Detection method, device, terminal and storage medium

Abstract

The present disclosure provides a detection method, device, terminal and storage medium, wherein the detection method comprises: obtaining target test data corresponding to a to-be-tested device in a radio frequency parameter index test production line; wherein the radio frequency parameter index test production line comprises a comprehensive test device and a calibration device, and the to-be-tested device is the comprehensive test device or the calibration device of the radio frequency parameter index test production line; determining a first feature sequence according to the target test data; and determining the running state of the to-be-tested device according to a running state recognition model and the first feature sequence. In the present disclosure, the detection of the comprehensive test device or the calibration device can be realized by using the running state recognition model, which can avoid relying on the experience of technical personnel, reduce the labor cost, facilitate the timely discovery of problems, and improve the detection efficiency.

Description

Technical Field

[0001] The present disclosure relates to the field of radio frequency parameter index testing, and in particular to a detection method, device, terminal and storage medium. Background Art

[0002] Products with wireless RF functionality typically undergo factory-tested RF parameter tests for both the transmitter and receiver to verify performance. RF parameter testing is typically performed on the RF parameter test line, but factors such as cable loss, poor fixture and connector contact on integrated test instruments, and improper antenna matching can often affect qualified products, causing them to fail testing. In severe cases, this can lead to batch rework, impacting production capacity and yield.

[0003] Due to the numerous factors that affect RF parameter index testing, manual troubleshooting of problems with integrated test equipment and calibration equipment in the RF parameter index testing production line is still the mainstream method during the current RF parameter index testing process. This leads to problems such as delayed problem discovery and heavy reliance on the work experience of technicians, resulting in low detection efficiency and high labor costs for integrated test equipment and calibration equipment. Summary of the Invention

[0004] To overcome the problems existing in the related art, the present disclosure provides a detection method, device, terminal and storage medium.

[0005] According to a first aspect of an embodiment of the present disclosure, a detection method is provided, which is applied to a terminal. The detection method includes:

[0006] Obtain target test data corresponding to the device under test in a radio frequency parameter index test production line; wherein the radio frequency parameter index test production line includes comprehensive testing equipment and calibration equipment, and the device under test is the comprehensive testing equipment or the calibration equipment of the radio frequency parameter index test production line;

[0007] Determining a first feature sequence according to the target test data;

[0008] The operating state of the device under test is determined according to the operating state recognition model and the first feature sequence.

[0009] Optionally, the operating state recognition model is trained in the following manner:

[0010] Constructing a first training sample set; wherein the first training sample set includes a plurality of first training sample pairs, the first training sample pairs include mutually corresponding first input samples and first output samples, the first output samples include state identification information samples representing the operating state of the device under test, the first input samples include first feature sequence samples, the first feature sequence samples are determined based on first test data samples, and the first test data samples include test data under the operating state of the device under test represented by the state identification information samples;

[0011] The first original model is trained according to the first training sample set to obtain the operating state recognition model.

[0012] Optionally, the first characteristic sequence sample is determined by at least one of the following methods:

[0013] performing normalization processing on the first test data sample to determine the first feature sequence sample;

[0014] Performing feature learning processing on the first test data sample to determine the first feature sequence sample having a derived feature; wherein the derived feature refers to a feature characterizing the operating state of the device under test.

[0015] Optionally, determining the operating state of the device under test according to the operating state recognition model and the first feature sequence includes:

[0016] The first feature sequence is input into the operating state recognition model. If the operating state recognition model outputs first state identification information indicating that the operating state of the device under test is normal, it is determined that the operating state of the device under test is normal.

[0017] Optionally, determining the operating state of the device under test according to the operating state model and the first feature sequence includes:

[0018] inputting the first feature sequence into the operating state recognition model, and if the operating state recognition model outputs second state identification information indicating that the operating state of the device under test is abnormal, determining a second feature sequence according to the target test data;

[0019] The fault type of the device under test is determined according to the fault type judgment model and the second feature sequence.

[0020] Optionally, determining the fault type of the device under test according to the fault type judgment model and the second feature sequence includes:

[0021] The second feature sequence is input into the fault type judgment model to determine the fault type of the device under test.

[0022] Optionally, the fault type judgment model is trained in the following manner:

[0023] Constructing a second training sample set; wherein the second training sample set includes a plurality of second training sample pairs, the second training sample pairs include mutually corresponding second input samples and second output samples, the second output samples include fault identification information samples representing the fault type of the device under test, the second input samples include second feature sequence samples, the second feature sequence samples are determined based on second test data samples, and the second test data samples include test data under the fault type of the device under test represented by the fault identification information samples;

[0024] The second original model is trained according to the second training sample set to obtain the fault type judgment model.

[0025] Optionally, the second characteristic sequence sample is determined by at least one of the following methods:

[0026] Determine an associated feature sequence representing a fault type based on the second test data sample, and determine the second feature sequence sample based on a feature sequence obtained by normalizing the second test data sample and the associated feature sequence;

[0027] An associated feature sequence characterizing a fault type is determined based on the second test data sample, and the second feature sequence sample is determined based on a feature sequence obtained by performing feature learning processing on the second test data sample and the associated feature sequence.

[0028] According to a second aspect of an embodiment of the present disclosure, a detection device is provided, applied to a terminal, the detection device including:

[0029] An acquisition module is configured to acquire target test data corresponding to a device under test in a radio frequency parameter index test production line; wherein the radio frequency parameter index test production line includes integrated test equipment and calibration equipment, and the device under test is the integrated test equipment or the calibration equipment of the radio frequency parameter index test production line;

[0030] A determination module, configured to determine a first feature sequence according to the target test data;

[0031] It is also used to determine the operating state of the device under test according to the operating state recognition model and the first feature sequence.

