81 results about "Performance curves" patented technology

The invention relates to an IGBT module junction temperature estimation method and system, and the method comprises the following steps: obtaining a performance curve set of an IGBT module under different working conditions, and carrying out the fitting optimization, and obtaining an optimal performance curve set; constructing a transient thermal resistance curve model according to a transient thermal resistance principle, and constructing a second-order Foster thermal network model based on the transient thermal resistance curve model; performing Laplace transformation and discretization processing based on a backward difference method on the second-order Foster thermal network model to obtain a junction temperature estimation model; collecting a phase current instantaneous value, and calculating the total power loss of the IGBT module in real time in combination with the optimal performance curve set; and finally, the junction temperature of the IGBT module is obtained in real time by using the total power loss of the IGBT module obtained through real-time calculation and the junction temperature estimation model, and junction temperature estimation is completed. According to the method, the effect of improving the power loss precision of the IGBT module is achieved by optimizing the performance curve, and then the junction temperature estimation precision is improved.

The invention relates to a defect analysis and process optimization method based on artificial intelligence multi-mode fusion, and the method comprises the steps: respectively inputting virtual data of a chip or device design layout, an expected process flow and parameters, and an expected electrical performance curve into an image encoder, a process encoder, and an electrical encoder, extracting a corresponding feature vector; performing cross-modal information interaction on the corresponding feature vectors according to a cross attention fusion mechanism; determining a representation image, a defect category, a root cause category and a process parameter correction suggestion; and based on a manufacturing execution system (MES), an electronic design automation (EDA) tool or an equipment interface, process parameters are corrected, and a closed-loop control result from defect analysis to process optimization is obtained. Therefore, the problems of single data dimension, mutual coupling of various defects, influence of background noise, limited machine identification accuracy, overlarge workload of artificial expert judgment and low efficiency are solved.

The invention discloses a vibration fatigue life prediction method and system based on a working condition adaptive spectrum. According to the method, load information of a target structure under various service conditions is obtained, and frequency domain load descriptions corresponding to different working conditions are subjected to weighted fusion according to working condition weights, so that a working condition adaptive load spectrum capable of reflecting real service characteristics is constructed; on the basis, frequency domain stress response of the structure under the random vibration condition is obtained in combination with a structural dynamics analysis model, equivalent stress description of dominant fatigue damage is further determined through a multi-axis equivalent criterion, and fatigue damage distribution is calculated through a frequency domain fatigue analysis method; and finally, predicting the vibration fatigue life of the structure in combination with a material fatigue performance curve and an environment-related performance correction relation. Compared with the prior art, the method has the advantages that the multi-working-condition load superposition effect and the service environment influence can be effectively considered, and the accuracy and engineering applicability of vibration fatigue life prediction are improved.

The application discloses a simple high-precision wind tunnel compressor variable speed aerodynamic performance parameterization modeling method, relates to the field of wind tunnel operation characteristic digital modeling, and comprises the following steps: determining and obtaining a plurality of wind tunnel compressor constant speed total pressure ratio performance curves; increasing a virtual working point to obtain a plurality of virtual constant speed total pressure ratio performance curves; constructing a first parameterization model to obtain parameters in the first parameterization model; obtaining a mapping relationship; constructing a second parameterization model corresponding to a given speed; calculating and obtaining a constant speed total pressure ratio performance curve of a wind tunnel compressor corresponding to the given speed; and calculating and obtaining a total pressure ratio of the wind tunnel compressor corresponding to the equivalent mass flow of the given wind tunnel compressor based on the constant speed total pressure ratio performance curve of the wind tunnel compressor corresponding to the given speed. The method solves the problem that conventional wind tunnel compressor parameterization modeling methods cannot realize simple high-precision wind tunnel compressor variable speed aerodynamic performance digital modeling and depend on massive real aerodynamic performance data.

The application discloses a water pump performance curve correction method based on a similarity principle, relates to the technical field of water pump working condition analysis, and comprises the following steps: acquiring a conversion working condition point corresponding to a model pump test working condition point converted to a prototype pump and a theoretical performance curve of the prototype pump; confirming a similarity parabola passing through the field measured working condition point and mapping the similarity parabola to the theoretical performance curve of the prototype pump to obtain an intersection point as a similarity working condition point of the field measured working condition point according to a similarity parabola relationship and field measured working condition point data; obtaining a corresponding point under a theoretical rotating speed according to the similarity relationship and combining the working condition point on the theoretical curve of the prototype pump; replacing the theoretical rotating speed with a measured rotating speed, converting to a corresponding point under a rated rotating speed, and fitting to obtain an actual flow-head curve; and acquiring a theoretical efficiency value and an optimal efficiency value corresponding to the similarity working condition point to obtain an actual flow-efficiency curve, so that the actual performance curve of the prototype pump can be efficiently and accurately corrected.

