A PBL mode-based national gate safety popular science research experiment intelligent management method

Abstract

The application relates to the fields of experimental teaching and intelligent control technology, and discloses a national gate safety popular science research experiment intelligent management method based on a PBL mode, which comprises the following steps: analyzing and exploring a hypothesis vector and experiment initialization data, constructing a static safety envelope and performing access verification based on physical limit parameters; collecting multi-modal sensing data, shrinking the static safety envelope by using a nonlinear coupling attenuation factor of a secondary parameter change rate mapping, and generating a dynamic safety threshold; performing hierarchical fuse control according to the comparison result of real-time main parameters, the static safety envelope and the dynamic safety threshold; after the experiment is finished, calculating a trajectory deviation degree, combining with environmental interference average weight for compensation, and generating attribution analysis data for distinguishing environmental factors and operation errors. Through dynamic risk control and multi-dimensional attribution evaluation, the application solves the problems of open exploration and safety management in popular science experiments.

Description

Technical Field

[0001] This invention relates to the field of experimental teaching and intelligent control technology, specifically to an intelligent management method for border security science popularization and research experiments based on the PBL (Problem-Based Learning) model. Background Technology

[0002] In border security science popularization and research education, problem-based learning is widely used. This model encourages students to independently propose research hypotheses and operate experimental equipment to verify them, thereby gaining a deeper understanding of the principles of physical interception and detection. Such experiments typically involve sophisticated sensors and complex signal processing logic, placing high demands on the safety control of the experimental process and the evaluation of results.

[0003] Current experimental management systems mostly employ static threshold control strategies for safety, which pre-set fixed upper and lower limits for physical parameters. When monitored values ​​exceed these limits, alarms or power outages are triggered. However, in open-ended exploratory experiments, a single static threshold is often insufficient to handle complex and variable experimental conditions. On the one hand, the exploratory parameters set by students may not theoretically exceed their range, but due to the nonlinear coupling between various physical quantities, system instability may be induced before the hard threshold is reached. On the other hand, existing protection mechanisms typically only monitor core primary parameters, neglecting the cumulative impact of secondary parameters such as environmental vibrations and temperature fluctuations on system stability. This makes dynamic intervention impossible in the early stages of risk formation, easily leading to damage to delicate experimental samples.

[0004] Furthermore, in the experimental evaluation and feedback phase, existing technologies typically calculate the deviation between student experimental data and standard theoretical values ​​directly, using this as the basis for scoring. This single-dimensional evaluation method ignores the interference of objective environmental factors on experimental results. In the field of research and study, uncontrollable factors such as external electromagnetic interference or mechanical vibration often cause fluctuations in experimental data. Existing management systems lack effective attribution analysis methods and cannot accurately separate environmental noise from human error. This leads to the system incorrectly judging student operation as improper when the environment is poor, or failing to provide targeted improvement suggestions for specific causes of deviation, thus affecting teaching effectiveness and the objectivity of evaluation. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an intelligent management method for border security science popularization and research experiments based on the PBL model. This method solves the problems that static protection mechanisms in existing open-ended inquiry experiments are unable to cope with nonlinear coupling risks, and that experimental evaluation cannot effectively distinguish between environmental interference and human operational deviations.

[0006] To achieve the above objectives, the present invention provides the following technical solution: This invention provides an intelligent management method for border security science popularization and research experiments based on the Problem-Based Learning (PBL) model. This method is applied to a distributed architecture intelligent management system, executed by an edge control unit, and enables dynamic risk control and multi-dimensional attribution evaluation of the experimental process.

[0007] The method first performs experimental initialization configuration and hardware access control. The edge control unit receives the investigation hypothesis vector through an interactive terminal, parses the experimental initialization data to determine the experimental type, and uses an RFID reader to read the physical limit parameter set of the experimental sample. Based on the investigation hypothesis vector, the edge control unit calculates the theoretical response principal parameter, constructs a static safety envelope by combining it with a preset static tolerance coefficient, and simultaneously constructs a hardware detection topology using a control signal topology isolation matrix. Subsequently, parameter compliance verification is performed. The edge control unit compares the expected parameters mapped by the investigation hypothesis vector with the physical limit parameter set. The specific comparison logic is as follows: the external excitation parameters are obtained by parsing the investigation hypothesis vector; the excitation tolerance threshold and response tolerance threshold of the sample are extracted; and it is determined whether the external excitation parameter and the theoretical response principal parameter are less than or equal to the corresponding tolerance threshold. If both are true, the edge control unit outputs an unlock signal to the physical protection interlock device; if any limit is exceeded, the locked state is maintained and a parameter conflict is indicated.

[0008] During the experimental operation phase, this method employs a multimodal sensing mechanism for dynamic protection. The edge control unit synchronously acquires real-time primary parameters from the primary parameter sensor and secondary parameter data from the secondary parameter sensor via a multimodal sensor array. For the secondary parameter data, the edge control unit calculates its real-time rate of change and maps it to a nonlinear coupling attenuation factor. This factor is then used to shrink the static safety envelope to generate a dynamic safety threshold. In this process, the edge control unit performs the following two judgments in parallel: first, it determines whether the real-time primary parameter exceeds the static safety envelope to identify theoretical preset errors; second, it determines whether the real-time primary parameter exceeds the dynamic safety threshold to identify gradient coupling risks. Based on the judgment results, the control subsystem is activated.

[0009] The specific calculation method for the nonlinear coupling attenuation factor and dynamic safety threshold is as follows: The collected secondary parameter data is filtered to obtain smoothed secondary parameters. Discrete-time difference operations and time dimension normalization are performed to obtain the real-time change rate of the secondary parameter data. A nonlinear activation mechanism based on the Sigmoid function is adopted, and an exponential operation is used to construct a mapping relationship between gradient and risk, mapping the real-time change rate of the secondary parameter data to a nonlinear coupling attenuation factor with a value between zero and one. This mapping relationship is configured with a sensitive case coefficient and a critical gradient threshold, which are used to adjust the steepness of the response curve and define the inflection point of the risk response, respectively. If there are multi-channel sensors, the maximum factor calculated by each channel is selected as the comprehensive attenuation factor. Furthermore, the edge control unit, centered on the theoretical response master parameter, uses the difference between one and the comprehensive attenuation factor as a compression ratio coefficient to reduce the allowable fluctuation range defined by the static safety envelope, and calculates the dynamic safety upper limit and dynamic safety lower limit constituting the dynamic safety threshold.