[0032] Optionally, the operating state recognition model is trained in the following manner:

[0033] Constructing a first training sample set; wherein the first training sample set includes a plurality of first training sample pairs, the first training sample pairs include mutually corresponding first input samples and first output samples, the first output samples include state identification information samples representing the operating state of the device under test, the first input samples include first feature sequence samples, the first feature sequence samples are determined based on first test data samples, and the first test data samples include test data under the operating state of the device under test represented by the state identification information samples;

[0034] The first original model is trained according to the first training sample set to obtain the operating state recognition model.

[0035] Optionally, the first characteristic sequence sample is determined by at least one of the following methods:

[0036] performing normalization processing on the first test data sample to determine the first feature sequence sample;

[0037] Performing feature learning processing on the first test data sample to determine the first feature sequence sample having a derived feature; wherein the derived feature refers to a feature characterizing the operating state of the device under test.

[0038] Optionally, the determining module is configured to:

[0039] The first feature sequence is input into the operating state recognition model. If the operating state recognition model outputs first state identification information indicating that the operating state of the device under test is normal, it is determined that the operating state of the device under test is normal.

[0040] Optionally, the determining module is configured to:

[0041] inputting the first feature sequence into the operating state recognition model, and if the operating state recognition model outputs second state identification information indicating that the operating state of the device under test is abnormal, determining a second feature sequence according to the target test data;

[0042] The fault type of the device under test is determined according to the fault type judgment model and the second feature sequence.

[0043] Optionally, the determining module is configured to:

[0044] The second feature sequence is input into the fault type judgment model to determine the fault type of the device under test.

[0045] Optionally, the fault type judgment model is trained in the following manner:

[0046] Constructing a second training sample set; wherein the second training sample set includes a plurality of second training sample pairs, the second training sample pairs include mutually corresponding second input samples and second output samples, the second output samples include fault identification information samples representing the fault type of the device under test, the second input samples include second feature sequence samples, the second feature sequence samples are determined based on second test data samples, and the second test data samples include test data under the fault type of the device under test represented by the fault identification information samples;

[0047] The second original model is trained according to the second training sample set to obtain the fault type judgment model.

[0048] Optionally, the second characteristic sequence sample is determined by at least one of the following methods:

[0049] Determine an associated feature sequence representing a fault type based on the second test data sample, and determine the second feature sequence sample based on a feature sequence obtained by normalizing the second test data sample and the associated feature sequence;

[0050] An associated feature sequence characterizing a fault type is determined based on the second test data sample, and the second feature sequence sample is determined based on a feature sequence obtained by performing feature learning processing on the second test data sample and the associated feature sequence.

[0051] According to a third aspect of an embodiment of the present disclosure, a terminal is provided, comprising:

[0052] processor;

[0053] a memory for storing instructions executable by the processor;

[0054] Wherein, the processor is configured to execute the detection method as described in the first aspect.

[0055] According to a fourth aspect of an embodiment of the present disclosure, a non-transitory computer-readable storage medium is provided. When instructions in the storage medium are executed by a processor of a terminal, the terminal is enabled to perform the detection method as described in the first aspect.

[0056] The technical solution provided by the embodiments of the present disclosure may include the following beneficial effects: In the present disclosure, an operating status recognition model can be used to realize the detection of comprehensive testing equipment or calibration equipment, which can avoid relying on the experience of technical personnel, reduce labor costs, facilitate timely discovery of problems, and improve detection efficiency.

[0057] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0059] Figure 1 is a flow chart of a detection method according to an exemplary embodiment.

[0060] Figure 2 is a block diagram of a detection device according to an exemplary embodiment.

[0061] Figure 3 is a block diagram of a terminal according to an exemplary embodiment. DETAILED DESCRIPTION

[0062] Exemplary embodiments are described in detail herein, with examples illustrated in the accompanying drawings. When the following description refers to the drawings, identical numerals in different figures represent identical or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present disclosure. Rather, they are merely examples of apparatus and methods of some embodiments of the present disclosure.

[0063] The present disclosure provides a detection method for a terminal. This method utilizes an operational status recognition model to detect integrated testing equipment or calibration equipment. This method avoids reliance on technician experience, reduces labor costs, facilitates timely problem detection, and improves detection efficiency.

[0064] In an exemplary embodiment, a detection method is provided, which is applied to a terminal. The detection method is used to detect problems with integrated testing equipment or calibration equipment in a radio frequency parameter index test production line. Figure 1 As shown, the method may include:

[0065] S110. Obtain target test data corresponding to the device under test in a radio frequency parameter index test production line; wherein the radio frequency parameter index test production line includes integrated test equipment and calibration equipment, and the device under test is the integrated test equipment or the calibration equipment of the radio frequency parameter index test production line;

[0066] S120, determining a first feature sequence according to target test data;

[0067] S130: Determine the operating state of the device under test according to the operating state recognition model and the first feature sequence.

[0068] In step S110, the target test data refers to test data related to determining the operating status of the device under test in the RF parameter index test production line. The target test data includes test data of the device under test itself, and may also include test data of other devices in the RF parameter index test production line, without limitation.

[0069] It should be noted that the RF parameter index test production line may include more than one comprehensive test device and more than one calibration device. The device to be tested may be any comprehensive test device or any calibration device in the RF parameter index test production line. The target test data may include the test data of the device to be tested itself, and may also include test data from other comprehensive test devices and calibration devices.

[0070] The types of test data may include test item name, test value, test spec (specification), test item CPK (process capability index), loss value (attenuation value of RF line / probe to test values ​​such as RF power and RF receiving signal strength), standard, frequency, transmission signal power detection data, receiving signal function detection data, error detection rate, antenna number, etc., without limitation.

[0071] In this step, the target test data can be manually input into the terminal so that the terminal obtains the target test data; the target test data can also be sent to the terminal by the RF parameter index test production line so that the terminal obtains the target test data; of course, the terminal can also obtain the target test data through other methods, which are not limited to this.