The application provides a method for measuring flow of different size nozzles of a ventilator, in order to meet the test of complete performance curve of a small ventilator which has a relatively wide flow range, a small pressure rise and often needs to run in a small flow range, two nozzles with different throat diameters are used to test the performance curve of a ventilator respectively, then the test results of the two curves are considered together to fit a complete aerodynamic performance curve. The method can overcome the defects of the prior art, i.e. the single fixed diameter nozzle flow measuring device cannot meet the test requirements of the same ventilator with a wide flow range, and the test precision is difficult to meet the application requirements.

This invention discloses a method for calculating the flow rate of a single compressor in a continuous wind tunnel parallel compressor system, relating to the field of supersonic wind tunnel operation and commissioning. The invention uses the mass flow rate of the wind tunnel nozzle as a benchmark for the total flow rate; pre-determines the constant-speed total pressure ratio performance curves of each individual compressor and parameterizes them into a continuous constant-speed total pressure ratio performance mathematical model; during wind tunnel operation, the rotational speed and total pressure ratio of each compressor are measured in real time, and the reduced flow rate is obtained by solving inverse functions or using iterative methods. The initial mass flow rate of each compressor is calculated by combining the inlet total pressure and inlet total temperature; finally, corrections are made based on the deviation between the nozzle mass flow rate and the sum of the initial mass flow rates of each compressor to obtain the actual mass flow rate of each compressor. This achieves accurate determination of the flow rate of each individual compressor in parallel operation, providing a guarantee for the establishment of the target flow field and operational safety in the wind tunnel test section.

The invention relates to the technical field of nonmetal pipeline performance detection, in particular to a multifunctional experimental device and method for nonmetal pipeline detection, and the method comprises the steps that a tensile stress test platform heats an experimental environment temperature to a set temperature, and a first to-be-detected nonmetal pipe section is stretched to be fractured; the composite test end obtains a tensile property curve of the nonmetal pipeline according to the real-time traction force value returned by the tensile stress test platform; the internal pressure test platform heats the experimental environment temperature to a set temperature, and inflates the second to-be-tested non-metal pipe section until bursting; and the composite test control end obtains an internal pressure resistance curve of the non-metal pipeline according to the real-time internal pressure value returned by the internal pressure test platform. According to the device and the method, a traction performance test and an internal pressure performance test can be simultaneously performed on the non-metal pipeline at different experimental temperatures, so that a tensile resistance performance curve and an internal pressure resistance performance curve of the non-metal pipeline are obtained, and the comprehensive performance detection efficiency of the non-metal pipeline is improved.

A method for selecting a booster pump for circulating cooling water of a bypass piping system is provided in the disclosure, which includes: acquiring booster pump data of the booster pump for the circulating cooling water of the bypass piping system, and constructing and updating a database for selecting the booster pump for the circulating cooling water of the bypass piping system; performing primary booster pump selection according to final user's selection requirements and constructing a booster pump preselection set; calculating water resistance of the circulating cooling water in the bypass piping system, determining a selection score according to user's selection evaluation standard for performing secondary booster pump selection, and determining a finally selected booster pump for the circulating cooling water of the bypass piping system; and visually outputting and displaying performance parameters, performance curves and installation data graphs.

The invention discloses a reinforced training method for improving the middle-section text analysis capability of a large language model, and relates to the field of natural language processing and generative large language models.The method comprises the steps that middle-section double-window construction is conducted according to a dynamic scale factor, extraction and splicing are conducted from original long text corpora based on window positions, and the middle-section text corpora are obtained; obtaining a preliminary training sample; performing middle text information density evaluation on the preliminary training sample, and verifying and optimizing the preliminary training sample to obtain an optimized sample; and inputting the optimized sample and the updated position identification sequence into a large language model, and realizing iterative training of the large language model through a position-aware middle loss weighting mechanism to obtain an optimized large language model. According to the method, the retrieval and understanding accuracy of the model in the process of processing the middle position information of the sequence can be remarkably improved in a balanced manner, the U-shaped performance curve which is sunken originally is successfully leveled, and the training performance bottleneck is overcome.