[0010] When the fuse mechanism is triggered, the edge control unit performs tiered protection. If the real-time main parameter exceeds the dynamic safety threshold, a dynamic fuse trigger flag is generated; if it exceeds the response tolerance threshold in the physical limit parameter set, a hard fuse trigger flag is generated. Based on the above flags, a power enable signal and a shield unlock signal are generated. When the power enable signal becomes invalid, a hybrid disconnection sequence is triggered: the solid-state relay control signal is cut off first, and the electromechanical relay drive signal is cut off after a preset delay.

[0011] After the experiment, an attribution evaluation based on data backtracking was performed. By extracting the effective data segment from the start of the experiment to the moment the circuit breaker was triggered, the average absolute percentage error (ASE) algorithm was used to calculate the average ratio of the absolute value of the difference between the real-time main parameter and the theoretical response main parameter to the normalized reference value, thus obtaining the trajectory deviation. The selection logic for the normalized reference value is as follows: the amplitude of the theoretical response main parameter is compared with a preset validity judgment threshold; when the amplitude of the theoretical response main parameter is greater than the validity judgment threshold, the normalized reference value is the theoretical response main parameter; when the amplitude of the theoretical response main parameter is less than or equal to the validity judgment threshold, the normalized reference value is the full-scale value of the sensor.

[0012] Finally, the edge control unit generates evaluation data based on environmental interference. The arithmetic mean of the nonlinear coupling attenuation factor within the experimental period is calculated to obtain the average weight of environmental interference. A weighted compensation logic is then used to calculate the net operational deviation index. The net operational deviation index equals the product of the trajectory deviation degree and the correction coefficient, which is the difference between the product of the environmental tolerance coefficient and the average weight of environmental interference. Attribution type is determined by comparison: if the average weight of environmental interference exceeds the environmental significance threshold, it is classified as an external environment-dominated deviation, and environmental improvement suggestions are generated; if environmental interference does not exceed the limit but the net operational deviation index exceeds the operational deviation threshold, it is classified as an operational error-dominated deviation, and the standard operation video is retrieved. The environmental tolerance coefficient is configured according to the teaching attributes of the experimental project: a first preset value is configured for introductory-level experiments, and a second preset value is configured for advanced assessment experiments.

[0013] This invention provides an intelligent management method for border security science popularization and research experiments based on the Problem-Based Learning (PBL) model. It has the following beneficial effects: 1. This invention achieves pre-emptive prevention of experimental risks by constructing a physical access verification mechanism based on an inquiry hypothesis vector. Before the experiment starts, the edge control unit maps the theoretical inquiry parameters input by the user to the expected response and compares it with the physical limit parameter set of the experimental sample. A physical protection interlock device forcibly isolates experimental requests with conflicting parameters. This mechanism ensures that the experimental parameters must be within the safe range allowed by physical laws, effectively preventing students' theoretical cognitive biases from directly transforming into destructive physical stimuli, thereby guaranteeing the basic safety of open-ended inquiry experiments.

[0014] 2. This invention introduces a dynamic safety envelope contraction mechanism based on multimodal secondary parameters, enhancing protection against nonlinear coupling risks. By monitoring the real-time rate of change of secondary parameters and mapping it to a nonlinear coupling attenuation factor, the edge control unit can dynamically contract the safety threshold range using the attenuation factor when the experimental environment stability has already decreased before the primary parameter reaches the hard tolerance threshold. This mechanism can identify and respond to gradient coupling risks that traditional fixed thresholds cannot capture, triggering a circuit breaker in advance at the initial stage of experimental instability, preventing the equipment from evolving from a critical state to an irreversible damage state.

[0015] 3. This invention establishes an attribution analysis model based on environmental interference weights, solving the evaluation problem of distinguishing between objective environmental interference and subjective operational errors in science popularization and research experiments. By quantifying the degree of environmental interference throughout the experiment and calculating correction coefficients, the edge control unit can perform weighted compensation for operational trajectory deviations and accurately distinguish whether the source of deviation is dominated by the external environment or operational errors based on threshold judgment logic. This not only eliminates the adverse effects of environmental fluctuations on assessment results but also generates environmental improvement suggestions or operational correction guidance for different causes, improving the objectivity and pertinence of experimental teaching evaluation. Attached Figure Description

[0016] Figure 1 This is a diagram of the overall system architecture of the present invention; Figure 2 This is the main flowchart of the method of the present invention.

[0017] Among them, 10 is the central management server; 20 is the intelligent experimental terminal; 30 is the edge control unit; 40 is the multimodal sensor array; 41 is the main parameter sensor; 42 is the secondary parameter sensor; 50 is the execution control subsystem; 51 is the programmable power controller; 52 is the signal topology isolation matrix; 53 is the physical protection interlock device; 60 is the human-machine interaction subsystem; 61 is the interactive terminal; and 62 is the radio frequency identification reader. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see the appendix Figure 1-2 This invention provides an intelligent management method for border security science popularization and research experiments based on the PBL (Problem-Based Learning) model. This method is applied to a distributed architecture intelligent management system, which mainly includes a central management server 10 and multiple distributed intelligent experimental terminals 20. The central management server 10 and each intelligent experimental terminal 20 establish a bidirectional data communication connection via industrial Ethernet or wireless local area network.

[0020] The central management server 10 is equipped with a physical model database and an experimental topology configuration library, used to store standard physical law models, chemical characteristic parameters of hazardous materials, and basic hardware configuration information of experimental platforms for various customs security experiments. The central management server 10 is used to send basic configuration parameters to the intelligent experimental terminal 20 and receive experimental process data for storage and traceability.

[0021] The intelligent experimental terminal 20 is a physical unit that performs experimental control and safety protection. The intelligent experimental terminal 20 includes an edge control unit 30, a multimodal sensor array 40, an execution control subsystem 50, and a human-computer interaction subsystem 60.

[0022] The edge control unit 30, as the core computing node of the intelligent experimental terminal 20, integrates an embedded microprocessor and a floating-point arithmetic unit. The edge control unit 30 is configured to perform local high-frequency data acquisition, gradient calculation, nonlinear coupling factor calculation, and millisecond-level interrupt control. The edge control unit 30 is connected to the multimodal sensor array 40, the execution control subsystem 50, and the human-machine interaction subsystem 60 via an internal bus.

[0023] The multimodal sensor array 40 is used to acquire physical parameters in the experimental environment in real time and convert them into digital signals for transmission to the edge control unit 30. Depending on the monitored object, the multimodal sensor array 40 is specifically divided into primary parameter sensors 41 and secondary parameter sensors 42.

[0024] The main parameter sensor 41 is used to collect physical quantities directly related to the core research objective of the current experiment. In the electrochemical experiment scenario, the main parameter sensor 41 includes a high-precision current transformer and a voltage sensor; in the thermal experiment scenario, the main parameter sensor 41 includes a contact thermocouple.