[0072] In step S120, after the terminal obtains the target test data, it can process it to obtain a first feature sequence that can be used by the operating state recognition model. The first feature sequence can be determined by at least one of the following methods:

[0073] Method 1: normalize the target test data to obtain a first feature sequence; wherein the first feature sequence may be composed of numbers;

[0074] In method 2, feature learning processing may be performed on the target test data to obtain a first feature sequence having derived features; wherein the derived features refer to features that can characterize the operating status of the device under test.

[0075] It should be noted that in computer science, derived features refer to new features obtained by performing feature learning on raw data. Sometimes, derived features can better reflect the relationships between data features. In this embodiment, the first feature sequence obtained through feature learning can better reflect the relationships between data features, allowing for more accurate determination of the operating state corresponding to the target test data based on the first feature sequence.

[0076] In some embodiments,

[0077] The product for performing the radio frequency parameter index detection can be a radio frequency product (for example, a mobile phone), and a production line for performing the radio frequency parameter index detection on the radio frequency product can include multiple comprehensive testing devices and calibration devices. The target test data can include multiple types of test data. The test data of the radio frequency product that passes through the same calibration device and the same comprehensive testing device can be recorded as a first group of data. The test data of the radio frequency product that passes through the same comprehensive testing device and different calibration devices can be recorded as a second group of data. The test data of the radio frequency product that passes through the same calibration device and different comprehensive testing devices can be recorded as a third group of data. Then, a CPK calculation model of each test item is established based on the first group of data, and the CPK index and the standard deviation are calculated, which are recorded as a first group of features. A CPK calculation model of each test item is established based on the second group of data, and the CPK index of different calibration devices, the average value of the CPK index, and the standard deviation are calculated, which are recorded as a second group of features. A CPK calculation model of each test item is established based on the third group of data, and the CPK index of different comprehensive testing devices, the average value of the CPK index, and the standard deviation are calculated, which are recorded as a third group of features. Finally, a first feature sequence is formed based on the first group of features, the second group of features, and the third group of features.

[0078] In this embodiment, the first feature sequence can better reflect the running state of the to-be-tested device, thereby improving the reliability of the method.

[0079] In step S130, the first feature sequence can be input into the running state recognition model, and the running state recognition model can output state identification information that can represent the running state of the to-be-tested device. The state identification information can be a digital label or other types of information, which are not limited.

[0080] The state identification information can include first state identification information and second state identification information. The first state identification information represents that the running state of the to-be-tested device is normal. The second state identification information represents that the running state of the to-be-tested device is abnormal. For example, the first state identification information is a first digital label "1", representing that the running state of the to-be-tested device is "normal", and the second state identification information is a second digital label "0", representing that the running state of the to-be-tested device is "abnormal".

[0081] In this step, the first feature sequence can be input into the running state recognition model. If the running state recognition model outputs the first state identification information, it is determined that the running state of the to-be-tested device is normal. If the running state recognition model outputs the second state identification information, it is determined that the running state of the to-be-tested device is abnormal, that is, the to-be-tested device has a fault.

[0082] In this method, the running state recognition model can be used to detect the comprehensive testing device or the calibration device, which can avoid relying on the experience of technicians, reduce the labor cost, and facilitate the timely discovery of problems, thereby improving the detection efficiency.

[0083] In an exemplary embodiment, a detection method is provided, which is applied to a terminal. In this method, the operating state recognition model can be trained in the following manner:

[0084] S210. Construct a first training sample set; wherein the first training sample set includes a plurality of first training sample pairs, the first training sample pairs including mutually corresponding first input samples and first output samples, the first output samples including state identification information samples representing the operating state of the device under test, the first input samples including first feature sequence samples, the first feature sequence samples determining, based on the first test data samples, that the first test data samples include test data under the operating state of the device under test represented by the operating state identification information samples;

[0085] S220: Train the first original model according to the first training sample set to obtain an operating status recognition model.

[0086] In step S210, the first characteristic sequence sample may be determined by referring to the first characteristic sequence. The first characteristic sequence sample may be determined by at least one of the following methods:

[0087] Method 1: Normalizing the first test data sample to determine the first feature sequence sample;

[0088] Method 2: Feature learning processing can be performed on the first test data sample to determine a first feature sequence sample with derived features.

[0089] Of course, the first characteristic sequence sample may also be determined by other methods, which are not limited thereto.

[0090] It should be noted that the first feature sequence sample is used as a first input sample to train the operation status recognition model, and the first feature sequence is used as input information to be input into the operation status recognition model to determine the operation status of the device under test.

[0091] Among them, the first test data sample and its corresponding operating status can be obtained from the Internet, or according to the work experience or experimental records of technical personnel, or through IOT (Internet of Things) technology. The first test data sample and its corresponding operating status can be obtained in other ways, and there is no limitation on this.

[0092] In some embodiments,

[0093] Each set of first test data samples may include test data representing the operating status of the integrated test equipment. In this embodiment, the first test data samples may be normalized to obtain first characteristic sequence samples corresponding to the first test data samples, and the first characteristic sequence samples may be composed of numbers.

[0094] In addition, the operating state corresponding to the first test data sample is marked to obtain a corresponding state identification information sample, wherein the digital label "0" can be used to mark the "abnormal" operating state, and the digital label "1" can be used to mark the "normal" operating state.

[0095] Finally, the first feature sequence samples and the state identification information samples corresponding to each other are determined as a set of first training sample pairs, wherein the first feature sequence samples are used as first input samples, and the state identification information samples are used as first output samples.

[0096] Through the above method, a first training sample set consisting of a plurality of first training sample pairs in running states can be constructed.

