Optical module automatic detection system and method based on numerical control robot

Abstract

The application relates to the technical field of automatic detection of optical modules, in particular to an optical module automatic detection system and method based on a numerical control robot, which specifically comprises the following steps: connecting judgment factors are constructed by analyzing the distribution of abnormal points in optical signal intensity, optical signal extinction ratio, Q factor and current data, and the data fluctuation near each abnormal point; a connection risk index during optical module detection is constructed by analyzing abnormal data changes of the current and using a t-test algorithm to reflect statistical significance differences between the current data and other parameters; the connection quality of the optical module is monitored in real time through the connection risk index; when the connection is seriously poor, a control signal is sent through a control module to instruct a six-axis robot to re-plug the optical module, so that detection errors caused by connection problems are reduced, the optical module is tested in the best state, and the accuracy of performance judgment of the optical module is improved.

Description

Technical Field

[0001] This application relates to the field of automatic detection technology for optical modules, specifically to an automatic detection system and method for optical modules based on CNC robots. Background Technology

[0002] The trend towards miniaturization, high speed, low power consumption, and low cost of optical modules has made them play an increasingly important role in modern data centers, significantly impacting data transmission efficiency and energy consumption reduction. During the automatic detection of optical modules by CNC robots, the mechanical structure of the robotic arm may loosen due to long-term operation and high-intensity working environments. Furthermore, environmental factors such as confined spaces and dim lighting can increase errors during positioning and insertion / removal, leading to poor contact between the optical module and the detection equipment.

[0003] However, whether it is poor contact or a performance problem of the optical module itself, it may cause abnormal parameters of the optical module during testing. For example, key performance indicators such as light signal intensity, extinction ratio and eye diagram may become abnormal. The abnormal parameters caused by the two are similar. In addition, during the testing process, the positioning and insertion / removal errors of the robotic arm increase, making it more difficult for existing technology to distinguish between poor contact and performance problems. This can easily lead to good products being judged as defective products, thereby reducing the accuracy of automatic testing of optical modules. Summary of the Invention

[0004] To address the aforementioned technical problems, the purpose of this application is to provide an automatic detection system and method for optical modules based on CNC robots. The specific technical solution adopted is as follows:

[0005] In a first aspect, embodiments of this application provide an automatic detection method for optical modules based on CNC robots, the method comprising the following steps:

[0006] The optical module collects various parameters at different times during detection and assembles these parameters into sequences; among these parameters is current.

[0007] Anomalies are identified in each sequence using an anomaly detection algorithm. Anomalies are assessed based on the anomalies in each sequence. The degree of anomaly is then determined for each anomaly in each sequence. Based on the distribution of anomalies in each sequence and the fluctuations near each anomaly, and combined with the degree of anomaly, connection determination factors for various parameters are constructed.

[0008] Based on the correlation between the current sequence and the other sequences, and combined with the corresponding connection determination factors, a comprehensive influence factor for the parameters of the other sequences is constructed. Combined with the connection determination factor of the current, a connection risk index is constructed for optical module detection.

[0009] The connection risk index is used to identify connection problems in optical modules.

[0010] In one embodiment, the various parameters further include: optical signal intensity, optical signal extinction ratio, and Q factor of the output eye diagram.

[0011] In one embodiment, the process of obtaining the anomaly degree of each anomaly point is as follows:

[0012] Calculate the mean of all elements in each sequence; calculate the difference between each outlier in each sequence and the mean, and denote it as the first difference;

[0013] If the first difference of each outlier in each sequence is greater than or equal to the preset outlier threshold, then the outlier degree of each outlier in each sequence is set to 0; otherwise, the outlier degree of each outlier in each sequence is set to the first difference.

[0014] In one embodiment, the process of obtaining the connection determination factor is as follows:

[0015] Each sequence is used as input to the time series statistical model, and the output is the conditional variance of each data point at the time of collection in each sequence; the product of the standard deviation of the collection time of all outliers in each sequence and the conditional variance of each outlier in each sequence is calculated and denoted as the first product.

