Material detection method and system based on big data

Abstract

The invention provides a big data-based material detection method and system, relates to the field of material quality detection, and solves the technical problems of insufficient detection precision and poor supply chain quality control linkage in an existing material detection method. The method comprises the following steps: transmitting sweep-frequency sound waves to warehouse-in materials of a peripheral warehouse by using a sound wave transmitting device; collecting vibration velocity spectrum data and sound pressure gradient data of the warehouse-in material after the warehouse-in material receives the frequency sweep sound wave; according to the vibration velocity spectrum data and the sound pressure gradient data, calculating a defect index of the warehouse-in material; and when the defect index is greater than a preset threshold value, marking the warehousing material as a defective material, sending a traceability detection instruction to a previous node warehouse, and starting a grading detection process of the previous node warehouse. The method and the device are used for warehousing detection, quality traceability and full-chain management and control processes of the materials in the supply chain.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of material quality detection, and specifically relates to a material detection method and system based on big data. BACKGROUND

[0002] In material supply chain management, material quality detection is a key link for guaranteeing stable operation of the supply chain and avoiding defective materials from flowing into terminal application or sales links. However, the existing material detection method usually adopts manual visual inspection combined with traditional instrument sampling inspection, which is prone to insufficient detection accuracy due to technical limitations of the detection means: manual visual inspection is limited by the resolution of the human eye, and the identification accuracy of microscopic defects such as millimeter-level fine cracks and uneven material density is insufficient, and critical-state defects are easily misjudged as qualified; traditional instruments mostly rely on preset thresholds for judgment, and it is difficult to capture small gray-scale differences or structural stress changes on the surface of the material, and the identification accuracy of quality fluctuations at the edge of the qualified is low; the limitation of sampling sample size makes it difficult for local detection results to reflect the true quality level of the whole batch, especially when the defects are distributed discretely, and sample deviation easily causes missed detection, resulting in a significant deviation between the detection accuracy and the actual quality state. SUMMARY

[0003] The application provides a material detection method and system based on big data, which solves the technical problem of insufficient detection accuracy in the prior art.

[0004] To achieve the above-mentioned purpose, the application adopts the following technical solutions: In a first aspect, a material detection method based on big data is provided, comprising: emitting a sweep frequency sound wave to the incoming material of the terminal warehouse by using a sound wave emitting device; collecting vibration velocity spectrum data and sound pressure gradient data of the incoming material after receiving the sweep frequency sound wave; the vibration velocity spectrum data represents the distribution of the vibration velocity of each point on the surface of the material with respect to the frequency, and the sound pressure gradient data represents the rate of change of the sound pressure value on the surface of the material with respect to the spatial position; calculating a defect index of the incoming material according to the vibration velocity spectrum data and the sound pressure gradient data; when the defect index is greater than a preset threshold, marking the incoming material as a defective material and sending a traceability detection instruction to the previous node warehouse; the traceability detection instruction is used to start a hierarchical detection process of the previous node warehouse.

[0005] Based on the above technical scheme, in the material detection method based on big data provided in the application, the high-precision defect screening of the warehousing materials is realized by means of the combination of sound wave detection and big data analysis. Specifically, by collecting vibration velocity spectrum data and sound pressure gradient data, the physical characteristics of the materials are described from different angles, which helps to accurately identify the defects of the materials; when the defect index exceeds the preset threshold, the defective materials are immediately marked and the traceability detection instruction is triggered, so that the problem materials can be processed in time, thereby improving the efficiency and accuracy of material quality control and providing guarantee for the stable operation of the supply chain.

[0006] Further, the vibration velocity spectrum data is obtained by a laser Doppler vibration tester, and the sound pressure gradient data is collected by a distributed microphone array, and the sampling frequencies of the two are the same.

[0007] Further, the calculation of the defect index of the target material according to the vibration velocity spectrum data and the sound pressure gradient data comprises: band-pass filtering and denoising the vibration velocity spectrum data to obtain preprocessed velocity spectrum data v(f); extracting an abnormal frequency band in the preprocessed velocity spectrum data v(f) whose deviation value from a reference spectrum v ref (f) is greater than a preset deviation threshold, to obtain a vibration abnormal zone; the reference spectrum represents the vibration velocity spectrum data of a standard material of the same type; extracting a spatial region in the sound pressure gradient data whose gradient value is greater than a preset gradient threshold, to obtain a sound pressure abnormal zone; marking the intersection of the vibration abnormal zone and the sound pressure abnormal zone as an abnormal vibration zone; obtaining the defect index of the target material according to a defect index calculation formula; the defect index calculation formula is used to represent the relationship among the preprocessed velocity spectrum data v(f), the reference spectrum v ref (f) and the abnormal vibration zone.

[0008] Further, the defect index calculation formula satisfies: ; wherein DI represents the defect index, Ω represents the frequency range of the swept frequency sound wave, A hs represents the area of the abnormal vibration zone, σ represents the standard deviation of the reference spectrum, and f represents the frequency of the swept frequency sound wave.

[0009] Further, before the hierarchical detection, the risk area division of the previous node warehouse is performed, comprising: obtaining a supplier code S and a production batch L of the defective material; screening an associated material set Q in the database of the previous node warehouse; the associated material set has the same supplier code as the defective material, and the production date of the production batch is within a preset time window; calculating the risk coefficient of the materials in the associated material set; According to the spatial position of the last node warehouse, the risk coefficient is clustered and analyzed to obtain a plurality of risk areas.

[0010] Further, the risk coefficient is calculated as follows: λ i =α×δ i +β×e -kΔTi ; wherein λ i represents the risk coefficient of the ith associated material, δ i represents the process similarity of the ith associated material and the defective material, which is obtained by calculating the proportion of the number of the same production equipment used by the ith material and the defective material to the total number of production equipment of the defective material, ΔTi represents the time difference of the production date of the ith associated material and the defective material, k represents the attenuation coefficient, and α and β represent the weight coefficients of each term, which are determined by experimental experience.

[0011] Further, the clustering analysis comprises: obtaining a three-dimensional model of the last node warehouse; mapping the materials based on the three-dimensional model, and creating a tuple containing spatial coordinates (x, y, z), supplier code, production batch, and risk coefficient λ for each material; wherein, if the material w is associated with the material set Q, then the risk coefficient λ w =0; using a spatial clustering algorithm to perform spatial clustering on the tuple of the material point set to obtain a risk area set {Z k |k=1,2,…,K}, wherein Z k represents the kth risk area, and K represents the number of risk areas.

[0012] Further, the hierarchical detection process of the last node warehouse comprises: calculating the average risk coefficient of all associated materials in the kth risk area ; When >T1, the risk area is marked as a first-level risk area, and a first-level detection is performed: emitting a sweep frequency wave to all materials in the first-level risk area in turn, calculating the defect index of each material in the first-level risk area according to the vibration velocity spectrum data and the sound pressure gradient data of the material, and marking the target material as a defective material when the defect index is greater than a preset threshold value; When T2 ≤T1, the risk area is marked as a second-level risk area, and a second-level detection is performed: randomly sampling all materials in the second-level risk area according to a first proportion to obtain first sampling materials; The sweep frequency radio waves are sequentially emitted to the first sampling materials, and defect indexes of the sampling materials are calculated according to the vibration velocity spectrum data and the sound pressure gradient data of the materials; when the defect indexes are greater than a preset threshold, the target material is marked as a defective material; If the proportion of defective materials in the first sampling materials exceeds a preset proportion, the second risk area is marked as a first risk area, and the first detection is performed; When T1≤T2, When T1≤T2, All materials in the third risk area are randomly sampled at a second proportion to obtain second sampling materials; The sweep frequency radio waves are sequentially emitted to the second sampling materials, and defect indexes of the sampling materials are calculated according to the vibration velocity spectrum data and the sound pressure gradient data of the materials; when the defect indexes are greater than a preset threshold, the target material is marked as a defective material; If the proportion of defective materials in the second sampling materials exceeds a preset proportion, the third risk area is marked as a second risk area, and the second detection is performed.