[0025] Secondary parametric sensor 42 is used to acquire physical quantities related to the current experimental environment or potential side reactions. Secondary parametric sensor 42 includes a metal oxide semiconductor gas sensor for monitoring organic volatiles, an infrared thermal imaging probe for monitoring the overall temperature distribution, and a vibration sensor for monitoring mechanical stability.

[0026] The execution control subsystem 50 is used to change the physical state of the experimental platform according to the instructions of the edge control unit 30. The execution control subsystem 50 includes a programmable power controller 51, a signal topology isolation matrix 52, and a physical protection interlock device 53.

[0027] The programmable power controller 51 is connected between the external power supply circuit and the experimental load to regulate the output voltage, limit the current, or cut off the power supply circuit.

[0028] A signal topology isolation matrix 52 is positioned on the signal transmission path between the edge control unit 30 and the multimodal sensor array 40. The signal topology isolation matrix 52 is composed of a multi-channel programmable relay array or an analog switch array. The signal topology isolation matrix 52 is used to physically connect or disconnect the signal connections of specific sensors according to control commands, thereby constructing a hardware detection topology that eliminates interference from irrelevant variables.

[0029] The physical protection interlock device 53 is installed at the physical protective cover or door of the experimental operation area. The physical protection interlock device 53 includes an electromagnetic lock and a position feedback switch, used to forcibly lock or open the experimental operation space.

[0030] The human-computer interaction subsystem 60 is used to realize information interaction between the user and the system. The human-computer interaction subsystem 60 includes an interactive terminal 61 and an RFID reader / writer 62.

[0031] The interactive terminal 61 includes a display screen and input devices for displaying the experimental status and receiving user-inputted hypothetical vectors. An RFID reader 62 is located in the sample placement area of ​​the experimental platform for reading information from RFID tags affixed to the experimental samples. These RFID tags store the sample's material properties and physical limit parameters.

[0032] See attached document Figure 2 This invention provides an intelligent management method for science popularization and research experiments on national border security based on the PBL (Problem-Based Learning) model, specifically including the following steps: The edge control unit 30 acquires experimental initialization data through the human-computer interaction subsystem 60. The interactive terminal 61 receives the research hypothesis vector input by the user, which includes the user-preset core experimental variables and their expected trends. Simultaneously, the RFID reader 62 reads the RFID tag data of the samples placed on the experimental platform to obtain the physical fingerprint information of the samples.

[0033] The edge control unit 30 constructs the experimental operating environment based on the investigation hypothesis vector. The edge control unit 30 calls the physical model stored on the central management server 10 or locally, maps the investigation hypothesis vector to the theoretical response principal parameter, and generates a static safety envelope based on a preset tolerance coefficient. Simultaneously, the edge control unit 30 sends a switching command to the signal topology isolation matrix 52. The signal topology isolation matrix 52 physically disconnects the signal loops of sensors unrelated to the investigation hypothesis vector according to the command, retaining only the conduction state of the principal parameter sensor 41 and related secondary parameter sensors 42, thereby constructing a single-variable hardware detection topology.

[0034] The edge control unit 30 performs a physical access consistency check. The edge control unit 30 compares the expected parameters in the user-input investigation hypothesis vector with the physical limit parameters in the RFID tag data. If the expected parameters are within the range of the sample's physical limit parameters, the edge control unit 30 controls the physical protection interlock device 53 to unlock, allowing the user to operate or power on the device; if the parameters are outside the range, the device remains locked and prompts for correction.

[0035] The edge control unit 30 activates the multimodal sensor array 40 and enters a real-time dynamic monitoring loop. During the experimental operation, the edge control unit 30 synchronously acquires the primary parameter data of the primary parameter sensor 41 and the secondary parameter data of the secondary parameter sensor 42 at a preset sampling frequency.

[0036] The edge control unit 30 performs nonlinear multi-parameter gradient coupling operations. The edge control unit 30 calculates the rate of change of secondary parameter data over time and calculates a nonlinear coupling attenuation factor based on this rate of change. Subsequently, the edge control unit 30 uses this nonlinear coupling attenuation factor to correct the upper limit of the static safety envelope in real time, generating the current dynamic safety threshold.

[0037] The edge control unit 30 performs multi-level circuit breaker decision-making and control. First, the edge control unit 30 determines whether the main parameter data exceeds the static safety envelope; if it does, it is determined to be a cognitive bias error. If the main parameter data does not exceed the static safety envelope, the edge control unit 30 further determines whether the main parameter data exceeds the dynamic safety threshold. If the main parameter data exceeds the dynamic safety threshold, it is determined to be a gradient coupling risk.

[0038] The edge control unit 30 controls the execution control subsystem 50 based on the decision result. When a cognitive bias error or gradient coupling risk is determined, the edge control unit 30 immediately sends a cut-off command to the programmable power controller 51 to block the experimental power supply. After the experiment ends or the power is cut off, the edge control unit 30 calculates the Euclidean distance between the measured trajectory of the principal parameter data and the theoretical response principal parameter, generating quantified cognitive bias evaluation data.

[0039] During the construction of the experimental operating environment in the edge control unit 30, the analysis of the hypothesis vector and the calculation of the theoretical value are performed. The specific processing steps include the following: The edge control unit 30 parses the received investigation hypothesis vector. The interactive terminal 61 transmits the investigation hypothesis vector. This is a data structure containing multidimensional experimental parameters. The edge control unit 30 unpacks this data structure and extracts three key components: experimental type identifier, external stimulus parameters, and so on. and the core physical parameters expected by users .

[0040] The experiment type identifier indicates the category of the current experiment, such as constant resistance characteristic testing, lithium-ion battery charge / discharge testing, or exothermic reaction rate testing. External excitation parameters. This is used to characterize the active control variables that will be applied to the load on the experimental platform. In electrical experiments, this is the applied voltage; in thermal experiments, it is the heating power or the ambient temperature setting. The core physical parameters expected by the user... This is used to characterize the properties of the measured object predicted by the user based on their theoretical understanding, such as the resistance value calculated by the user or the equilibrium temperature after the reaction. The unpacking and field extraction of data packets are standard techniques in the field of data processing and will not be elaborated upon here.

[0041] The edge control unit 30 retrieves and calls up the matching physical model. Using the parsed experiment type identifier, the edge control unit 30 performs an index match in local memory or the configuration library issued by the central management server 10 to determine the physical law function followed by the current experiment. .

[0042] This physical law function These are pre-constructed mathematical mappings used to describe the deterministic relationship between input excitation and system response under ideal physical conditions. For different experimental scenarios, physical law functions... They have different specific forms. When the experiment type is a linear resistance test, the physical law function... Calling the Ohm's law model; when the experiment type is nonlinear component testing, the physical law function... Call a pre-stored lookup table of volt-ampere characteristic curves or a polynomial fitting function; when the experiment type is a chemical reaction, use the physical law function. Call upon the reaction kinetics equation or the thermodynamic equilibrium equation.