[0016] The connection determination factor for each parameter is determined based on the first product and the degree of abnormality of each outlier point in each sequence.

[0017] In one embodiment, the process for determining the connection determination factors of each parameter is as follows:

[0018] Calculate the result of the exponential function with the natural constant as the base and the negative of the first product as the exponent; calculate the product of the calculation result and the degree of anomalousness of each outlier in each sequence, and record it as the second product; use the mean of the second products of all outliers in each sequence as the connection determination factor of the corresponding parameters of each sequence.

[0019] In one embodiment, the process of obtaining the comprehensive impact factor is as follows:

[0020] For each sequence other than the current sequence, the remaining sequences and the current sequence are used as inputs to the t-test algorithm, and the output is the P-value between the corresponding parameters of the remaining sequences and the current. Based on the P-value and the connection determination factor of each parameter other than the current, the comprehensive influence factor of each parameter is obtained, wherein the comprehensive influence factor of each parameter is positively correlated with the connection determination factor of each parameter and negatively correlated with the P-value.

[0021] In one embodiment, the process for obtaining the comprehensive influence factor of the remaining parameters is as follows:

[0022] The exponential function with the natural constant as the base and the negative of the P value as the exponent is denoted as the first exponential function. The product of the first exponential function and the connection determination factor of the remaining parameters is taken as the comprehensive influence factor of the remaining parameters.

[0023] In one embodiment, the expression for the connection risk index during optical module detection is:

[0024] In the formula, B is the connection risk index during optical module detection. This is the connection determination factor for the current. This represents the average of the combined influence factors of all parameters except current. This is the normalization function.

[0025] In one embodiment, the identification of poor connectivity of the optical module based on the connectivity risk index specifically involves:

[0026] If the connection risk index is greater than or equal to the preset connection failure threshold, the optical module has a connection failure problem, and the judgment result is fed back to the PLC control system for adjustment; otherwise, the optical module connection is normal.

[0027] Secondly, embodiments of this application also provide an automatic detection system for optical modules based on CNC robots, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of any of the methods described above.

[0028] The embodiments of this application have at least the following beneficial effects:

[0029] This application proposes an automatic optical module inspection system and method based on CNC robots. Addressing the confusion between optical module connection performance and its inherent performance, the system analyzes the distribution of anomalies in optical signal intensity, extinction ratio, Q-factor, and current data, as well as data fluctuations near these anomalies, to construct a connection judgment factor. The degree of anomaly at each anomaly point filters for optical module detection parameter anomalies caused by connection performance issues, preventing the misclassification of high-performance optical modules as defective. Analyzing abnormal current data directly reflects the impact of connection defects on current. A t-test algorithm is used to reflect the statistical significance difference between current data and other parameters, constructing a connection risk index for optical module inspection. This quantifies the impact of connection defects on optical module parameters, improving the accuracy of optical module performance judgment. Real-time monitoring of optical module connection quality is achieved by monitoring connection risks and setting thresholds. When connection defects are severe, a control signal is issued via the control module, instructing a six-axis robot to re-insert the optical module, reducing detection errors caused by connection problems and ensuring the optical module is tested in optimal condition. Attached Figure Description

[0030] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] Figure 1 A flowchart illustrating the steps of an automatic detection method for optical modules based on a CNC robot, provided in one embodiment of this application;

[0032] Figure 2 This is a schematic diagram of the automatic optical module detection device.

[0033] Figure 3 This is a schematic diagram of the anomaly detection results for the intensity sequence. Detailed Implementation

[0034] To further illustrate the technical means and effects adopted by this application to achieve the intended inventive purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the automatic detection system and method for optical modules based on CNC robots proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0035] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0036] The following description, in conjunction with the accompanying drawings, details the specific scheme of the automatic detection system and method for optical modules based on CNC robots provided in this application.