[0013] Further, when the hierarchical detection is performed in the last node warehouse: If new defective materials are detected in the last node warehouse, a trace detection instruction is sent to a superior node warehouse of the last node warehouse, and the risk area division and the hierarchical detection process are performed in the superior node warehouse.

[0014] In a second aspect, the application provides a material detection system based on big data, comprising: a sound wave emitting module, a data acquisition module and a data analysis module; wherein, The sound wave emitting module is configured to emit sweep frequency sound waves to the warehouse-in materials in the terminal warehouse by using a sound wave emitting device; The data acquisition module is configured to acquire vibration velocity spectrum data and sound pressure gradient data of the warehouse-in materials after the warehouse-in materials receive the sweep frequency sound waves; the vibration velocity spectrum data represents the distribution of the vibration velocity of each point on the surface of the materials with the change of frequency, and the sound pressure gradient data represents the rate of change of the sound pressure value of the materials with the spatial position; The data analysis module is configured to calculate a defect index of a target material according to the vibration velocity spectrum data and the sound pressure gradient data, mark the warehouse-in materials as defective materials when the defect index is greater than a preset threshold, and send a trace detection instruction to a last node warehouse; the trace detection instruction is used to start a hierarchical detection process of the last node warehouse.

[0015] In a third aspect, a material detection device based on big data is provided, comprising: a communication unit and a processing unit; wherein, The communication unit is configured to perform data transmission and communication with a sound wave emitting device, a data acquisition device and a last node warehouse. The processing unit is configured to control the sound wave emitting device to emit a sweep frequency sound wave, process the collected vibration velocity spectrum data and sound pressure gradient data, calculate a defect index of the target material, determine whether the material has a defect according to the defect index, and generate and send a traceability detection instruction when the defect exists.

[0016] In a fourth aspect, the present application provides a material detection device based on big data, comprising a processor and a storage medium; the storage medium comprises instructions, and the processor is configured to execute the instructions to implement the method described in the first aspect and any possible implementation manner of the first aspect. The material detection device based on big data can be an electronic device or a chip in an electronic device.

[0017] In a fifth aspect, the present application provides a computer readable storage medium, which stores instructions, and when the instructions are executed on the material detection device based on big data, the material detection device based on big data executes the method described in the first aspect and any possible implementation manner of the first aspect.

[0018] In a sixth aspect, the present application provides a computer program product comprising instructions, and when the computer program product is executed on the material detection device based on big data, the material detection device based on big data executes the method described in the first aspect and any possible implementation manner of the first aspect.

[0019] The present application provides a material detection method and system based on big data, which realizes high-precision defect recognition through a multi-dimensional collaborative detection mechanism, and improves the quality control efficiency of the supply chain by combining traceability detection. In the aspect of defect detection, a laser Doppler vibrometer and a distributed microphone array are used to synchronously collect vibration velocity spectrum data and sound pressure gradient data, which are processed by band-pass filtering and denoising, and then compared with a reference spectrum to extract abnormal vibration areas. The intersection of the abnormal vibration areas and the spatial sound pressure gradient abnormal areas is determined to determine the abnormal vibration area. Then, a defect index is calculated through a quantitative formula, which can improve the accuracy of defect recognition.

[0020] In the aspect of defect traceability, a risk area division-classification detection-multistage linkage traceability analysis mechanism is constructed: according to the supplier code and production batch of the defective material, a set of associated materials is screened, a risk coefficient is calculated by combining the process similarity and the production date difference, and a spatial clustering is performed based on the warehouse three-dimensional model to divide the risk area; the classification detection is implemented according to the average risk coefficient, and the detection is dynamically adjusted from full detection to sampling detection, and when a problem is detected at a node, the traceability is performed to the upper node to form a progressive traceability network. This mechanism can quickly locate the risk source and provide full-process guarantee for the quality control of the supply chain.

[0021] It should be understood that the description of technical features, technical solutions, advantages or similar language in this application does not imply that all features and advantages can be realized in any single embodiment. On the contrary, it can be understood that the description of a feature or advantage means that the specific technical feature, technical solution or advantage is included in at least one embodiment. Therefore, the description of technical features, technical solutions or advantages in this specification does not necessarily refer to the same embodiment. Further, the technical features, technical solutions and advantages described in this embodiment can be combined in any appropriate manner. Those skilled in the art will understand that the embodiments can be implemented without one or more specific technical features, technical solutions or advantages of a specific embodiment. In other embodiments, additional technical features and advantages can be identified in specific embodiments that do not embody all embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the drawings required in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings from these drawings without creative labor.

[0023] Figure 1 A system architecture diagram of a material detection system based on big data is provided for the embodiments of the present application; Figure 2 A flowchart of a material detection method based on big data is provided for the embodiments of the present application; Figure 3 A flowchart of another material detection method based on big data is provided for the embodiments of the present application; Figure 4 A flowchart of another material detection method based on big data is provided for the embodiments of the present application; Figure 5 A structural diagram of a material detection device based on big data is provided for the embodiments of the present application; Figure 6 A hardware structure diagram of a material detection device based on big data is provided for the embodiments of the present application. DETAILED DESCRIPTION

[0024] In the description of the present application, unless otherwise specified, " / " means "or", for example, A / B can mean A or B. "And / or" herein is only a description of the relationship between the associated objects, which means that there can be three relationships, for example, A and / or B, which can mean that A exists alone, A and B exist together, and B exists alone. In addition, "at least one" means one or more, and "multiple" means two or more. "First", "second", and the like do not limit the quantity and execution order, and "first", "second", and the like do not necessarily mean different.

[0025] It should be noted that in the present application, "exemplary" or "for example" is used to mean an example, illustration, or description. Any embodiment or design scheme described as "exemplary" or "for example" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the use of "exemplary" or "for example" is intended to present the relevant concept in a specific manner.

[0026] The material detection method based on big data provided by the embodiments of the present application can be applied to a material detection system based on big data as shown in Figure 1 as shown in Figure 1 The communication system includes a sound wave emitting module, a data acquisition module, and a data analysis module; wherein, The sound wave emitting module is configured to emit a sweep frequency sound wave to the incoming material of the terminal warehouse by using a sound wave emitting device; The data acquisition module is configured to acquire vibration velocity spectrum data and sound pressure gradient data of the incoming material after receiving the sweep frequency sound wave; the vibration velocity spectrum data represents the distribution of the vibration velocity of each point on the surface of the material with respect to the frequency, and the sound pressure gradient data represents the rate of change of the sound pressure value on the surface of the material with respect to the spatial position; The data analysis module is configured to calculate a defect index of the target material according to the vibration velocity spectrum data and the sound pressure gradient data, mark the incoming material as a defective material when the defect index is greater than a preset threshold, and send a traceability detection instruction to the previous node warehouse; the traceability detection instruction is used to start a hierarchical detection process of the previous node warehouse.

[0027] To solve the technical problems of low material detection accuracy and poor supply chain quality control linkage in the prior art, the embodiments of the present application provide a material detection method and system based on big data, which comprises: emit a sweep frequency sound wave to the incoming material of the terminal warehouse by using a sound wave emitting device; acquire vibration velocity spectrum data and sound pressure gradient data of the incoming material after receiving the sweep frequency sound wave; the vibration velocity spectrum data represents the distribution of the vibration velocity of each point on the surface of the material with respect to the frequency, and the sound pressure gradient data represents the rate of change of the sound pressure value on the surface of the material with respect to the spatial position; Calculate the defect index of the warehousing material according to the vibration velocity spectrum data and the sound pressure gradient data; When the defect index is greater than a preset threshold, the warehousing material is marked as a defective material, and a traceability detection instruction is sent to the previous node warehouse; the traceability detection instruction is used to start the hierarchical detection process of the previous node warehouse.