[0097] In some embodiments,

[0098] Each set of first test data samples may include test data representing the operating status of the integrated test equipment. The operating status corresponding to the first test data samples is labeled to obtain corresponding status identification information samples. A digital label of "0" may be used to indicate an "abnormal" operating status, while a digital label of "1" may be used to indicate a "normal" operating status.

[0099] Furthermore, feature learning processing may be performed on the first test data sample to obtain a first feature sequence sample having a derived feature, and the first feature sequence sample may be composed of numbers.

[0100] In this embodiment, the product used for radio frequency parameter index detection can be recorded as a radio frequency product (such as a mobile phone). The first test data sample may include multiple test data. Among them, the test data of the radio frequency product that has passed the same calibration equipment and the same comprehensive testing equipment can be recorded as the first sample data; the test data of the radio frequency product that has passed the same comprehensive testing equipment but different calibration equipment can be recorded as the second sample data; the test data of the radio frequency product that has passed the same calibration equipment but different comprehensive testing equipment can be recorded as the third sample data. Then, based on the first sample data, a CPK calculation model for each test item is established, and the CPK index and standard deviation are calculated, which are recorded as the first set of features; based on the second sample data, a CPK calculation model for each test item is established, and the CPK index of different calibration equipment, the average value and standard deviation of the CPK index are calculated, which are recorded as the second set of features; based on the third sample data, a CPK calculation model for each test item is established, and the CPK index of different comprehensive testing equipment, the average value and standard deviation of the CPK index are calculated, which are recorded as the third set of features.

[0101] Then, based on the first, second, and third sets of features, a set of feature sequences representing the operating status of the integrated test equipment is formed, recorded as first feature sequence samples. Finally, the corresponding first feature sequence samples and state identification information samples are determined as a set of first training sample pairs. The first feature sequence samples serve as the first input samples, and the state identification information samples serve as the first output samples.

[0102] Through the above method, a first training sample set consisting of a plurality of first training sample pairs in running states can be constructed.

[0103] In step S220, the first original model may be a classification model, such as a decision tree model, a support vector machine model, etc., which is not limited here. In this step, the first original model may be trained using the first training sample set to obtain a final operating state recognition model.

[0104] It should be noted that when the first feature sequence sample for training the operating state recognition model is determined using Method 1, when the operating state recognition model is used, the first feature sequence used as input must also be determined using Method 1. When the first feature sequence sample for training the operating state recognition model is determined using Method 2, when the operating state recognition model is used, the first feature sequence used as input must also be determined using Method 2.

[0105] In this method, a running status recognition model is trained on a large amount of data to determine its reliability. This model can then be used to test integrated testing or calibration equipment, eliminating reliance on technician experience, reducing labor costs, facilitating timely problem detection, and improving testing efficiency.

[0106] In an exemplary embodiment, a detection method is provided for use in a terminal. In the method, determining the operating state of the device under test based on an operating state recognition model and a first feature sequence may include:

[0107] S310. Input the first feature sequence into the operation state recognition model. If the operation state recognition model outputs second state identification information indicating that the operation state of the device under test is abnormal, determine the second feature sequence according to the target test data, and determine the fault type of the device under test according to the fault type judgment model and the second feature sequence.

[0108] Among them, since the operating state recognition model outputs the second state identification information, and the second state identification information indicates that the operating state of the device under test is abnormal, it means that the device under test may have a fault, and the fault type of the device under test can be further determined.

[0109] A second feature sequence can be determined based on the target test data, and then the fault type of the device under test can be determined based on the second feature sequence. For example, the second feature sequence can be input into a fault type determination model, which can then output corresponding fault identification information, which is used to characterize the fault type of the device under test.

[0110] Fault types may include, but are not limited to, probe pressure failure, test value deviation from the upper limit, test value deviation from the lower limit, and line loss. Fault identification information may be a text label corresponding to one of the aforementioned fault types, used to indicate that the device under test has the fault indicated by the text label. For example, if the fault identification information is a text label reading "probe pressure failure," it indicates that the device under test has a probe pressure failure.

[0111] It should be noted that fault identification information can also be a digital label. For example, the digital label "01" indicates that the device under test has a probe pressing fault, the digital label "02" indicates that the device under test has a test value fault that is biased toward the upper limit, the digital label "03" indicates that the device under test has a test value fault that is biased toward the lower limit, and the digital label "04" indicates that the device under test has a line loss fault. This is not limited to this.

[0112] The correlation features representing the fault type of the device under test can be first determined based on the target test data, and then the second feature sequence can be determined based on the correlation features and the first feature sequence. The correlation features can be determined through feature learning processing, which will not be described in detail.

[0113] In some embodiments,

[0114] Products used for RF parameter index testing can be denoted as RF products (e.g., mobile phones). The production line used for RF product RF parameter index testing can include multiple integrated testing equipment and calibration equipment. Target test data can include multiple types of test data. Test data for RF products that have undergone the same calibration equipment and the same integrated testing equipment can be denoted as a first set of data; test data for RF products that have undergone the same integrated testing equipment but different calibration equipment can be denoted as a second set of data; and test data for RF products that have undergone the same calibration equipment but different integrated testing equipment can be denoted as a third set of data. Then, based on the first set of data, a CPK calculation model is established for each test item, and the CPK index and standard deviation are calculated, which are denoted as a first set of features. Based on the second set of data, a CPK calculation model is established for each test item, and the CPK index, average value, and standard deviation of the CPK index for different calibration equipment are calculated, which are denoted as a second set of features. Based on the third set of data, a CPK calculation model is established for each test item, and the CPK index, average value, and standard deviation of the CPK index for different integrated testing equipment are calculated, which are denoted as a third set of features. The first, second, and third sets of features can constitute a first feature sequence.