[0037] Please see Figure 1 The diagram illustrates a flowchart of an automatic detection method for optical modules based on a CNC robot, according to an embodiment of this application. The method includes the following steps:

[0038] Step S1: Collect various parameters at different times during the detection of the optical module, and assemble each type of parameter into a sequence; among these parameters, current is included.

[0039] In this embodiment, the automatic optical module detection device consists of an automatic optical module insertion / removal structure, an optical module connection performance judgment structure, and an automatic optical module detection structure. A schematic diagram of the specific device structure is shown below. Figure 2 As shown, the automatic insertion and removal structure for optical modules uses an automatic insertion and removal technology based on a six-axis robot to automate the automatic insertion and removal process of semi-finished printed circuit boards (PCBAs). The automatic insertion and removal technology based on a six-axis robot utilizes the high flexibility and precision of the six-axis robot and uses 3D vision technology to achieve precise positioning of the target object, thereby ensuring accurate gripping of the optical module and connecting the optical module to the automatic detection structure. After the automatic insertion and removal steps of the optical module are completed, the six-axis robot is reset and waits for the next control signal input.

[0040] For the automatic detection structure connected to the optical module, a test device group is installed at the optical module interface of the detection structure. Specifically, the test device group includes an optical power meter, an optical modulation amplitude tester, a digital communication analyzer, and a current sensor. The optical power meter is used to collect the intensity data of the optical signal emitted by the optical module, the optical modulation amplitude tester is used to obtain the extinction ratio data of the optical signal, the digital communication analyzer is used to collect the Q factor of the output eye diagram of the optical module, and the current sensor is used to collect the current data at the connection point.

[0041] The data acquisition frequency of the four instruments was set to 10kHz to ensure synchronous acquisition of the four types of data. It should be noted that, in other embodiments of this application, the implementer can set the data acquisition frequency according to actual conditions. All data acquired by the four instruments were subjected to overall normalization processing to eliminate the influence of dimensions.

[0042] The sequence of normalized values ​​of all optical signal intensity data arranged in ascending order of time is called the intensity sequence; the sequence of normalized values ​​of all optical signal extinction ratio data arranged in ascending order of time is called the extinction ratio sequence; the sequence of normalized values ​​of all Q factor data arranged in ascending order of time is called the Q factor sequence; and the sequence of normalized values ​​of all current data arranged in ascending order of time is called the current sequence.

[0043] Step S2: Obtain outliers in each sequence using an anomaly detection algorithm; construct the degree of anomaly for each outlier in each sequence based on the anomaly status of each outlier in each sequence; and construct connection determination factors for various parameters based on the distribution of outliers in each sequence and the fluctuations near each outlier, combined with the degree of anomaly.

[0044] In automated testing of optical modules, if the connection between the optical module and the testing structure is poor during the connection process by the six-axis robot (e.g., a successful connection but poor connection due to improper operation), it will cause abnormal parameters during the optical module testing. Poor performance of the optical module will also lead to abnormal parameters during testing, making it impossible to accurately distinguish whether the abnormal parameters are caused by poor connection or by the performance of the optical module itself. This may result in a good optical module being misjudged as a defective product, or a poor-performing optical module being misjudged as a good product.

[0045] Compared to anomalies caused by poor performance, anomalies caused by poor connectivity are persistent and stable. That is, as long as the connection status is not improved, the anomaly will persist, and anomalies caused by connectivity problems are stable under unchanged connection conditions, without being improved or worsened by minor changes in time or environment. In contrast, anomalies caused by poor performance are random and variable. That is, module performance problems may exhibit randomness due to aging or damage of internal components, potentially intermittently functioning normally and abnormally over a period of time. Furthermore, performance problems may vary with usage time, temperature changes, or other environmental factors.

