the inert gas medium transmission pipeline 401 to the main control unit 600. The main control unit 600 controls the pressure control valve 402 to change the amount and speed of inert gas entering the explosion-proof chamber 100, thereby changing the pressure environment inside the explosion-proof chamber 100. This can create a slight positive pressure inside the explosion-proof chamber 100, preventing air from entering the explosion-proof chamber 100. The environmental monitoring unit 500 monitors the hydrogen concentration, ambient pressure, and ambient temperature inside the explosion-proof chamber and sends the monitoring results to the main control unit 600. The main control unit generates control signals based on the monitoring results from each sensor and the environmental monitoring unit, thereby controlling the cooling control valve 202, the heating control valve 302, and the pressure control valve 402. This ensures that the hydrogen concentration, ambient pressure, and ambient temperature inside the explosion-proof chamber 100 are maintained within their respective preset ranges, guaranteeing that the testing environment meets national standards while ensuring safety. Through the above-described solution in this embodiment, combined with… Figure 2 As shown, the hydrogen supply terminal 10 is located outside the explosion-proof chamber 100. After receiving the hydrogen supplied by the hydrogen supply terminal 10, the hydrogen refueling machine 20 adds it to each vehicle-mounted gas storage cylinder 30, thereby ensuring that both the hydrogen refueling machine 20 and the vehicle-mounted gas storage cylinder 30 are located in the test environment. The various units in this solution work together to achieve functions such as sealing, heat preservation, explosion protection, and high and low temperature environment creation, solving various difficulties faced in the performance testing of the hydrogen refueling machine.

[0078] In the above scheme, the inner explosion-proof plate 101 and the outer explosion-proof plate 102 are made of steel plates, and the thickness of the steel plates meets the explosion-proof requirements of hydrogen environments. The flame-retardant interlayer 103 is filled with flame-retardant materials, such as nitrile rubber (NBR), fluororubber (FKM), polytetrafluoroethylene (PTFE), and polyphenylene sulfide (PPS). The cooling actuation pipeline 204 and the heating actuation pipeline 304 can be arranged alternately inside the explosion-proof compartment 100, preferably in the form of coils. In order to achieve cooling and heating efficiency, the larger the arrangement area of ​​the coils, the better, and there should be a certain interval between the cooling coils and the heating coils to avoid affecting the cooling or heating efficiency. Figure 1 The example described is based on the refrigeration actuator 204 and the heating actuator 304 being located on the inside sides of the explosion-proof enclosure 100, respectively. In actual application, the refrigeration actuator 204 and the heating actuator 304 can also be added to the upper or lower parts. If a fixed support assembly is installed in the hollow position of the explosion-proof enclosure 100, the refrigeration actuator 204 and the heating actuator 304 can also be placed in the hollow position.

[0079] In practical implementation, the refrigerant and heating medium can be selected from existing publicly available media, either liquid or gaseous. In this application, liquid nitrogen or nitrogen gas is preferred as the refrigerant, and steam is preferred as the heating medium. If liquid nitrogen is used as the refrigerant, combined with coil heat exchange to cool the internal space of the explosion-proof cabin 100, no other power equipment is required. Moreover, liquid nitrogen, after vaporization, is an inert gas and can be used simultaneously in the pressure regulation unit 400 for gas replacement and maintaining a slight positive pressure in the sealed environment within the explosion-proof cabin 100. Cooling and pressure regulation can be supplied from the same medium source, making the system configuration simpler and easier to implement. In practical implementation, such as... Figure 2As shown, in the refrigeration control unit 200, the refrigeration control valve 202 includes a first refrigeration valve 2021, a second refrigeration valve 2022, a first flow controller 2023, and a third refrigeration valve 2024. The refrigeration pipeline status sensor 203 includes a first pressure sensor 2031 and a first temperature sensor 2032. The refrigeration medium transmission pipeline 201 is connected to a storage tank, which stores the refrigeration medium, which is a liquid state medium of the inert gas. In this scheme, it is liquid nitrogen. The first end of the first refrigeration valve 2021 is connected to the outlet of the storage tank; the second end of the first refrigeration valve 2021 is connected to the vaporizer 40, and the second end of the vaporizer 40 is connected to the pressure control unit 400; the first end of the second refrigeration valve 2022 is connected to the second end of the vaporizer 40, and the second end of the second refrigeration valve 2022 is connected to the first end of the first flow controller 2023; the second end of the first flow controller 2023 is connected to the first end of the refrigeration execution pipeline 204, and the second end of the refrigeration execution pipeline 204 is connected to the inert gas recovery device 205 via the third refrigeration valve 2024; the first pressure sensor 2031 and the first temperature sensor 2032 are disposed between the second end of the first flow controller 2023 and the first end of the refrigeration execution pipeline 204. More preferably, the pressure regulating valve 402 includes a pressure regulating valve 4021 and a fifth flow controller 4022, and the inert gas medium transmission pipeline status sensor 403 includes a fourth pressure sensor 4031 and a fourth temperature sensor 4044, wherein: the first end of the pressure regulating valve 4021 is connected to the second end of the vaporizer 40, the second end of the pressure regulating valve 4021 is connected to the first end of the fifth flow controller 4022, and the second end of the fifth flow controller 4022 is connected to the interior of the explosion-proof chamber 100 via a pipeline; the fourth pressure sensor 4031 and the fourth temperature sensor 4032 are disposed between the second end of the fifth flow controller 4022 and the explosion-proof chamber 100. (Comparison) Figure 2 and Figure 1 As shown, Figure 2 The system only requires a storage tank to store liquid nitrogen, along with a vaporizer 40, to provide the refrigeration control unit 200 and the pressure control unit 400 with the required medium, thus simplifying the overall system structure.