[0115] In addition, according to the experience of technical personnel, the absolute value of the difference between the mean test value of the equipment to be tested and the upper limit is strongly correlated with line loss faults, and the absolute value of the difference between the mean test value of the equipment to be tested and the lower limit is strongly correlated with line loss faults. Therefore, the absolute value of the difference between the mean test value of the equipment to be tested and the upper and lower limits can be used as a correlation feature.

[0116] That is, in this embodiment, the difference between the mean and twice the standard deviation of the CPK indicators of different devices under test, and the difference between the mean and twice the standard deviation of the CPK indicators of different calibration devices, can be used as the lower limit of the CPK indicator; and the sum of the mean and twice the standard deviation of the CPK indicators of different devices under test, and the sum of the mean and twice the standard deviation of the CPK indicators of different calibration devices, can be used as the upper limit of the CPK indicator. Then, the mean and standard deviation of the line loss data of different devices under test are calculated. Then, the difference between the mean and three times the standard deviation of the line loss data of the device under test can be used as the lower limit of the line loss of the device under test, and the sum of the mean and three times the standard deviation of the line loss data of the device under test can be used as the upper limit of the line loss of the device under test. Then, the absolute value of the difference between the mean of the test value of the device under test and the upper and lower limits can be calculated.

[0117] Finally, the upper limit of the CPK indicator, the lower limit of the CPK indicator, the mean and standard deviation of the line loss data of different devices under test, the lower limit of the line loss of the device under test, the upper limit of the line loss of the device under test, and the absolute value of the difference between the mean of the test value of the device under test and the upper and lower limits can be combined into a set of feature sequences that characterize the fault type of the device under test, recorded as an associated feature sequence, and then the above associated feature sequence is combined with the first feature sequence to form a second feature sequence.

[0118] In this implementation, the second characteristic sequence can better reflect the fault type of the device under test, thereby improving the reliability of the method.

[0119] In this method, the operating status identification model and the fault type judgment model can be used to realize the detection of comprehensive testing equipment or calibration equipment, so as to determine the fault type more quickly and accurately. This can avoid relying on the experience of technicians, reduce labor costs, facilitate timely detection of problems, and improve detection efficiency.

[0120] In an exemplary embodiment, a detection method is provided for use in a terminal. In this method, a fault type determination model can be trained in the following manner:

[0121] S410: Construct a second training sample set; wherein the second training sample set includes a plurality of second training sample pairs, each of the second training sample pairs including a second input sample and a second output sample corresponding to each other, the second output sample including a fault identification information sample representing a fault type of the device under test, the second input sample including a second feature sequence sample, the second feature sequence sample being determined based on a second test data sample, the second test data sample including test data under the fault type of the device under test represented by the fault identification information sample;

[0122] S420: Train the second original model according to the second training sample set to obtain a fault type judgment model.

[0123] In step S410, the second characteristic sequence sample may be determined by referring to the second characteristic sequence. The second characteristic sequence sample may be determined by at least one of the following methods:

[0124] Mode a, determining a correlation feature sequence representing the fault type based on the second test data sample, and determining a second feature sequence sample based on a feature sequence obtained by normalizing the second test data sample and the correlation feature sequence;

[0125] Mode b: determining an associated feature sequence representing the fault type based on the second test data sample, and determining a second feature sequence sample based on a feature sequence obtained by performing feature learning processing on the second test data sample and the associated feature sequence.

[0126] Of course, the second characteristic sequence sample may also be determined by other methods, which are not limited thereto.

[0127] It should be noted that the second feature sequence sample is used as a second input sample to train the fault type judgment model, and the second feature sequence is used as input information to be input into the fault type judgment model to determine the fault type of the device under test.

[0128] The method for obtaining the second test data sample may refer to the first test data sample, and will not be described in detail.

[0129] In method a, the method of normalizing the second test data sample to obtain the feature sequence is the same as the method of normalizing the first test data sample to obtain the first feature sequence sample, which is not described in detail. The associated feature sequence can be determined by feature learning processing, which is not described in detail either.

[0130] In mode b, the method of performing feature learning processing on the second test data sample to obtain the feature sequence is the same as the method of performing feature learning processing on the first test data sample to obtain the first feature sequence sample, which is not described in detail. The associated feature sequence can be determined by feature learning processing, which is not described in detail either.

[0131] In some embodiments,

[0132] Each set of second test data samples may include test data indicating abnormal operation of the integrated test equipment. The fault types corresponding to the second test data samples are marked to obtain corresponding fault identification information samples.

[0133] Furthermore, feature learning processing may be performed on the second test data sample to obtain a feature sequence having derived features. The feature sequence sample may be composed of numbers.

[0134] In this embodiment, the product used for radio frequency parameter index detection can be recorded as a radio frequency product (such as a mobile phone). The first test data sample may include multiple test data. Among them, the test data of the radio frequency product that has passed the same calibration equipment and the same comprehensive testing equipment can be recorded as the first sample data; the test data of the radio frequency product that has passed the same comprehensive testing equipment but different calibration equipment can be recorded as the second sample data; the test data of the radio frequency product that has passed the same calibration equipment but different comprehensive testing equipment can be recorded as the second sample data. Then, based on the first sample data, a CPK calculation model for each test item is established, and the CPK index and standard deviation are calculated, which are recorded as the first set of features; based on the second sample data, a CPK calculation model for each test item is established, and the CPK index of different calibration equipment, the average value and standard deviation of the CPK index are calculated, which are recorded as the second set of features; based on the third sample data, a CPK calculation model for each test item is established, and the CPK index of different comprehensive testing equipment, the average value and standard deviation of the CPK index are calculated, which are recorded as the third set of features.