The invention provides a simplified design system for a damping structure. Non-damping structure data calculated through a vibration mode decomposition reaction spectrum method can be obtained through a data importing and preprocessing module; performing calculation and CQC correction on a displacement spectrum according to the obtained acceleration response spectrum through an anti-seismic standard response spectrum calculation module to obtain a displacement spectrum corresponding to the equivalent single-degree-of-freedom system; the performance curve generation module is used for generating a performance curve graph of the system ductility coefficient relative to the rigidity ratio, identifying an optimal performance design point, and generating a damper design scheme according to the number of dampers in each layer; the parameters of the damper are calculated layer by layer according to a preset additional rigidity distribution strategy through a multi-degree-of-freedom system design module, and the matching degree of the deformation capacity of the damper and the interlayer displacement angle requirement is verified; and parameter output and visual display of each layer of damper are realized through a result output and visual module, so that rapid estimation of damper parameters is realized, and preliminary design of a damping structure is completed.

The invention discloses a high-temperature alloy component optimization method and system based on reverse tracing of a performance track. The method comprises the following steps: establishing a standardized data set containing high-temperature alloy components, full-temperature-range performance data and thermodynamic state parameters; performing numerical analysis on the performance curve to extract track geometric features and key instability point temperature; generating a dynamic mechanism boundary function based on the thermodynamic state parameters and the key instability points, and dividing a performance orbit into subintervals dominated by different physical mechanisms; constructing a ternary tensor sensitivity model by using a deep neural network, and generating a ternary sensitivity tensor representing a coupling mapping relationship among elements, a microscopic phase and a strengthening mechanism; constructing a composite loss function containing a physical mechanism consistency penalty term, and correcting a mathematical gradient direction by using a ternary sensitivity tensor to invert an optimal component; and performing perturbation screening and experimental iteration on the candidate components through an active learning closed-loop feedback module.

The application provides a calibration method, device, storage medium, compressor and computer program product. The method comprises: controlling the operation frequency of a compressor to be sequentially raised to different preset calibration frequencies, and performing calibration of a group of calibration points when the operation frequency is raised to one preset calibration frequency; wherein when calibration of the calibration points at each preset calibration frequency is performed, the parameter value of each calibration parameter of each calibration point is recorded as calibration data of the calibration point, starting from the compressor outlet pressure corresponding to a preset starting calibration point and every preset pressure value being one calibration point; and when the parameter value of each calibration parameter of each calibration point at each preset calibration frequency is recorded, calibration of the compressor performance curve reference point is completed. The scheme provided by the application can reduce the need for manual intervention and improve the accuracy of test data.

This invention discloses an interpolation fitting method, apparatus, and computer equipment for centrifugal pump performance curves. The method includes: setting the input frequency of a frequency converter, adjusting the valve opening, and acquiring test sample data of shaft power P, rotational speed N, and flow rate Q of centrifugal pumps with different flow rates; acquiring multiple performance test sample data sets of the centrifugal pump under the same impeller diameter and rotational speed of the same pump unit; each performance test sample data set includes the power parameter value and other performance parameter values ​​of the centrifugal pump. This method significantly improves accuracy, reducing the error of traditional linear interpolation, least squares polynomial fitting, and empirical formula methods from >5% to ≤2.5%, meeting high-precision control requirements; significantly reducing cost: requiring only a few measurement points, greatly shortening the experimental cycle and testing cost; strong stability: no oscillation, no overshoot, physically compliant; adaptable to simulation and control; wide scenario adaptability: supporting fitting of wide rotational speeds, variable operating conditions, non-uniform measurement points, and missing points.

The invention discloses a heat pump system fault diagnosis method and system, and relates to the technical field of water pump fault diagnosis. According to the method, the preset proportion of the rated water flow of the heat pump system is set to be the minimum water flow, end equipment is closed, the real-time power of different rotating speeds of a variable frequency water pump is obtained, the real-time water flow is obtained based on a performance curve of the corresponding rotating speed, and when the real-time water flow reaches the minimum water flow, the real-time power of the variable frequency water pump is cut off. The rotating speed corresponding to the real-time water flow is set as the minimum rotating speed; different combinations of the end equipment are set, and the minimum power and the maximum power of the variable frequency water pump from the minimum rotating speed to the maximum rotating speed under the different combinations are obtained; and when the heat pump system stably operates, the real-time power of the variable frequency water pump is obtained every preset time interval, the real-time power is compared with the minimum power and the maximum power, and whether the heat pump system breaks down or not is judged. According to the method, fault diagnosis can be realized at a relatively low cost.

