Physical experiment examination judgment method, device and equipment based on digital instrument

Abstract

The invention relates to a physical experiment examination judgment method, device and equipment based on a digital instrument, and belongs to the field of digital processing, experimental physical parameters and operation behavior data are collected in real time through a sensor, and after signal noise reduction and unit standardization preprocessing, the data are matched with a preset experimental operation standard library to extract key features; constructing an experimental environment model by using a digital twinning technology, performing time-space alignment analysis on key features and a standard process, identifying operation abnormality and data deviation, and generating sub-items and comprehensive scores by combining a pre-trained intelligent analysis model with a teaching outline scoring reference; and finally, generating a visual report containing an operation track graph, a knowledge mastering degree graph and personalized suggestions, and performing non-tampering evidence storage on experimental data and a scoring process through a block chain technology. The method solves the problems of large operation error, high subjectivity, process non-traceability and the like in the traditional physical experiment examination.

Description

Technical Field

[0001] The present invention relates to the technical field of digital processing, and particularly relates to a method, device, and equipment for evaluating physical experiment examinations based on digital instruments. Background Art

[0002] Traditional physical experiment examination evaluation mainly relies on manual operation and subjective scoring, and there are significant technical bottlenecks. In the data acquisition link, when manually measuring physical quantities such as temperature and voltage using mechanical instruments, it is easily affected by the operator's reading habits and environmental light factors, resulting in large data deviations. For example, when manually reading the ammeter value in a circuit experiment, an error of ±0.5 mA may occur; in terms of evaluating operation standardization, it is difficult for teachers to track every step of the students' operations throughout the process. For example, the angular deviation of lens adjustment in an optical experiment and the position error of weight placement in a mechanics experiment are often ignored, and the scoring criteria are easily affected by subjective experience. The evaluation of the same operation by different teachers may have a score difference of 10%-20%. In addition, traditional evaluation lacks systematic recording of the experimental process and cannot trace the time series of data acquisition and operation trajectories. For example, it is difficult to accurately capture the behavior of students not preheating the instrument according to the standard process in a thermal experiment, resulting in the evaluation only focusing on the results and ignoring the process. At the same time, data processing relies on manual calculation. When facing complex experiments such as alternating current signal analysis, the efficiency of manually drawing waveform diagrams and calculating frequency errors is low, and it is difficult to discover hidden laws in the data, such as the cumulative impact of periodic noise on experimental results. Although there are some applications of digital instruments in the prior art, there is a lack of a systematic solution that combines digital twin technology, intelligent analysis models, and blockchain evidence storage, and it is impossible to achieve full-process automation and objectivity from data acquisition, analysis to scoring, and it is difficult to meet the current experimental teaching requirements for accurate evaluation, process traceability, and personalized guidance. Summary of the Invention

[0003] The main object of the present invention is to provide a method, device, and equipment for evaluating physical experiment examinations based on digital instruments, which solves the problems of large operation errors, strong subjectivity, and non-traceable processes in traditional physical experiment examinations.

[0004] To achieve the above object, the method for evaluating physical experiment examinations based on digital instruments provided by the present invention includes the following steps: Real-time collect physical parameter data and operation behavior data during the experiment process. The physical parameter data includes at least one of spatial position parameters, circuit signal parameters, optical phenomenon parameters, and mechanical action parameters of the experimental device, and the operation behavior data includes instrument operation action sequences and experimental step execution trajectories; Preprocess the collected data to form standardized data through signal denoising and unit standardization; match the standardized data with a preset experimental operation standard library to extract key features such as the integrity of experimental steps, the standardization of instrument operation, and the accuracy of data collection; Build an experimental environment model based on digital twin technology, perform spatio-temporal alignment analysis on the key features and the standard experimental procedures in the model to identify operation anomaly points and data deviation values; calculate the weights of the key features through a pre-trained intelligent analysis model, and generate item scores and comprehensive scores in combination with the scoring benchmarks set by the experimental teaching syllabus; Generate a visual analysis report according to the scoring results, and the report includes an operation trajectory map, a knowledge mastery map, and personalized improvement suggestions; use blockchain evidence storage technology to store the original experimental data and scoring process data in an immutable manner to form a traceable exam record.

[0005] Further, the steps of collecting physical parameter data and operation behavior data during the experiment in real time include: Real-time sample the physical parameters of the experimental device through sensors, and convert physical quantities such as temperature, pressure, current, and voltage into electrical signals; convert the electrical signals into digital signals through an analog-to-digital conversion module, continuously collect the digital signals at a preset frequency, and store timestamps; Use image acquisition devices arranged at multiple angles to record videos of the experimental operation area, and synchronously record the spatial trajectory and time series of operation actions; synchronize and calibrate the digital signals collected by the sensors and the video stream according to the timestamps to form an original data set with spatio-temporal tags.

[0006] Further, the steps of preprocessing the collected data to form standardized data through signal denoising and unit standardization include: Adopt a wavelet transform filtering algorithm to remove noise from the digital signals collected by the sensors, decompose the signal frequency band by setting thresholds, and filter out high-frequency interference noise and low-frequency drift noise; for signals with periodic interference, use a Kalman filtering algorithm for dynamic noise suppression, and establish a signal state space model to iteratively optimize the signal estimated value; Uniformly convert temperature data to Kelvin units, voltage / current data to volt / ampere standard units, and mechanical parameters to newton / pascal units; batch-convert non-standard physical quantity units through a preset conversion coefficient table to form data in the International System of Units standard; Calibrate the timestamps of the operation behavior data in the video stream to eliminate the timing error caused by the clock deviation between the camera and the sensor; synchronize multi-source data with inconsistent sampling frequencies through an interpolation algorithm to ensure that the time axes of the physical parameter data and the operation behavior data are aligned.

[0007] Further, the steps of matching the standardized data with a preset experimental operation standard library and extracting the key features of the integrity of experimental steps, the standardization of instrument operation, and the accuracy of data collection include: Load the preset experimental operation standard library, which includes the standard operation procedures disassembled based on the physical experiment teaching syllabus, the timing logic relationship of each step, and the key parameter thresholds; decompose the standardized data into operation action segments according to the time series, perform semantic matching with the standard steps in the standard library, and identify the degree of coincidence between the actual operation steps and the standard process and the missing steps; Compare the instrument operation action sequence in the standardized data with the preset operation postures, force thresholds, and time interval standard parameters in the standard library, and mark the illegal operations exceeding the thresholds; calculate the similarity between the actual operation trajectory and the standard operation trajectory through the dynamic time warping algorithm to generate the operation standardization scoring feature; Regarding the feature of data collection accuracy, compare the standardized physical parameter data with the theoretical data curve and the error tolerance range of the corresponding experiment in the standard library, and extract the feature parameters of data deviation and fluctuation amplitude; analyze the difference between the frequency spectrum feature of the periodic experimental data and the standard frequency spectrum through Fourier transform to identify the systematic error in the data collection process.