[0080] Furthermore, such as Figure 2As shown, in the system for achieving the performance testing conditions of the hydrogen refueling machine, the heating control unit 300 includes a first heating valve 3021, a second flow controller 3022, a second heating valve 3023, and a third flow controller 3024. The heating pipeline status sensor 303 includes a second pressure sensor 3031, a second temperature sensor 3032, a third pressure sensor 3033, and a third temperature sensor 3034. The heating medium transmission pipeline is connected to a heat source box, which stores the heating medium. The first end of the first heating valve 3021 is connected to the outlet of the heat source box, and the second end of the first heating valve 3021 is connected to the first end of the heating execution pipeline 304 via the second flow controller 3022. The first end of the second heating valve 3023 is connected to the second end of the heating execution pipeline 304, and the second end of the second heating valve 3023 is connected to the inlet of the heat source box via the third flow controller 3024. The second pressure sensor 3031 and the second temperature sensor 3032 are located between the heat source box and the first end of the heating execution pipeline 304. The third pressure sensor 3033 and the third temperature sensor 3034 are located between the second end of the second heating valve 3022 and the inlet of the heat source box. The heating medium in the heat source box can be any known medium capable of achieving rapid heat release, such as steam. After the steam enters the heating execution pipeline 304, it comes into contact with the internal environment of the explosion-proof chamber 100 and is cooled to form a liquid, which then flows directly back to the heat source box. Steam heating is a mature and low-cost technology.

[0081] Furthermore, the heat source box includes a box body 3011, a first speed control valve 3012 disposed at the box body outlet, a heater 3013, and a second speed control valve 3014 disposed at the box body inlet; the heating control unit 300 further includes a speed control pump 305, which is disposed between the heater 3013 and the first heating valve 3021. The main control unit 600 can control the amount of steam entering the heating execution pipeline 304 and the steam flow rate by controlling the speed control pump 305, the first speed control valve 3012, and the second speed control valve 3014, thereby adjusting the rate at which the internal ambient temperature of the explosion-proof cabin 100 changes.

[0082] Preferably, the environmental monitoring unit 500 includes a hydrogen concentration sensing probe, a pressure sensor probe, and a temperature sensor probe disposed inside the explosion-proof chamber. The hydrogen concentration sensing probe detects the hydrogen concentration inside the explosion-proof chamber and feeds it back to the hydrogen concentration detector 501. The pressure sensor probe detects the ambient pressure inside the explosion-proof chamber and feeds it back to the pressure detector 502. The temperature sensor probe detects the ambient temperature inside the explosion-proof chamber and feeds it back to the temperature detector 503. Explosion-proof sealed interfaces are provided at the positions where the hydrogen concentration sensing probe, the pressure sensor probe, and the temperature sensor probe contact the explosion-proof chamber. The hydrogen concentration detector 501, the pressure detector 502, and the temperature detector 503 are all communicatively connected to the main control unit 600 and can send corresponding data to the main control unit 600.