[0135] In addition, according to the experience of technical personnel, the absolute value of the difference between the mean test value of the equipment to be tested and the upper limit is strongly correlated with line loss faults, and the absolute value of the difference between the mean test value of the equipment to be tested and the lower limit is strongly correlated with line loss faults. Therefore, the absolute value of the difference between the mean test value of the equipment to be tested and the upper and lower limits can be used as a correlation feature.

[0136] In this embodiment, the difference between the mean and twice the standard deviation of the CPK indicators of different devices under test, and the difference between the mean and twice the standard deviation of the CPK indicators of different calibration devices, can be used as the lower limit of the CPK indicator; and the sum of the mean and twice the standard deviation of the CPK indicators of different devices under test, and the sum of the mean and twice the standard deviation of the CPK indicators of different calibration devices, can be used as the upper limit of the CPK indicator. Then, the mean and standard deviation of the line loss data of different devices under test are calculated. Then, the difference between the mean and three times the standard deviation of the line loss data of the device under test can be used as the lower limit of the line loss of the device under test, and the sum of the mean and three times the standard deviation of the line loss data of the device under test can be used as the upper limit of the line loss of the device under test. Then, the absolute value of the difference between the mean of the test value of the device under test and the upper and lower limits can be calculated.

[0137] Then, the upper limit of the CPK indicator, the lower limit of the CPK indicator, the mean and standard deviation of the line loss data of different devices under test, the lower limit of the line loss of the device under test, the upper limit of the line loss of the device under test, and the absolute value of the difference between the mean of the test value of the device under test and the upper and lower limits can be combined into a set of characteristic sequences that characterize the fault type of the device under test, which is recorded as an associated characteristic sequence.

[0138] Then, based on the first, second and third sets of features, a set of feature sequences characterizing the operating status of the comprehensive testing equipment is formed. Finally, the above feature sequences are combined with the associated feature sequences to form the final second feature sequence sample.

[0139] Finally, the second feature sequence samples and the fault identification information samples corresponding to each other are determined as a set of second training sample pairs, wherein the second feature sequence samples serve as second input samples and the fault identification information samples serve as second output samples.

[0140] Through the above method, a second training sample set consisting of second training sample pairs under multiple fault types can be constructed.

[0141] In step S420, the second original model can be similar to the first original model. For example, the second original model can be a classification model, such as a decision tree model, a support vector machine model, etc., without limitation. In this step, the second original model can be trained using the second training sample set to obtain a final operating state recognition model.

[0142] It should be noted that when the second feature sequence samples for training the fault type judgment model are determined using method a, the second feature sequence used as input must also be determined using method a when the fault type judgment model is used. When the second feature sequence samples for training the fault type judgment model are determined using method b, the second feature sequence used as input must also be determined using method b when the fault type judgment model is used.

[0143] In this method, a fault type determination model is trained on a large amount of data to determine its reliability. This model can then be used to detect fault types in integrated testing or calibration equipment. This eliminates reliance on technicians' experience, reduces labor costs, facilitates timely problem detection, and improves detection efficiency.

[0144] In an exemplary embodiment, a detection device is provided, which is applied to a terminal. The detection device is used to implement the above detection method. For example, referring to Figure 2 As shown, the device may include an acquisition module 101 and a determination module 102. In the process of implementing the above method, the device

[0145] An acquisition module 101 is configured to acquire target test data corresponding to a device under test (DUT) in a radio frequency parameter index test production line. The radio frequency parameter index test production line includes integrated test equipment and calibration equipment, and the DUT is either the integrated test equipment or the calibration equipment of the radio frequency parameter index test production line.

[0146] A determination module 102 is configured to determine a first feature sequence based on target test data;

[0147] It is also used to determine the operating state of the device under test according to the operating state identification model and the first feature sequence.

[0148] In an exemplary embodiment, a detection device is provided, which is applied to a terminal. In the device, the operating state recognition model is trained in the following manner:

[0149] Constructing a first training sample set; wherein the first training sample set includes a plurality of first training sample pairs, the first training sample pairs including mutually corresponding first input samples and first output samples, the first output samples including state identification information samples representing the operating state of the device under test, the first input samples including first feature sequence samples, the first feature sequence samples being determined based on first test data samples, the first test data samples including test data under the operating state of the device under test represented by the state identification information samples;

[0150] The first original model is trained according to the first training sample set to obtain an operating state recognition model.

[0151] In an exemplary embodiment, a detection device is provided, which is applied to a terminal. In the device, the first characteristic sequence sample is determined by at least one of the following methods:

[0152] Performing normalization processing on the first test data sample to determine a first feature sequence sample;

[0153] A feature learning process is performed on the first test data sample to determine a first feature sequence sample having a derived feature; wherein the derived feature refers to a feature that characterizes the operating state of the device under test.

[0154] In an exemplary embodiment, a detection device is provided, which is applied to a terminal. Figure 2 As shown, in the device, the determination module 102 is used to:

[0155] The first feature sequence is input into the operation state recognition model. If the operation state recognition model outputs first state identification information indicating that the operation state of the device under test is normal, it is determined that the operation state of the device under test is normal.

[0156] In an exemplary embodiment, a detection device is provided, which is applied to a terminal. Figure 2As shown, in the device, the determination module 102 is used to:

[0157] Inputting the first feature sequence into the operating state recognition model, and if the operating state recognition model outputs second state identification information indicating that the operating state of the device under test is abnormal, determining the second feature sequence according to the target test data;

[0158] The fault type of the device under test is determined according to the fault type judgment model and the second characteristic sequence.

[0159] In an exemplary embodiment, a detection device is provided, which is applied to a terminal. Figure 2 As shown, in the device, the determination module 102 is used to:

[0160] The second characteristic sequence is input into the fault type judgment model to determine the fault type of the device under test.

