Non-destructive testing system and method for density and compactness of solid materials

Abstract

The application discloses a kind of solid material density and the nondestructive testing system and method of density, it is related to intelligent sensing system technical field;Method includes: based on the workpiece geometry data obtained in advance, workpiece surface temperature distribution data is collected;Record multiple time periods of infrared thermal imager frame;Based on multiple time periods of infrared thermal imager frame, determine the region where workpiece appears nitrogen desorption heat release effect, and calculate to obtain nitrogen desorption heat characteristic data;Based on workpiece surface temperature distribution data, workpiece geometry data and nitrogen desorption heat characteristic data, simulation calculation obtains workpiece internal temperature field data;Signal detection is carried out to workpiece, and detection signal data is obtained;Nonlinear temperature compensation is carried out using detection signal data, and fusion density calculation is carried out, to obtain the density and density of workpiece;The application realizes that detection can be completed without waiting for workpiece cooling, and improves the quality control level and production efficiency of special-shaped sintered part.

Description

Technical Field

[0001] This invention relates to the field of intelligent sensing system technology, and more specifically, to a non-destructive testing system and method for the density and compactness of solid materials. Background Technology

[0002] In industrial settings where solid sintered parts are continuously produced in a nitrogen-based atmosphere double-pass pusher kiln, in order to match the continuous operation rhythm of the production line and avoid efficiency loss caused by testing after cooling, it is necessary to conduct online non-destructive testing on the sintered parts that have just come out of the furnace and are in a hot state, so as to achieve real-time quantitative assessment of density and compactness. Among them, a large number of sintered parts to be tested are designed with irregular shapes containing hidden structures due to actual application requirements. Such structures generally have characteristics such as abrupt changes in wall thickness, internal chambers or deep extension channels, and nitrogen adsorption will occur on the surface and inside of the material during nitrogen-based atmosphere sintering.

[0003] Existing online non-destructive testing technologies for hot workpieces are mostly based on physical effects such as ultrasound and electromagnetics. They utilize the intrinsic relationship between physical parameters such as sound velocity, conductivity, and magnetic permeability and material density to construct quantitative models. Simultaneously, to address the temperature influence of hot workpieces, surface temperature measurement combined with linear correction algorithms is used to compensate for temperature interference with the detection signal. The core logic is to assume a linear mapping relationship between the detection signal and temperature. By acquiring single-point or multi-point temperature data on the workpiece surface, the detection signal is uniformly corrected, and the material density is inferred. However, the heat dissipation conditions of concealed and exposed areas of irregularly shaped structures differ significantly. Impeded heat dissipation in concealed areas forms local heat traps, resulting in a three-dimensional temperature gradient across the entire workpiece. Different regions exhibit significant temperature differences and varying cooling rates, deviating from a simple two-dimensional temperature distribution between the surface and interior. Furthermore, nitrogen adsorbed during sintering desorbs during the furnace cooling stage. This desorption process, accompanied by a small amount of heat release, creates a dynamic thermal effect, not only slowing down the cooling process in concealed areas but also causing non-monotonic temperature changes in the workpiece, resulting in localized short-term temperature fluctuations or abrupt changes in cooling rates, thus forming a dynamic coupling between temperature and desorption heat.

[0004] Existing surface temperature measurement technologies can only acquire temperature data of the exposed area, failing to reflect the true distribution of the three-dimensional temperature gradient. Furthermore, the linear correction algorithm does not consider the dynamic time-varying characteristics of temperature and the influence of desorption heat, making it impossible to accurately compensate for multi-path propagation detection signals. This results in a non-linear deviation in the mapping relationship between the detection signal and density, making it difficult for existing technologies to achieve accurate quantification of density and compacted density. This not only fails to provide reliable data support for production line quality control but may also lead to the misjudgment of qualified products as unqualified products or the failure to detect unqualified products, which may then flow into subsequent processes. At the same time, it cannot provide accurate feedback for sintering process optimization, thereby increasing the scrap rate and production costs on the production line, and affecting the quality control and production efficiency improvement of irregularly shaped sintered parts in nitrogen-based atmosphere double-pass pusher kiln production lines.

[0005] In view of this, the present invention proposes a non-destructive testing system and method for the density and compactness of solid materials to solve the above problems. Summary of the Invention

[0006] To overcome the aforementioned deficiencies of the prior art and achieve the above objectives, the present invention provides the following technical solution: a non-destructive testing method for the density and compactness of solid materials, comprising:

[0007] Based on the pre-acquired workpiece geometry data, collect workpiece surface temperature distribution data;

[0008] Record infrared thermal imager frames for multiple time periods; based on the infrared thermal imager frames for multiple time periods, determine the area of ​​the workpiece where nitrogen desorption exothermic effect occurs, and calculate the nitrogen desorption thermal characteristic data.

[0009] Based on the workpiece surface temperature distribution data, workpiece geometric structure data, and nitrogen desorption heat characteristic data, the internal temperature field data of the workpiece is simulated and calculated.

[0010] Perform ultrasonic and electromagnetic testing on the workpiece to obtain ultrasonic testing signal data and electromagnetic testing signal data;

[0011] By using ultrasonic and electromagnetic detection signal data, nonlinear temperature compensation is performed on the internal temperature field data of the workpiece, and fusion density calculation is performed to obtain the density and compactness of the workpiece.

[0012] Furthermore, the nitrogen desorption thermal characteristic data are the local heat release characteristic parameters of the workpiece caused by nitrogen desorption, including the location of heat release, the start time of heat release, the duration of heat release, and the equivalent heat release intensity.

[0013] Furthermore, based on infrared thermal imager frames from multiple time periods, time-series data of workpiece surface temperature are generated;

[0014] The workpiece surface is divided into regions based on the workpiece's geometric structure data, and the surface of the structurally hidden areas is determined.

[0015] Furthermore, the method for determining the surface of the structurally concealed area is as follows: extract the spatial boundary between the internal cavity and the depth extension channel from the workpiece geometric data, and project the spatial boundary along the normal direction onto the outer surface of the workpiece to obtain the outer surface mapping area corresponding to the internal structure; calculate the shortest distance from each point of the outer surface mapping area to the nearest internal cavity or depth extension channel in the workpiece geometric data, and merge the outer surface points whose shortest distance is less than a preset distance threshold into the surface of the structurally concealed area.

[0016] Furthermore, temperature change curves over time are extracted for all areas of the workpiece surface, and the rate of temperature change is calculated.