[0083] In the above solution, the explosion-proof cabin 100 is equipped with an explosion-proof door 104; the explosion-proof door 104 includes a door body, and a fixing plate is pressed onto the edge of the door body. The fixing plate is pressed between the outer explosion-proof plate and the inner explosion-proof plate of the explosion-proof cabin 100. Through this solution, most of the area of ​​the fixing plate is sandwiched between the inner and outer explosion-proof plates, which can ensure the sealing performance and reliability of the explosion-proof door installation structure.

[0084] More preferably, such as Figure 2 As shown, the system for achieving the performance testing conditions of the hydrogen refueling machine further includes a venting control unit, comprising a gas venting pipeline and a venting control valve 701; one end of the gas venting pipeline is connected to the interior of the explosion-proof chamber 100, and the venting control valve 701 is disposed between the gas venting pipeline and the explosion-proof chamber 100; the main control unit 600 is also used to control the operation of the venting control valve 701, for example, controlling the venting control valve to open when the gas inside the explosion-proof chamber 100 needs to be vented after the test is completed. Preferably, the system further includes an exhaust pipeline, comprising an exhaust valve 801 and a blower 802, the exhaust valve 801 being connected to the explosion-proof chamber 100, and the blower 802 being connected to the exhaust valve 801. Both can be controlled by the main control unit 600 to form an emergency venting pipeline. This system is equivalent to having multiple venting structures. The exhaust valve 801 and the fan 802 serve as an emergency venting system, while the venting control valve 701 can ultimately release all the gas inside the explosion-proof cabin 100. The main control unit 600 controls and adjusts the system according to different needs such as overpressure, hydrogen concentration exceeding the limit, and normal venting.

[0085] Combination Figure 2 This application provides a specific testing process, which includes:

[0086] The pressure control unit 200 controls the slight positive pressure inside the explosion-proof chamber 100 to prevent outside air from entering: the first refrigeration valve 2021 and the pressure regulating valve 4021 are opened, the vaporizer 40 is started, and the fifth flow controller 4022 is turned on. The nitrogen gas obtained from the vaporization of liquid nitrogen enters the interior of the explosion-proof chamber 100 through this pipeline, displacing the original gas inside the explosion-proof chamber 100 and creating a slight positive pressure, such as 5 Pa.

[0087] The system described in this application can be implemented as follows:

[0088] Construction and monitoring of the positive pressure cryogenic test environment: The first refrigeration valve 2021 and the second refrigeration valve 2022 are opened, the vaporizer 40 is started, and the first flow controller 2023 is activated. The cryogenic nitrogen gas obtained from the vaporization of liquid nitrogen enters the refrigeration execution pipeline 204 coiled within the explosion-proof chamber 100, lowering and maintaining the temperature inside the explosion-proof chamber 100 to the required low temperature for the test, such as -40℃ for hydrogen refueling machine performance testing. During this process, the hydrogen concentration, pressure, and temperature inside the explosion-proof chamber 100 are monitored in real time using a hydrogen concentration detector 501, a pressure detector 502, and a temperature detector 503, and transmitted in real time to the main control unit 600, which then issues corresponding control commands.

[0089] Specifically, it may include:

[0090] The main control unit 600 makes adjustment commands to adjust the opening and closing status of the pressure regulating valve 4021 and the opening degree of the fifth flow controller 4022 based on the monitored pressure inside the explosion-proof chamber 100, so as to maintain a slightly positive pressure state inside the explosion-proof chamber 100.

[0091] Based on the temperature inside the explosion-proof cabin 100, the main control unit 600 regulates the flow rate of low-temperature nitrogen entering the refrigeration execution pipeline 204 by adjusting the opening size of the first flow controller 2023, thereby maintaining the low-temperature environment inside the explosion-proof cabin 100.

[0092] Based on the temperature inside the explosion-proof chamber 100, the main control unit 600 opens the first heating valve 3021, the second flow controller 3022, the second heating valve 3023, and the fourth flow controller 3024. At the same time, it opens the chamber 3011, the first speed control valve 3012, starts the heater 3013, the speed control pump 305, and the second speed control valve 3013. The water in the chamber 3011 passes through the heater 3013 and generates steam, which is then powered by the speed control pump 305 and enters the heating execution pipeline 304 coiled around the bottom and sides of the high and low temperature environment chamber. This heats and maintains the temperature inside the explosion-proof chamber 100 at the high temperature required for the test, such as the upper limit of 40°C for hydrogen refueling machine performance testing.