The application discloses a device log and rehabilitation scale law deduction method and system based on machine learning, and relates to the technical field of medical data processing and machine learning. The device log and rehabilitation scale law deduction method based on machine learning collects phototherapy equipment operation logs, prescription plan values and patient rehabilitation scale data, constructs a lag evaluation window according to rehabilitation scale recording time for multiple treatment sessions, extracts an execution deviation feature vector and trains a multi-task learning model, generates a constrained disturbance feature group under the condition that the prescription plan value is fixed, deduces a dose continuity performance curve, and generates a prescription parameter compensation suggestion accordingly. The application correlates device logs, prescription plan values and rehabilitation scale data, deduces the mapping relationship between execution deviation and rehabilitation effect, provides a basis for judging dose deficiency, treatment interruption, interval deviation and other problems, and reduces blind adjustment of prescription parameters.

The application discloses a kind of compressor simulation control methods, comprising: obtaining the valve opening of anti-surge valve and Cv value curve;Valve opening and Cv value curve are polynomial fitted, and the valve opening of anti-surge valve and Cv value polynomial are obtained;Combining valve opening and Cv value polynomial, valve before and after pressure and valve before and after temperature, establish valve opening and valve flow polynomial;Obtain compressor performance curve, and the polynomial fitting of compressor performance curve is carried out, and the inlet flow and outlet pressure polynomial corresponding to each preset speed based on compressor are obtained;According to valve opening and valve flow polynomial and inlet flow and outlet pressure polynomial, compressor simulation control is carried out.The application shortens the verification time of compressor project factory program, improves work efficiency, saves cost.The application also discloses a kind of compressor simulation control device, equipment and storage medium, with corresponding technical effects.

A computer-implemented method for detecting mechanical damage to a machine based on a performance indicator present at a rotation speed of the machine is provided, including the steps of: receiving a number of calibration data points during a calibration period, each comprising the rotation speed of the machine and a performance characteristic value associated with the rotation speed, calibrating an AI-based Gaussian process regressor, by inputting the calibration data points received during the calibration period and outputting an estimated performance curve and an uncertainty range assigned to each value of the performance curve for all rotation speed values, receiving a data point to be evaluated, and comparing the data point to be evaluated with the estimated performance curve, and outputting an anomaly message if the data point to be evaluated is outside a specified limit value, which is dependent on the uncertainty range of the performance characteristic value.

This application discloses a method, apparatus, and electronic device for determining a device performance curve, relating to the field of electronic equipment technology. The method includes: acquiring the power input range and voltage input range of an ultrasonic transducer; determining an initial performance curve of the ultrasonic transducer based on the power input range and voltage input range; acquiring N expected power values ​​set for the ultrasonic transducer, wherein the expected power values ​​all fall within the power input range, and N is an integer greater than or equal to 1; and iteratively updating the initial performance curve based on the target power actually received by the ultrasonic transducer when each expected power value is expected to be received, to obtain the target performance curve of the ultrasonic transducer. This application solves the technical problem in the prior art where the performance evaluation of ultrasonic transducers is often poor due to reliance on a uniform performance curve described at the time of manufacture, which fails to determine the actual performance curve of the ultrasonic transducer in a timely manner.

The invention discloses a high-fidelity identification method and system for a pump and a pipe network system based on active resistance modulation, and the method comprises the steps: firstly, manually setting at least two pipeline resistance states with obvious differences through adjusting devices such as valves, carrying out the frequency conversion operation of a pump in each state, and collecting a plurality of groups of pressure and frequency data; then, an integrated system physical model is constructed, and the model forcibly uses uniquely shared pump performance parameters and pipe network static lift parameters to synchronously fit all data clusters in different resistance states; and finally, by solving the global optimization problem of the model, identifying a real performance curve of the pump, the static lift of the pipe network and the independent resistance coefficient in each state at one time with high precision. According to the method, a flow sensor is not needed, the ill-conditioned problem of a passive data analysis method is solved, on-site in-situ calibration comparable with laboratory precision is achieved, and a reliable basis is provided for construction of a digital twin model and a virtual flowmeter and system health assessment.