[0043] The edge control unit 30 calculates the theoretical response master parameters. The edge control unit 30 then extracts the external excitation parameters. and users Expected core physical parameters Substitute the retrieved physical law function In the process, the theoretical response principal parameters are calculated. The calculation process follows the formula below: ; in: Indicates the principal parameters of the theoretical response; Functions representing physical laws; Indicates external excitation parameters; This represents the core physical parameters expected by the user.

[0044] Through the above steps, the edge control unit 30 transforms the user's subjective input of the research hypothesis into a digital theoretical benchmark value that it can recognize and process.

[0045] For physical law functions For nonlinear scenarios stored in lookup tables, if the input external excitation parameters... Located between two discrete points in the lookup table, the edge control unit 30 is configured to calculate the corresponding output value using linear interpolation to ensure the theoretical response master parameter. Continuity and accuracy.

[0046] This process enables the transformation from teaching cognitive language to industrial control language, providing a numerical reference benchmark for subsequent static envelope generation and dynamic safety monitoring.

[0047] After calculating the theoretical response principal parameters, the edge control unit 30 further performs the construction of a static safety envelope, a process designed to provide an initial operating permission range for the experiment based on cognitive assumptions. The specific process includes the following steps.

[0048] The edge control unit 30 determines the static tolerance coefficient of the current experiment. Using the experiment type identifier as an index key, the edge control unit 30 accesses the experiment configuration lookup table stored in local non-volatile memory and reads the corresponding static tolerance coefficient. The static tolerance coefficient Used to define the maximum relative deviation of measured data from theoretical expectations allowed during normal investigation. Static tolerance coefficient for experiments with different risk levels. The values ​​differ. For low-risk basic circuit experiments, the static tolerance coefficient read by the edge control unit 30 is... Configured to a value in the range of 0.20 to 0.30 to allow for fluctuations in contact resistance during operation; for high-risk chemical or battery experiments, the edge control unit 30 reads the static tolerance coefficient. Configure values ​​in the range of 0.05 to 0.10 to tighten the safety margin.

[0049] The edge control unit 30 calculates the boundary values ​​of the static safety envelope. The edge control unit 30 then uses the theoretical response master parameters calculated in the preceding steps. and the determined static tolerance coefficient Calculate the numerical range within which the main parameter sensor 41 can fluctuate, i.e., the static safety envelope. The calculation follows the formula: ; in: This represents the static safety envelope, which is a closed interval containing a lower limit and an upper limit. Indicates the principal parameters of the theoretical response; This represents the static tolerance coefficient.

[0050] The edge control unit 30 performs a nonnegativity constraint check after the calculation is complete. This checks the calculated static safety envelope. When the lower limit value is less than the physical zero point and the current physical parameter does not have a negative physical meaning (e.g., in absolute pressure or resistance measurement scenarios), the edge control unit 30 is configured to force the lower limit value to be set to zero to conform to physical reality.

[0051] The edge control unit 30 deploys static security monitoring logic. The edge control unit 30 calculates the static security envelope. The upper and lower limits are written into the high-speed hardware comparator register of the internal microprocessor or the threshold variable of the monitoring interrupt service routine, respectively. Afterward, the edge control unit 30 enters a standby monitoring state. During subsequent experimental operation, if the main parameter data collected at any time exceeds this static safety envelope... Within the defined range, the edge control unit 30 marks the state as a severe cognitive deviation and triggers a first-level fuse protection command.

[0052] After the static security envelope is constructed, and before the test bench is officially powered on or the physical protection is unlocked, the edge control unit 30 performs physical verification based on the physical sample. The specific process includes the following steps.

[0053] The edge control unit 30 acquires the physical characteristic data of the sample. When the experimental sample is placed within the sensing area of ​​the experimental platform, the RFID reader 62 excites and demodulates the RFID tag affixed to the sample surface. The edge control unit 30 receives the data frames uploaded by the RFID reader 62 through a serial communication interface, decodes the data frames, and extracts the physical limit parameter set of the sample. .

[0054] Physical Limit Parameter Set This is a read-only data block storing the material properties of this specific sample. For battery-type samples, it is a set of physical limiting parameters. The included data items include maximum withstand voltage, maximum permissible charging current, and thermal runaway critical temperature; for chemical reagent samples, the physical limiting parameter set... The included data items include flash point temperature and compatibility category code. For data encryption and anti-collision reading technologies for RFID tags, those skilled in the art can implement them using existing ISO 14443 or ISO 15693 protocol standards, which will not be elaborated upon here.

[0055] The edge control unit 30 constructs a consistency verification criterion. The edge control unit 30 calls the external excitation parameters parsed in the preceding steps. and the calculated theoretical response principal parameters Meanwhile, the edge control unit 30 obtains data from the physical limit parameter set. The index retrieves two key thresholds corresponding to the current experiment type: stimulus tolerance threshold. and response tolerance threshold

[0056] The edge control unit 30 performs numerical comparison logic operations. The edge control unit 30 determines whether the experimental parameters set by the user are within the physical tolerance range of the current sample. This determination process follows the following logical expression: ; in: Indicates the Boolean value of the verification result; Indicates external excitation parameters; This represents the excitation tolerance threshold (i.e., the maximum input excitation allowed by the sample). Indicates the principal parameters of the theoretical response; Indicates the response tolerance threshold; Represents the logical AND operation.

[0057] If the user sets the external stimulus parameters This leads to the theoretical response main parameter Exceeded the sample's response tolerance threshold That is, the Boolean value of the verification result. If the result is false, the edge control unit 30 determines that there is a risk of physical damage to the current operation.

[0058] The edge control unit 30 controls the physical protection interlock device 53 based on the verification result. If the verification result is a Boolean value... If true, the edge control unit 30 outputs a high-level unlock signal to the drive circuit of the physical protection interlock device 53, driving the electromagnetic lock to release and allowing the user to open the protective cover for subsequent operations. If the verification result is a Boolean value... If false, the edge control unit 30 maintains the locked state of the physical protection interlock device 53 and displays the specific parameter conflict values ​​through the interactive terminal 61, prompting the user to change the sample or modify the research hypothesis vector.

[0059] After entering the real-time dynamic monitoring loop, the edge control unit 30 performs gradient extraction operations on the secondary parameters. This process aims to separate dynamic features that can characterize the precursors of sudden risks from the environmental background noise. The specific processing includes the following steps.