[0017] The change in the rate of temperature change is calculated to identify three abnormal patterns: slowing temperature decline, local rebound, and plateau.

[0018] If there is a region that meets any of the three abnormal modes' criteria, then the corresponding region is determined to have a nitrogen desorption exothermic effect.

[0019] Furthermore, the method for determining the slowdown in temperature decrease is as follows: select two adjacent time periods within the same region, calculate the average value of the temperature change rate for each period, and if the absolute value of the temperature change rate in the latter period is smaller than that in the former period and the magnitude of the decrease exceeds the slowdown threshold, then it is determined that the temperature decrease has slowed down.

[0020] The method for determining local rebound is as follows: if there is a segment on the filtered temperature change curve over time where the rate of temperature change changes from negative to positive and continues to exceed the rebound threshold, and the temperature increment corresponding to this segment exceeds the minimum rebound temperature difference threshold, then a local rebound is determined to have occurred.

[0021] The method for determining a plateau period is as follows: if the absolute value of the temperature change rate is less than a preset plateau threshold and the duration exceeds a preset minimum plateau duration threshold within a continuous time period, then a plateau period is determined to have occurred.

[0022] Furthermore, the temperature difference residual sequence between the actual cooling curves and the theoretical pure cooling curves of all regions is calculated, and the equivalent heat release intensity and duration of all regions are inversely determined by minimizing the residual energy; the actual cooling curve is the temperature change curve of the region after filtering over time.

[0023] Furthermore, the ultrasonic path for ultrasonic testing and the electromagnetic measuring point for electromagnetic testing are selected based on the workpiece's geometric structure data.

[0024] The original flight time of the ultrasonic path is extracted from the ultrasonic detection signal data, and the original flight time is corrected by piecewise temperature integration to obtain the ultrasonic propagation time characteristics.

[0025] The original attenuation features are extracted from the ultrasonic detection signal data, and the temperature-induced internal friction attenuation contribution is subtracted from the original attenuation features to obtain the ultrasonic attenuation residual features.

[0026] The observed impedance amplitude and observed phase of each measuring point are extracted from the electromagnetic detection signal data. The temperature compensation function is then used to solve the observed impedance amplitude and observed phase to obtain the electromagnetic impedance amplitude characteristics and electromagnetic phase characteristics.

[0027] Furthermore, the ultrasonic propagation time characteristics, ultrasonic attenuation residual characteristics, electromagnetic impedance amplitude characteristics, and electromagnetic phase characteristics are input into a pre-constructed density evaluation model, and the density value and compaction value are output.

[0028] When the equivalent exothermic intensity is greater than or equal to the preset effectiveness threshold, the density assessment model will use nitrogen desorption heat characteristic data as additional input.

[0029] A non-destructive testing system for the density and packing density of solid materials, comprising:

[0030] The temperature acquisition module collects surface temperature distribution data of the workpiece based on pre-acquired workpiece geometric data.

[0031] The thermal feature module is used to record infrared thermal imager frames for multiple time periods; based on the infrared thermal imager frames for multiple time periods, the area where the nitrogen desorption exothermic effect occurs on the workpiece is determined, and the nitrogen desorption thermal feature data is calculated.

[0032] The internal temperature module simulates and calculates the internal temperature field data of the workpiece based on the workpiece surface temperature distribution data, workpiece geometric structure data, and nitrogen desorption heat characteristic data.

[0033] The signal detection module is used to perform ultrasonic and electromagnetic testing on the workpiece and obtain ultrasonic and electromagnetic test signal data.

[0034] The results calculation module uses ultrasonic and electromagnetic detection signal data to perform nonlinear temperature compensation on the internal temperature field data of the workpiece and performs fusion density calculation to obtain the density and compactness of the workpiece.

[0035] Compared with the prior art, the technical effects and advantages of the non-destructive testing system and method for the density and compactness of solid materials of the present invention are as follows:

[0036] This invention obtains workpiece geometric structure data in advance, collects workpiece surface temperature distribution data, records infrared thermal imager frames for multiple time periods to determine the nitrogen desorption exothermic effect area and calculate related characteristic data, combines surface temperature distribution, geometric structure and desorption heat characteristic data to simulate the internal temperature field data of the workpiece, then performs ultrasonic and electromagnetic detection on the workpiece to obtain corresponding signal data, uses the signal data to perform nonlinear temperature compensation on the internal temperature field data and then performs fusion density calculation to finally obtain the density and compactness of the workpiece.

[0037] This invention effectively solves the problem of nonlinear deviation in detection signals caused by the dynamic coupling of three-dimensional temperature gradient and nitrogen desorption exothermic reaction during hot online detection of irregularly shaped sintered parts in continuous production in a nitrogen-based atmosphere dual-track pusher kiln. It overcomes the shortcomings of existing technologies, such as incomplete surface temperature measurement and poor linear correction, achieving accurate quantitative determination of density and compactness. This invention enables detection without waiting for workpiece cooling, matching the continuous operation rhythm of the production line, avoiding efficiency losses, providing reliable data support for production line quality control, reducing misjudgments of qualified products and omissions of unqualified products, and providing accurate feedback for sintering process optimization. This reduces scrap rates and production costs, improving the quality control level and production efficiency of irregularly shaped sintered parts. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of a non-destructive testing system for the density and compactness of solid materials according to an embodiment of the present invention;

[0039] Figure 2 This is a flowchart of a non-destructive testing method for the density and compactness of solid materials according to an embodiment of the present invention;

[0040] Figure 3 This is a flowchart illustrating a method for determining whether a nitrogen desorption exothermic effect occurs in a corresponding region, according to an embodiment of the present invention. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be described in detail, clearly, and completely below with reference to the accompanying drawings. It should be particularly noted that the specific embodiments described below are only for better illustrating and explaining the technical solutions of the present invention, and are intended to enable those skilled in the art to better understand and implement the present invention, and should not be construed as limiting the scope of protection of the present invention. Without departing from the spirit and substance of the present invention, those skilled in the art can modify, adjust, or make equivalent substitutions based on the content disclosed in the present invention, and these should all be considered within the scope of protection of the present invention.

[0042] Example 1:

[0043] Please see Figure 1 As shown in the figure, this embodiment discloses a non-destructive testing system for the density and compactness of solid materials, including a temperature acquisition module, a thermal characteristic module, an internal temperature module, a signal detection module, and a result calculation module. Each module is connected by wires and / or wirelessly to realize data transmission.

[0044] The temperature acquisition module collects surface temperature distribution data of the workpiece based on pre-acquired workpiece geometric data.