[0093] Based on the hydrogen concentration inside the explosion-proof chamber 100, the main control unit 600 issues a command to activate the exhaust valve 801 and the fan 802 to release the gas inside the explosion-proof chamber 100 when the hydrogen concentration exceeds the specified safety limit. When the test is completed and the nitrogen inside the explosion-proof chamber 100 needs to be released, the release control valve 701 is opened to release the gas inside the explosion-proof chamber 100 through the gas release pipeline.

[0094] During this period, nitrogen is used to maintain a slight positive pressure inside the explosion-proof chamber 100. The opening of the pressure regulating valve 4021 and the fifth flow controller 4022 are adjusted according to the real-time pressure inside the explosion-proof chamber 100.

[0095] This application also provides a method for implementing hydrogen refueling machine performance testing conditions, applied to the main control unit of the system for implementing hydrogen refueling machine performance testing conditions as described in any of the above embodiments, such as... Figure 3 As shown, it includes the following steps:

[0096] S100: Acquire the working status data of the refrigerant transmission pipeline monitored by the refrigerant pipeline status sensor, the working status data of the heating medium transmission pipeline monitored by the heating pipeline status sensor, and the working status data of the inert gas medium transmission pipeline monitored by the inert gas medium transmission pipeline status sensor within a historical time period as sample input data.

[0097] S200: Obtain the hydrogen concentration, environmental pressure, and environmental temperature inside the explosion-proof cabin monitored by the environmental monitoring unit corresponding to the sample data within the historical time period as sample output data.

[0098] S300: The sample input data and the sample output data are merged according to the condition that the collection time is the same to obtain the key variable data grouping result of the internal environment control of the explosion-proof cabin. Each group is a time series data.

[0099] S400: Divide the grouping results into a training dataset and a test dataset to obtain the training dataset of the initial learning model, wherein the initial learning model includes an anomaly detection initial model and an anomaly prediction initial model;

[0100] S500: Input the training dataset into the anomaly monitoring initial model to learn and train the monitoring results of the refrigeration pipeline status sensor, the heating pipeline status sensor, the inert gas medium transmission pipeline status sensor, and the environmental monitoring unit to detect anomalies; input the training dataset into the anomaly prediction initial model to learn and train the monitoring results of the environmental monitoring unit to predict anomalies.

[0101] S600: The initial model for anomaly monitoring and the initial model for anomaly prediction, which have completed learning and training, will be used as the control model for the hydrogen refueling machine test environment.

[0102] The technical solution provided in this application can establish a system for judging whether there are any abnormalities in the hydrogen refueling machine performance testing environment based on the data monitored by each sensor in the system that realizes the hydrogen refueling machine performance testing conditions through learning and training. At the same time, it can also predict abnormal situations that have not yet occurred, thereby enabling timely feedback of abnormal situations to the staff and further improving the safety of testing while ensuring that the testing conditions are met.

[0103] Furthermore, such as Figure 4 As shown, the initial model for anomaly monitoring includes:

[0104] The system employs a two-time-block structure. The input training dataset is processed through these two time-block structures sequentially, decomposing the one-dimensional time-series data into a two-dimensional time-series data matrix that includes a trend component and short-term fluctuations, according to a predetermined period. In practice, the time blocks can be used to extract short-term and long-term trends from the time-series data through Fast Fourier Transform.

[0105] A convolutional neural network, set in the convolutional block shown in the figure, extracts the time trend features in the two-dimensional time series data matrix output by two time block structures; the convolutional neural network may include a 2D convolutional layer, a Layer Norm layer, a 2D convolutional layer, an Elu activation function, and then the residuals are summed after Dropout.

[0106] The deep learning algorithm structure, set within the fully connected block shown in the diagram, extracts all time-trend features from the convolutional neural network output and sums them to obtain the time-series prediction result. The anomaly detection output structure obtains a reasonable threshold based on the test dataset from the training dataset, uses this reasonable threshold and the time-series prediction result to perform anomaly detection and output the anomaly monitoring result. The deep learning algorithm can be implemented using an LSTM-SRU neural network. The LSTM-SRU neural network assigns different weights to the input data, calculates forgetting, input, and output gating, and combines an improved LSTM network and a SRU network. The output of the LSTM network layer becomes the input of the SRU network layer; this structure helps the SRU layer fit the data more quickly. To prevent overfitting, a dropout layer is introduced to randomly discard some data, ensuring the generalization ability of the network model. To maximize the continuity of the network, the hidden state at the last moment of the LSTM neural network training is extracted as the initial hidden state of the SRU neural network. This effectively improves the learning efficiency of the network and achieves higher accuracy. Finally, through unsupervised anomaly detection, the prediction error is used as the classification criterion, and the maximum value of the prediction loss on the validation dataset is used as the anomaly detection threshold. In the test set, the predicted classification result is obtained by comparing the error between the predicted result and the actual result with the threshold obtained in the previous step. For example, the prediction of anomaly is labeled as 1, and the prediction of normality is labeled as 0.