[0060] The edge control unit 30 performs data acquisition and buffering operations. The edge control unit 30 reads raw digital signals from the secondary parameter sensor 42 at a fixed sampling period. The edge control unit 30 stores the raw digital signals in chronological order into a pre-allocated first-in-first-out (FIFO) circular buffer in its internal RAM. The depth of this circular buffer is configured to store 20 to 50 of the most recent historical data points; when new data is written, the oldest data is removed to ensure real-time updates of the data sequence.

[0061] The edge control unit 30 performs digital filtering preprocessing. Since directly performing differential calculations on the raw signal containing random noise would amplify the differential noise, the edge control unit 30 is configured to perform digital low-pass filtering on the data sequence in the circular buffer. In this embodiment, the edge control unit 30 uses a moving average filtering algorithm to calculate the current sampling time. forward sampling points ( The smoothing secondary parameter is obtained by taking the arithmetic mean of values ​​from 5 to 10. For embedded environments with higher real-time requirements and limited computing resources, the edge control unit 30 can also be configured to use a first-order hysteresis filtering algorithm, the calculation formula of which is:

[0062] in: This is the current original sample value; , which is the filter coefficient, and its value ranges from 0.6 to 0.8.

[0063] ; The edge control unit 30 calculates the real-time rate of change of the secondary parameters. The edge control unit 30 performs discrete-time difference operations based on the smoothed secondary parameters. The edge control unit 30 calculates the difference between the smoothed secondary parameters at the current time step and the smoothed secondary parameters at the previous time step, and normalizes it to the time dimension. This calculation follows the formula: ; In the formula, Indicates the real-time rate of change of the secondary parameter; Indicates the current sampling time Smoothed secondary parameters after filtering; Indicates the previous sampling time Smoothed secondary parameters after filtering; This indicates the sampling time interval of the edge control unit 30.

[0064] Through this step, the edge control unit 30 converts static physical quantity readings into dynamic trend quantities. This secondary parameter's real-time rate of change... It can reflect the intensity of changes in physicochemical reactions earlier than a simple static threshold, providing a predictive input variable for the subsequent calculation of nonlinear coupling factors.

[0065] After obtaining the real-time rate of change of the secondary parameter, the edge control unit 30 performs a nonlinear mapping from the physical gradient to the dimensionless control factor. This process aims to quantify the degree of influence of environmental instability on the safety state of the primary parameter. The specific processing includes the following steps.

[0066] The edge control unit 30 constructs a gradient risk mapping model. The edge control unit 30 reads the real-time rate of change of the secondary parameters calculated in the previous step. To match the nonlinear characteristics of experimental risk accumulation, the edge control unit 30 employs a nonlinear activation mechanism based on the Sigmoid function. This nonlinear activation mechanism is configured to maintain a low gain when the real-time rate of change of the secondary parameter is low, and rapidly increase the gain when the real-time rate of change of the secondary parameter approaches a preset threshold, thereby achieving an early and sensitive response to potential thermal or chemical runaway.

[0067] The edge control unit 30 calculates the nonlinear coupling attenuation factor. The edge control unit 30 uses its built-in floating-point unit to perform exponential operations to calculate the nonlinear coupling attenuation factor at the current moment. The calculation follows the formula: ; in: This represents the nonlinear coupling attenuation factor, which takes values ​​in the open interval (0, 1). The closer the value is to 1, the higher the environmental instability. The base of the natural logarithm; Indicates the real-time rate of change of the secondary parameter; This indicates the absolute value operation; Indicates the sensitivity coefficient; This represents the critical gradient threshold.

[0068] The edge control unit 30 configures the model control parameters. For different secondary parameter sensors 42, the edge control unit 30 calls the corresponding parameters from the local database. Value and The value is used in the calculation.

[0069] For the critical gradient threshold This is defined as the turning point in risk response. When equal At that time, the nonlinear coupling attenuation factor The value is 0.5. The edge control unit 30 configures this parameter according to the sensor type; for example, for a temperature sensor, it sets... The speed is 0.5℃ / s; for VOCs gas sensors, the setting is... It is 50 ppm / s.

[0070] For the sensitivity coefficient It is defined as the steepness of the control system's response near the critical point, and is configured as a positive real number. Sensitivity case coefficient. The larger the value, the steeper the function curve, and the more drastic the system's response to gradient changes exceeding the threshold, resembling a step signal. This is suitable for high-risk chemical experiments (the value range is typically 3.0 to 5.0); Sensitivity coefficient The smaller the value, the flatter the function curve, and the more gradual the system response, which is suitable for basic thermal experiments (the value range is usually 0.5 to 2.0).

[0071] Through the above steps, the edge control unit 30 uniformly maps the real-time change rates of secondary parameters with different physical dimensions to a normalized nonlinear coupling attenuation factor. This factor directly reflects the quantitative value of environmental instability at the current moment, providing a mathematical basis for the subsequent dynamic adjustment of the safety threshold of the main parameters.

[0072] After calculating the nonlinear coupling attenuation factor, the edge control unit 30 performs a shrinkage transformation from the static safety envelope to the dynamic threshold. The specific processing includes the following steps.

[0073] The edge control unit 30 determines the overall system attenuation factor. When the multimodal sensor array 40 contains multiple secondary parametric sensors 42 of different types, the edge control unit 30 calculates the corresponding nonlinear coupling attenuation factor for each sensor channel. The edge control unit 30 executes the maximum value filtering logic, selecting the factor with the largest value among all channels as the comprehensive attenuation factor acting on the system at the current moment. The calculation relationship is as follows This logic ensures that the edge control unit 30 always responds based on the most obvious risk precursors in the current environment.

[0074] The edge control unit 30 acquires static boundary parameters. The edge control unit 30 reads the static safety envelope generated during the initialization phase from the register. upper limit and lower limit value and theoretical response principal parameters These parameters constitute the basic permissible range when there is no environmental disturbance.

[0075] The edge control unit 30 calculates the dynamic safety threshold. The edge control unit 30 utilizes a comprehensive attenuation factor. The static allowable deviation is compressed and corrected. The edge control unit 30 responds to the theoretical main parameter. Centered on, the allowable fluctuation range is determined according to The proportion was reduced.

[0076] The calculation process follows the formula below: ; in: This represents the dynamic safety limit at the current moment; This represents the dynamic safety lower bound at the current moment; Indicates the principal parameters of the theoretical response; This represents the upper limit of the static safety envelope; This represents the lower limit of the static safety envelope; This represents the overall attenuation factor.

[0077] According to the above formula, when the environment is stable ( When the dynamic safety upper and lower limits approach 0, they degenerate into the original boundaries of the static safety envelope, allowing for the maximum experimental error; when the environmental risk is extremely high ( When approaching 1), the dynamic safety upper bound and dynamic safety lower bound converge to the theoretical response principal parameter. This means that the edge control unit 30 compresses the allowable deviation range to a minimum, achieving adaptive adjustment of the safety boundary according to the environmental risk gradient.