[0028] As Figure 2 indicated, the material detection method based on big data provided by the embodiments of the present application comprises: S1, emitting a swept frequency sound wave to the warehousing material in the terminal warehouse by a sound wave emitting device.

[0029] Among them, the terminal warehouse refers to the warehouse at the end of the supply chain, which directly interfaces with the terminal demand and is used for temporarily storing the warehousing material to be distributed to the terminal. The swept frequency sound wave is used to excite the warehousing material to vibrate, and the internal or surface defect condition of the warehousing material is reflected through the vibration response and sound field change of the warehousing material. The frequency range is usually 20Hz-1MHz (covering audio to ultrasonic frequency band, which can be adjusted according to the material quality and detection requirements).

[0030] In some implementations, the sound wave emitting device can use a loudspeaker, an ultrasonic transducer, etc., and adjust the swept frequency rate (such as linear sweep or logarithmic sweep) through a control device to ensure that the sound wave covers the frequency interval in which the warehousing material may resonate.

[0031] It should be noted that the emission of the swept frequency sound wave should maintain stability to avoid interference with the vibration response collection of the warehousing material due to the vibration of the device itself or external noise.

[0032] For example, for metal material, a swept frequency sound wave of 50kHz-500kHz can be used to more easily excite vibration abnormalities at internal defects.

[0033] S2, collecting vibration velocity spectrum data and sound pressure gradient data of the warehousing material after receiving the swept frequency sound wave.

[0034] Among them, the vibration velocity spectrum data represents the distribution of the vibration velocity of each point on the surface of the warehousing material with respect to the frequency change, and the sound pressure gradient data represents the rate of change of the sound pressure value on the surface of the warehousing material with respect to the spatial position.

[0035] In some implementations, the vibration velocity spectrum data can be collected by a vibration sensor (Vibration Sensor, VS), such as a piezoelectric vibration sensor array; and the sound pressure gradient data can be collected by a microphone array (Microphone Array, MA), and the rate of change of the sound pressure with respect to the position is calculated through the spatial distribution of multiple microphones.

[0036] It should be noted that the layout of the vibration sensor and the microphone array should cover the key area of the material surface, and the sampling frequency of the two should be consistent (such as 1 kHz-100 kHz) to ensure the matching of the data in the time dimension.

[0037] S3, calculating the defect index of the warehousing material according to the vibration velocity spectrum data and the sound pressure gradient data.

[0038] The defect index is used to represent the possibility and severity of defects in the warehousing material, and the higher the value, the greater the risk of defects.

[0039] In some implementations, the defect index can be calculated by a data fusion algorithm, for example, the proportion of frequency components deviating from the normal range in the vibration velocity spectrum data and the area proportion of abnormal gradient regions in the sound pressure gradient data are weighted and summed, and the weight of the weighted sum can be dynamically adjusted according to the type of the material.

[0040] It should be noted that since the healthy object has uniform vibration velocity distribution at resonance, the sound field form is symmetrical. Once there is a crack, a stress concentration area will be formed at the crack tip, resulting in three abnormalities: local vibration velocity mutation, vibration mode phase shift, and sound pressure gradient distortion. Therefore, by using vibration velocity spectrum data and sound pressure gradient data to judge the defect condition of the warehousing material, the abnormal characteristics of the two kinds of data are combined, which can effectively reduce the misjudgment rate of single data detection.

[0041] For example, if 30% of the frequency components in the vibration velocity spectrum data deviate from the normal range, and 20% of the regions in the sound pressure gradient data have abnormal gradients, and the vibration weight is set to 0.6 and the sound pressure weight is set to 0.4, then the defect index can be calculated as 30% x 0.6 + 20% x 0.4 = 26%.

[0042] S4, when the defect index is greater than a preset threshold, marking the warehousing material as a defective material and sending a traceability detection instruction to the previous node warehouse.

[0043] The traceability detection instruction is used to start the hierarchical detection process of the previous node warehouse.

[0044] In some implementations, the hierarchical detection process can be: dividing the detection range according to the risk level of the material, such as full detection in high-risk areas, sampling detection in medium-risk areas, and regular patrol in low-risk areas.

[0045] It should be noted that the setting of the preset threshold needs to be determined by statistical analysis of historical data judged as defective materials by humans, combined with an acceptable misjudgment rate (such as 5%), for example, taking the mean or 90th percentile of the defect index of historical defective materials as the initial threshold, and then dynamically optimizing through actual application.

[0046] For example, if the average of the defect index of the historical defective materials is 40%, the preset threshold can be set to 35%. When the defect index of a certain material is detected to be 42%, the material is marked as a defective material and a traceability instruction is sent.

[0047] Based on the above technical solution, the material detection method based on big data provided in the application realizes accurate detection from the warehousing link of the terminal warehouse by exciting the material vibration through the swept frequency sound wave and collecting multi-dimensional response data, and combining the defect index quantification to evaluate the material quality; at the same time, the traceability instruction is sent to the upper node warehouse to promote the collaborative control of each node of the supply chain, effectively improving the comprehensiveness of the material quality detection and the traceability efficiency of the supply chain quality problem, and providing technical support for the stable operation of the supply chain.

[0048] In a possible implementation manner of the embodiment of the application, the S1 can be implemented through the following S101, S102 and S103, which are specifically described as follows: S101, determine the emission parameters of the swept frequency sound wave, including the frequency range, the swept frequency mode and the sampling frequency.

[0049] The frequency range needs to cover the frequency band that can excite the vibration response of the warehousing material, the swept frequency mode is used to control the frequency variation law, and the sampling frequency needs to be consistent with the equipment for collecting the vibration velocity spectrum data and the sound pressure gradient data subsequently.

[0050] In some implementation manners, the frequency range can be determined according to the material quality, such as 20 kHz-1 MHz for metal materials, and 20 Hz-20 kHz for non-metal materials, the swept frequency mode can adopt linear sweep (frequency changes linearly with time) or logarithmic sweep (frequency changes logarithmically with time), and the sampling frequency is determined by matching the parameters of the laser Doppler vibrometer and the distributed microphone array.

[0051] It should be noted that the sampling frequency needs to be higher than 2 times the highest frequency of the swept frequency sound wave, that is, the Nyquist sampling theorem is followed, so as to avoid signal aliasing affecting the accuracy of subsequent data.

[0052] For example, if a plastic material warehousing material is detected, the frequency range is determined to be 50 Hz-10 kHz, the linear sweep mode is adopted, and the sampling frequency of the laser Doppler vibrometer is 20 kHz, then the sampling frequency of the swept frequency sound wave is set to 20 kHz.

[0053] S102, input the determined emission parameters into the control module of the sound wave emission device to complete the device debugging.

[0054] The control module is used to receive the parameters and control the running state of the sound wave emission device to ensure that the emission parameters are accurately executed.

[0055] In some implementations, the control module can be implemented by a programmable logic controller (PLC) or a microprocessor, supporting real-time modification and storage of parameters, facilitating quick switching of parameters when detecting different batches of materials.

[0056] It should be noted that the output power stability of the sound wave emitting device needs to be detected during the debugging process to avoid inconsistent material vibration responses caused by power fluctuations.

[0057] For example, the control module sets the sweep rate to 1 kHz / s (i.e., the frequency increases by 1 kHz per second) and starts the power detection function to ensure that the output power fluctuation range does not exceed ±5%.

[0058] S103, controlling the sound wave emitting device to emit sweep frequency sound waves with set parameters to the incoming materials in the terminal warehouse.

[0059] During the emission process, it is necessary to ensure that the sound waves cover the key detection areas of the materials, such as the surface vulnerable parts and the structural stress concentration areas.