The application discloses a time sequence data curve rendering method, system, device and computer medium, relates to the technical field of computer graphics and data visualization, and obtains performance time sequence data of an industrial equipment; based on the performance time sequence data, a current performance data point to be rendered and a first performance data point of adjacent sequence are determined; a denoising threshold is obtained, and the denoising threshold is generated according to a horizontal sampling interval of a target equipment performance curve; whether a first vertical difference value is less than the denoising threshold is detected, and the first vertical difference value is a vertical difference value between the current performance data point and the first performance data point; in response to the first vertical difference value being less than the denoising threshold, a vertical value of a first control point is set as a vertical value of the current performance data point, and the first control point is located between the current performance data point and the first performance data point; and a curve segment is generated on the performance curve of the target equipment according to the current performance data point, the first control point and the first performance data point. There is no pseudo-trend and overshoot phenomenon, and the accuracy is high.

The invention provides a performance evaluation method and system for a differential pressure plugging material for leakage control, and belongs to the field of petroleum engineering plugging materials.The method comprises the steps that the physical and chemical performance of the differential pressure plugging material is evaluated, particle size distribution is obtained, and density, acid solubility and dispersion stability tests are conducted; testing the mechanical performance of the differential pressure plugging material to obtain a laboratory test performance curve of stretching and friction; testing and evaluating the critical starting pressure in a laboratory; performing laboratory testing and mechanism analysis on the ultimate bearing pressure to obtain a performance evaluation result; according to the method, the evaluation results of laboratory testing, theoretical simulation and mechanism analysis of the plugging material on the performance of the plugging material are comprehensively considered, the plugging material is evaluated to have the characteristics of low-pressure activation, self-adaptive plugging and high-pressure resistance, the preparation process is simple and controllable, the method can be effectively applied to micro-leakage plugging of petroleum and natural gas downhole pipe columns, and the application prospect is wide. And a reliable technical scheme is provided for shaft integrity repair.

The invention relates to the technical field of thermal power generating unit analysis and evaluation, in particular to a thermal power generating unit output performance evaluation method based on time-sharing energy consumption fitting, which comprises the following steps of: dividing an operation cycle of a thermal power generating unit into a plurality of time-sharing periods, collecting independent energy consumption data in each time-sharing period, synchronously collecting actual load, and calculating the output performance of the thermal power generating unit; calculating the response time lag between the energy consumption change and the load change of each component, and forming a cross-component time-sharing energy consumption response track; establishing a time-sharing energy consumption decoupling fitting frame comprising a phase adjustment function and a response elastic window, and outputting quasi-steady-state energy consumption and dynamic additional energy consumption after purification; and generating an output performance curve after phase purification, identifying an abnormal response stage with high adjustment cost based on the proportion of the two types of energy consumption and performance deviation characteristics, and outputting a comprehensive evaluation result including a performance grade, a response load ratio and an energy consumption deviation source. According to the invention, purification and fine decomposition of energy consumption signals are realized, and the problem of confusion of normal fluctuation and abnormal consumption in traditional evaluation is solved.

The invention discloses a design method of a centrifugal compressor impeller case structure. The working stability of a compressor is enhanced. A design method of a centrifugal compressor impeller casing structure comprises the steps that S1, a prototype impeller performance curve is calculated; s2, multiple jet flow hole schemes are designed on the cartridge receiver, jet flow holes are all located in the incoming flow direction of the front edge of the impeller, the centers of the jet flow holes all face the positions where shock waves and gap leakage flow are generated on the impeller, jet flow control needs to be externally connected with an air source, and multiple new impeller performance curves are calculated; and comparing the prototype impeller performance curve, the various new impeller performance curves and the impeller airflow angle under each scheme, and selecting an optimal jet hole scheme.

The present application belongs to the technical field of radar target detection, and discloses a weak target detection method under sea clutter background based on graph entropy, comprising the following steps: extracting information entropy features from the adjacent matrix of graph signals in the frequency domain by using echo data received by a radar; and obtaining target detection results for comparative analysis based on the realization of a weak target detector of graph entropy. The present application has the beneficial effect of further exploring the characteristics between clutter and targets, and proposes information entropy features based on graphs. In the experimental analysis of actual sea clutter data combined with the performance curve analysis of other entropy feature detection, it is shown that the weak target detection device based on graph entropy proposed by the present application has better detection performance.