[0078] The edge control unit 30 updates the hardware monitoring registers. The edge control unit 30 then applies the calculated dynamic security limit. and dynamic safety lower limit The threshold configuration register of the digital comparator is written in real time, overwriting the original static threshold. This operation is repeated in each control cycle, thus forming a dynamic safety envelope that changes continuously over time.

[0079] After generating a dynamic safety threshold that changes in real time with environmental risks, the edge control unit 30 performs real-time numerical comparison and adjudication to determine whether to trigger a protection action. The specific process includes the following steps.

[0080] The edge control unit 30 acquires real-time principal parameters. The edge control unit 30 obtains the current physical state data of the experimental platform through the principal parameter sensor 41 at a high-frequency sampling rate of 1kHz to 10kHz. To eliminate signal glitches caused by transient electromagnetic interference, the edge control unit 30 performs sliding median filtering on the raw data, selects 3 to 5 consecutive sampling points, sorts them, and takes the median value to obtain the real-time principal parameters. This real-time main parameter It represents the actual voltage, current, temperature, or pressure values ​​applied to the load at the current moment.

[0081] The edge control unit 30 performs dynamic limit violation determination. The edge control unit 30 acquires real-time main parameters. With the calculated dynamic safety upper limit and dynamic safety lower limit A comparison is performed. This comparison logic aims to determine whether the current physical state exceeds the dynamically permissible range compressed by environmental risk factors. The decision logic follows the Boolean expression: ; in: This indicates the dynamic circuit breaker trigger flag; a true value indicates that an over-limit has occurred. Indicates the real-time main parameter; Indicates the dynamic safety limit; Indicates the dynamic safety lower limit; Represents a logical OR operation.

[0082] The edge control unit 30 performs physical limit decisions. The edge control unit 30 performs hardware physical limit-based decisions in parallel. The edge control unit 30 reads the set of physical limit parameters parsed from the RFID tag in the preceding steps. Extract the response tolerance threshold. The edge control unit 30 determines real-time key parameters. Whether the hardware limit has been exceeded. The decision logic follows the following Boolean expression: ; in: ) indicates the hard fuse trigger flag; Indicates the real-time main parameter; This indicates the response tolerance threshold.

[0083] This step serves as redundant protection logic independent of the dynamic algorithm, ensuring device protection even when the dynamic algorithm is not triggered but the absolute value of the physical quantity is too large.

[0084] The edge control unit 30 performs graded protection actions. The edge control unit 30 determines the action based on the dynamic fuse trigger flag. and hard fuse trigger flag The logic state output control signal.

[0085] When the dynamic circuit breaker trigger flag is set True and hard circuit breaker triggered flag When the result is false, the edge control unit 30 determines that the system is in a dynamic limit violation state. The edge control unit 30 pulls down the enable pin of the logic control switch through the general purpose input / output interface, blocks the pulse width modulation drive signal of the power device, cuts off the main power output of the experimental platform, and keeps the physical protection interlock device 53 in a locked state. At this time, the interactive terminal 61 displays a dynamic boundary violation prompt.

[0086] When the hard fuse trigger flag is set When true, the edge control unit 30 determines that a physical-level over-limit state has been reached. The edge control unit 30 immediately triggers a non-maskable interrupt of the microcontroller, forcibly disconnects the main relay of the logic control switch in the interrupt service routine, and drives the buzzer to emit a continuous alarm sound. In this state, the physical protection interlock device 53 remains in a deadlock state until a reset command with administrator privileges is received.

[0087] The edge control unit 30 controls the physical protection interlock device 53 and the logic control switch through a drive circuit containing opto-isolators to realize the conversion from algorithm decision to physical action. The specific execution process includes the following steps.

[0088] The edge control unit 30 generates a hardware drive signal. The edge control unit 30 then generates a dynamic fuse trigger flag based on the steps described above. and hard fuse trigger flag Based on the user's manual operation commands, the final hardware control signal is calculated. The edge control unit 30 performs the following Boolean algebra logic operations to generate a power enable signal for controlling the power loop. and the unlocking signal for the shield used to control the electromagnetic lock :

[0089] in: A high level indicates that power output is enabled; A high level indicates that the electromagnetic lock is open; Indicates the logical NOT operation; Represents a logical OR operation; This represents the logical AND operation; This indicates the status of the running command (i.e., the status of the start button pressed by the user). Indicates the real-time main parameter; Indicates the safe zero energy threshold; This indicates a request to open the lid.

[0090] The edge control unit 30 drives the logic control switch to perform power on / off operations. The logic control switch adopts a hybrid switch topology, and its main circuit consists of a solid-state relay and an electromechanical relay connected in series. The edge control unit 30 is connected to the control terminal of the solid-state relay and the coil drive transistor of the electromechanical relay through optocouplers.

[0091] When the edge control unit 30 outputs the power enable signal When the level changes from high to low, the edge control unit 30 performs timing control: first, it pulls down the control level of the solid-state relay to cut off the load current within microseconds; then, it starts the internal hardware timer and, after a delay of 20ms to 50ms, pulls down the drive level of the electromechanical relay to open the physical contacts without current flowing through them.

[0092] The edge control unit 30 controls the physical protection interlock device 53 to perform mechanical locking. The physical protection interlock device 53 includes a normally closed electromagnetic latch and a position feedback sensor (such as a micro limit switch or a Hall sensor).

[0093] When the experiment is in operation (power enable signal) When either the fuse is true or a fuse protection circuit is activated (either the fuse flag is true), the shield unlock signal is activated. When set to low level, the electromagnetic latch is de-energized, and the latch pops out under the action of spring force, putting the protective cover into a mechanically locked state.

[0094] The system logic will only make a judgment if the following conditions are met simultaneously: no risk alarm, power is off, and real-time main parameters are available. The value has naturally decayed to the safe zero energy threshold. At this time, the edge control unit 30 outputs a high-level shield unlock signal. The electromagnetic coil is energized, retracting the bolt and allowing the user to open the protective cover. The position feedback sensor reads the physical position of the bolt in real time and feeds it back to the edge control unit 30. If the control signal is detected as locked but the physical position shows as unlocked, the edge control unit 30 will trigger a mechanical fault alarm.

[0095] After the experiment ends or is terminated by the circuit breaker mechanism, the edge control unit 30 enters the post-processing stage to perform backtracking and analysis of the experimental process data. The specific processing includes the following steps.