[0008] Further, the steps of constructing an experimental environment model based on digital twin technology and performing spatio-temporal alignment analysis on the key features and the standard experimental process in the model to identify operation abnormal points and data deviation values include: Based on the geometric parameters, physical characteristics, and dynamic behaviors of the physical experimental device, construct a 1:1 scale digital twin experimental environment model through 3D modeling technology, and embed the timing nodes and parameter thresholds of the standard experimental process in the model; map the key features extracted from the standardized data to the corresponding entity objects and time axis of the digital twin model; Through the timestamp synchronization mechanism, align the time series of the key features with the theoretical time nodes of the standard experimental process in the digital twin model, and calculate the time deviation value between the actual operation steps and the standard process; use the space coordinate mapping algorithm to compare the instrument operation trajectory and the spatial characteristics of the experimental phenomenon occurrence position with the standard operation space area in the digital twin model to identify the abnormal operations exceeding the preset space range; Perform dynamic fitting analysis on the theoretical parameter curve of the standard experimental process in the digital twin model and the actually collected physical parameter data, and calculate the data deviation parameter; through the preset abnormal rule library in the model, automatically mark the abnormal points of incorrect timing of operation steps, over-limit of instrument state parameters, and data collection result exceeding the difference, and generate a detailed record of the abnormal type and deviation value.

[0009] Further, perform weight calculation on the key features through a pre-trained intelligent analysis model, and generate the steps for sub-item scoring and comprehensive scoring in combination with the scoring benchmarks set in the experimental teaching syllabus, including: Load an intelligent analysis model trained based on historical experimental data. The model ranks the importance of key features such as the integrity of experimental steps, the standardization of instrument operation, and the accuracy of data acquisition through the gradient boosting algorithm, and generates a feature weight matrix. Input the key features of operation anomaly points and data deviation values extracted from the digital twin model analysis into the model, and calculate the influence coefficient of each feature on the experimental result based on the weight matrix; Retrieve the preset scoring benchmarks in the experimental teaching syllabus. The benchmarks include the score distribution standards for each experimental session, the scoring thresholds for operation specifications, and the scoring levels for data accuracy; Map the influence coefficient output by the model to the score distribution rules in the scoring benchmark to generate sub-item scores. Calculate the step scores for the compliance of the experimental steps with the standard process, calculate the operation scores for the standardization of instrument operation according to the number of violations and the severity, and calculate the data scores for the accuracy of data acquisition according to the level where the deviation value is located; Integrate each sub-item score through the weighted summation algorithm. The weight parameters are set according to the importance of the experimental session in the teaching syllabus to generate the experimental comprehensive score. Synchronously mark the deduction points and the basis for deduction of each scoring item to form a score generation record with detailed scoring logic.

[0010] Further, generate a visual analysis report according to the scoring results. The steps for the report including the operation trajectory map, the knowledge mastery degree map, and personalized improvement suggestions include: Generate an operation trajectory map based on the time series data of the experimental operation and the time series comparison result with the standard process: use the time axis as the horizontal axis, visualize the compliance of the actual operation steps with the standard steps and the time deviation value in the form of a heat map, mark the missing steps and time sequence error anomaly points with different colors, and superimpose and display the overlap degree of the spatial coordinate curve of the instrument operation trajectory and the standard trajectory; Construct a knowledge mastery degree map, associate the knowledge points involved in the experiment with the scoring results of the key features, and display the mastery degree of each knowledge point in the form of a node network. The size of the node corresponds to the weight of the knowledge point, and the depth of the color reflects the scoring deviation value; According to the deduction points in the sub-item scoring, match the preset teaching suggestion library, generate specific improvement measures for each deduction point, and organize them into a stepped learning path according to the knowledge modules.

[0011] Further, perform tamper-proof storage on the original experimental data and scoring process data through blockchain evidence storage technology to form a traceable exam record. The steps include: Classify and package the original experimental data and the scoring process data to generate a data set; calculate a unique hash value for each data set, and construct a data digest through the Merkle tree algorithm to ensure data integrity verification; Package the data digest and the timestamp information to form a block to be chained. Verify the legality of the block through the blockchain consensus mechanism, and link the legal blocks to the main chain of the blockchain in chronological order; embed the experiment number and student identification association information in the block header to establish a mapping relationship between the data and the examination subject; Generate an immutable blockchain certification identifier for each certified block, which includes block height and hash value information; associate and bind the certification identifier with the visualization analysis report to form an electronic examination record with a blockchain traceability path; Distributively store the certified data through the blockchain node network, support real-time query of the data's chaining time and modification record information through the block hash value or the certification identifier, and realize the full-process traceability of the original experimental data and the scoring process; when it is necessary to verify the data integrity, re-calculate the data hash value and compare it with the digest value stored in the blockchain to ensure the immutability of the evidence chain.

[0012] A physical experiment examination and evaluation device based on digital instruments proposed by the present invention includes: An acquisition unit for real-time acquisition of physical parameter data and operation behavior data during the experiment process. The physical parameter data includes at least one of spatial position parameters, circuit signal parameters, optical phenomenon parameters, and mechanical action parameters of the experimental device, and the operation behavior data includes an instrument operation action sequence and an experimental step execution trajectory; A standard unit for preprocessing the acquired data, forming standardized data through signal denoising and unit standardization; matching the standardized data with a preset experimental operation standard library, and extracting key features of experimental step integrity, instrument operation standardization, and data acquisition accuracy; A simulation unit for constructing an experimental environment model based on digital twin technology, performing spatio-temporal alignment analysis on the key features and the standard experimental process in the model, and identifying operation abnormal points and data deviation values; calculating weights for the key features through a pre-trained intelligent analysis model, and generating item scores and comprehensive scores in combination with the scoring benchmarks set by the experimental teaching syllabus; A report unit for generating a visualization analysis report according to the scoring results. The report includes an operation trajectory map, a knowledge mastery degree map, and personalized improvement suggestions; storing the original experimental data and the scoring process data in an immutable manner through blockchain certification technology to form a traceable examination record.

[0013] The present invention also provides a computer device, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps of the above-mentioned physical experiment examination evaluation method based on digital instruments are implemented.

[0014] The physical experiment examination evaluation method, device and equipment based on digital instruments provided by the present invention have the following beneficial effects: By using digital sensors to collect physical parameters (such as temperature, current) in real time and combining wavelet transform noise reduction preprocessing technology, artificial measurement errors and environmental interferences are eliminated, and subjective biases in traditional manual scoring are avoided.

[0015] By using a digital twin model and a pre-trained intelligent analysis model, automatic matching of operation trajectories, identification of abnormal points and calculation of weights are completed, realizing full-process automation from data collection to comprehensive scoring.

[0016] Through blockchain evidence storage technology, the original experimental data and the scoring process are stored in an unalterable manner. Combined with a visual analysis report, detailed traceable scoring basis and personalized improvement suggestions are provided for teachers and students, helping to improve teaching quality. Description of the Drawings

[0017] Figure 1 is a schematic flowchart of a physical experiment examination evaluation method based on digital instruments in an embodiment of the present invention; Figure 2 is a structural block diagram of a physical experiment examination evaluation device based on digital instruments in an embodiment of the present invention; Figure 3 is a schematic structural block diagram of a computer device in an embodiment of the present invention.