[0107] As a preferred option, such as Figure 5 The anomaly prediction initial model includes:

[0108] The historical time series construction module extracts the historical time series features of the time series prediction results; a two-period Dlinear model (set in the Dlinear module in the figure) can be used, which is simple to operate and can effectively extract the historical time series features of the predicted sequence.

[0109] The relevant time series construction module concatenates the historical time features, uses a scaled reset matrix and performs scaled position encoding, and then passes through a multi-head attention module and a bidirectional cross-attention module to extract inter-time series features and intra-time series features respectively, determining the degree of influence of each time series feature on the relevant time series features. In specific implementation, a convolutional neural network set in the convolutional dimensionality reduction block can be used to reduce the dimensionality of the relevant time series data, and then a matrix concatenation block is used to concatenate the historical time series. In the periodic decomposition block, a scaled reset matrix is ​​used and scaled position encoding is performed, and then through a multi-head attention module and a bidirectional cross-attention module respectively, inter-time series features and intra-time series features are extracted respectively.

[0110] The relevant real-time construction module performs real-time correction on the internal features of each time series based on the introduced real-time relevant sensor data. In specific implementation, dimensionality reduction can be achieved through the convolutional neural network in the convolutional dimensionality reduction block, and then real-time linear correction can be performed by introducing real-time relevant sensor data in the Linear module.

[0111] The prediction result output module sums the output results of the historical time series construction module, the relevant time series construction module, and the relevant real-time construction module and outputs the sum. It then determines the prediction result, performs anomaly monitoring on the prediction result, and outputs the anomaly prediction result.

[0112] The learning and training methods described in this application primarily target whether the monitoring results of different sensors are abnormal, the mutual influence between monitoring results of correlated sensors, and the prediction of whether a particular sensor's monitoring result will soon become abnormal. For example, inside an explosion-proof enclosure, pressure and temperature are correlated. If the pressure is within the normal range but the temperature is in a high range, the monitoring results inside the explosion-proof enclosure will only detect temperature anomalies. However, if only the cooling control unit is controlled at this time, the efficiency may be limited. But if the gas venting pipeline is controlled in conjunction to reduce the pressure inside the explosion-proof enclosure slightly, while still maintaining a slightly positive pressure (e.g., from 5 Pa to 1.5 Pa), it can still prevent outside air from entering the explosion-proof enclosure, but it greatly improves the safety of the explosion-proof enclosure. Once the temperature inside the explosion-proof enclosure reaches a safe range, the pressure value can be adjusted back to the initial 5 Pa.

[0113] The above-described scheme in this application utilizes a pre-trained initial model for anomaly monitoring and anomaly prediction to process data monitored by various sensors in a system that implements the performance testing conditions of a hydrogen refueling machine, thereby achieving anomaly monitoring and prediction. This scheme extracts the interrelationships and dependencies between the monitoring results of multiple sensors to ensure the accuracy of the monitoring and prediction results.

[0114] As needed, the above technical solutions can be combined to achieve the best technical effect.

[0115] The above are merely the principles and preferred embodiments of this application. It should be noted that, for those skilled in the art, several other modifications can be made based on the principles of this application, and these modifications should also be considered within the scope of protection of this application.