[0161] In an exemplary embodiment, a detection device is provided for use in a terminal. In the device, a fault type determination model is trained in the following manner:

[0162] Constructing a second training sample set; wherein the second training sample set includes a plurality of second training sample pairs, the second training sample pairs include mutually corresponding second input samples and second output samples, the second output samples include fault identification information samples representing the fault type of the device under test, the second input samples include second feature sequence samples, the second feature sequence samples are determined based on second test data samples, and the second test data samples include test data under the fault type of the device under test represented by the fault identification information samples;

[0163] The second original model is trained according to the second training sample set to obtain a fault type judgment model.

[0164] In an exemplary embodiment, a detection device is provided, which is applied to a terminal. In the device, the second characteristic sequence sample is determined by at least one of the following methods:

[0165] Determine a correlation feature sequence representing the fault type according to the second test data sample, and determine a second feature sequence sample according to the feature sequence obtained by normalizing the second test data sample and the correlation feature sequence;

[0166] An associated feature sequence characterizing the fault type is determined according to the second test data sample, and a second feature sequence sample is determined according to a feature sequence obtained by performing feature learning processing on the second test data sample and the associated feature sequence.

[0167] In an exemplary embodiment, a terminal is provided, such as a desktop computer, a laptop computer, a tablet computer, an all-in-one computer, a mobile phone, etc., without limitation thereto.

[0168] refer to Figure 3 As shown, terminal 400 may include one or more of the following components: a processing component 402 , a memory 404 , a power component 406 , a multimedia component 408 , an audio component 410 , an input / output (I / O) interface 412 , a sensor component 414 , and a communication component 416 .

[0169] Processing component 402 generally controls the overall operation of device 400, such as operations associated with display, phone calls, data communications, camera operation, and recording operations. Processing component 402 may include one or more processors 420 to execute instructions to perform all or part of the steps of the above-described method. In addition, processing component 402 may include one or more modules to facilitate interaction between processing component 402 and other components. For example, processing component 402 may include a multimedia module to facilitate interaction between multimedia component 408 and processing component 402.

[0170] The memory 404 is configured to store various types of data to support operations on the device 400. Examples of such data include instructions for any application or method operating on the device 400, contact data, phone book data, messages, pictures, videos, etc. The memory 404 can be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0171] The power component 406 provides power to the various components of the device 400. The power component 406 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the device 400.

[0172] The multimedia component 408 includes a screen that provides an output interface between the device 400 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive input instructions from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor can not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 408 includes a front camera and / or a rear camera. When the terminal 400 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera can receive external multimedia data. Each front camera and rear camera can be a fixed optical lens system or have a focal length and optical zoom capability.

[0173] The audio component 410 is configured to output and / or input audio instructions. For example, the audio component 410 includes a microphone (MIC), which is configured to receive external audio instructions when the device 400 is in an operating mode, such as a call mode, a recording mode, and a speech understanding mode. The received audio instructions can be further stored in the memory 404 or sent via the communication component 416. In some embodiments, the audio component 410 also includes a speaker for outputting audio instructions.

[0174] I / O interface 412 provides an interface between processing component 402 and peripheral interface modules, such as a keyboard, click wheel, buttons, etc. These buttons may include but are not limited to: a home button, volume buttons, a start button, and a lock button.

[0175] The sensor assembly 414 includes one or more sensors for providing various aspects of status assessment for the terminal 400. For example, the sensor assembly 414 can detect the open / closed state of the terminal 400, the relative positioning of components, such as the display and keypad of the terminal 400. The sensor assembly 414 can also detect changes in the position of the device 400 or a component of the terminal 400, the presence or absence of user contact with the device 400, the orientation or acceleration / deceleration of the device 400, and temperature changes of the device 400. The sensor assembly 414 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 414 may also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 414 may also include an accelerometer, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0176] The communication component 416 is configured to facilitate wired or wireless communication between the device 400 and other devices. The device 700 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 416 receives broadcast instructions or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 416 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0177] In an exemplary embodiment, the device 400 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above methods.

[0178] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as memory 404 including instructions. The instructions can be executed by processor 420 of device 400 to perform the above-described method. For example, the non-transitory computer-readable storage medium can be ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device. When the instructions in the storage medium are executed by the processor of the terminal, the terminal is able to perform the method in the above-described embodiment.

[0179] Other embodiments of the present disclosure will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the present disclosure being indicated by the claims.

[0180] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes can be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A detection method, applied to a terminal, characterized in that: The detection method comprises: Obtain target test data corresponding to the device under test in a radio frequency parameter index test production line; wherein the radio frequency parameter index test production line includes comprehensive testing equipment and calibration equipment, and the device under test is the comprehensive testing equipment or the calibration equipment of the radio frequency parameter index test production line; Determining a first feature sequence according to the target test data; determining the operating state of the device under test according to the operating state recognition model and the first feature sequence; The step of determining the operating state of the device under test according to the operating state recognition model and the first feature sequence includes: inputting the first feature sequence into the operating state recognition model, and if the operating state recognition model outputs second state identification information indicating that the operating state of the device under test is abnormal, determining a second feature sequence according to the target test data; Determining the fault type of the device under test according to the fault type judgment model and the second feature sequence; Wherein, determining the second characteristic sequence according to the target test data includes: Determine, based on the target test data, a correlation feature sequence characterizing a fault type of the device under test, wherein the correlation feature sequence includes an absolute value of a difference between a mean value of a test value of the device under test and an upper limit and a lower limit; The second feature sequence is determined according to the associated feature sequence and the first feature sequence.

2. The detection method according to claim 1, wherein The operating state recognition model is trained in the following way: Constructing a first training sample set; wherein the first training sample set includes a plurality of first training sample pairs, the first training sample pairs include mutually corresponding first input samples and first output samples, the first output samples include state identification information samples representing the operating state of the device under test, the first input samples include first feature sequence samples, the first feature sequence samples are determined based on first test data samples, and the first test data samples include test data under the operating state of the device under test represented by the state identification information samples; The first original model is trained according to the first training sample set to obtain the operating state recognition model.