[0018] The realization, functional features and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. Detailed Embodiments

[0019] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

[0020] Referring to Figure 1 , which is a schematic flowchart of a physical experiment examination evaluation method based on digital instruments proposed by the present invention, the method includes the following steps: S1. Collect physical parameter data and operation behavior data in the experiment process in real time. The physical parameter data includes at least one of spatial position parameters, circuit signal parameters, optical phenomenon parameters, and mechanical action parameters of the experimental device. The operation behavior data includes an instrument operation action sequence and an experimental step execution trajectory; S2. Preprocess the collected data, form standardized data through signal denoising and unit standardization; match the standardized data with a preset experimental operation standard library, and extract the key features of the integrity of experimental steps, the standardization of instrument operation, and the accuracy of data collection. S3. Build an experimental environment model based on digital twin technology, perform spatio-temporal alignment analysis on the key features and the standard experimental process in the model, and identify operation anomaly points and data deviation values; calculate the weights of the key features through a pre-trained intelligent analysis model, and generate sub-item scores and comprehensive scores in combination with the scoring benchmarks set in the experimental teaching syllabus. S4. Generate a visual analysis report according to the scoring results. The report includes an operation trajectory map, a knowledge mastery degree map, and personalized improvement suggestions; use blockchain evidence storage technology to store the original experimental data and scoring process data in an immutable manner to form a traceable examination record.

[0021] Specifically: In the experimental bench circuit, a high-precision current sensing device is connected in series and a voltage sensing device is connected in parallel to collect the loop current and the voltage signal across the resistor in real time. Among them, the current sensing device can convert the physical quantity of current from 0 - 500 mA into an electrical signal, which is converted into a digital signal at a frequency of 100 Hz by an analog-to-digital conversion module and marked with a millisecond-level timestamp; the voltage sensing device synchronously collects the 0 - 10 V voltage signal and stores it synchronously in the edge computing node. At the same time, a binocular image acquisition device is arranged above the experimental bench to record the operation process of students adjusting the sliding rheostat and connecting the circuit at a frame rate of 30 fps, and extract the operation action sequence (such as the wire connection sequence, the adjustment direction of the rheostat knob) and the step execution trajectory (such as the change curve of the rheostat resistance over time) through image recognition technology. The digital signal output by the sensor and the video stream of the image acquisition device are calibrated through a hardware clock synchronization module, so that the voltage data at a certain moment (such as 2.35 V) is accurately bound to the operation action of "closing the switch" in the video frame, forming an original data set with spatio-temporal marks to ensure the integrity and temporal consistency of the experimental process data; Remove the temperature fluctuations caused by environmental interference through filtering. For example, for a heating curve with ±0.5°C random noise, decompose the signal frequency band by setting a threshold, filter out high-frequency interference and low-frequency drift, reducing the noise level of the temperature data by more than 70%. Subsequently, convert the temperature data (original unit: degree Celsius) to Kelvin units and the length data (original unit: millimeter) to meter units according to physical formulas, and batch process through a preset unit conversion table to form a standard data set in the International System of Units. In the key feature extraction link, load the standard operation library constructed based on the physical experiment teaching syllabus. This library contains standard processes such as "preheat the instrument for 30 minutes" and "record the temperature every 5 minutes". Compare the standardized temperature-time data with the standard library. If it is found that the student did not perform the preheating step, it is identified as a missing step. By comparing the degree of coincidence between the actual heating rate curve and the standard curve, if the heating rate exceeds the standard range, it is marked as an operation violation. Compare the finally measured linear expansion amount data of the metal wire with the theoretical calculated value. If the deviation exceeds the allowable range (such as ±2%), extract it as a data accuracy feature. For example, when the actual expansion amount deviation reaches 3.5%, automatically record this data deviation as the scoring basis.

[0022] Build a 1:1 digital twin experimental environment model based on the three-dimensional geometric parameters of the interferometer (such as the adjustable range of the mirror spacing 0 - 100mm, the angular accuracy of the beam splitter ±0.1°). The timing nodes of the standard adjustment process (such as "coarse-tune the mirror inclination first and then fine-tune") and parameter thresholds (the adjustable range of the inclination ±0.5°) are embedded in the model. Map the preprocessed key features (such as the mirror adjustment angle sequence, the clarity parameter of the interference fringes) to the corresponding entity objects in the digital twin model. Through the timestamp synchronization mechanism, compare the time deviation between the actual operation and the standard process. For example, it is found that the student prematurely terminated the coarse-tuning step (the standard time-consuming is 5 minutes, but actually only 2 minutes), and through the spatial coordinate mapping algorithm, identify the abnormal operation where the mirror inclination adjustment amount reaches 1.2° (exceeding the threshold by 0.7°). In the intelligent scoring link, based on the scoring benchmarks set in the teaching syllabus (such as the total score is 100 points, the step integrity accounts for 40%, the operation standardization accounts for 35%, and the data accuracy accounts for 25%), calculate the weights of the identified abnormal points. For example, the influence coefficient of the missing step is 0.3, corresponding to a deduction of 10 points; the influence coefficient of the operation exceeding the standard is 0.25, corresponding to a deduction of 5 points; the influence coefficient of the data exceeding the standard is 0.2, corresponding to a deduction of 8 points. Generate a comprehensive score through weighted summation: (100 - 10)×40% + (100 - 5)×35% + (100 - 8)×25% = 92.25 points, and synchronously mark the specific reasons for each deduction point (such as "did not complete the coarse-tuning step", "the inclination adjustment exceeded the range") to ensure the traceability of the scoring process.

[0023] The operation trajectory map uses the time axis as the horizontal axis and shows the degree of compliance of the steps of "installing the pendulum ball" and "adjusting the pendulum length" with the standard process in a heat map. Abnormal points are marked with different colors (for example, when the pendulum length is adjusted to 101.2 cm, exceeding the standard value of 100±0.5 cm), and the overlap curve of the pendulum ball swing trajectory and the standard trajectory is superimposed; the knowledge mastery map associates the knowledge point of the "simple pendulum period formula" involved in the experiment with the scoring results and shows it in the form of a node network. The size of the node reflects the weight of the knowledge point, and the shade of the color indicates the degree of mastery (for example, a 5% error in period measurement corresponds to a relatively red color of the node, indicating a knowledge shortcoming); personalized improvement suggestions match the teaching suggestion library according to the deduction points to generate a step-by-step learning path of "using a vernier caliper to measure the pendulum length multiple times" and "reviewing the error synthesis method". In the blockchain evidence storage link, the original experimental data (such as the pendulum angle - time series, operation video) and the scoring process data (sub-item scoring calculation log) are classified and encapsulated. After calculating the unique hash value, a data digest is constructed through the Merkle tree algorithm and packaged into a block with a timestamp (such as 2025-06-23 14:30:22). After being verified by the blockchain consensus mechanism, it is stored on the chain. The block header embeds the experiment number and student identification to generate an evidence storage identifier containing the block height and hash value, which is bound to the visualization report. Through the blockchain node network, distributed storage is realized. Users can query the data upload time and modification records through the evidence storage identifier. When verifying, the data hash value is recalculated and compared with the on-chain digest to ensure that the data cannot be tampered with, forming an electronic exam record that can be traced throughout the process.