[0060] In some implementations, the installation position of the sound wave emitting device can be adjusted, such as being at an angle of 30°-60° with the surface of the material, or the number of emitting devices can be increased, such as forming an array arrangement with multiple emitting devices, to improve the excitation effect of the sound waves on the materials.

[0061] For example, for cuboid-shaped incoming materials, one sound wave emitting device is deployed on each of the two opposite sides, and sweep frequency sound waves are emitted synchronously to ensure that the entire material can be effectively excited.

[0062] Based on the above technical solutions, through the step-by-step operation of clear parameter setting, device debugging, and accurate emission, it is ensured that the sweep frequency sound waves can stably and effectively excite the vibration response of the incoming materials, laying a foundation for accurate collection of subsequent vibration velocity spectrum data and sound pressure gradient data, and improving the reliability of the pre-data of material defect detection.

[0063] In one possible implementation of the embodiments of the present application, S2 can be implemented by S201, S202, and S203 as follows, which are described in detail below: S201, deploying vibration velocity spectrum data acquisition equipment and sound pressure gradient data acquisition equipment, and aligning with the incoming material detection area of the terminal warehouse.

[0064] The vibration velocity spectrum data acquisition device adopts a laser Doppler vibrometer (LDV) for non-contact measurement of the vibration velocity of each point on the surface of the material; and the sound pressure gradient data acquisition device adopts a distributed microphone array (DMA) formed by a plurality of microphones arranged at a preset spatial interval for acquisition of the spatial distribution of the sound pressure around the material.

[0065] In some implementations, the laser Doppler vibrometer can be installed on a three-axis displacement platform, and the scanning range is adjusted through program control to ensure that the key areas of the surface of the material, such as the corners and the joints, are covered; the distributed microphone array can be fixed within a range of 1-3 meters around the material, arranged in a circular or rectangular shape, and the number of microphones is set according to the detection accuracy requirement, which can usually be 8-64.

[0066] It should be noted that the equipment deployment needs to avoid obstruction to ensure that the laser beam of the laser Doppler vibrometer is perpendicular or approximately perpendicular to the surface of the material, and the pickup direction of the microphone array is towards the material to reduce signal attenuation.

[0067] It should be noted that in the present application, the sound wave emitting device is used to cause forced vibration of the surface of the material through air or structural conduction. When the frequency of the sound wave matches the natural frequency of the material, resonance occurs, and the local stiffness change caused by the defect area causes abnormal vibration response. The "vibration velocity spectrum data" generated by the vibration of the material caused by the sound wave is obtained by the laser Doppler vibrometer, that is, when the laser irradiates the surface of the vibrating object, the frequency of the reflected light will be Doppler shifted due to the vibration of the object, and the vibration velocity of the object can be calculated by measuring this frequency shift.

[0068] For example, for a warehouse material with a surface area of 2m x 1m, the scanning step of the laser Doppler vibrometer is set to 5mm to cover the entire surface of the material; 16 microphones are arranged around the material to form a 32-channel distributed array with a spacing of 30cm between adjacent microphones.

[0069] S202, set the sampling parameters of the acquisition equipment, and perform time synchronization calibration.

[0070] The sampling parameters include the sampling frequency and the sampling duration, and the sampling frequency of the laser Doppler vibrometer and the distributed microphone array needs to be the same, and the sampling duration needs to be no shorter than the emission duration of the swept-frequency sound wave, so as to record the response of the material during the entire swept-frequency process.

[0071] In some implementations, a high-precision clock synchronization module, such as Global Positioning System (GPS) synchronization or IEEE 1588 Precision Time Protocol, is used to calibrate the time of the two devices to ensure that the timestamp error of the data acquisition does not exceed 10 microseconds.

[0072] It should be noted that the sampling frequency setting must match the highest frequency of the swept sound wave, usually 2-5 times the highest frequency of the swept wave, in order to satisfy the Nyquist sampling theorem and avoid signal aliasing.

[0073] For example, if the highest frequency of the swept sound wave is 1MHz, the sampling frequency is uniformly set to 2MHz, and the sampling duration is set to the sweep duration plus 2 seconds to ensure that the attenuated vibration signal after the sweep ends is included.

[0074] S203. While transmitting the sweeping sound wave, start the acquisition equipment to simultaneously acquire the vibration velocity spectrum data and sound pressure gradient data of the materials entering the warehouse.

[0075] In some implementations, the acquisition device can be connected to a data caching server to store the real-time acquired data in a time series, generating a data segment every second for easy subsequent verification and traceability.

[0076] It should be noted that unnecessary noise sources (such as broadcasts and fans) in the warehouse should be turned off during the data collection process, and the interference of environmental noise on the microphone array should be reduced by using soundproof covers or sound-absorbing materials to ensure that the signal-to-noise ratio of the sound pressure gradient data is not less than 20dB.

[0077] For example, when the sound wave emitting device starts emitting frequency sweeping sound waves, it triggers the synchronous start signal of the acquisition device. The laser Doppler vibrometer scans the surface of the material point by point with a step size of 5mm, and collects 1024 data samples at each point. The distributed microphone array synchronously records the sound pressure signal, with each microphone generating 2 million sampling points per second, and continues to collect until 1 second after the frequency sweep ends.

[0078] Based on the above technical solution, the accuracy, synchronization and integrity of vibration velocity spectrum data and sound pressure gradient data are ensured, providing reliable raw data support for subsequent calculation of the defect index, and improving the effectiveness of data correlation analysis.

[0079] In one possible implementation of the embodiments of this application, combined with Figure 2 ,like Figure 3 As shown, the above S3 can be implemented through the following S301, S302 and S303, which are explained in detail below: S301. Preprocess the vibration velocity spectrum data to obtain the preprocessed velocity spectrum data.

[0080] The pre-processing includes band-pass filtering and de-noising, aiming to eliminate irrelevant signals such as environmental noise and equipment interference, and retain effective vibration information related to the defects of the materials.

[0081] In some implementations, the band-pass filtering can use a Butterworth filter or a Chebyshev filter, and the filtering frequency band is set according to the resonance frequency range corresponding to the material quality, such as 10 kHz-500 kHz for metal materials; the de-noising can use wavelet transform de-noising, mean filtering or median filtering algorithm to reduce the influence of random noise on the data.

[0082] For example, the collected vibration velocity spectrum data is filtered by an 8th order Butterworth band-pass filter (passband 100 kHz-300 kHz), and then de-noised by wavelet transform (db4 wavelet basis is selected, and 3 layers are decomposed) to obtain smooth pre-processed velocity spectrum data v(f).

[0083] S302, extract the vibration abnormal area and the sound pressure abnormal area, and determine the intersection thereof as the abnormal vibration area.

[0084] The vibration abnormal area refers to a frequency band in the pre-processed velocity spectrum data that deviates from the reference spectrum by more than a preset threshold; the sound pressure abnormal area refers to a spatial region in the sound pressure gradient data whose gradient value exceeds a preset gradient threshold; the abnormal vibration area is the intersection of the two, which is used to focus on the area where defects may exist. The reference spectrum (RS) refers to the vibration velocity spectrum data of a standard material (without defects) of the same type, which can be obtained by taking the mean value after detecting a plurality of standard materials in advance.

[0085] In some implementations, the preset deviation threshold can be determined by statistics of the velocity spectrum deviation of historical standard materials and defective materials (such as 30%); and the preset gradient threshold can be set according to the spatial sound pressure gradient distribution range of the standard material of the same type (such as 1.2 times the maximum gradient value of the standard material).

[0086] It should be noted that the reference spectrum v ref (f) needs to be updated regularly, such as recalculated after detecting each batch of standard materials, to avoid reference deviation caused by slight changes in production process; and the extraction of the sound pressure abnormal area needs to be combined with the spatial coordinates of the material to ensure the accuracy of the area positioning.