[0096] The edge control unit 30 extracts historical time-series data. The edge control unit 30 accesses its internal non-volatile memory via the file system, indexes and reads the real-time master parameters of the entire process based on the start and end timestamps of this experiment. The data is recorded. This data sequence consists of discrete numerical points recorded at a fixed sampling frequency in the preceding steps. Simultaneously, the edge control unit 30 retrieves the theoretical response master parameters set for this experiment from the configuration register. .

[0097] The edge control unit 30 performs data sequence construction. The edge control unit 30 organizes the read real-time master parameter data into a one-dimensional time series vector and confirms the total number of valid sampling points. If the experiment terminates prematurely due to the triggering of the circuit breaker, the edge control unit 30 will capture the data from the start of the experiment to the triggering time of the circuit breaker. Valid data segments between circuit breakers are used in subsequent calculations, while invalid data after the circuit breaker is triggered is discarded.

[0098] The edge control unit 30 calculates the trajectory deviation. The edge control unit 30 uses the mean absolute percentage error algorithm to calculate the average deviation between the actual trajectory of the real-time master parameter and the theoretical response master parameter. This calculation follows the formula: ; in: This represents the trajectory deviation, which is a percentage value. The smaller the value, the closer the experimental process is to the theoretical set value. This represents the total number of sampling points involved in the calculation; Indicates the serial number of the discrete sampling point ( ; Indicates the first The time points corresponding to each discrete sampling point; Indicates at time The collected real-time values ​​of the main parameters; Indicates the principal parameters of the theoretical response; This indicates the absolute value operation; This represents the normalized baseline value.

[0099] To avoid division by zero errors in mathematical calculations and to adapt to different experimental settings, the edge control unit 30 selects the normalized reference value according to the following logic. Under normal circumstances Values when When equal to 0 The value is the full-scale value of the sensor's range. The calculated trajectory deviation It will be stored in the user's historical experiment database as a quantitative weight parameter for the intelligent management system to subsequently evaluate the accuracy of experimental operations.

[0100] After calculating the trajectory deviation, the edge control unit 30 further analyzes the source of the deviation and generates targeted feedback data by decoupling objective environmental factors and human operation factors. The specific processing includes the following steps.

[0101] The edge control unit 30 calculates the average weight of environmental interference. The edge control unit 30 retrieves the comprehensive attenuation factor recorded during this experiment. The time series. The edge control unit 30 uses the arithmetic mean method to calculate the mean environmental instability over the entire experimental period. This calculation follows the formula: ; in: The value range is

[01] . The larger the value, the higher the instability of the external environment during the experiment, indicating that the average amplitude of the dynamic shrinkage safety threshold of the edge control unit 30 is larger. Indicates the total number of sampling points; Indicates the sampling point number; Indicates the first The combined attenuation factor at each sampling time.

[0102] The edge control unit 30 calculates the net operational deviation index. To objectively evaluate the user's experimental operational accuracy, the edge control unit 30 executes a weighted compensation algorithm to separate the passive deviation component caused by the environmental risk compression threshold from the overall trajectory deviation. This calculation follows the formula: ; in: The net operational deviation index reflects the degree of trajectory deviation caused solely by improper user operation control after removing the weight of environmental interference. Indicates the degree of trajectory deviation; This represents the average weight of environmental disturbances; This represents the environmental tolerance coefficient.

[0103] The edge control unit 30 configures the environmental tolerance coefficient according to the preset attributes of the experimental project. Its value range is constrained to [0, 1]. For introductory experiments, set... A value of 1.0 indicates maximum compensation for environmental factors, meaning that when environmental disturbances are significant, Significantly smaller than For advanced assessment experiments, set up A value of 0 indicates that no environmental compensation is performed. Directly equal to Users are required to maintain precise operation in any environment.

[0104] The edge control unit 30 generates an attribution determination result. The edge control unit 30 will calculate the result... and Each judgment threshold is logically compared, and differentiated feedback actions are executed based on the comparison results: like If the deviation exceeds a significant environmental threshold (e.g., 0.4), the edge control unit 30 determines that the experimental deviation is primarily due to external environmental factors. The edge control unit 30 generates a diagnostic log containing environmental fluctuation curves and displays maintenance suggestions for improving the experimental environment (e.g., enhancing heat dissipation, closing vents) on the interactive terminal 61.

[0105] like Less than or equal to the environmental significance threshold, and If the deviation exceeds the operational deviation threshold (e.g., 10%), the edge control unit 30 determines that the deviation in this experiment is primarily due to operational error. The edge control unit 30 retrieves the standard operation video stream of this experiment from the server via the communication module and pushes it to the interactive terminal 61 for playback.

[0106] like and If all values ​​are below their respective thresholds, the edge control unit 30 determines that the experimental process is qualified and marks the experimental data as valid samples and stores them in the historical database.

Claims

1. A smart management method for national border security science popularization and research experiments based on the PBL (Problem-Based Learning) model, wherein the method is executed by a distributed edge control system, characterized in that, Specifically, the following steps are included: The experimental hypothesis vector is received through the interactive terminal (61), the experimental initialization data is parsed to determine the experimental type, and the physical limit parameter set of the experimental sample is read through the radio frequency identification reader (62). Based on the proposed hypothesis vector, the theoretical response principal parameter is calculated, and a static safety envelope is constructed by combining the preset static tolerance coefficient. The hardware detection topology is constructed by controlling the signal topology isolation matrix (52). The expected parameters of the exploration hypothesis vector mapping are compared with the physical limit parameter set, and the physical protection interlock device (53) is controlled according to the comparison result. The real-time main parameters of the main parameter sensor (41) and the secondary parameter data of the secondary parameter sensor (42) are synchronously acquired by the multimodal sensor array (40). The real-time rate of change of the secondary parameter data is calculated and mapped to a nonlinear coupling attenuation factor. The nonlinear coupling attenuation factor is then used to shrink the static safety envelope to generate a dynamic safety threshold. Determine whether the real-time main parameter exceeds the preset error of the static safety envelope judgment theory, and determine whether the real-time main parameter exceeds the gradient coupling risk of the dynamic safety threshold judgment, and control the execution control subsystem (50) to act according to the judgment result; After the experiment, the trajectory deviation of the real-time principal parameter relative to the theoretical response principal parameter was calculated, and attribution analysis data including operational evaluation suggestions was generated by combining the average weight of environmental disturbances.

2. The method according to claim 1, characterized in that, The specific methods for constructing the static security envelope include: The external excitation parameters and the core physical parameters expected by the user are extracted by analyzing the research hypothesis vector, and the matching physical law function is retrieved according to the experimental type; Substitute the external excitation parameters and the user-expected core physical parameters into the physical law function to calculate the theoretical response principal parameters. Read the static tolerance coefficient corresponding to the current experimental risk level, and calculate the allowable fluctuation range according to the relative proportion determined by the static tolerance coefficient, based on the theoretical response master parameter, to obtain the upper and lower limits of the static safety envelope. When the lower limit of the static safety envelope is less than the physical zero point and the current physical parameter does not have a negative physical meaning, the lower limit is forcibly set to zero.