[0024] In one embodiment, the steps of collecting physical parameter data and operation behavior data during the experiment in real time include: The physical parameters of the experimental device are sampled in real time through sensors, and physical quantities such as temperature, pressure, current, and voltage are converted into electrical signals; the electrical signals are converted into digital signals through an analog-to-digital conversion module, and the digital signals are continuously collected at a preset frequency and the timestamps are stored. The experimental operation area is video-recorded by image acquisition devices arranged at multiple angles, and the spatial trajectory and time series of the operation actions are synchronously recorded; the digital signals collected by the sensors and the video stream are synchronously calibrated according to the timestamps to form an original data set with space-time marks.

[0025] First, install a mechanical sensor and an angle sensor on the pendulum support. The mechanical sensor samples the tension of the pendulum string in real time (range 0 - 5N), and the angle sensor monitors the swing angle of the pendulum bob (accuracy ±0.5°). The two types of sensors convert the physical quantities of tension and angle into 0 - 5V electrical signals, which are converted into digital signals at a frequency of 50Hz through an analog-to-digital conversion module. Each data point is marked with a millisecond-level timestamp (such as 2025-06-23 10:15:30.256ms) and stored in the edge computing unit. At the same time, arrange two high-definition cameras in front of and on the side of the experimental bench to record the operation process of students installing the pendulum bob, adjusting the pendulum length, and releasing the pendulum bob at a frame rate of 25fps. Through image recognition technology, track the movement trajectory of the pendulum bob and the hand operation actions of students (such as the rotation direction when adjusting the pendulum length and the timing of releasing the pendulum bob), and synchronously record the spatial coordinates of the operation actions (such as the three-dimensional position when the pendulum length is adjusted to 100cm) and the time series (such as the pendulum length adjustment is completed at 10:15:20). The digital signals collected by the sensors (such as the pendulum angle is 5° at 10:15:30) and the camera video stream are calibrated through a hardware clock synchronization module to eliminate the device clock deviation, so that the tension data, angle data, and the swing state of the pendulum bob in the corresponding video frames are accurately matched. Finally, a raw dataset with spatio-temporal markers containing timestamps, physical parameter values, and operation video segments is formed, providing a complete and temporally consistent data basis for subsequent experimental analysis.

[0026] In one embodiment, the steps of preprocessing the collected data to form standardized data through signal denoising and unit standardization include: Adopt the wavelet transform filtering algorithm to remove noise from the digital signals collected by the sensors, decompose the signal frequency band by setting a threshold, and filter out high-frequency interference noise and low-frequency drift noise; for signals with periodic interference, use the Kalman filtering algorithm for dynamic noise suppression, and establish a signal state space model to iteratively optimize the signal estimated value; Uniformly convert the temperature data to the Kelvin unit, the voltage / current data to the volt / ampere standard unit, and the mechanical parameters to the newton / pascal unit; batch-convert non-standard physical quantity units through a preset conversion coefficient table to form data in the International System of Units (SI); Calibrate the timestamps of the operation behavior data in the video stream to eliminate the timing error caused by the clock deviation between the camera and the sensors; synchronize multi-source data with inconsistent sampling frequencies through an interpolation algorithm to ensure that the time axes of the physical parameter data and the operation behavior data are aligned.

[0027] Noise processing is performed on the original signals collected by the temperature sensor. By setting an appropriate threshold value, the signals are decomposed according to the frequency range, effectively filtering out high-frequency fluctuations (such as short-term temperature jumps caused by device heat dissipation) and low-frequency drifts (such as the influence of slow changes in the overall temperature and humidity in the laboratory) generated by environmental interference. For the regular temperature oscillations caused by the periodic fluctuations in the power of the heating device, a dynamically adjusted signal optimization mechanism is adopted. A state model is established based on the real-time collected data, and the temperature estimation value is continuously corrected through iterative calculations, reducing the noise impact by more than 70%. In the unit standardization process, the temperature data (original unit is Celsius) is uniformly converted to the Kelvin unit according to the physical formula (for example, 20°C is converted to 293.15 K), the voltage / current data is converted to the standard units of volts / amps, and the mechanical parameters (such as pressure) are converted to the units of newtons / pascals. For the non-standard unit of millimeters of mercury, batch conversion is performed through a preset conversion table. For example, 760 millimeters of mercury is converted to the pascal unit according to the coefficient 133.322, forming a standard data set that conforms to the International System of Units (SI). Regarding the time synchronization problem between the video stream and the sensor data, first, the clock deviation between the camera and the sensor is calibrated through the hardware clock synchronization module to eliminate the timing error caused by the asynchronous device clocks (such as an operation actually occurring at 10:15:30, but the camera records it as 10:15:30.5). Then, for the data with different sampling frequencies (such as the sensor samples at 100 Hz and the camera records at 25 fps), the interpolation calculation method is used to insert intermediate values on the time axis to accurately align the current change data with the corresponding operation video frames, ensuring the time series consistency of the physical parameters and the operation behavior.

[0028] In one embodiment, the steps of matching the specification data with a preset experimental operation standard library and extracting the key features of the integrity of the experimental steps, the standardization of instrument operations, and the accuracy of data collection include: Load the preset experimental operation standard library, which contains the standard operation procedures disassembled based on the physical experiment teaching syllabus, the timing logic relationships of each step, and the key parameter thresholds. Decompose the specification data into operation action segments according to the time series, and perform semantic matching with the standard steps in the standard library to identify the compliance degree of the actual operation steps with the standard process and the missing steps. Compare the instrument operation action sequence in the specification data with the preset operation postures, force thresholds, and time interval specification parameters in the standard library, and mark the violation operations that exceed the thresholds. Calculate the similarity between the actual operation trajectory and the standard operation trajectory through the dynamic time warping algorithm to generate the operation standardization scoring features. Regarding the accuracy characteristics of data acquisition, compare the standardized physical parameter data with the theoretical data curve and the allowable error range of the corresponding experiment in the standard library, and extract the characteristic parameters of data deviation and fluctuation amplitude. For periodic experimental data, analyze the difference between its spectral characteristics and the standard spectrum through Fourier transform to identify systematic errors in the data acquisition process.