[0087] For example, the reference spectrum v0(f) of a certain type of standard material has a vibration speed of 0.5 m / s in the 100 kHz frequency band, and the v(f) of a certain warehousing material after pretreatment is 0.8 m / s in the frequency band, with a deviation of 60% (more than 30% of the preset threshold), and the 100 kHz frequency band is marked as a vibration abnormal area; in the sound pressure gradient data, the gradient value of a certain 20 cm x 20 cm area on the surface of the material exceeds 5 Pa / m (the preset gradient threshold is 3 Pa / m), and the area is a sound pressure abnormal area, and the intersection of the two is the abnormal vibration area.

[0088] It should be noted that the intersection of the frequency band and the area is not a direct dimensional intersection, but refers to the coincidence and correlation of "vibration abnormalities corresponding to a specific abnormal frequency band" and "sound pressure abnormalities of a specific spatial area" on the material, that is, the spatial area simultaneously appears vibration abnormalities and sound pressure gradient abnormalities in the abnormal frequency band.

[0089] Specifically, the vibration abnormal area is defined from the frequency dimension - when the vibration speed spectrum data of a certain frequency band deviates from the reference spectrum by more than the preset threshold, it indicates that the vibration of the material at that frequency is abnormal; and the sound pressure abnormal area is defined from the spatial dimension - when the sound pressure gradient value of a certain spatial area exceeds the preset threshold, it indicates that the sound field distribution of the area is abnormal.

[0090] In actual detection, if the material has defects (such as cracks), it will cause local vibration abnormalities (corresponding to a certain abnormal frequency band) at a certain frequency, and the spatial area where the defect is located will cause sound pressure gradient distortion due to stress concentration (corresponding to a certain spatial sound pressure abnormal area). Therefore, the "intersection" refers to: in the abnormal frequency band, a certain spatial area on the surface of the material simultaneously appears vibration speed abnormalities and sound pressure gradient abnormalities, that is, the vibration and sound field of the area in the frequency band both exhibit abnormal characteristics.

[0091] That is, in the example scenario, the 100 kHz frequency band is a vibration abnormal area, indicating that the material vibrates abnormally at 100 kHz frequency; the 20 cm x 20 cm area is a sound pressure abnormal area, indicating that the sound pressure gradient of the area is abnormal. The intersection of the two is "in the 100 kHz frequency band, the 20 cm x 20 cm area simultaneously appears vibration speed abnormalities and sound pressure gradient abnormalities", so it can be determined that the area is an abnormal vibration area, that is, the area where the defect may exist.

[0092] S303, calculating the defect index of the warehousing material according to the defect index calculation formula.

[0093] The defect index calculation formula is used to quantify the correlation between the characteristics of the abnormal vibration area and the speed spectrum deviation degree, represent the severity of the material defect, and the formula involves the speed spectrum data v(f) after pretreatment, the reference spectrum v0(f), and the area A of the abnormal vibration area. ​

[0094] In some implementations, the calculation formula of the defect index DI is: ; wherein Ω represents the frequency range of the swept acoustic wave, A hs represents the area of the abnormal vibration zone, σ represents the standard deviation of the reference spectrum, v ref (f) represents the reference spectrum.

[0095] In some implementations, the standard deviation σ of the reference spectrum can include the following steps: First, collect the vibration velocity spectrum data of the same type of standard materials. In this application, the reference spectrum can be obtained by collecting the vibration velocity spectrum of at least 3 pieces of the same type of standard materials without defects (each piece of material is repeated 5-10 times, and the total sample size is ≥15 groups) by a laser Doppler vibration tester. For example, for a batch of standard metal parts, within the frequency range of 100 kHz-300 kHz, the surface of each material is scanned at a step of 5 mm, and the vibration velocity-frequency curve of each point is obtained.

[0096] Then calculate the standard deviation of a single frequency point. For each frequency point f, extract the vibration velocity values of all standard samples at this frequency {v 01 (f), v 02 (f), …, v 0n (f)} at this frequency, and calculate the standard deviation of a single frequency point according to the overall standard deviation formula: ; wherein μ(f) is the mean value of the frequency point, and n is the total number of standard samples.

[0097] Finally, construct a frequency-standard deviation mapping table. Associate σ(f) of each frequency point with the corresponding frequency f to form the standard deviation distribution of the reference spectrum, covering the full frequency range of the swept acoustic wave. This mapping table is used as a reference parameter for defect index calculation, to quantify the dispersion degree of the vibration velocity spectrum of the material to be tested and the standard spectrum.

[0098] It should be noted that the frequency range F in the formula needs to be consistent with the frequency range of the swept acoustic wave in S1, and the calculation of the area A needs to be based on the actual size of the material surface, such as through spatial coordinate conversion, to ensure that the numerical value has clear physical meaning. Moreover, the standard deviation σ needs to be calculated separately for each type of material to avoid confusion between different types of reference.

[0099] It should be noted that, since the application is based on the vibration velocity spectrum data and the sound pressure gradient data collected after the incoming warehouse materials receive the swept frequency sound wave, the vibration velocity spectrum data is pre-processed by band-pass filtering, denoising and the like, and then the pre-processed velocity spectrum data is compared with the vibration velocity spectrum data (i.e. the reference spectrum) of the same type of standard materials, and the frequency band with a deviation greater than the preset deviation threshold is selected as the vibration abnormal area. In this process, the reference spectrum is a special reference based on the vibration characteristics of the same type of defect-free materials. Even if the measured materials have uneven surfaces and uneven textures, the vibration rules of the defect-free state will remain consistent with the same type of standard materials. Therefore, by comparing the deviation from the reference spectrum, the vibration abnormalities caused by defects in the measured materials can be accurately captured, rather than being disturbed by the non-uniform characteristics of the materials.

[0100] Meanwhile, the application will analyze the collected sound pressure gradient data, and in combination with the preset gradient threshold set based on the spatial sound pressure gradient distribution range of the same type of standard materials, the spatial area with a sound pressure gradient value greater than the threshold is located as the sound pressure abnormal area. The preset gradient threshold is not a unified fixed standard, but is determined according to the sound pressure distribution characteristics of the same type of materials in a normal state, which can adapt to the natural sound pressure gradient differences of materials with different structures and different material distributions, and only identify the sound pressure gradient distortion caused by defects.

[0101] On this basis, the application determines the intersection of the vibration abnormal area and the sound pressure abnormal area as the abnormal vibration area, and in combination with the deviation of the pre-processed velocity spectrum data from the reference spectrum, the standard deviation of the reference spectrum, and the area of the abnormal vibration area, calculates the defect index through a quantitative formula to determine whether the materials have defects. The entire detection logic does not depend on whether the surface morphology of the materials is flat or whether the texture is uniform. As long as the characteristics of the same type of standard materials are used as a reference, the defect detection of various materials can be realized, whether it is a uniform material with a regular surface or a material with an irregular surface and uneven texture. As long as the reference data of the same type of standard materials can be obtained, the defects can be accurately identified by this method.

[0102] Based on the above technical solution, step S3 improves the signal quality through data preprocessing, accurately locates the possible defect area through the abnormal area, and then calculates the defect index through a quantitative formula, realizing the conversion from the original data to the quantitative defect degree, providing an objective and comparable index for subsequent defect judgment, and improving the accuracy and consistency of the material defect detection.

[0103] In one possible implementation of the embodiment of the application, in combination with Figure 2 As shown in Figure 4 S4 specifically includes the following S401 to S403: S401, compare the calculated defect index of the incoming material with a preset threshold value, if the defect index is greater than the preset threshold value, mark the incoming material as defective material.

[0104] The preset threshold value is a critical value for determining whether the material has defects, which is set based on the detection data of historical defective materials to ensure effective differentiation between normal materials and defective materials.