[0046] (1) To filter out optical module parameter anomalies caused by connection problems, the intensity sequence, extinction ratio sequence, Q factor sequence, and current sequence are used as inputs to the LOF anomaly detection algorithm for anomaly detection. Preferably, in this embodiment, the k-distance is set to k=10, and the anomaly point ratio is set to 0.01. As other embodiments of this application, the implementer can set the k-distance and anomaly point ratio according to the actual situation. The output of the LOF anomaly detection algorithm is all the anomalies in the intensity sequence, extinction ratio sequence, Q factor sequence, and current sequence. A schematic diagram of the anomaly detection results of the intensity sequence is shown below. Figure 3 As shown. Among them, the LOF anomaly detection algorithm is a well-known technology, and the specific process will not be described in detail.

[0047] It should be noted that this application only provides one anomaly detection method for anomaly detection in sequences. There are many existing anomaly detection methods, and implementers may also use other anomaly detection algorithms to monitor anomalies in sequences. This application does not impose any specific restrictions.

[0048] (2) Taking the intensity sequence as an example, the degree of anomalousness of each outlier in the intensity sequence is set according to the difference between each outlier in the intensity sequence and the overall data:

[0049] The number of outliers detected in the intensity sequence is denoted as N. The mean of all elements in the intensity sequence is calculated and denoted as the intensity mean. The difference between each outlier in the intensity sequence and the intensity mean is calculated and denoted as the first difference. An outlier threshold T is set. Preferably, in this embodiment, the outlier threshold is set to 0.1. As other embodiments of this application, implementers can set the outlier threshold according to actual conditions.

[0050] The degree of anomalousness of each anomaly is determined based on its first difference, expressed as follows:

[0051] In the formula, The degree of anomalousness of the nth anomalous point in the intensity sequence. The first difference is the nth anomaly point in the intensity sequence, and T is a preset anomaly threshold.

[0052] If the first difference A value greater than or equal to T indicates a high degree of abnormality in the outlier value, and the outlier is not determined to be caused by a connection problem. Then it is 0; if the difference If the value is less than T, it indicates that the anomaly level is low, and the anomaly is determined to be caused by a connection problem. Then it is .

[0053] (3) Outliers represent data points in the intensity sequence that deviate significantly from other data. The intensity sequence is used as the input of the generalized autoregressive conditional heteroscedasticity model (GARCH model), and the output is the conditional variance of each data point in the intensity sequence at the time of collection, which is denoted as the first conditional variance.

[0054] Based on the above analysis, the connection determination factor for optical signal intensity is calculated, and the expression is:

[0055] In the formula, is the connection determination factor for optical signal intensity, and N is the number of anomalies detected in the intensity sequence. Let be the degree of anomalousness of the nth outlier in the intensity sequence, and b be the standard deviation of all outlier points in the intensity sequence at the time of data collection. Let V be the first conditional variance of the nth outlier in the intensity sequence. It is an exponential function with base e. Where, The first product, This is the second product.

[0056] It not only identified connectivity issues but also characterized the degree of deviation of outliers. The larger the value, the greater the deviation of the outlier, which means the more serious the connection problem. The larger the value of b, the more random the distribution of outliers, and the less likely the outliers are caused by connection problems. The smaller; This indicates the volatility around the time of anomaly point collection. The larger the value, the greater the fluctuation in the local optical signal intensity centered on the anomaly point, and the less likely the anomaly point is caused by a connectivity problem. The smaller.

[0057] The quality of the connection between the optical module and the detection structure is determined by the change in the light signal intensity. The larger the value, the more likely a connection problem is to be the main cause of abnormal optical signal strength. A smaller value indicates that the abnormal optical signal strength may be caused by a performance problem of the optical module itself.

[0058] (4) Based on the outliers in the extinction ratio sequence, Q factor sequence, and current sequence, respectively, the connection determination factor of the optical signal extinction ratio is calculated using the same acquisition method as the connection determination factor of the optical signal intensity. Q-factor connection determination factor and the current connection determination factor .