[0045] The workpiece geometry data includes the sintered part's three-dimensional shape, dimensions, internal cavities, wall thickness, and other structural information. The structural model of the workpiece is obtained from the product design's CAD model or 3D scanning, and the geometric data corresponding to the workpiece number is retrieved from the database. This workpiece geometry data is used to simulate heat distribution and plan the location of detection sensors.

[0046] The workpiece surface temperature distribution data represents the surface temperature field distribution of the sintered part immediately after it exits the furnace and during the cooling process. The data is acquired by using a high-resolution infrared thermal imager to capture real-time images of the entire surface temperature of the workpiece after it exits the furnace. The infrared camera needs to be blackbody calibrated and an appropriate emissivity set to ensure accurate temperature acquisition of each exposed area. Infrared sensors are arranged at multiple angles to cover any surfaces that may be exposed in concealed areas of the irregularly shaped part. The workpiece surface temperature distribution data is represented as a matrix or thermal image of the temperature values ​​of each exposed surface. For example, a temperature matrix is ​​output, where each element corresponds to the temperature on the model's surface grid. This output will be used for subsequent internal temperature calculations and signal temperature compensation.

[0047] The thermal feature module is used to record infrared thermal imager frames for multiple time periods; based on the infrared thermal imager frames for multiple time periods, the area where nitrogen desorption exothermic effect occurs on the workpiece is determined, and the nitrogen desorption thermal feature data is calculated.

[0048] Nitrogen desorption thermal characteristic data is a set of characteristic parameters of local heat release in the workpiece caused by nitrogen desorption. Its contents are uniformly defined as the location of heat release, the start time of heat release, the duration and heat estimation. The heat estimation is characterized by the equivalent heat release intensity and is used together with the duration for subsequent heat source modeling.

[0049] After the workpiece exits the furnace, continuous frames are recorded by an infrared thermal imager to form a time-series data of the workpiece surface temperature. The recording duration is not limited to tens of seconds, but is limited to a time window covering the observable impact of nitrogen desorption exothermic effects on surface temperature changes. The logic for determining this time window is to cover at least one cooling phase of the workpiece from the moment it exits the furnace, without reducing the production line cycle time, and to ensure that the temperature change rate of each area to be analyzed in the workpiece surface temperature time-series data exhibits a stable decreasing segment and possible abnormal segments. In engineering implementation, the time window is determined through an online adaptive method. That is, when the global average temperature change rate of the workpiece remains monotonically decreasing over several consecutive frames and its change amplitude is lower than a preset convergence threshold, the observable dynamics caused by desorption heat are considered to have decayed, and recording ends. The preset convergence threshold is jointly determined by the infrared thermal imager's temperature measurement noise and sampling frequency, specifically ensuring that the temperature change rate fluctuation caused by noise does not exceed the convergence threshold, thereby avoiding misjudging noise as an exothermic anomaly.

[0050] Based on workpiece geometric data, the workpiece surface is divided into regions, and the surfaces of structurally concealed areas are determined. The method for determining the surfaces of structurally concealed areas is as follows: The spatial boundaries of internal cavities and deep extension channels are extracted from the workpiece geometric data, and these spatial boundaries are projected onto the workpiece's outer surface along the normal direction, obtaining the outer surface mapping region corresponding to the internal structure. The shortest distance from each point in the outer surface mapping region to the nearest internal cavity or deep extension channel is calculated from the workpiece geometric data. Outer surface points with shortest distances less than a preset distance threshold are grouped into the surfaces of structurally concealed areas. The preset distance threshold is set to ensure that the region covers the area where the internal structure has the greatest impact on heat dissipation from the outer surface. Its value is determined jointly based on the thermal diffusion length of the workpiece material and the spatial resolution of the infrared thermal imager. Specifically, it is not less than the entity length corresponding to one spatial pixel of the infrared thermal imager and not greater than the thermal diffusion length of the material within the recording time, thus ensuring that the region division can be effectively distinguished by the infrared thermal imager while also covering the area affected by the heat trap caused by the internal structure.

[0051] Please see Figure 3 As shown, after obtaining the region division, the temperature change curve over time is extracted for all regions of the workpiece surface, and the temperature change rate is calculated to avoid missing the exothermic effect of nitrogen desorption at unexpected locations. The surfaces of structurally concealed areas are prioritized to improve the confidence level of anomaly detection. Specifically, for each region, the pixel temperatures in the workpiece surface temperature time series data are spatially averaged to obtain the temperature change curve over time for that region. A time-smoothing filter is applied to this temperature change curve to suppress measurement noise. The filter window length is determined by the infrared thermal imager frame rate, ensuring that the random fluctuation of the filtered temperature change rate is less than a preset convergence threshold. The temperature change rate is calculated on the filtered temperature change curve over time, and the change in the temperature change rate is further calculated to identify three types of anomaly patterns: slowing temperature decrease, local rebound, or plateau.

[0052] The method for determining a slowdown in temperature decrease is to select two adjacent time periods within the same region, calculate the average temperature change rate for each period, and determine if a slowdown in temperature decrease has occurred if the absolute value of the temperature change rate in the latter period is smaller than that in the former period and the decrease exceeds a slowdown threshold. The slowdown threshold is set based on a lower limit of temperature change rate fluctuations caused by infrared thermal imager noise and an upper limit of the maximum natural decay of the temperature change rate under normal convection cooling conditions. A threshold that can reliably distinguish between natural decay and abnormal slowdown is selected between these two. The method for determining a local rebound is to find segments on the filtered temperature-time curve where the temperature change rate changes from negative to positive and continuously exceeds a rebound threshold, and the temperature increment corresponding to this segment exceeds a minimum rebound temperature difference threshold. Both the rebound threshold and the minimum rebound temperature difference threshold are determined based on the statistical amplitude of the infrared thermal imager's temperature measurement noise, ensuring that the probability of the above determination under pure noise conditions is lower than a preset false alarm rate. The plateau period is determined when the absolute value of the temperature change rate is less than a plateau threshold and the duration exceeds a minimum plateau duration threshold within a continuous time period. The plateau threshold is also determined based on the infrared thermal imager noise level, while the minimum plateau duration threshold is determined based on the production line cycle time and the infrared thermal imager frame rate, ensuring that the plateau period determination spans a sufficient number of consecutive frames to eliminate random noise disturbances. Once any of the above three abnormal modes meets any of the determination conditions in a certain area, it is determined that a nitrogen desorption exothermic effect has occurred in that area, and that area is included in the subsequent equivalent exothermic intensity estimation.