[0105] In some implementations, the preset threshold value can be determined by statistical analysis of the defect indexes of historical defective materials confirmed by manual review, for example, taking the 75th percentile of the defect indexes of historical defective materials as the initial threshold value, and subsequently dynamically adjusting according to the misjudgment rate in actual detection (such as controlling the misjudgment rate within 5%).

[0106] It should be noted that the preset threshold value needs to be set separately for each type of material. Due to the difference in physical properties, the defect index range of different materials and models is different, and cannot be used universally (for example, the threshold value of metal materials is usually higher than that of plastic materials).

[0107] For example, the average defect index of a certain type of metal material in the historical data is 60, and the 75th percentile is 70. After testing, the misjudgment rate is 3% when the threshold value is set to 65. Therefore, the preset threshold value of this type of material is determined to be 65. If the defect index of a certain incoming metal material is 72 (greater than 65), it is marked as defective material.

[0108] S402, generate a trace detection instruction according to the information of the defective material.

[0109] The trace detection instruction needs to contain key information sufficient to start the previous node warehouse grading detection process, and provide a basis for the risk area division and detection operation of the previous node.

[0110] In some implementations, the trace detection instruction can include the supplier code (S), production batch (L), defect index, and abnormal vibration area characteristics (such as area, corresponding frequency range) of the defective material. These information is obtained through the database of the material management system.

[0111] For example, the supplier code of a certain defective material is "S001", the production batch is "L202305", the defect index is 72, the abnormal vibration area is 0.04 m², and the corresponding frequency is 100 kHz-150 kHz. Therefore, the trace detection instruction explicitly contains this information, and then it is generated in a structured data format (such as JSON) for use by the previous node warehouse.

[0112] S403, after the previous node warehouse receives the instruction, perform risk area division, specifically including: (1) Screening associated materials set: According to the supplier code S and the production batch L of the defective material, the materials with the same supplier code and the production date within the preset time window (such as ± 15 days) in the last node database are screened out to form an associated material set.

[0113] (2) Calculate risk coefficient: For the ith material in the associated material set, the risk coefficient λi is calculated according to the formula i = α × δ i + β × e -kΔTi , where δ i is the process similarity of the material and the defective material, which is obtained by calculating the proportion of the number of the same production equipment used by the ith material and the defective material to the total number of production equipment of the defective material, ΔTi is the time difference of the production date of the ith associated material and the defective material, k is the attenuation coefficient, usually taking 0.1, and α and β represent the weight coefficients of each term.

[0114] (3) Divide risk area by spatial clustering: Obtain the three-dimensional model of the last node warehouse; create a tuple containing spatial coordinates, supplier code, production batch, and risk coefficient for each associated material; use a density clustering algorithm such as DBSCAN to perform spatial clustering on the tuples, and divide the materials with a spatial distance less than a preset neighborhood radius and a risk coefficient difference within a preset range into the same risk area.

[0115] It should be noted that the values of α and β can be determined by experimental experience. For example, for materials with high production process stability and equipment significantly affecting quality (such as precision mechanical parts), the weight of process similarity is higher, and α = 0.6 and β = 0.4 can be set; for materials with strong time effectiveness and large fluctuations with production time (such as electronic components), the weight of production date difference is higher, and α = 0.4 and β = 0.6 can be set. At the same time, α and β must satisfy α + β = 1 to ensure that the value range of the risk coefficient λ i is between 0 and 1, facilitating horizontal comparison of the risk levels of different materials. The preset neighborhood radius is set according to the warehouse storage density (such as 1-3 meters) to ensure that materials with similar physical locations are included in the same analysis unit; the risk coefficient difference range (such as ± 0.1) is used to screen materials with similar risk levels, so that the materials in the same area have similar quality risk characteristics.

[0116] S404, the last node warehouse performs a hierarchical detection process based on the risk area: (1) Calculate the average risk coefficient of all materials in the kth risk area = (λ1+ λ2+... + λ n ) / n, where n is the number of materials in the area, and the risk coefficient of non-associated materials is 0.

[0117] (2) According to λ kThe comparison result with the threshold T1, T2 (T1>T2) is graded detection: If >T1 (such as T1=0.6), marked as a first-level risk area, and all materials are fully detected, and the full detection method is as shown in the first aspect; If T2 ≤T1 (such as T2=0.3), marked as a second-level risk area, and sampling detection is performed at a first proportion (such as 30%), and if the defect proportion exceeds a preset proportion (such as 10%), the first-level detection is upgraded; If ≤T2, marked as a third-level risk area, and sampling detection is performed at a second proportion (such as 10%), and if the defect proportion exceeds a preset proportion, the second-level detection is upgraded.

[0118] In some implementations, T1 and T2 are preset grading thresholds, which are determined by statistics of defect rates of historical risk areas; for example, T1 is set to a value that makes the defect detection rate of the first-level area ≥95%; and the sampling proportions (including the first proportion and the second proportion) are adjusted according to the area size, and the larger the area is, the lower the proportion can be.

[0119] It should be noted that when the defect materials are detected by the previous node, a traceability instruction is sent to the upper node to form a multi-level linkage.

[0120] For example, the average risk coefficient of the risk area Z1 =0.5 (T1=0.6, T2=0.3), determined as a second-level risk area, 10 pieces are sampled at a proportion of 30%, and if 2 pieces are defective (a proportion of 20%>10%), Z1 is upgraded to a first-level area and full detection is performed.

[0121] Based on the above technical solution, step S4 realizes full-chain quality tracing from the terminal warehouse to the previous node through defect marking, traceability instruction generation, risk area division, and graded detection. Both the accurate marking ensures that the defective materials do not flow into the downstream, and the traceability instruction promotes the targeted detection of the previous node, reduces the cost of invalid detection, improves the synergy and efficiency of supply chain quality control, and provides support for rapid positioning and solving of full-chain quality problems.

[0122] It should be noted that the present application adopts the way of tracing from the terminal warehouse to the previous node to locate defects, in order to accurately trace the responsibility and distinguish whether the cause of the material defect is the original quality problem provided by the supplier or the damage caused in the supply chain circulation (such as transportation, handling, and storage).

[0123] Specifically, if the defect is caused by the quality problem of the supplier, the associated materials of the same production batch or similar process usually have common defects. By tracing back to the upper node through the terminal warehouse, screening the associated material set according to the supplier code and production batch, calculating the risk coefficient and dividing the risk area, it can be found that there are defects in the associated material set of multiple nodes. The traceability process will extend to the upper level one by one, and finally locate to the initial supplier node. If the defect is caused by transportation or manual handling, such damage usually has contingency and only exists in a small number of materials in a specific flow link. At this time, when tracing back to the upper node, the risk area division will find that the risk coefficient of the associated material set is generally low, and the number of defective materials in the hierarchical detection is small. The traceability process will be interrupted in a node warehouse and will not continue to extend upwards.

[0124] Therefore, the detection results of the terminal warehouse can reverse drive the upstream quality control through the traceability mechanism, and as the last checkpoint before the terminal, it can extend the quality guarantee to the front end of actual application and sales, which not only meets the actual needs of the terminal quality control of the supply chain, but also provides protection for the application and sales of products in the market.

[0125] The above describes the scheme of the embodiments of the present application mainly from the perspective of device implementation. It can be understood that, in order to implement the above functions, each device, for example, the material detection device based on big data, contains at least one of the corresponding hardware structure and software module for executing each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present application can be realized in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed in the form of hardware or computer software driven hardware depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0126] The embodiments of the present application can divide the functional units of the material detection device based on big data according to the above method examples. For example, each functional unit can be divided according to each function, or two or more functions can be integrated in one processing unit. The integrated unit can be realized in the form of hardware or software functional unit. It should be noted that the division of units in the embodiments of the present application is illustrative, and is only a logical function division. Actual implementation can have another division method.