Claims

1. A system for achieving performance testing conditions for a hydrogen refueling machine, characterized in that, include: An explosion-proof chamber is installed outside the hydrogen refueling machine to be tested. The explosion-proof chamber includes an inner explosion-proof plate, an outer explosion-proof plate, and a flame-retardant interlayer disposed between the inner explosion-proof plate and the outer explosion-proof plate. An explosion-proof valve is installed on the explosion-proof chamber. The refrigeration control unit includes a refrigeration medium transmission pipeline, a refrigeration control valve, a refrigeration pipeline status sensor, and a refrigeration execution pipeline, wherein the refrigeration execution pipeline is a refrigeration pipeline installed inside the explosion-proof cabin. The heating control unit includes a heating medium transmission pipeline, a heating control valve, a heating pipeline status sensor, and a heating execution pipeline, wherein the heating execution pipeline is a heating pipeline installed inside the explosion-proof cabin. The pressure control unit includes an inert gas medium transmission pipeline, an inert gas medium transmission pipeline status sensor, and a pressure control valve. The outlet end of the inert gas transmission pipeline is connected to the interior of the explosion-proof chamber. An environmental monitoring unit is used to monitor the hydrogen concentration, ambient pressure, and ambient temperature inside the explosion-proof cabin. The main control unit receives monitoring results from the environmental monitoring unit, the refrigeration pipeline status sensor, the heating pipeline status sensor, and the inert gas medium transmission pipeline status sensor. Based on the monitoring results, it controls at least one of the refrigeration control valve, the heating control valve, and the pressure control valve to maintain the hydrogen concentration, environmental pressure, and environmental temperature inside the explosion-proof cabin within their respective preset ranges.

2. The system for achieving the performance testing conditions of a hydrogenation machine according to claim 1, characterized in that, In the refrigeration control unit, the refrigeration control valves include a first refrigeration valve, a second refrigeration valve, a first flow controller, and a third refrigeration valve; the refrigeration pipeline status sensors include a first pressure sensor and a first temperature sensor, wherein: The refrigerant transmission pipeline is connected to a storage tank, which stores the refrigerant, which is a liquid medium of the inert gas. The first end of the first refrigeration valve is connected to the liquid outlet of the storage tank; The second end of the first refrigeration valve is connected to the vaporizer, and the second end of the vaporizer is connected to the pressure control unit; The first end of the second refrigeration valve is connected to the second end of the vaporizer, and the second end of the second refrigeration valve is connected to the first end of the first flow controller; The second end of the first flow controller is connected to the first end of the refrigeration execution pipeline, and the second end of the refrigeration execution pipeline is connected to the inert gas recovery device via the third refrigeration valve; The first pressure sensor and the first temperature sensor are disposed between the second end of the first flow controller and the first end of the refrigeration actuation pipeline.

3. The system for achieving the performance testing conditions of a hydrogen refueling machine according to claim 1, characterized in that, In the heating control unit, the heating control valve includes a first heating valve, a second flow controller, a third flow controller, and the heating pipeline status sensor includes a second pressure sensor, a third pressure sensor, a second temperature sensor, and a third temperature sensor, wherein: The heating medium transmission pipeline is connected to the heat source box, which is used to store the heating medium. The first end of the first heating valve is connected to the liquid outlet of the heat source box, and the second end of the first heating valve is connected to the first end of the heating execution pipeline after passing through the second flow controller; The first end of the second heating valve is connected to the second end of the heating execution pipeline, and the second end of the second heating valve is connected to the liquid inlet of the heat source box after passing through the third flow controller; The second pressure sensor and the second temperature sensor are disposed between the heat source box and the first end of the heating execution pipeline; the third pressure sensor and the third temperature sensor are disposed between the second end of the second heating valve and the liquid inlet of the heat source box.

4. The system for achieving the performance testing conditions of a hydrogen refueling machine according to claim 3, characterized in that: The heat source box includes a box body, a first speed control valve located at the box body outlet, a heater, and a second speed control valve located at the box body inlet; The heating control unit also includes a speed regulating pump, which is disposed between the heater and the first heating valve.

5. The system for achieving the performance testing conditions of a hydrogenation machine according to claim 2, characterized in that, In the pressure control unit, the pressure control valve includes a pressure regulating valve and a fifth flow controller, and the inert gas medium transmission pipeline status sensor includes a fourth pressure sensor and a fourth temperature sensor, wherein: The first end of the pressure regulating valve is connected to the second end of the vaporizer, the second end of the pressure regulating valve is connected to the first end of the fifth flow controller, and the second end of the fifth flow controller is connected to the interior of the explosion-proof chamber via a pipeline; The fourth pressure sensor and the fourth temperature sensor are located between the second end of the fifth flow controller and the explosion-proof chamber.