[0053] The equivalent heat release intensity is estimated by comparing the actual cooling curve and the theoretical pure cooling curve of the region. The actual cooling curve is the filtered temperature change curve over time in that region. The theoretical pure cooling curve is obtained by establishing a three-dimensional transient heat conduction model based on the workpiece geometry data and material thermophysical parameters without introducing an internal heat source. The initial frame of workpiece surface temperature distribution data is used as the initial boundary temperature. Simulation is performed using convective heat transfer boundary conditions consistent with the actual cooling environment to obtain the temperature change over time in that region without an internal heat source, which is then used as the theoretical pure cooling curve. To avoid introducing systematic errors into the theoretical pure cooling curve due to the uncertainty of the convection coefficient, the convective heat transfer parameters are calibrated before simulation by selecting a reference region that is not determined to have nitrogen desorption heat release effect. This minimizes the error between the theoretical pure cooling curve and the actual cooling curve of the reference region in the time period before the anomaly occurs, thus ensuring that the theoretical pure cooling curve reflects the real cooling conditions.

[0054] After obtaining the actual cooling curve and the theoretical pure cooling curve, the residual temperature difference sequence between the two is calculated, and the equivalent heat release intensity and duration are inversely determined by minimizing the residual energy. Specifically, the nitrogen desorption heat release effect is equivalent to a time function heat source acting on the corresponding internal volume of the region. Its spatial location is determined by the internal mapping of the region in the workpiece geometry data. The heat source time function adopts the form of rising from zero to the equivalent heat release intensity at the heat release start time and maintaining it during the duration. Using the equivalent heat release intensity, heat release start time, and duration as parameters to be estimated, a transient heat conduction model is called to generate a predicted cooling curve after adding the equivalent heat source, and the mean square error between the predicted cooling curve and the actual cooling curve is calculated. The parameter set that minimizes the mean square error and satisfies the parametric physical constraints through iterative search is used as the estimation result. The parametric physical constraints include the equivalent heat release intensity being non-negative, the duration being not less than the minimum platform duration threshold and not exceeding the recorded duration, and the heat release start time being not earlier than the furnace exit time and not later than the time when the anomaly is first detected, thereby ensuring that the estimation result has uniqueness and interpretability.

[0055] The final nitrogen desorption heat characteristic data includes location coordinates, exothermic onset time, duration, and equivalent exothermic intensity. The location coordinates are determined by the coordinates of the center point of the surface region where the nitrogen desorption exothermic effect occurs in the workpiece geometry data and the coordinates of the center point of the corresponding internal mapped volume. The exothermic onset time and duration are obtained through the aforementioned inverse calculation process, and the equivalent exothermic intensity is given by the inverse calculation result minimizing the mean square error. This nitrogen desorption heat characteristic data is used in the next step to input the nitrogen desorption exothermic effect as a time-varying internal heat source into the calculation of the internal temperature field, thereby improving the accuracy of the internal temperature field inversion and enhancing the reliability of subsequent quantitative assessments of density and compactness.

[0056] The internal temperature module simulates and calculates the internal temperature field data of the workpiece based on the workpiece surface temperature distribution data, workpiece geometric structure data, and nitrogen desorption heat characteristic data.

[0057] First, a three-dimensional transient heat conduction model of the workpiece is constructed based on its geometric data. The three-dimensional geometry in the workpiece geometric data is then discretized into a set of computational units for numerical solution, allowing the temperature field to be expressed and updated using these computational units. The moment of exiting the furnace is set as the initial condition, which includes the initial internal temperature and the initial external surface temperature of the workpiece. The initial internal temperature is set to a uniform initial value close to the sintering temperature based on the sintering exit state, while the initial external surface temperature is assigned based on the initial frame of the workpiece surface temperature distribution data. This ensures that the temperature of each external surface region at the moment of exiting the furnace is consistent with the measured temperature in the model, thereby avoiding initial errors caused by localized cooling of the surface.

[0058] Regarding boundary conditions, convective heat transfer boundary conditions are applied to the three-dimensional transient heat conduction model to characterize the cooling environment after exiting the furnace. These convective heat transfer boundary conditions include ambient temperature and convective heat transfer coefficient, and are set according to the actual operating conditions of the production line cooling station, enabling the model to depict the temperature gradient formation process caused by the difference in heat dissipation conditions between exposed and concealed areas. Regarding the internal heat source setting, nitrogen desorption heat characteristic data is introduced as an internal heat source. The position coordinates in the nitrogen desorption heat characteristic data are mapped to a set of computational units obtained by discretizing the workpiece geometric structure data to determine the corresponding internal action volume. The heat release start time, duration, and equivalent heat release intensity in the nitrogen desorption heat characteristic data are converted into time-dependent volumetric heat source terms, and these volumetric heat source terms are allocated to the computational units corresponding to the internal action volume, thus ensuring that the desorption heat release is consistent with the identification results in both time and space. The time function of the heat source term is triggered by the heat release start time. During the duration, heat is continuously injected into the internal working volume at an equivalent heat release intensity, ensuring that the desorption heat can be reflected as a dynamic thermal effect in the three-dimensional transient heat conduction model and change the local cooling rate.

[0059] Subsequently, transient numerical solutions were run, progressing from the furnace exit time to the detection time, updating the temperature of each computational unit hourly and obtaining the three-dimensional temperature distribution at the detection time. To improve the consistency between the simulation results and actual working conditions, the time-varying results of the external surface temperature calculated by the three-dimensional transient heat conduction model were aligned and compared with the measured time series data of the workpiece surface temperature during the solution process. The comparison object was the temperature curve of the same external surface region on the same time series. If there was a deviation between the two, iterative correction was performed: the convective heat transfer coefficient was adjusted first to reduce the global cooling rate error, and when the deviation was mainly concentrated in the region where the nitrogen desorption exothermic effect occurred, the equivalent exothermic intensity of the volume heat source term was fine-tuned in conjunction, so that the mean square error of the simulated external surface temperature curve and the measured external surface temperature curve continuously decreased and converged to the allowable error range. The logic for setting the allowable error range is based on the temperature measurement uncertainty and surface emissivity calibration error of the infrared thermal imager. The combined temperature error is taken as the lower limit, and the upper limit is set in combination with the production line's requirements for temperature compensation accuracy. This ensures that when the mean square error is lower than the allowable error range, further iterations will no longer significantly improve the density assessment accuracy, thus achieving a balance between accuracy and real-time calculation.