3. The method according to claim 1, characterized in that, The steps of comparing the expected parameters of the exploration hypothesis vector mapping with the physical limit parameter set, and controlling the physical protection interlock device (53) according to the comparison result, specifically include: The external excitation parameters are obtained by analyzing the research hypothesis vector. The maximum allowable input excitation of the sample, i.e., the excitation tolerance threshold, and the maximum allowable response value of the sample, i.e., the response tolerance threshold, are extracted from the set of physical limit parameters. Determine whether the external excitation parameter is less than or equal to the excitation tolerance threshold, and determine whether the theoretical response principal parameter is less than or equal to the response tolerance threshold; If all judgment results are true, an unlock signal is output to the physical protection interlock device (53); if any judgment result is false, the locked state of the physical protection interlock device (53) is maintained and a parameter conflict is indicated.

4. The method according to claim 1, characterized in that, The specific methods for calculating the real-time rate of change of the secondary parameter data and mapping it to a nonlinear coupling decay factor include: The collected secondary parameter data are processed by moving average filtering or first-order lag filtering to obtain smoothed secondary parameters. Perform discrete-time difference operations on the smoothed secondary parameters and normalize them to the time dimension to obtain the real-time rate of change of the secondary parameter data; A nonlinear activation mechanism based on the Sigmoid function is adopted, and the mapping relationship between gradient and risk is constructed by using exponential operation. The real-time rate of change of the secondary parameter data is mapped to the nonlinear coupling decay factor with a value between zero and one. The mapping relationship includes a sensitive situation coefficient and a critical gradient threshold. The sensitive situation coefficient adjusts the steepness of the response curve, and the critical gradient threshold defines the inflection point of the risk response.

5. The method according to claim 4, characterized in that, When the multimodal sensing array (40) contains multiple secondary parametric sensors (42) of different types, the method for calculating the comprehensive attenuation factor is as follows: Calculate the corresponding nonlinear coupling attenuation factor for each sensor channel; Execute the maximum value filtering logic and select the factor with the largest value among all channels as the comprehensive attenuation factor at the current moment; The specific method for generating the dynamic safety threshold using the comprehensive attenuation factor is as follows: Centered on the theoretical response principal parameter, the value obtained by subtracting the comprehensive attenuation factor is used as the compression ratio coefficient to reduce the allowable fluctuation range defined by the static safety envelope, and the dynamic safety upper limit and dynamic safety lower limit constituting the dynamic safety threshold are calculated respectively.

6. The method according to claim 1, characterized in that, Determine whether the real-time main parameter exceeds the preset error of the static safety envelope judgment theory, and determine whether the real-time main parameter exceeds the gradient coupling risk of the dynamic safety threshold judgment. Based on the judgment result, control the execution control subsystem (50) to perform actions, specifically including: The real-time main parameter is compared with the dynamic security upper limit and dynamic security lower limit that constitute the dynamic security threshold. If the value exceeds the range, a dynamic circuit breaker trigger flag is generated. The response tolerance threshold of the physical limit parameter set is read in parallel to determine whether the real-time main parameter exceeds the response tolerance threshold. If it exceeds the threshold, a hard fuse trigger flag is generated. A power enable signal and a shield unlock signal are generated based on the dynamic fuse trigger flag and the hard fuse trigger flag. When the power enable signal changes from an active state to an inactive state, the logic control switch in the execution control subsystem (50) is driven to perform a hybrid disconnection sequence: first, the control signal of the solid-state relay in the series circuit is cut off, and after a preset delay time, the drive signal of the electromechanical relay in the series circuit is cut off.

7. The method according to claim 6, characterized in that, The method further includes the step of controlling the physical protection interlock device (53) to perform mechanical locking: When the power supply enable signal is valid or any fuse trigger flag is valid, the physical protection interlock device (53) is controlled to be in a mechanically locked state. Only when the following conditions are met simultaneously: no risk alarm, power is cut off, and the value of the real-time main parameter naturally decays to below the safe zero energy threshold, will a valid shield unlocking signal be output to drive the electromagnetic coil to release the mechanical lock. The physical position of the lock is monitored in real time by a position feedback sensor. If the control signal indicates that the lock is locked but the physical position of the lock indicates that it is not locked, a mechanical fault alarm is triggered.

8. The method according to claim 1, characterized in that, The steps for calculating the trajectory deviation of the real-time principal parameter relative to the theoretical response principal parameter specifically include: Extract the valid data segment from the start of the experiment to the moment the circuit breaker is triggered; The average absolute percentage error algorithm is used to calculate the average ratio of the absolute value of the difference between the real-time main parameter and the theoretical response main parameter to the normalized reference value, thereby obtaining the trajectory deviation. The selection logic for the normalized benchmark value is as follows: The magnitude of the theoretical response main parameter is compared with a preset validity threshold. When the magnitude of the theoretical response principal parameter is greater than the validity determination threshold, the normalized benchmark value is taken as the theoretical response principal parameter; When the amplitude of the theoretical response master parameter is less than or equal to the validity determination threshold, the normalized reference value is taken as the full-scale value of the sensor.

9. The method according to claim 8, characterized in that, The step of generating attribution analysis data containing operational evaluation recommendations by combining the average weight of environmental disturbances specifically includes: The arithmetic mean of the nonlinear coupling attenuation factor during the experimental period is calculated to obtain the average weight of the environmental disturbance. The net operational deviation index is calculated based on the weighted compensation logic. The net operational deviation index is equal to the trajectory deviation multiplied by the correction coefficient. The correction coefficient is the difference between the value "1" and the product of the environmental tolerance coefficient and the average weight of the environmental disturbance. The average weight of the environmental disturbance is compared with the environmental significance threshold, and the net operating deviation index is compared with the operating deviation threshold. If the average weight of the environmental disturbance exceeds the environmental significance threshold, it is determined to be an external environment-dominated deviation and environmental improvement suggestions are generated. If the average weight of the environmental interference does not exceed the environmental significance threshold and the net operational deviation index exceeds the operational deviation threshold, it is determined to be an operational error-driven deviation and the standard operation video is retrieved.

10. The method according to claim 9, characterized in that, The environmental tolerance coefficient is configured according to the teaching attributes of the experimental project: For introductory-level experiments, the environmental tolerance coefficient is configured to a first preset value greater than or equal to 0.8; For advanced assessment experiments, the environmental tolerance coefficient is configured to a second preset value that is less than or equal to 0.2.

Images