[0127] In the case of integrated units, Figure 5A possible structural diagram of the big data-based material detection device (denoted as big data-based material detection device 50) involved in the above embodiments is shown, which includes a processing unit 501 and a communication unit 502, and can also include a storage unit 503. Figure 5 The structural diagram shown can be used to illustrate the structure of the big data-based material detection device involved in the above embodiments.

[0128] When Figure 5 When the structural diagram shown is used to illustrate the structure of the big data-based material detection device involved in the above embodiments, the processing unit 501 is used to control and manage the operation of the big data-based material detection device, the communication unit 502 is used for communication between the big data-based material detection device and other devices, and the storage unit 503 is used to store the program code and data of the big data-based material detection device.

[0129] For example, the communication unit 502 is configured to receive vibration velocity spectrum data and sound pressure gradient data, send a traceability detection instruction to the previous node warehouse, and interact with the sound wave emitting device and the data acquisition device.

[0130] The processing unit 501 is configured to control the sound wave emitting device to emit a sweep frequency sound wave to the warehoused material, process the collected vibration velocity spectrum data and sound pressure gradient data to calculate a defect index, determine whether the material is a defective material according to the defect index, and generate a traceability detection instruction when it is determined that the material is a defective material.

[0131] In a possible implementation, the processing unit 501 is further configured to preprocess the vibration velocity spectrum data, extract vibration abnormal zones and sound pressure abnormal zones and determine abnormal vibration zones, and divide risk areas and perform a hierarchical detection process according to a risk coefficient.

[0132] In a possible implementation, the communication unit 502 is further configured to receive a hierarchical detection result fed back by the previous node warehouse; and the processing unit 501 is further configured to adjust a preset threshold and a risk coefficient calculation parameter according to the feedback result, and optimize a subsequent detection process.

[0133] The storage unit 503 is configured to store reference spectrum data of the same type of standard material, defect material data of historical detection, a preset threshold, program code, and three-dimensional model information of each node warehouse, etc., to provide data support for data analysis and decision-making of the processing unit

[0134] The processing unit 501 can be a processor or a controller, and the communication unit 502 can be a communication interface, a transceiver, a transceiver, a transceiver circuit, a transceiver device, etc. The communication interface is collectively referred to, and can include one or more interfaces. The storage unit 503 can be a memory. When the big data-based material detection device 50 is a chip, the processing unit 501 can be a processor or a controller, and the communication unit 502 can be an input interface and / or an output interface, a pin or a circuit, etc. The storage unit 503 can be a storage unit (for example, a register, a cache, etc.) within the chip, or can be a storage unit (for example, a read-only memory (ROM), a random access memory (RAM), etc.) located outside the chip.

[0135] The communication unit can also be referred to as a transceiver unit. The antenna and control circuit with transceiver function in the big data-based material detection device 50 can be regarded as the communication unit 502 of the big data-based material detection device 50, and the processor with processing function can be regarded as the processing unit 501 of the big data-based material detection device 50. Optionally, the device for realizing the receiving function in the communication unit 502 can be regarded as a communication unit, and the communication unit is used to execute the receiving steps in the embodiments of the present application, and the communication unit can be a receiver, a receiver, a receiving circuit, etc. The device for realizing the sending function in the communication unit 502 can be regarded as a sending unit, and the sending unit is used to execute the sending steps in the embodiments of the present application, and the sending unit can be a transmitter, a sender, a sending circuit, etc.

[0136] Figure 5 The integrated units in the above embodiments can be stored in a computer readable storage medium if they are realized in the form of software function modules and sold or used as independent products. Based on such understanding, the technical solutions of the embodiments of the present application or the whole or part of the technical solutions that make contributions to the prior art can be embodied in the form of software products. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the method described in the embodiments of the present application. The storage medium for storing computer software products includes: a U disk, a mobile hard disk, a read-only memory, a random access memory, a magnetic disk or an optical disk, and various media that can store program codes.

[0137] Figure 5 The units in the above embodiments can also be referred to as modules, for example, the processing unit can be referred to as a processing module.

[0138] The embodiment of the present application further provides a hardware structure schematic diagram of the material detection device based on big data (denoted as a material detection device based on big data 60). Referring to Figure 6 The material detection device based on big data 60 comprises a processor 601, and optionally further comprises a memory 602 connected with the processor 601.

[0139] In the first possible implementation, referring to Figure 6 The material detection device based on big data 60 further comprises a transceiver 603. The processor 601, the memory 602 and the transceiver 603 are connected through a bus. The transceiver 603 is used for communicating with other devices or communication networks. Optionally, the transceiver 603 can comprise a transmitter and a receiver. The device for realizing the receiving function in the transceiver 603 can be regarded as a receiver, and the receiver is used for executing the receiving steps in the embodiment of the present application. The device for realizing the sending function in the transceiver 603 can be regarded as a transmitter, and the transmitter is used for executing the sending steps in the embodiment of the present application.

[0140] Based on the first possible implementation, Figure 6 The structure schematic diagram shown can be used for illustrating the structure of the material detection device based on big data involved in the above embodiment.

[0141] Among them, Figure 6 The system chip in the material detection device based on big data can also be illustrated. In this case, the actions performed by the above material detection device based on big data can be realized by the system chip, and the specific actions performed can be referred to the above, and will not be described here.

[0142] In the implementation process, each step in the method provided by the embodiment can be completed by the integrated logic circuit of hardware in the processor or the instructions in the form of software. The steps of the method disclosed by the embodiment of the present application can be directly embodied as the execution completed by the hardware processor, or executed by the combination of hardware and software modules in the processor.

[0143] The processor in the present application can include, but is not limited to, at least one of the following: a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a microcontroller unit (MCU), or an artificial intelligence processor, and the like, each of which is a computing device running software, and each of which can include one or more cores for executing software instructions to perform operations or processing. The processor can be a separate semiconductor chip, or can be integrated with other circuits as a semiconductor chip, for example, can be integrated with other circuits (such as coding and decoding circuits, hardware acceleration circuits, or various bus and interface circuits) to form a SoC (system on chip), or can be integrated as a built-in processor in an ASIC. The ASIC integrated with the processor can be packaged separately or packaged together with other circuits. In addition to including cores for executing software instructions to perform operations or processing, the processor can further include necessary hardware accelerators, such as field programmable gate arrays (FPGAs), PLDs (programmable logic devices), or logic circuits implementing special logic operations.

[0144] The memory in the embodiments of the present application can include at least one of the following types: a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM). In some scenarios, the memory can also be a compact disc read-only memory (CD-ROM) or other optical disk storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited thereto.

[0145] The embodiments of the present application also provide a computer readable storage medium including instructions, which, when executed on a computer, cause the computer to perform any of the above methods.

[0146] The embodiments of the present application also provide a computer program product including instructions, which, when executed on a computer, cause the computer to perform any of the above methods.

[0147] The embodiment of the present application further provides a chip, comprising a processor and an interface circuit, the interface circuit being coupled with the processor, the processor being used to run computer programs or instructions to realize the method described above, and the interface circuit being used to communicate with other modules outside the chip.

[0148] In the above embodiments, the implementation can be achieved by software, hardware, firmware or any combination thereof, entirely or partially. When implemented by software, the implementation can be achieved in the form of a computer program product, entirely or partially. The computer program product comprises one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the flow or function described in the embodiments of the present application is generated, entirely or partially. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable devices. The computer instructions can be stored in a computer readable storage medium or transferred from one computer readable storage medium to another, for example, the computer instructions can be transferred from one website, computer, server or data center to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (such as infrared, wireless, microwave, etc.) mode. The computer readable storage medium can be any available medium that can be accessed by a computer or data storage device including one or more servers, data centers, etc. integrated with the medium. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state disk (solid state disk, SSD)) and the like.