[0060] The final obtained internal temperature field data of the workpiece represents the temperature distribution in the three-dimensional space inside the workpiece at the time of detection, and provides the current temperature value of each calculation unit using the set of calculation units as an index. The internal temperature field data of the workpiece can reflect the temperature retention and three-dimensional temperature gradient characteristics of the hidden area, providing a basic input for subsequent nonlinear temperature compensation of ultrasonic and electromagnetic detection signal data by path and measurement point, thereby reducing the interference of temperature gradient and desorption thermal dynamic effects on the absolute quantification of density and compaction.

[0061] The signal detection module is used to perform ultrasonic and electromagnetic testing on the workpiece to obtain ultrasonic and electromagnetic test signal data.

[0062] In the ultrasonic testing section, the ultrasonic path is first determined based on the workpiece's geometric data. The process involves obtaining the spatial location, wall thickness distribution, and geometric boundaries of the internal cavities or depth-extending channels of the area to be tested from the workpiece's geometric data. This allows for the determination of the transmitter and receiver positions, the ultrasonic incident direction, and the propagation mode, enabling the ultrasonic waves to form a calculable spatial propagation trajectory within the workpiece. Propagation modes include bilateral direct propagation and unilateral reflection propagation. Bilateral direct propagation is used in areas where probes can be arranged opposite each other on opposite sides of the workpiece, while unilateral reflection propagation is used in areas where probes can only be placed on the same side and where interface reflection is required to achieve an effective propagation distance. After determining the ultrasonic path, the ultrasonic transducer is controlled to emit ultrasonic pulses according to preset excitation parameters, and echo signals are simultaneously acquired at the receiver to obtain the original waveform data corresponding to each ultrasonic path. For non-contact acquisition using electromagnetic ultrasonic transducers, the relative distance and incident angle between the transducer and the workpiece surface are geometrically calibrated based on the workpiece's geometric structure data to ensure the ultrasonic energy coupling efficiency remains within a stable range. During acquisition, the echo signal-to-noise ratio (SNR) is continuously monitored to guarantee data validity. The threshold setting logic for the echo SNR is based on the root mean square amplitude of the no-load background noise as the noise reference, combined with the requirement that flight time jitter during repeated measurements does not exceed the upper limit of the allowable error. Echo signals below this threshold are excluded from feature extraction, thus avoiding the introduction of low-confidence data into subsequent compensation errors. Subsequently, features are extracted from the raw waveform data of each ultrasonic path. Feature extraction includes determining the transmission trigger time and the arrival time of the main echo and calculating the flight time, as well as calculating the echo amplitude within a specified frequency band and combining it with the geometric propagation distance of the ultrasonic path to obtain attenuation features, thereby forming flight time features and attenuation features corresponding to each ultrasonic path. The obtained ultrasonic detection signal data is a set of original signal features organized according to the ultrasonic path. Its content includes the path identifier of each ultrasonic path and the corresponding time-of-flight characteristics and attenuation characteristics. It also retains the transmitter and receiver layout information associated with the ultrasonic path for subsequent temperature compensation and fusion calculation.

[0063] In the electromagnetic detection section, electromagnetic measurement points are determined based on the workpiece's geometric structure data. These measurement points are a set of spatial locations on the surface or shallow areas where density or compaction needs to be evaluated. The determination logic involves dividing the workpiece's geometric structure data into regions, prioritizing areas with abrupt changes in wall thickness, the outer surface mapping areas of internal cavities or depth extension channels, and geometric transition areas. These regions are then mapped to the workpiece's outer surface coordinate system to generate the spatial coordinates of the electromagnetic measurement points. Subsequently, the eddy current sensor probe is positioned near the surface corresponding to each electromagnetic measurement point. A stable gap is maintained using a non-contact method, and alternating excitation is applied to generate an alternating electromagnetic field in the coil. The response signal of the probe coil is collected, and electromagnetic response characteristics such as impedance amplitude and phase are calculated. To ensure the stability of the non-contact gap, a gap stability threshold is set, and online consistency verification is performed. The logic for setting the gap stability threshold is based on the maximum gap variation range caused by production line mechanical vibration and the statistical amplitude of the probe's impedance noise under no-load conditions. This ensures that short-term fluctuations in impedance amplitude and phase do not affect the repeatability of density evaluation when the gap variation does not exceed the threshold. When the gap variation exceeds the threshold, repositioning is triggered, or the measurement point data is marked for reduced weight during subsequent compensation. After data acquisition, the observed impedance amplitude and observed phase of each electromagnetic measuring point are output, forming a set of electromagnetic response features corresponding to each electromagnetic measuring point. The obtained electromagnetic detection signal data includes the measuring point identifier of each electromagnetic measuring point and its corresponding observed impedance amplitude and observed phase, and retains the excitation frequency and probe gap state information associated with the electromagnetic measuring point. This information is used in subsequent steps to perform reference temperature equivalent feature transformation in conjunction with the internal temperature field data of the workpiece and to participate in the fusion calculation of density and compaction.

[0064] The results calculation module uses ultrasonic and electromagnetic detection signal data to perform nonlinear temperature compensation on the internal temperature field data of the workpiece and performs fusion density calculation to obtain the density and compactness of the workpiece.

[0065] Through multi-parameter temperature compensation, the ultrasonic and electromagnetic detection signal data collected by the hot workpiece under the dynamic coupling conditions of three-dimensional temperature gradient and nitrogen desorption exothermic reaction can be nonlinearly compensated by path and measurement point. This restores the stable mapping relationship between the compensated detection features and the material density and compactness, and on this basis, outputs quantitative results that can be used for production line quality judgment and process closed loop.

[0066] First, ultrasonic signal compensation is performed. When acquiring ultrasonic detection signal data, the ultrasonic path corresponding to each measurement is determined based on the spatial coordinates of the transmitting and receiving ends, the incident direction, and the workpiece's outer boundary in the workpiece's geometric data. This ultrasonic path serves as the geometric basis for subsequent temperature compensation. When using a single-sided arrangement, the ultrasonic path corresponds to the equivalent propagation trajectory from the transmitting point along the incident direction into the workpiece, reaching the interface reflection point, and then returning to the receiving point. When using a double-sided arrangement, the ultrasonic path corresponds to the direct propagation trajectory from the transmitting point to the receiving point. For each ultrasonic path, the temperature distribution along the ultrasonic path is extracted from the workpiece's internal temperature field data. The extracted temperatures are the unit temperatures of the voxels or units traversed by the ultrasonic path at the current detection moment, and these unit temperatures constitute the temperature sequence of the ultrasonic path.