6. The system for achieving the performance testing conditions of a hydrogen refueling machine according to claim 1, characterized in that: The environmental monitoring unit includes a hydrogen concentration sensing probe, a pressure sensor probe, and a temperature sensor probe disposed inside the explosion-proof cabin. The hydrogen concentration sensing probe detects the hydrogen concentration inside the explosion-proof cabin, the pressure sensor probe detects the environmental pressure inside the explosion-proof cabin, and the temperature sensor probe detects the environmental temperature inside the explosion-proof cabin. The hydrogen concentration sensing probe, the pressure sensor probe, and the temperature sensor probe are provided with explosion-proof sealing interfaces at the positions where they contact the explosion-proof cabin.

7. The system for achieving the performance testing conditions of a hydrogen refueling machine according to any one of claims 1-6, characterized in that: The explosion-proof cabin is equipped with an explosion-proof door; The explosion-proof door includes a door body, and a fixing plate is pressed onto the edge of the door body. The fixing plate is pressed between the outer explosion-proof plate and the inner explosion-proof plate of the explosion-proof compartment.

8. The system for achieving the performance testing conditions of a hydrogen refueling machine according to claim 6, characterized in that, Also includes: The venting control unit includes a gas venting pipeline and a venting control valve; One end of the gas venting pipeline is connected to the interior of the explosion-proof cabin, and the venting control valve is located between the gas venting pipeline and the explosion-proof cabin; The main control unit is also used to control the operation of the discharge regulating valve.

9. A method for achieving performance testing conditions for a hydrogen refueling machine, characterized in that, The main control unit applied to the system for achieving the performance testing conditions of the hydrogen refueling machine as described in any one of claims 1-8 includes: The working status data of the refrigerant transmission pipeline monitored by the refrigeration pipeline status sensor, the working status data of the heating pipeline monitored by the heating pipeline status sensor, and the working status data of the inert gas medium transmission pipeline monitored by the inert gas medium transmission pipeline status sensor are obtained as sample input data within a historical time period. The hydrogen concentration, environmental pressure, and environmental temperature inside the explosion-proof cabin, as monitored by the environmental monitoring unit corresponding to the sample data during the historical time period, are obtained as sample output data. The sample input data and the sample output data are merged according to the condition that the collection time is the same to obtain the key variable data grouping result of the internal environment control of the explosion-proof cabin. Each group is a time series data. The grouping results are divided into a training dataset and a test dataset to obtain the training dataset of the initial learning model. The initial learning model includes an anomaly detection initial model and an anomaly prediction initial model. The training dataset is input into the anomaly monitoring initial model to learn and train the monitoring results of the refrigeration pipeline status sensor, the heating pipeline status sensor, the inert gas medium transmission pipeline status sensor, and the environmental monitoring unit to detect anomalies. The training dataset is input into the anomaly prediction initial model to learn and train the prediction of anomalies in the monitoring results of the environmental monitoring unit. The initial models for anomaly monitoring and anomaly prediction, which have completed the learning and training, will be used as the control models for the hydrogen refueling machine test environment.

10. The method for achieving the performance testing conditions of a hydrogenation machine according to claim 9, characterized in that: The initial model for anomaly monitoring includes: The input training dataset is processed through two time block structures in sequence. Based on a predetermined period, the one-dimensional time series data is decomposed into a two-dimensional time series data matrix that includes a trend component and a short-term fluctuation component. A convolutional neural network is used to extract the temporal trend features from the two-dimensional time-series data matrix output by two time-block structures. The deep learning algorithm structure extracts all time trend features from the output of the convolutional neural network and performs summation to obtain the time series prediction result. An anomaly detection output structure is generated by obtaining a reasonable threshold based on the test dataset in the training dataset, performing anomaly detection using the reasonable threshold and the time-series prediction result, and outputting the anomaly monitoring result. The initial model for anomaly prediction includes: The historical time series construction module extracts the historical time series features of the time series prediction results; The relevant time series construction module concatenates the historical time features, uses a scaled reset matrix and performs position encoding on a scaled basis, and then passes through a multi-head attention module and a bidirectional cross-attention module to extract inter-time series features and intra-time series features respectively, and determines the degree of influence of each time series feature on the relevant time series features. The relevant real-time construction module performs real-time correction on the internal features of each time series based on the introduced real-time relevant sensor data; The prediction result output module sums the output results of the historical time series construction module, the relevant time series construction module, and the relevant real-time construction module and outputs the sum. It then determines the prediction result, performs anomaly monitoring on the prediction result, and outputs the anomaly prediction result.

Images