[0029] Load the standard operation library constructed based on the "College Physics Experiment Teaching Syllabus". This library includes standard processes such as "check the instrument range before circuit connection" and "adjust the voltage point by point and record the current", as well as parameter thresholds such as "the voltage meter range should be set to 0 - 10V" and "the adjacent voltage adjustment interval should not be less than 0.5V". Decompose the pre - processed standardized data into operation action segments such as "connect the circuit", "adjust the voltage", and "record the data" in chronological order, and conduct semantic comparison with the steps in the standard library. For example, if it is identified that the student did not perform the step of "check the ammeter range", it is determined that there is a lack of operation integrity. In the link of extracting the standardization characteristics of instrument operation, compare the action sequence of the student adjusting the sliding rheostat with the standard parameter of "slowly rotate the knob, and the rotation angle per second does not exceed 15°" in the standard library. If the actual adjustment speed reaches 25° per second, it is marked as an illegal operation; by calculating the coincidence degree between the actual adjustment trajectory of the sliding rheostat and the standard trajectory (for example, the standard trajectory requires the voltage to rise uniformly from 0V to 5V, but the actual voltage jumps to 6V), generate the scoring characteristics of operation standardization. Regarding the accuracy of data acquisition, compare the standardized voltage - current data with the theoretical volt - ampere characteristic curve in the standard library (when the resistance value is 50Ω, the current should be linearly related to the voltage). If the measured data points deviate from the theoretical curve by more than ±3%, extract the characteristic parameter with a data deviation of 4.2%; for the periodically fluctuating current data, analyze whether its fluctuation amplitude exceeds the allowable range of ±0.1mA in the standard library, and identify the current fluctuation exceeding the standard due to poor contact (the actual fluctuation reaches ±0.3mA) as the basis for deducting points for data acquisition accuracy.

[0030] In one embodiment, the steps of constructing an experimental environment model based on digital twin technology and performing spatio - temporal alignment analysis of key features with the standard experimental process in the model to identify operation abnormal points and data deviation values include: Based on the geometric parameters, physical characteristics, and dynamic behavior of the physical experimental device, construct a 1:1 scale digital twin experimental environment model through 3D modeling technology. The time - sequence nodes and parameter thresholds of the standard experimental process are embedded in the model; map the key features extracted from the standardized data to the corresponding entity objects and time axis of the digital twin model; Through the timestamp synchronization mechanism, align the time series of key features with the theoretical time nodes of the standard experimental process in the digital twin model, and calculate the time deviation value between the actual operation steps and the standard process; use the spatial coordinate mapping algorithm to compare the spatial features of the instrument operation trajectory and the occurrence location of experimental phenomena with the standard operation space area in the digital twin model, and identify abnormal operations beyond the preset spatial range. Conduct dynamic fitting analysis on the theoretical parameter curve of the standard experimental process in the digital twin model and the actually collected physical parameter data, and calculate the data deviation parameter; through the preset abnormal rule library in the model, automatically mark abnormal points such as timing errors in operation steps, over-limit of instrument state parameters, and abnormal data acquisition results, and generate a detailed record of the abnormal type and deviation value.

[0031] Based on the actual geometric parameters of the interferometer (such as the adjustable range of the mirror spacing 0 - 100mm, the angular accuracy of the beam splitter plate ±0.1°) and physical characteristics (the relationship between the optical path difference and the interference fringes), construct a 1:1 scale digital twin model through 3D modeling technology. The model embeds the standard process timing nodes of "coarse-tune the mirror inclination first and then fine-tune", as well as parameter thresholds such as the inclination adjustment amount ±0.5° and the fringe clarity threshold. Map the preprocessed key features (such as the mirror adjustment angle sequence, the number of interference fringe movements) to the mirror entity object and time axis in the digital twin model, so that each adjustment action is dynamically associated with the corresponding component in the model. Through the timestamp synchronization mechanism, compare the start time of the "coarse-tune the mirror" step in the actual operation (such as 10:15:20) with the theoretical start time (10:15:00) of the standard process in the digital twin model, and calculate a time deviation value of 20 seconds; use the spatial coordinate mapping algorithm to compare the actual angle trajectory of the student's mirror adjustment (such as adjusting from 0° to 1.2°) with the standard space area (±0.5°) in the model, and identify an abnormal operation that exceeds the preset range by 0.7°. Conduct dynamic fitting analysis on the theoretical interference fringe movement curve (calculated based on the 632.8nm laser wavelength) in the digital twin model and the actually collected fringe movement data. If the deviation between the measured fringe movement amount and the theoretical value reaches 5.2% (the over-tolerance threshold is 5%), then through the preset abnormal rule library in the model (constructed based on the teaching syllabus and historical experimental data), automatically mark abnormal points such as "early termination of the coarse-tuning step", "over-range of inclination adjustment", and "over-tolerance of fringe movement amount", and generate a detailed record including the abnormal type (such as operation timing error) and deviation value (such as time deviation 20 seconds, angle over-tolerance 0.7°) to provide an objective basis for subsequent scoring.

[0032] In one embodiment, the steps of calculating the weights of key features through a pre-trained intelligent analysis model and generating sub-scores and comprehensive scores in combination with the scoring benchmarks set by the experimental teaching syllabus include: Load an intelligent analysis model trained based on historical experimental data. The model ranks the importance of key features such as the integrity of experimental steps, the standardization of instrument operation, and the accuracy of data collection through the gradient boosting algorithm, and generates a feature weight matrix. Input the key features of operation anomaly points and data deviation values extracted from the digital twin model analysis into the model, and calculate the influence coefficient of each feature on the experimental results based on the weight matrix. Retrieve the preset scoring benchmarks in the experimental teaching syllabus. The benchmarks include the score distribution standards for each experimental link, the scoring thresholds for operation specifications, and the scoring levels for data accuracy. Map the influence coefficient output by the model to the score distribution rules in the scoring benchmark to generate sub - item scores. Calculate the step score for the integrity of experimental steps according to the degree of compliance with the standard process, calculate the operation score for the standardization of instrument operation according to the number of violations and the severity, and calculate the data score for the accuracy of data collection according to the level where the deviation value is located. Integrate each sub - item score through the weighted summation algorithm. The weight parameters are set according to the importance of the experimental links in the teaching syllabus to generate an overall experimental score. Simultaneously mark the deduction points and deduction bases for each scoring item to form a score generation record with detailed scoring logic.

[0033] First, load the intelligent analysis model trained based on 1000 groups of historical experimental data. By learning a large number of qualified and unqualified experimental cases, this model ranks the importance of three key features, namely "completeness of experimental steps", "standardization of instrument operation", and "accuracy of data collection", to form a feature weight matrix (for example, the weight of step completeness is 40%, the weight of operation standardization is 35%, and the weight of data accuracy is 25%). Input the specific abnormal points extracted from the digital twin model analysis (such as "premature termination of the coarse adjustment step", "tilt adjustment exceeding the range by 0.7°", "deviation of the fringe movement amount by 5.2%") into the model. The model calculates the influence coefficient of each feature on the experimental result based on the weight matrix. For example, the influence coefficient of step omission is 0.3, the influence coefficient of operation exceeding the standard is 0.25, and the influence coefficient of data exceeding the tolerance is 0.2. Retrieve the scoring benchmarks preset in the experimental teaching syllabus, which clearly define rules such as "total score is 100 points, 10 points are deducted for each item of step omission", "5 points are deducted for each item of operation exceeding the standard", and "8 points are deducted for data exceeding the tolerance". Map the influence coefficients output by the model to the scoring benchmarks to generate sub-scores: the step completeness score is calculated according to the degree of coincidence between the actual operation and the standard process. If the "coarse adjustment step" is missing, 10 points will be deducted, and the score is 90 points; the operation standardization score is calculated based on the number and severity of violations. If the tilt deviation exceeds 0.7°, 5 points will be deducted, and the score is 95 points; the data accuracy score is calculated according to the level where the deviation value is located. If the deviation of 5.2% exceeds the tolerance, 8 points will be deducted, and the score is 92 points. Integrate the sub-scores through the weighted summation algorithm. The weight parameters are set according to the importance of the experimental links in the teaching syllabus (step completeness 40%, operation standardization 35%, data accuracy 25%). Calculate the comprehensive score as 90×40% + 95×35% + 92×25% = 92.25 points. Synchronously mark the deduction points and bases for each scoring item, such as "coarse adjustment step not completed (10 points deducted)", "tilt adjustment exceeding the standard range by 0.7° (5 points deducted)", "fringe movement amount deviation exceeding the threshold by 0.2% (8 points deducted)", to form a score generation record containing detailed scoring logic, ensuring that each deduction point can be traced back to specific operation anomalies or data deviations.