[0059] When the optical module and the detection structure are poorly connected, the increased resistance due to poor contact will cause abnormal current. In other words, current is a direct and important factor in judging the connection status between the optical module and the detection structure, because it is directly related to the working state and performance of the optical module. The optical signal intensity of the optical module is directly related to the emission power of the laser. In the optocoupler, the light-emitting diode (LED) is responsible for converting the input electrical signal into an optical signal. Its luminous intensity is proportional to the input current. When the connection status of the optical module is unstable, the current supply is unstable, which causes the laser emitter to switch between high and low levels unstably, thereby reducing the extinction ratio. Moreover, the change in current will directly affect the receiving performance of the optical module, thus affecting the Q factor.

[0060] The above analysis confirms that abnormal current directly affects the light intensity signal, extinction ratio, and Q factor. The connection determination factor of the current analyzed above... It can not only determine whether the current is affected by connection problems, but also reflect the degree of current abnormality caused by connection problems.

[0061] Step S3: Based on the correlation between the current sequence and the other sequences, and combined with the corresponding connection determination factors, construct the comprehensive influence factors of the parameters corresponding to the other sequences, and construct the connection risk index for optical module detection by combining the connection determination factors of the current.

[0062] (1) In order to analyze the effects of current on optical signal intensity, extinction ratio and Q factor, we first analyze the effect of current on optical signal intensity:

[0063] The intensity and current sequences are used as inputs to the t-test algorithm. The null hypothesis is that there is no significant difference between the means of the optical signal intensity and current data, indicating no significant influence between the two sequences. The alternative hypothesis is that there is a significant difference between the means of the optical signal intensity and current data, indicating a significant influence between the two sequences. In this embodiment, the significance level is set to 0.05. In other embodiments, the implementer can set the significance level according to the actual situation. The output is the p-value between the optical signal intensity and the current, denoted as P0. .

[0064] Furthermore, based on the extinction ratio sequence and current sequence, and the Q factor sequence and current sequence, respectively, using the same... Using the same calculation method, the P-value between the optical signal extinction ratio and the current is obtained, denoted as... And the P-value between the Q-factor and the current, denoted as .

[0065] (2) Further, based on the above analysis, the connection risk index B during optical module detection is calculated as follows:

[0066] First, the connection determination factor based on optical signal strength and The comprehensive influence factor of optical signal intensity was obtained, and it is positively correlated with the connection determination factor of optical signal intensity. They are negatively correlated.

[0067] It is understood that the positive and negative correlations in this application refer to the relationship between the independent and dependent variables. A positive correlation means that the independent variable increases (decreases) as the dependent variable increases (decreases); a negative correlation means that the independent variable decreases (increases) as the dependent variable increases (decreases). The specific positive and negative correlations can be determined according to the actual situation during application, and this application does not impose any special restrictions.

[0068] Preferably, in one embodiment of this application, the comprehensive influence factor of the optical signal intensity can be: obtained by using the natural constant e as the base, and... The result of the calculation of the exponential function with the opposite number of the exponent is used as the product of the calculation result and the connection determination factor of the optical signal intensity as the comprehensive influence factor of the optical signal intensity.

[0069] Preferably, in other embodiments of this application, the comprehensive influence factor of the optical signal intensity can be calculated by: calculating the natural number 1 and... The difference between the normalized values ​​is denoted as the first difference. The product of the first difference and the connection determination factor of the optical signal intensity is used as the comprehensive influence factor of the optical signal intensity.

[0070] Connection determination factors based on optical signal extinction ratio and Q-factor connection decision factor and Using the same acquisition method as the comprehensive influence factor of optical signal intensity, the comprehensive influence factors of optical signal extinction ratio and Q factor were obtained.