[0067] The ultrasonic propagation time characteristic is obtained by correcting the original flight time with piecewise temperature integration. Specifically, the original flight time of the ultrasonic path is first extracted from the ultrasonic detection signal data. The extraction method involves performing time-domain processing on the transmission trigger time and the received echo waveform, determining the arrival time of the main echo using an arrival time criterion consistent with the calibration, and subtracting the trigger time from this arrival time to obtain the original flight time. Subsequently, the coupling relationship between sound velocity and temperature and density is obtained using a material property and density relationship model. The material property and density relationship model is obtained by selecting a calibration sample of the same material system with a density determined by a benchmark method under nitrogen-based atmosphere hot conditions consistent with the production line, performing ultrasonic measurements at multiple controlled temperature points, establishing a corresponding dataset of sound velocity, temperature, and density, and using the continuous differentiability of sound velocity with respect to temperature and monotonicity with respect to density as a physical constraint, obtaining the sound velocity function through parametric regression fitting, so that the sound velocity can be expressed as a function of temperature and density. Based on the sound velocity function, each ultrasonic path is segmented and calculated, spatially discretized into several adjacent segments. Each segment corresponds to a unit temperature read from the workpiece's internal temperature field data, and this unit temperature is used as the temperature input for that segment. The temperature input is the unit temperature of the unit containing that segment at the current detection time. The density input uses the nominal density as the initial value to complete a forward correction, thereby obtaining a calculable segmented sound velocity without introducing unknown densities. Then, the propagation time of each segment is obtained by dividing the length of each segment by the corresponding segmented sound velocity and summed to obtain the temperature-compensated propagation time of the ultrasonic path. After consistency correction between this temperature-compensated propagation time and the original flight time in the ultrasonic detection signal data, the consistency correction result is taken as the ultrasonic propagation time feature. The consistency correction means minimizing the systematic deviation between the ultrasonic propagation time feature and the original flight time under the same temperature and density conditions of the calibrated sample, thereby eliminating the fixed errors caused by probe delay and acquisition link delay.

[0068] The ultrasonic attenuation residual characteristic is obtained by subtracting the contribution of temperature-induced internal friction attenuation from the original attenuation characteristic. First, the original attenuation characteristic is extracted from the ultrasonic detection signal data. This extraction method involves selecting at least two echo amplitudes with known propagation distances along the same ultrasonic path, or selecting a reference echo and a target echo with known distances along the same path. The amplitude ratio is calculated and combined with the propagation distance to obtain the equivalent attenuation coefficient. The propagation distance is determined by the workpiece geometry and the geometric length of the ultrasonic path. Subsequently, the attenuation coefficient is decomposed into two parts: a material internal friction term caused by temperature and a structural term caused by pore scattering. The material internal friction term caused by temperature is established on a calibrated sample using a material property and density relationship model. This model is established by measuring the attenuation curves of multi-band ultrasonic amplitudes with propagation distance under the same density but different temperature conditions, and fitting the curves to obtain the contribution function of temperature to the internal friction attenuation. During compensation, the contribution function is used to integrate the temperature sequence of the ultrasonic path piecewise, calculating the predicted attenuation caused by material friction under the current temperature distribution. This predicted attenuation is then subtracted from the original attenuation features to obtain the ultrasonic attenuation residual features. These residual features primarily reflect pore scattering information, thus improving the attenuation features' ability to distinguish density. A signal-to-noise ratio (SNR) screening threshold is set to ensure the reliability of the attenuation features. When the echo SNR of an ultrasonic path is lower than the threshold, that path is not included in subsequent fusion calculations. The SNR threshold is determined based on sensor background noise and repeatability stability, specifically by statistically analyzing the noise root mean square of the echo amplitude under no-load and stable thermal conditions. The threshold is set to the minimum SNR required for the jitter at the echo arrival time to exceed the allowable error limit in repeatability measurements when the threshold is lower, thus avoiding low-confidence data introducing density inversion bias and preventing noise dominance in the attenuation residuals.

[0069] Electromagnetic signal compensation is then performed. For each electromagnetic measuring point, the current temperature of the corresponding material volume is read from the internal temperature field data of the workpiece as the temperature input. This temperature input is the weighted average temperature within the effective electromagnetic penetration depth range after the electromagnetic measuring point is projected onto the workpiece surface. The weight of the weighted average is determined by the excitation frequency of the electromagnetic measuring point and the electromagnetic parameters of the material, ensuring that the temperature used for compensation is consistent with the volume sensitive area of ​​the electromagnetic measurement. The model used for electromagnetic compensation is also derived from the material property and density relationship model. It is obtained by selecting a calibration sample with the same material system and known density under nitrogen-based atmosphere hot conditions consistent with the production line, and collecting impedance amplitude and phase data at multiple controlled temperature points and multiple excitation frequencies. A calibration curve of impedance amplitude and phase changing with temperature is established, and continuity and monotonicity constraints are applied to the temperature sensitive terms of impedance amplitude and phase to form a reversible temperature compensation function. The meaning of the reversible temperature compensation function is that, under given excitation frequency and temperature input, the temperature sensitive term can be uniquely determined and separated from the observation, thereby obtaining the equivalent characteristics at the reference temperature.

[0070] The electromagnetic impedance amplitude and electromagnetic phase characteristics are obtained by inversely solving the observed impedance amplitude and observed phase using a temperature compensation function. Specifically, the observed impedance amplitude and observed phase of each electromagnetic measuring point are first extracted from the electromagnetic detection signal data. The extraction method involves acquiring the coil response at a set excitation frequency and obtaining the amplitude and phase through demodulation. Then, the observed impedance amplitude, observed phase, and temperature input are jointly input into a reversible temperature compensation function to calculate the corresponding temperature-sensitive term. The temperature-sensitive term is then separated from the observed impedance amplitude and observed phase to obtain the electromagnetic impedance amplitude and electromagnetic phase characteristics. The reference temperature is a unified reference temperature selected when establishing the temperature compensation function for the calibration sample, ensuring that the electromagnetic characteristics of different workpieces and different detection times are comparable under the same temperature reference.