The invention provides an intelligent optimization adjusting method and system for a fluorine pump air conditioner and a medium. The method comprises the steps that based on source data of a machine room, load prediction of a preset duration is conducted through an XGBoost-BiLSTMm model; combining a load prediction result and a pre-established performance curve of the fluorine pump air conditioner, taking the lowest total operation energy consumption of the fluorine pump air conditioner as a target, applying an actual operation constraint condition, establishing a group control optimization model of the whole fluorine pump air conditioner of the single machine room, and taking real-time operation parameters and real-time meteorological data of the machine room as input of the optimization model; and the optimal starting number of main and standby units, the air supply temperature set value, the operation mode and the partial load rate of the fluorine pump air conditioner are determined. According to the method, the lowest power consumption parameter setting of the air conditioner under different field conditions on the premise that the refrigerating capacity requirement is met can be obtained, the PUE index of the data center is optimized, and the power consumption cost is reduced.

This invention discloses a parallel pump group scheduling method and system based on dynamic programming, belonging to the technical field of intelligent pump group scheduling. The method involves data acquisition and preprocessing, discretizing and equivalently treating the variable frequency pumps as n equivalent fixed frequency pumps. Then, single-pump performance curves and pipeline network performance curves are fitted. Based on the single-pump performance curves, the high-efficiency operating range of each pump is determined. Then, using water supply flow demand, water supply head demand, and the high-efficiency operating range of each pump as hard constraints for pump group operation, candidate pump group combinations are selected. The start-up and shutdown costs and usage time of each pump are incorporated as soft constraints into the total cost, constructing a minimum objective function for the total cost. Finally, a dynamic programming algorithm is used to obtain the global optimal solution, generating the optimal pump group scheduling sequence within the scheduling cycle. This invention balances low energy consumption and equipment health, and is applicable to mixed pump group scenarios including fixed-frequency and variable frequency pumps, significantly reducing energy consumption in traditional scheduling and mitigating equipment wear.

The application discloses a kind of MLP-based centrifugal air compressor multi-working condition performance prediction method.The method comprises: constructing database comprising multiple groups of compressor initial performance curve data corresponding to different operating parameters;The initial performance curve data is standardized and gridded, to obtain standardized performance data;Based on the boundary points of each group of initial performance curve data, a physical boundary for constraining the prediction result is constructed;Based on the standardized performance data, a neural network model is trained, the target operating parameters and working condition parameters are input into the trained neural network model, and the corresponding performance parameter prediction value is output, and the prediction value is reasonably checked in combination with the physical boundary.The present application can accurately predict the performance curve of centrifugal air compressor under all working conditions, and provide a reliable basis for compressor optimization control and performance analysis.

The invention provides a fan operation control method and device, a ventilation system and a storage medium, the method is applied to the ventilation system comprising a plurality of fans, and the method comprises the steps that the environment static pressure of each fan in the multiple fans in the ventilation system is obtained; inputting the environment static pressure of each fan into a pre-established structure model to obtain a performance curve of each fan; according to the performance curve of each fan, taking the minimum total power consumption of the system as a target function, and solving to obtain the optimized rotating speed of each fan; and the optimized rotating speed is sent to a controller of each fan so as to control each fan to operate according to the optimized rotating speed. Through the scheme, the technical effect of dynamically optimizing the operation strategy of the fan based on the real-time environment parameters is achieved.

The invention relates to the technical field of chip testing, in particular to a communication chip reliability testing method based on big data analysis. Chip data and test data are preprocessed to obtain multiple pieces of preprocessed data, then a test model is created, the performance index of a target chip is calculated based on the test data through the test model, reliability judgment is conducted on the target chip based on the performance index, and multiple normal chips and abnormal chips are obtained. The method comprises the following steps: establishing a performance curve of a real-time chip, performing single classification learning based on a normal chip to obtain a trained test model, finally obtaining a real-time performance curve of the real-time chip, and judging the reliability of the real-time chip in combination with the real-time performance curve and a standard performance curve. The performance curve representing the chip is used as the standard performance curve, so that a user can judge the reliability of the real-time chip according to the standard performance curve, and the reliability test efficiency is improved.