[0071] The statistical significance levels of the differences between the current data and the optical signal intensity, optical signal extinction ratio, and Q factor respectively reflect the degree to which the optical signal intensity, extinction ratio, and Q factor are affected by the current. The smaller the value, the more significant the impact of the current on these parameters; that is, the greater the impact of current anomalies caused by connection problems on the performance parameters of the optical module. This reflects the degree of anomalies in optical signal intensity, extinction ratio, and Q factor. These anomalies directly reflect the impact of connection status on the operating status and performance of the optical module. The larger the value, the higher the degree of abnormality of the corresponding parameter, which means that the connection problem has a greater impact on these parameters.

[0072] Then, the expression for calculating the connection risk index during optical module detection is:

[0073] In the formula, B is the connection risk index during optical module detection. This is the connection determination factor for the current. This represents the average of the combined influence factors of optical signal intensity, optical signal extinction ratio, and Q factor. This is the normalization function.

[0074] Since current is a direct and important factor in determining the connection status between the optical module and the detection structure, using Make an overall assessment of connection risks, when When the current returns to zero, it indicates that the abnormal current is not caused by a connection problem. The larger the value of B, the greater the degree of abnormality in the current due to connection problems. In this case, the connection problems have a greater impact on the testing of the optical module, and the larger the value of B, the greater the impact. The larger the value of B, the greater the impact of abnormal current caused by connection problems on the performance parameters of the optical module.

[0075] B describes the overall quality risk of the connection between the optical module and the detection structure. The larger the B value, the more serious the problem is in the connection between the optical module and the detection structure.

[0076] Step S4: Identify connection problems of optical modules based on the connection risk index.

[0077] After quantifying the overall quality risk of the connection between the optical module and the testing structure using a connection risk index B, the connection risk index B is monitored during optical module testing. When the connection risk index exceeds a threshold, a six-axis robotic arm in the automatic insertion / removal structure re-inserts the optical module under test to avoid low testing accuracy due to connection problems with the optical module. Specifically:

[0078] Set a poor connection threshold Preferably, in this embodiment of the application, the poor connection threshold is... Set to 0.4. As another embodiment of this application, the implementer can set the connection failure threshold according to actual conditions. The connection risk index B during optical module detection is judged; if... If the signal is strong, it indicates a significant connection problem with the optical module, and the result is fed back to the PLC control system, which then translates it into specific control signals that act on the six-axis robot. Conversely, if the signal is weak, it indicates a weak connection problem with the optical module, has no strong impact on detection, and no control signals are output. The PLC (Programmable Logic Controller) control system is a well-known technology, and its details will not be elaborated upon here.

[0079] Based on the above, poor connection problems of optical modules can be accurately identified, and the poorly connected optical modules can be re-inserted to avoid low detection accuracy due to connection problems of optical modules. When there are no connection problems of optical modules, the optical power, bit error rate, eye diagram and receiver sensitivity of optical modules are detected by the optical module automatic detection structure.

[0080] Based on the same inventive concept as the above methods, this application also provides an automatic detection system for optical modules based on CNC robots, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-described automatic detection methods for optical modules based on CNC robots.

[0081] In summary, this application provides an automatic detection method for optical modules based on CNC robots. Addressing the confusion between optical module connection performance and its own performance, it constructs a connection judgment factor by analyzing the distribution of abnormal points in optical signal intensity, optical signal extinction ratio, Q factor, and current data, as well as data fluctuations near each abnormal point. Specifically, it filters out optical module detection parameter anomalies caused by connection performance based on the degree of anomaly at each abnormal point, avoiding misjudging well-performing optical modules as defective. By analyzing abnormal changes in current data, it directly reflects the impact of poor connection on current. A t-test algorithm is used to reflect the statistical significance difference between current data and other parameters, constructing a connection risk index for optical module detection. This quantifies the impact of poor connection on optical module parameters, improving the accuracy of optical module performance judgment. Real-time monitoring of optical module connection quality is achieved by monitoring connection risk and setting thresholds. When the degree of poor connection is severe, a control signal is issued through the control module, instructing a six-axis robot to re-insert the optical module, reducing detection errors caused by connection problems and ensuring that the optical module is tested in optimal condition.