[0071] After temperature stripping, to suppress amplitude and phase drift caused by changes in the gap between the probe and the workpiece, a consistency constraint correction is further applied to the electromagnetic impedance amplitude and electromagnetic phase characteristics. This consistency constraint correction is based on the principle that gap changes at adjacent moments at the same electromagnetic measurement point will not cause simultaneous abrupt changes in the electromagnetic impedance amplitude and electromagnetic phase characteristics in the same direction as the temperature change. If a simultaneous abrupt change occurs, it is determined to be a gap disturbance, and corrected through local time smoothing and neighborhood electromagnetic measurement point consistency regression. The window length of the local time smoothing is determined by the sampling frequency and the dominant frequency of the production line vibration, ensuring that instantaneous disturbances caused by vibration are smoothed without masking slow changes in material properties. The constraint of neighborhood electromagnetic measurement point consistency regression is defined by the workpiece geometric data, ensuring that neighborhood electromagnetic measurement points are in the same structural region or have the same wall thickness, thereby avoiding mistaking actual structural differences for gap disturbances. The threshold used for gap disturbance determination is set based on the statistical amplitude of probe impedance noise under no-load conditions and the maximum gap change range caused by production line mechanical vibration. The threshold setting logic is to ensure that the amplitude of synchronous mutation generated under no-load noise and maximum mechanical vibration conditions does not exceed the upper limit of the threshold. When the amplitude of synchronous mutation exceeds the upper limit of the threshold, it has a sufficiently high confidence level to determine it as a gap disturbance. Thus, the false alarm probability is lower than the preset false alarm rate when there is no gap disturbance, and the correction can be stably triggered when a gap mutation occurs.

[0072] After completing ultrasonic and electromagnetic signal compensation, a fused density calculation is performed. The density assessment model used in the fused calculation is implemented by a learning sub-model in the material property and density relationship model. Its inputs are ultrasonic propagation time characteristics, ultrasonic attenuation residual characteristics, electromagnetic impedance amplitude characteristics, and electromagnetic phase characteristics, and the outputs are density and compaction values. The reason for choosing the above input data is that ultrasonic propagation time characteristics mainly characterize the propagation speed information related to the elasticity and bulk density of the material; ultrasonic attenuation residual characteristics mainly characterize the pore scattering intensity and are strongly correlated with compaction; electromagnetic impedance amplitude characteristics and electromagnetic phase characteristics mainly characterize the changes in conductivity and permeability and are sensitive to porosity and microstructure. These four types of characteristics are complementary in physical mechanism, which can reduce the systematic bias of a single method under the conditions of propagation path differences caused by irregular structures and temperature non-uniformity caused by local heat traps, thereby solving the problem of absolute quantification of density and compaction under hot online conditions.

[0073] The training data acquisition environment for the density assessment model is a nitrogen-based atmosphere double-pass pusher kiln hot environment or equivalent environment consistent with actual production. The training samples include sintered parts with different geometries and wall thickness distributions. For each sample, while acquiring ultrasonic and electromagnetic detection signal data, the internal temperature field data of the workpiece is simultaneously acquired and the same compensation process is executed to ensure that the density assessment model learns the stable relationship between the compensated features and density and compactness, rather than environmental drift. Training labels are obtained by a benchmark method, which provides the true values ​​of density and compactness for the same batch of samples and records the nitrogen content changes of the samples to avoid mislearning nitrogen content fluctuations as density fluctuations. The density assessment model specifically employs a physically constrained multi-input regression neural network. Its network structure consists of a shared feature extraction layer and a dual-output regression head. The shared layer is used to learn the coupled representation of ultrasonic and electromagnetic features, and the dual-output regression heads output density and compactness values ​​respectively. During training, a physical consistency constraint is added to the loss function so that the density and compactness values ​​satisfy a monotonic relationship when other conditions remain unchanged. A penalty term is applied to the correlation between the temperature compensation residual and the output to reduce the influence of temperature residuals on the output.

[0074] Furthermore, the density assessment model incorporates nitrogen desorption heat characteristic data as additional input. This additional input includes only the equivalent heat release intensity and duration related to the heat release energy, enabling the density assessment model to distinguish between electromagnetic characteristic drift caused by nitrogen content in the pores and electromagnetic characteristic drift caused by changes in true density when encountering changes in desorption intensity. The activation of this additional input is controlled by an effectiveness threshold. When the equivalent heat release intensity is below the effectiveness threshold, the additional input is not activated. This effectiveness threshold is determined based on infrared thermography noise and desorption heat identification error, ensuring that equivalent heat release intensities below the effectiveness threshold lack reliable physical meaning, thereby avoiding noise feature perturbation affecting the fusion results.

[0075] The density value represents the actual absolute density obtained by inversion under hot online detection conditions, while the compaction value represents the degree of densification relative to the theoretical compact state. Because the compensation process uses internal workpiece temperature field data to perform nonlinear temperature compensation on ultrasonic and electromagnetic signals at different paths and measurement points, and because the fusion model training phase is constrained to a hot acquisition environment consistent with the production line and physical consistency constraints are introduced, the output density and compaction values ​​can reduce the systematic deviations caused by the three-dimensional temperature gradient and the exothermic dynamic effects of nitrogen desorption. These values ​​can be used for subsequent quality assessment and process feedback control.

[0076] Through the above steps, this embodiment achieves accurate online quantitative assessment of the hot density and compaction of irregularly shaped sintered parts. Multi-source data fusion and nonlinear dynamic compensation effectively solve the interference of three-dimensional temperature gradients and nitrogen desorption exothermic effects on the detection signal, restoring the density mapping relationship to linearity and stability, and controlling the measurement error within acceptable limits. This embodiment provides reliable quality data support for the production line, enabling timely detection of process deviations, reducing scrap rates, and improving the quality control level and production efficiency of irregularly shaped sintered parts in a continuous production environment of a nitrogen-based atmosphere dual-track pusher kiln.

[0077] Example 2:

[0078] Please see Figure 2 As shown, this embodiment provides a non-destructive testing method for the density and compactness of solid materials, including:

[0079] Based on the pre-acquired workpiece geometry data, collect workpiece surface temperature distribution data;

[0080] Record infrared thermal imager frames for multiple time periods; based on the infrared thermal imager frames for multiple time periods, determine the area of ​​the workpiece where nitrogen desorption exothermic effect occurs, and calculate the nitrogen desorption thermal characteristic data.

[0081] Based on the workpiece surface temperature distribution data, workpiece geometric structure data, and nitrogen desorption heat characteristic data, the internal temperature field data of the workpiece is simulated and calculated.