[0034] In one embodiment, a visual analysis report is generated according to the scoring results. The report includes steps for an operation trajectory map, a knowledge mastery map, and personalized improvement suggestions, including: Based on the time series data of the experimental operation and the time series comparison result with the standard process, generate an operation trajectory map: with the time axis as the horizontal axis, visualize the degree of coincidence between the actual operation steps and the standard steps and the time deviation value in the form of a heat map, mark the abnormal points of step omission and time series error with different colors, and superimpose and display the overlap degree of the spatial coordinate curve of the instrument operation trajectory and the standard trajectory; Construct a knowledge mastery graph, associate the scoring results of the knowledge points and key features involved in the experiment, and display the mastery degree of each knowledge point in the form of a node network. The size of the node corresponds to the weight of the knowledge point, and the depth of the color reflects the scoring deviation value. According to the deduction points in the itemized scoring, match the preset teaching suggestion library, generate specific improvement measures for each deduction point, and organize them into a stepped learning path according to knowledge modules.

[0035] Generate an operation trajectory graph based on the comparison results between the time series data of students' operations and the standard process: use the time axis from 0 to 30 minutes as the horizontal axis, and present the degree of coincidence between the actual operation steps such as "installing the pendulum bob", "adjusting the pendulum length", and "releasing the pendulum bob" and the standard process in the form of a heat map - the green area indicates that the step coincidence degree is higher than 90%, the yellow area indicates a coincidence degree of 70%-90%, and the red area marks the missing steps or timing errors (such as the standard requires adjusting the pendulum length to be completed in 10 minutes, but it is actually executed at the 15th minute). At the same time, superimpose and display the overlap degree between the spatial coordinate curve of the pendulum length adjustment (such as the trajectory from 90 cm to 101.2 cm) and the standard trajectory (100±0.5 cm), and mark the boundary of the standard range with a dotted line to visually display the operation deviation. In the construction of the knowledge mastery graph, associate knowledge points such as the "simple pendulum period formula T = 2π√(L / g)" and the "error propagation formula" involved in the experiment with the scoring results of key features: display in the form of a node network, the size of the node of the "simple pendulum period calculation" knowledge point corresponds to 30% of the weight in the teaching syllabus. If the student's period measurement error reaches 5% (exceeding the standard by 2%), the color of this node is marked as red, and the node edge blinks to prompt the knowledge shortcoming; the color of the node of the "pendulum length measurement method" knowledge point is displayed as orange according to the operation normality scoring (such as using a tape measure instead of a vernier caliper resulting in errors), and the size of the node corresponds to 25% of the weight. The connection thickness reflects the association strength between knowledge points (such as the pendulum length measurement error directly affects the period calculation result). For the deduction points in the itemized scoring (such as "the pendulum length measurement error exceeds the standard by 3.2%" and "there is an initial velocity when releasing the pendulum bob"), the system automatically matches the preset teaching suggestion library: for the pendulum length measurement problem, generate specific improvement measures such as "it is recommended to use a vernier caliper to repeat the measurement 3 times and take the average value" and "review the accuracy selection of length measurement instruments"; for the initial velocity problem, generate suggestions such as "watch the standard operation video of releasing the pendulum bob" and "practice the skill of releasing without an initial velocity", and organize them into a stepped learning path according to knowledge modules such as "measurement error analysis" and "operation norm training". Each suggestion is attached with a learning resource link (such as relevant experimental demonstration videos and knowledge point explanation documents), forming an interactive electronic report page that supports clicking on the node to jump to the corresponding experimental video clip or knowledge point analysis content.

[0036] In one embodiment, the steps of using blockchain evidence storage technology to immutably store the original experimental data and scoring process data to form a traceable exam record include: Classify and encapsulate the original experimental data and scoring process data to generate a data set; calculate a unique hash value for each data set, and construct a data digest through the Merkle tree algorithm to ensure data integrity verification; Package the data digest and timestamp information to form a block to be chained, prove the legitimacy of the block through the blockchain consensus mechanism, and link the legitimate blocks to the main blockchain in chronological order; embed the experiment number and student identification association information in the block header to establish a mapping relationship between the data and the exam subject; Generate an immutable blockchain evidence storage identifier for each evidence storage block, which includes block height and hash value information; associate and bind the evidence storage identifier with the visualization analysis report to form an electronic exam record with a blockchain traceability path; Distributively store the evidence storage data through the blockchain node network, support real-time query of the blockchain time and modification record information of the data through the block hash value or evidence storage identifier, and realize the full-process traceability of the original experimental data and scoring process; when it is necessary to verify the data integrity, re-calculate the data hash value and compare it with the digest value of the blockchain evidence storage to ensure the immutability of the evidence chain.

[0037] Classify and package the original experimental data and the scoring process data. The original data includes the swing angle-time series collected by the swing angle sensor (such as the swing angle of 5° at 10:15:30), the operation video recorded by the binocular camera (resolution 1920×1080, frame rate 25fps). The scoring data includes the scoring record of step integrity (such as "missing the pendulum length adjustment step deducts 10 points"), the comprehensive scoring calculation log (the weighted summation process of 92.25 points). Package the two types of data separately to generate independent data sets. Calculate a unique digital fingerprint (such as the SHA-256 hash value) for each set, and construct a data digest through a hierarchical data digest structure (similar to a folder tree structure) to ensure that any data modification will cause the digest value to change. Then, package the data digest and a timestamp accurate to milliseconds (such as 2025-06-23 14:30:22.156) into a block to be chained. Through the consensus mechanism of the blockchain network (such as node voting to verify the data legality), link the legal blocks to the main chain in chronological order. Embed the experiment number (20250623001) and the student ID number (S2023001) in the block header to establish the corresponding relationship between the data and the exam subject. Generate an immutable identifier for each stored block, including the block height (such as the 123456th block) and the hash value (abc123def456), and embed this identifier in the first page of the visual analysis report in the form of a QR code. Scanning the code can display the blockchain traceability path (such as the link sequence from the genesis block to the current block). Store the stored data distributedly through the blockchain node network (multiple servers distributed in the school data center). Teachers or students can query information such as the data chaining time and whether it has been modified in real time in the blockchain browser through the block hash value or the stored identifier. When it is necessary to verify the data integrity, recalculate the hash value of the original data and compare it with the digest value stored on the chain. If the two are consistent, it proves that the data has not been tampered with. For example, if a student questions the scoring result, the original data of "pendulum length measurement error exceeding the tolerance" can be retrieved, the hash value is calculated and compared with the digest on the chain to ensure the reliability of the evidence chain and form a fully traceable electronic file from the experimental operation to the scoring record.