[0082] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this application. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0083] The various embodiments in this application are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0084] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. An automatic detection method for optical modules based on CNC robots, characterized in that, The method includes the following steps: The optical module collects various parameters at different times during detection and assembles these parameters into sequences; among these parameters is current. Anomalies are identified in each sequence using an anomaly detection algorithm. Anomalies are assessed based on the anomalies in each sequence. The degree of anomaly is then determined for each anomaly in each sequence. Based on the distribution of anomalies in each sequence and the fluctuations near each anomaly, and combined with the degree of anomaly, connection determination factors for various parameters are constructed. Based on the correlation between the current sequence and the other sequences, and combined with the corresponding connection determination factors, a comprehensive influence factor for the parameters corresponding to the other sequences is constructed. Combined with the connection determination factor of the current, a connection risk index is constructed for optical module detection. Identify connection problems of optical modules based on the connection risk index; The process of obtaining the connection determination factor is as follows: Each sequence is used as input to the time series statistical model, and the output is the conditional variance of each data point at the time of collection in each sequence; the product of the standard deviation of the collection time of all outliers in each sequence and the conditional variance of each outlier in each sequence is calculated and denoted as the first product. The connection determination factor for each parameter is determined based on the first product and the degree of abnormality of each abnormal point in each sequence. The process for obtaining the comprehensive impact factor is as follows: For each sequence other than the current sequence, the remaining sequences and the current sequence are used as inputs to the t-test algorithm, and the output is the P-value between the corresponding parameters of the remaining sequences and the current. Based on the P-value and the connection determination factor of each parameter other than the current, the comprehensive influence factor of each parameter is obtained, wherein the comprehensive influence factor of each parameter is positively correlated with the connection determination factor of each parameter and negatively correlated with the P-value. The expression for the connection risk index during optical module detection is: In the formula, B is the connection risk index during optical module detection. This is the connection determination factor for the current. This represents the average of the combined influence factors of all parameters except current. This is the normalization function.

2. The automatic detection method for optical modules based on CNC robots as described in claim 1, characterized in that, The parameters also include: optical signal intensity, optical signal extinction ratio, and Q factor of the output eye diagram.

3. The automatic detection method for optical modules based on CNC robots as described in claim 1, characterized in that, The process for obtaining the degree of abnormality for each abnormal point is as follows: Calculate the mean of all elements in each sequence; calculate the difference between each outlier in each sequence and the mean, and denote it as the first difference; If the first difference of each outlier in each sequence is greater than or equal to the preset outlier threshold, then the outlier degree of each outlier in each sequence is set to 0; otherwise, the outlier degree of each outlier in each sequence is set to the first difference.

4. The automatic detection method for optical modules based on CNC robots as described in claim 1, characterized in that, The process for determining the connection determination factors of each parameter is as follows: Calculate the result of the exponential function with the natural constant as the base and the negative of the first product as the exponent; calculate the product of the calculation result and the degree of anomalousness of each outlier in each sequence, and denote it as the second product; The mean of the second product of all outliers in each sequence is used as the connection determination factor for the corresponding parameters of each sequence.

5. The automatic detection method for optical modules based on CNC robots as described in claim 1, characterized in that, The process for obtaining the comprehensive influence factor of the remaining parameters is as follows: The exponential function with the natural constant as the base and the negative of the P value as the exponent is denoted as the first exponential function. The product of the first exponential function and the connection determination factor of the remaining parameters is taken as the comprehensive influence factor of the remaining parameters.

6. The automatic detection method for optical modules based on CNC robots as described in claim 1, characterized in that, The identification of poor connectivity issues in optical modules based on a connectivity risk index is specifically as follows: If the connection risk index is greater than or equal to the preset connection failure threshold, the optical module has a connection failure problem, and the judgment result is fed back to the PLC control system for adjustment; otherwise, the optical module connection is normal.

7. An automatic detection system for optical modules based on CNC robots, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1-6.

Images