3. The detection method according to claim 2, characterized in that The first characteristic sequence sample is determined by at least one of the following methods: performing normalization processing on the first test data sample to determine the first feature sequence sample; Performing feature learning processing on the first test data sample to determine the first feature sequence sample having a derived feature; wherein the derived feature refers to a feature characterizing the operating state of the device under test.

4. The detection method according to any one of claims 1 to 3, characterized in that The step of determining the operating state of the device under test according to the operating state recognition model and the first feature sequence includes: The first feature sequence is input into the operating state recognition model. If the operating state recognition model outputs first state identification information indicating that the operating state of the device under test is normal, it is determined that the operating state of the device under test is normal.

5. The detection method according to claim 1, wherein The determining the fault type of the device under test according to the fault type judgment model and the second feature sequence includes: The second feature sequence is input into the fault type judgment model to determine the fault type of the device under test.

6. The detection method according to claim 5, characterized in that The fault type judgment model is trained in the following way: Constructing a second training sample set; wherein the second training sample set includes a plurality of second training sample pairs, the second training sample pairs include mutually corresponding second input samples and second output samples, the second output samples include fault identification information samples representing the fault type of the device under test, the second input samples include second feature sequence samples, the second feature sequence samples are determined based on second test data samples, and the second test data samples include test data under the fault type of the device under test represented by the fault identification information samples; The second original model is trained according to the second training sample set to obtain the fault type judgment model.

7. The detection method according to claim 6, characterized in that The second characteristic sequence sample is determined by at least one of the following methods: Determine an associated feature sequence representing a fault type based on the second test data sample, and determine the second feature sequence sample based on a feature sequence obtained by normalizing the second test data sample and the associated feature sequence; An associated feature sequence characterizing a fault type is determined based on the second test data sample, and the second feature sequence sample is determined based on a feature sequence obtained by performing feature learning processing on the second test data sample and the associated feature sequence.

8. A detection device, applied to a terminal, characterized in that: The detection device comprises: An acquisition module is configured to acquire target test data corresponding to a device under test in a radio frequency parameter index test production line; wherein the radio frequency parameter index test production line includes integrated test equipment and calibration equipment, and the device under test is the integrated test equipment or the calibration equipment of the radio frequency parameter index test production line; A determination module, configured to determine a first feature sequence according to the target test data; further configured to determine the operating state of the device under test according to the operating state recognition model and the first feature sequence; The determining module is further configured to: inputting the first feature sequence into the operating state recognition model, and if the operating state recognition model outputs second state identification information indicating that the operating state of the device under test is abnormal, determining a second feature sequence according to the target test data; Determining the fault type of the device under test according to the fault type judgment model and the second feature sequence; Wherein, determining the second characteristic sequence according to the target test data includes: Determine, based on the target test data, a correlation feature sequence characterizing a fault type of the device under test, wherein the correlation feature sequence includes an absolute value of a difference between a mean value of a test value of the device under test and an upper limit and a lower limit; The second feature sequence is determined according to the associated feature sequence and the first feature sequence.

9. The detection device according to claim 8, characterized in that The operating state recognition model is trained in the following way: Constructing a first training sample set; wherein the first training sample set includes a plurality of first training sample pairs, the first training sample pairs include mutually corresponding first input samples and first output samples, the first output samples include state identification information samples representing the operating state of the device under test, the first input samples include first feature sequence samples, the first feature sequence samples are determined based on first test data samples, and the first test data samples include test data under the operating state of the device under test represented by the state identification information samples; The first original model is trained according to the first training sample set to obtain the operating state recognition model.

10. The detection device according to claim 9, characterized in that: The first characteristic sequence sample is determined by at least one of the following methods: performing normalization processing on the first test data sample to determine the first feature sequence sample; Performing feature learning processing on the first test data sample to determine the first feature sequence sample having a derived feature; wherein the derived feature refers to a feature characterizing the operating state of the device under test.

11. The detection device according to any one of claims 8 to 10, characterized in that: The determining module is configured to: The first feature sequence is input into the operating state recognition model. If the operating state recognition model outputs first state identification information indicating that the operating state of the device under test is normal, it is determined that the operating state of the device under test is normal.

12. The detection device according to claim 8, characterized in that The determining module is configured to: The second feature sequence is input into the fault type judgment model to determine the fault type of the device under test.

13. The detection device according to claim 12, characterized in that: The fault type judgment model is trained in the following way: Constructing a second training sample set; wherein the second training sample set includes a plurality of second training sample pairs, the second training sample pairs include mutually corresponding second input samples and second output samples, the second output samples include fault identification information samples representing the fault type of the device under test, the second input samples include second feature sequence samples, the second feature sequence samples are determined based on second test data samples, and the second test data samples include test data under the fault type of the device under test represented by the fault identification information samples; The second original model is trained according to the second training sample set to obtain the fault type judgment model.

14. The detection device according to claim 13, characterized in that: The second characteristic sequence sample is determined by at least one of the following methods: Determine an associated feature sequence representing a fault type based on the second test data sample, and determine the second feature sequence sample based on a feature sequence obtained by normalizing the second test data sample and the associated feature sequence; An associated feature sequence characterizing a fault type is determined based on the second test data sample, and the second feature sequence sample is determined based on a feature sequence obtained by performing feature learning processing on the second test data sample and the associated feature sequence.

15. A terminal, characterized in that: The terminal includes: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the detection method according to any one of claims 1 to 7.

16. A non-transitory computer-readable storage medium, characterized in that When the instructions in the storage medium are executed by a processor of a terminal, the terminal is enabled to execute the detection method according to any one of claims 1 to 7.

Images