[0038] Reference appendix Figure 2 The following is the structural block diagram of a physical experiment exam evaluation device proposed by the present invention. The device includes: An acquisition unit for real-time collecting physical parameter data and operation behavior data during the experiment. The physical parameter data includes at least one of the spatial position parameters, circuit signal parameters, optical phenomenon parameters, and mechanical action parameters of the experimental device. The operation behavior data includes the instrument operation action sequence and the experimental step execution trajectory; A standard unit for preprocessing the collected data, forming standardized data through signal denoising and unit standardization; matching the standardized data with a preset experimental operation standard library to extract key features of the integrity of experimental steps, the standardization of instrument operation, and the accuracy of data collection; A simulation unit for constructing an experimental environment model based on digital twin technology, performing spatio-temporal alignment analysis on the key features and the standard experimental procedures in the model to identify operation anomaly points and data deviation values; calculating the weights of the key features through a pre-trained intelligent analysis model, and generating sub-item scores and comprehensive scores in combination with the scoring benchmarks set by the experimental teaching syllabus; A report unit for generating a visual analysis report according to the scoring results, the report including an operation trajectory map, a knowledge mastery degree map, and personalized improvement suggestions; storing the original experimental data and the scoring process data in an immutable manner through blockchain deposit technology to form a traceable examination record.

[0039] Refer to Figure 3 , in an embodiment of the present invention, a computer device is further provided. The computer device may be a server, and its internal structure may be as Figure 3 shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected through a system bus. Among them, the processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0040] Those skilled in the art can understand that Figure 3 the structure shown in is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0041] In summary, a physical experiment exam evaluation method based on digital instruments proposed by the present invention collects experimental physical parameters and operation behavior data in real time through sensors. After signal denoising and unit standardization preprocessing, key features are extracted by matching with a preset experimental operation standard library. The digital twin technology is used to construct an experimental environment model, and the key features are analyzed by aligning with the standard process in time and space to identify operation anomalies and data deviations. The pre-trained intelligent analysis model is combined with the teaching syllabus scoring benchmark to generate itemized and comprehensive scores. Finally, a visual report including an operation trajectory map, a knowledge mastery degree map, and personalized suggestions is generated, and the experimental data and the scoring process are stored in an immutable manner through blockchain technology. This method solves the problems of large operation errors, strong subjectivity, and non-traceability in traditional physical experiment exams, realizes the automated and objective evaluation of the experimental process, improves the evaluation accuracy and teaching feedback efficiency, and provides technical support for the standardization and intelligentization of experimental exams.

[0042] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above method embodiments. Among them, any reference to a memory, storage, database, or other medium provided by the present invention and used in the embodiments can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0043] It should be noted that in this text, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article or method comprising a series of elements not only includes those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article or method. Without further limitation, an element defined by the phrase "comprising an..." does not exclude the presence of additional identical elements in the process, apparatus, article or method comprising such element.

[0044] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be similarly included in the patent protection scope of the present invention.

Claims

1. A physical experiment exam evaluation method based on digital instruments, characterized in that It includes the following steps: Collect physical parameter data and operation behavior data during the experiment in real time. The physical parameter data includes at least one of the spatial position parameters, circuit signal parameters, optical phenomenon parameters, and mechanical action parameters of the experimental device. The operation behavior data includes the instrument operation action sequence and the experimental step execution trajectory. Preprocess the collected data, and form standardized data through signal denoising and unit standardization. Match the standardized data with the preset experimental operation standard library, and extract the key features of the integrity of experimental steps, the standardization of instrument operation, and the accuracy of data collection. Build an experimental environment model based on digital twin technology, perform spatio-temporal alignment analysis on the key features and the standard experimental process in the model, and identify operation abnormal points and data deviation values. Calculate the weights of the key features through a pre-trained intelligent analysis model, and generate sub-item scores and comprehensive scores in combination with the scoring benchmarks set in the experimental teaching syllabus. Generate a visual analysis report according to the scoring results. The report includes an operation trajectory map, a knowledge mastery degree map, and personalized improvement suggestions. Use blockchain forensics technology to store the experimental original data and scoring process data in an immutable manner to form a traceable exam record.

2. The physical experiment exam evaluation method based on a digital instrument according to claim 1, wherein The step of collecting physical parameter data and operation behavior data during the experiment in real time includes: Real-time sample the physical parameters of the experimental device through sensors, and convert physical quantities such as temperature, pressure, current, and voltage into electrical signals. Convert the electrical signals into digital signals through an analog-to-digital conversion module, continuously collect the digital signals at a preset frequency, and store the timestamps. Use image acquisition devices arranged at multiple angles to record videos of the experimental operation area, and synchronously record the spatial trajectory and time series of the operation actions. Synchronize and calibrate the digital signals collected by the sensors and the video stream according to the timestamps to form a raw data set with spatio-temporal tags.

3. The physical experiment examination evaluation method based on a digital instrument according to claim 2, characterized in that The step of preprocessing the collected data and forming standardized data through signal denoising and unit standardization includes: Adopt the wavelet transform filtering algorithm to eliminate noise from the digital signals collected by the sensors, decompose the signal frequency band by setting thresholds, and filter out high-frequency interference noise and low-frequency drift noise. For signals with periodic interference, adopt the Kalman filtering algorithm for dynamic noise suppression, and establish a signal state space model to iteratively optimize the signal estimated value. Unify the conversion of temperature data into Kelvin units, voltage / current data into Volt / Ampere standard units, and mechanical parameters into Newton / Pascal units. Batch-convert non-standard physical quantity units through a preset conversion coefficient table to form standard data in the International System of Units. Calibrate the timestamps of the operation behavior data in the video stream to eliminate the timing error caused by the clock deviation between the camera and the sensor. Synchronize the multi-source data with inconsistent sampling frequencies through the interpolation algorithm to ensure the alignment of the time axes of the physical parameter data and the operation behavior data.

4. The physical experiment examination evaluation method based on a digital instrument according to claim 3, characterized in that The step of matching the standardized data with the preset experimental operation standard library and extracting the key features of the integrity of experimental steps, the standardization of instrument operation, and the accuracy of data collection includes: Load the preset standard library of experimental operations, where the standard library contains standard operation procedures disassembled based on the physics experiment teaching syllabus, the timing logic relationship of each step, and the key parameter thresholds; decompose the specification data into operation action segments according to the time series, perform semantic matching with the standard steps in the standard library, and identify the compliance degree of the actual operation steps with the standard process and the missing steps; Compare the instrument operation action sequence in the specification data with the preset operation postures, force thresholds, and time interval specification parameters in the standard library, and mark the non-compliant operations that exceed the thresholds; calculate the similarity between the actual operation trajectory and the standard operation trajectory through the dynamic time warping algorithm, and generate the operation standardization scoring features; For the data acquisition accuracy features, compare the standardized physical parameter data with the theoretical data curve and the error tolerance range of the corresponding experiment in the standard library, and extract the feature parameters of data deviation degree and fluctuation amplitude; analyze the difference between the frequency spectrum features and the standard frequency spectrum of the periodic experimental data through Fourier transform to identify the systematic errors in the data acquisition process.