[0149] Although the present application is described herein in conjunction with various embodiments, other variations of the disclosed embodiments can be understood and implemented by those skilled in the art through viewing the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "one" does not exclude a plurality. A single processor or other unit can implement several functions listed in the claims. Some measures are described in mutually different dependent claims, but this does not mean that these measures cannot be combined to produce good results.

[0150] Although the present application has been described in connection with certain specific features and embodiments thereof, it is to be understood that it is intended to cover all modifications and variations of this application which are within the scope of the appended claims and their equivalents. Accordingly, the description and drawings are to be regarded as illustrative in nature and not as restrictive. It is intended that all such modifications and variations are included within the scope of the present application as defined by the following claims and their equivalents.

Claims

1. A big data-based material detection method, characterized in that, The application relates to a method for detecting defects of goods in a warehouse, and belongs to the technical field of warehouse management. The method comprises the following steps: a sweep frequency acoustic wave is emitted to the goods in the warehouse by using an acoustic wave emitting device; vibration velocity spectrum data and sound pressure gradient data of the goods after the goods receive the sweep frequency acoustic wave are collected; the vibration velocity spectrum data represent the distribution of vibration velocity of each point on the surface of the goods with frequency variation, and the sound pressure gradient data represent a set of variation rates of sound pressure values at the spatial position points of the surface of the goods in the sound field; a defect index of the goods is calculated according to the vibration velocity spectrum data and the sound pressure gradient data; 2. The method of claim 1, wherein, when the defect index is greater than a preset threshold value, the goods are marked as defective goods, and a trace detection instruction is sent to a previous node warehouse; the trace detection instruction is used for instructing the node warehouse to execute a hierarchical detection process. The method for calculating the defect index of the target goods according to the vibration velocity spectrum data and the sound pressure gradient data comprises the following steps: extracting a reference spectrum v ref an abnormal frequency band whose deviation value of (f) is greater than a preset deviation threshold value, to obtain a vibration abnormal area; the reference spectrum represents a vibration velocity spectrum data of a standard material of the same type. band-pass filtering and denoising processing are performed on the vibration velocity spectrum data to obtain preprocessed velocity spectrum data v (f); a spatial region with a gradient value greater than a preset gradient threshold value in the sound pressure gradient data is extracted to obtain a sound pressure abnormal region; The defect index of the target material is obtained according to a defect index calculation formula; the defect index calculation formula is used to represent the relationship between the pre-processed velocity spectrum data v(f), the reference spectrum v ref (f), and the abnormal vibration region.

3. The method of claim 2, wherein, The defect index calculation formula satisfies: ; wherein DI represents a defect index, Ω represents a frequency range of a sweep sound wave, A hs represents an area of an abnormal vibration region, σ represents a standard deviation of a reference spectrum, and f represents a frequency of the sweep sound wave.

4. The method of claim 1, wherein, an intersection of the vibration abnormal region and the sound pressure abnormal region is marked as an abnormal vibration region; the hierarchical detection process comprises the following steps: a supplier code S and a production batch L of the defective goods are obtained; a related goods set Q in a database of the previous node warehouse is screened; the related goods set has the same supplier code as the defective goods, and the production date of the production batch is within a preset time window; a risk coefficient of the goods in the related goods set is calculated; a clustering analysis is performed on the risk coefficient according to the spatial position of the previous node warehouse to obtain a plurality of risk regions; 5. The method of claim 4, wherein, The calculation formula of the risk coefficient is: λ i = α × δ i + β × e -kΔTi ; wherein, λ i represents the risk coefficient of the i th associated material in the associated material concentration, δ i represents the process similarity of the i th associated material and the defective material, which is obtained by calculating the proportion of the number of the same production equipment used by the i th material and the defective material in the total number of production equipment of the defective material, Δ Ti represents the time difference of the production date of the i th associated material and the defective material, k represents the attenuation coefficient, and α and β represent the weight coefficients of each term.

6. The big data-based material detection method according to claim 4, characterized in that, a hierarchical detection is performed on the plurality of risk regions. The clustering analysis comprises the following steps: Based on the three-dimensional model, a material space mapping is performed, creating a tuple for each material containing spatial coordinates (x, y, z), a supplier code, a production batch, and a risk coefficient λ; where, if the material w is associated with the set Q, then the risk coefficient λ w = 0; The tuples of the material point set are spatially clustered by using a spatial clustering algorithm to obtain a risk region set {Z k |k=1,2,…,K} wherein Z k represents the kth risk region, and K represents the number of risk regions.

7. The method of claim 5, wherein, a three-dimensional model of the previous node warehouse is obtained; calculating an average risk factor for all associated resources within the kth risk zone ; When When T1, the risk area is marked as a first risk area, and a first detection is performed, which includes: sequentially emitting a sweep frequency wave to all goods in the first risk area, calculating a defect index of each good in the first risk area according to the vibration velocity spectrum data and the sound pressure gradient data of the goods; when the defect index is greater than a preset threshold, marking the target good as a defective good; When T2 < T1 ≤ T1, the risk region is marked as a second-level risk region, and a second-level detection is performed, the second-level detection comprising: the hierarchical detection on the plurality of risk regions comprises the following steps: all the goods in a secondary risk region are randomly sampled at a first proportion to obtain first sampling goods; sweep frequency acoustic waves are emitted to the first sampling goods in sequence, and a defect index of each sampling good is calculated according to the vibration velocity spectrum data and the sound pressure gradient data of the goods; when the defect index is greater than a preset threshold value, the target goods are marked as defective goods; When when T2, marking the risk region as a third risk region and performing a third detection including: if the proportion of the defective goods in the first sampling goods exceeds a preset proportion, the secondary risk region is marked as a primary risk region, and the primary detection is performed; all the goods in a tertiary risk region are randomly sampled at a second proportion to obtain second sampling goods; and the second proportion is less than the first proportion; sweep frequency acoustic waves are emitted to the second sampling goods in sequence, and a defect index of each sampling good is calculated according to the vibration velocity spectrum data and the sound pressure gradient data of the goods; when the defect index is greater than a preset threshold value, the target goods are marked as defective goods; 8. The big data-based material detection method according to claim 7, characterized in that, if the proportion of the defective goods in the second sampling goods exceeds a preset proportion, the tertiary risk region is marked as a secondary risk region, and the secondary detection is performed. When the hierarchical detection is performed in the previous node warehouse: if new defective goods are detected in the hierarchical detection on the plurality of risk regions, a trace detection instruction is sent to a higher node warehouse of the previous node warehouse, and the risk region division and the hierarchical detection process are performed in the higher node warehouse.

9. The method of claim 1, wherein, The vibration velocity spectrum data is acquired by a laser Doppler vibrometer, the sound pressure gradient data is collected by a distributed microphone array, and the sampling frequency of the vibration velocity spectrum data and the sound pressure gradient data is the same. 10.A big data based material detection system, characterized in that, Comprise: An acoustic wave emitting module, a data collecting module and a data analyzing module; wherein, The acoustic wave emitting module is configured to emit a sweep acoustic wave to the incoming goods of the terminal warehouse by using an acoustic wave emitting device; The data collecting module is configured to collect vibration velocity spectrum data and sound pressure gradient data of the incoming goods after the incoming goods receive the sweep acoustic wave; the vibration velocity spectrum data represents the distribution of the vibration velocity of each point on the surface of the goods with respect to the frequency, and the sound pressure gradient data represents the rate of change of the sound pressure value on the surface of the goods with respect to the spatial position; The data analyzing module is configured to calculate a defect index of the target goods according to the vibration velocity spectrum data and the sound pressure gradient data, mark the incoming goods as defective goods when the defect index is greater than a preset threshold, and send a traceability detection instruction to the previous node warehouse; the traceability detection instruction is configured to start a hierarchical detection process of the previous node warehouse.

Images