[0082] Perform ultrasonic and electromagnetic testing on the workpiece to obtain ultrasonic testing signal data and electromagnetic testing signal data;

[0083] By using ultrasonic and electromagnetic detection signal data, nonlinear temperature compensation is performed on the internal temperature field data of the workpiece, and fusion density calculation is performed to obtain the density and compactness of the workpiece.

[0084] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0085] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A non-destructive testing method for the density and packing density of solid materials, characterized in that, include: Based on the pre-acquired workpiece geometry data, collect workpiece surface temperature distribution data; Record infrared thermal imager frames over multiple time periods; Based on infrared thermal imager frames from multiple time periods, the region where nitrogen desorption exothermic effect occurs on the workpiece is determined, and nitrogen desorption thermal characteristic data is calculated. Based on the workpiece surface temperature distribution data, workpiece geometric structure data, and nitrogen desorption heat characteristic data, the internal temperature field data of the workpiece is simulated and calculated. Perform ultrasonic and electromagnetic testing on the workpiece to obtain ultrasonic testing signal data and electromagnetic testing signal data; By using ultrasonic and electromagnetic detection signal data, nonlinear temperature compensation is performed on the internal temperature field data of the workpiece, and fusion density calculation is performed to obtain the density and compactness of the workpiece.

2. The non-destructive testing method for the density and packing density of a solid material according to claim 1, characterized in that, The nitrogen desorption thermal characteristic data are the local heat release characteristic parameters of the workpiece caused by nitrogen desorption, including the location of heat release, the start time of heat release, the duration and the equivalent heat release intensity.

3. The non-destructive testing method for the density and compactness of solid materials according to claim 2, characterized in that, Based on infrared thermal imager frames with multiple time periods, time series data of workpiece surface temperature are generated. The workpiece surface is divided into regions based on the workpiece's geometric structure data, and the surface of the structurally hidden areas is determined.

4. The non-destructive testing method for the density and packing density of a solid material according to claim 3, characterized in that, The method for determining the surface of the structurally concealed area is as follows: extract the spatial boundary of the internal cavity and the depth extension channel from the workpiece geometric data, and project the spatial boundary along the normal direction onto the outer surface of the workpiece to obtain the outer surface mapping area corresponding to the internal structure; calculate the shortest distance from each point of the outer surface mapping area to the nearest internal cavity or depth extension channel in the workpiece geometric data, and classify the outer surface points whose shortest distance is less than a preset distance threshold into the surface of the structurally concealed area.

5. The non-destructive testing method for the density and compactness of solid materials according to claim 3, characterized in that, Extract the temperature change curves over time for all areas of the workpiece surface and calculate the rate of temperature change. The change in the rate of temperature change is calculated to identify three abnormal patterns: slowing temperature decline, local rebound, and plateau. If there is a region that meets any of the three abnormal modes' criteria, then the corresponding region is determined to have a nitrogen desorption exothermic effect.

6. The non-destructive testing method for the density and compactness of solid materials according to claim 5, characterized in that, The method for determining a slowdown in temperature decrease is as follows: Select two adjacent time periods within the same region, calculate the average temperature change rate for each period, and if the absolute value of the temperature change rate in the latter period is smaller than that in the former period and the magnitude of the decrease exceeds the slowdown threshold, then it is determined that a slowdown in temperature decrease has occurred. The method for determining local rebound is as follows: if there is a segment on the filtered temperature change curve over time where the rate of temperature change changes from negative to positive and continues to exceed the rebound threshold, and the temperature increment corresponding to this segment exceeds the minimum rebound temperature difference threshold, then a local rebound is determined to have occurred. The method for determining a plateau period is as follows: if the absolute value of the temperature change rate is less than a preset plateau threshold and the duration exceeds a preset minimum plateau duration threshold within a continuous time period, then a plateau period is determined to have occurred.

7. The non-destructive testing method for the density and packing density of a solid material according to claim 5, characterized in that, Calculate the temperature difference residual sequence between the actual cooling curves and the theoretical pure cooling curves for all regions, and inversely determine the equivalent heat release intensity and duration for all regions by minimizing the residual energy; the actual cooling curve is the temperature change curve over time after filtering for the region.

8. The non-destructive testing method for the density and packing density of a solid material according to claim 1, characterized in that, Select the ultrasonic path for ultrasonic testing and the electromagnetic measuring point for electromagnetic testing based on the workpiece's geometric structure data. The original flight time of the ultrasonic path is extracted from the ultrasonic detection signal data, and the original flight time is corrected by piecewise temperature integration to obtain the ultrasonic propagation time characteristics. The original attenuation features are extracted from the ultrasonic detection signal data, and the temperature-induced internal friction attenuation contribution is subtracted from the original attenuation features to obtain the ultrasonic attenuation residual features. The observed impedance amplitude and observed phase of each measuring point are extracted from the electromagnetic detection signal data. The temperature compensation function is then used to solve the observed impedance amplitude and observed phase to obtain the electromagnetic impedance amplitude characteristics and electromagnetic phase characteristics.

9. A non-destructive testing method for the density and packing density of a solid material according to claim 8, characterized in that, The ultrasonic propagation time characteristics, ultrasonic attenuation residual characteristics, electromagnetic impedance amplitude characteristics, and electromagnetic phase characteristics are input into a pre-constructed density evaluation model, and the density value and compaction value are output. When the equivalent exothermic intensity is greater than or equal to the preset effectiveness threshold, the density assessment model will use nitrogen desorption heat characteristic data as additional input.

10. A non-destructive testing system for the density and packing density of solid materials, used to implement the non-destructive testing method for the density and packing density of solid materials according to any one of claims 1-9, characterized in that, include: The temperature acquisition module collects surface temperature distribution data of the workpiece based on pre-acquired workpiece geometric structure data; The thermal feature module is used to record infrared thermal imager frames over multiple time periods. Based on infrared thermal imager frames from multiple time periods, the region where nitrogen desorption exothermic effect occurs on the workpiece is determined, and nitrogen desorption thermal characteristic data is calculated. The internal temperature module simulates and calculates the internal temperature field data of the workpiece based on the workpiece surface temperature distribution data, workpiece geometric structure data, and nitrogen desorption heat characteristic data. The signal detection module is used to perform ultrasonic and electromagnetic testing on the workpiece and obtain ultrasonic and electromagnetic test signal data. The results calculation module uses ultrasonic and electromagnetic detection signal data to perform nonlinear temperature compensation on the internal temperature field data of the workpiece and performs fusion density calculation to obtain the density and compactness of the workpiece.

Images