5. The physical experiment exam evaluation method based on a digital instrument according to claim 1, characterized in that, Construct an experimental environment model based on digital twin technology, and perform spatio-temporal alignment analysis on the key features and the standard experimental process in the model to identify the steps of operation abnormal points and data deviation values, including: Based on the geometric parameters, physical properties, and dynamic behaviors of the physical experimental device, construct a 1:1 scale digital twin experimental environment model through 3D modeling technology, and embed the timing nodes and parameter thresholds of the standard experimental process in the model; map the key features extracted from the specification data to the corresponding entity objects and time axis of the digital twin model; Through the timestamp synchronization mechanism, align the time series of the key features with the theoretical time nodes of the standard experimental process in the digital twin model, and calculate the time deviation value between the actual operation steps and the standard process; use the spatial coordinate mapping algorithm to compare the instrument operation trajectory and the spatial features of the experimental phenomenon occurrence location with the standard operation space area in the digital twin model to identify the non-compliant operations that exceed the preset space range; Perform dynamic fitting analysis on the theoretical parameter curve of the standard experimental process in the digital twin model and the actually collected physical parameter data, and calculate the data deviation degree parameter; through the preset abnormal rule library in the model, automatically mark the abnormal points of operation step timing errors, instrument state parameter overlimits, and data acquisition result over-differences, and generate a detailed record of the abnormal type and deviation value.

6. The physical experiment examination evaluation method based on a digital instrument according to claim 5, characterized in that, Calculate the weights of the key features through a pre-trained intelligent analysis model, and generate the sub-item scores and comprehensive scores in combination with the scoring benchmarks set by the experiment teaching syllabus, including: Load the intelligent analysis model trained based on historical experimental data. The model ranks the importance of the key features of experiment step integrity, instrument operation standardization, and data acquisition accuracy through the gradient boosting algorithm to generate a feature weight matrix; input the operation abnormal points and data deviation value key features extracted from the digital twin model analysis into the model, and calculate the influence coefficient of each feature on the experimental result based on the weight matrix; Retrieve the preset scoring benchmarks in the experimental teaching syllabus, where the benchmarks include the score distribution criteria for each experimental link, the scoring thresholds for operation specifications, and the data accuracy scoring levels; Map the influence coefficients output by the model to the score distribution rules in the scoring benchmarks to generate sub-item scores. Calculate the step scores for the compliance of the experimental step integrity with the standard process, calculate the operation scores for the instrument operation standardization according to the number of violations and the severity, and calculate the data scores for the data acquisition accuracy according to the level where the deviation value is located; Integrate each sub-item score through the weighted summation algorithm. The weight parameters are set according to the importance of the experimental links in the teaching syllabus to generate the comprehensive experimental score; simultaneously mark the deduction points and the deduction basis for each scoring item to form a score generation record with detailed scoring logic.

7. The physical experiment examination evaluation method based on a digital instrument according to claim 1, wherein Generate a visual analysis report according to the scoring results. The report includes the steps of the operation trajectory map, the knowledge mastery degree map, and personalized improvement suggestions, including: Generate an operation trajectory map based on the time series comparison results of the experimental operation data and the standard process: use the time axis as the horizontal axis, visualize the compliance between the actual operation steps and the standard steps and the time deviation value in the form of a heat map, mark the missing steps and abnormal time sequence error points with different colors, and superimpose and display the overlap degree of the spatial coordinate curve of the instrument operation trajectory and the standard trajectory; Construct a knowledge mastery degree map, associate the scoring results of the knowledge points and key features involved in the experiment, and display the mastery degree of each knowledge point in the form of a node network. The size of the node corresponds to the weight of the knowledge point, and the depth of the color reflects the scoring deviation value; According to the deduction points in the sub-item scoring, match the preset teaching suggestion library, generate specific improvement measures for each deduction point, and organize them into a ladder-like learning path according to the knowledge modules.

8. The physical experiment exam evaluation method based on a digital instrument according to claim 7, characterized in that, The steps of storing the original experimental data and the scoring process data in an immutable manner through blockchain evidence storage technology include: Classify and encapsulate the original experimental data and the scoring process data to generate data sets; calculate the unique hash value for each data set, and construct a data digest through the Merkle tree algorithm to ensure data integrity verification; Package the data digest and the timestamp information to form a block to be chained. Verify the legality of the block through the blockchain consensus mechanism, and link the legal blocks to the main blockchain in chronological order; embed the experimental number and student identification association information in the block header to establish the mapping relationship between the data and the examination subject; Generate an immutable blockchain evidence storage identifier for each stored block, where the identifier includes the block height and hash value information; associate and bind the evidence storage identifier with the visual analysis report to form an electronic examination record with a blockchain traceability path; Distributively store the stored data through the blockchain node network, support real-time query of the blockchain time and modification record information of the data through the block hash value or the evidence storage identifier, and realize the full-process traceability of the original experimental data and the scoring process; when it is necessary to verify the data integrity, ensure the immutability of the evidence chain by recalculating the data hash value and comparing it with the digest value stored in the blockchain.

9. A physical experiment examination evaluation device based on a digital instrument, characterized in that, Including: The acquisition unit is used to collect physical parameter data and operation behavior data during the experiment in real time. The physical parameter data includes at least one of the spatial position parameters, circuit signal parameters, optical phenomenon parameters, and mechanical action parameters of the experimental device. The operation behavior data includes the instrument operation action sequence and the experimental step execution trajectory; The standard unit is used to preprocess the collected data, and form standardized data through signal denoising and unit standardization; match the standardized data with the preset experimental operation standard library, and extract the key features of the integrity of the experimental steps, the standardization of instrument operation, and the accuracy of data collection; The simulation unit is used to construct an experimental environment model based on digital twin technology, perform spatio-temporal alignment analysis on the key features and the standard experimental process in the model, and identify operation abnormal points and data deviation values; calculate the weights of the key features through a pre-trained intelligent analysis model, and generate item scores and comprehensive scores in combination with the scoring benchmarks set by the experimental teaching syllabus; The reporting unit is used to generate a visual analysis report according to the scoring results. The report includes an operation trajectory map, a knowledge mastery degree map, and personalized improvement suggestions; store the experimental original data and scoring process data in an immutable manner through blockchain evidence storage technology to form a traceable examination record.

10. A computer device, comprising a memory and a processor, wherein a computer program is stored in the memory, characterized in that When the processor executes the computer program, it implements the steps of the physical experiment examination and evaluation method based on digital instruments described in any one of claims 1 to 8.

Images