Method for realizing bonding copper wire equipment control based on Internet of Things and 5G network

Abstract

The embodiment of the invention provides a method for controlling bonding copper wire equipment based on the Internet of Things and a 5G network, and belongs to the technical field of data processing. The method comprises the following steps: obtaining a target control result of target equipment, and obtaining a similar control result corresponding to the target control result through the Internet of Things; obtaining first sensing data of the target device under the target sensor under the similar control result, and carrying out abnormity identification on the first sensing data to obtain target abnormal data of the target sensor; performing data elimination on the first sensing data according to the target abnormal data to obtain second sensing data; performing correlation analysis according to the second sensing data to obtain a corresponding target relationship between the target sensors; obtaining current sensing data corresponding to the target sensor of the target equipment under the target control result according to the 5G network; and determining a target control strategy corresponding to the target equipment according to the current sensing data and the target relationship, and performing equipment control on the target equipment according to the target control strategy to obtain a target control result.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and particularly relates to a method for realizing control of a bonding copper wire device based on an Internet of Things and a 5G network. BACKGROUND

[0002] In today's industrial production field, the bonding copper wire device plays a crucial role in many manufacturing processes, and its stable operation is of great importance to product quality and production efficiency. In the prior art, the abnormality identification method only focuses on the abnormality of a single parameter, ignoring the correlation and collaborative changes between multiple parameters. A slight change in one parameter may not directly indicate an abnormality, but when multiple related parameters simultaneously exhibit abnormal changes, it may indicate potential problems with the device. However, existing data abnormality identification techniques are difficult to capture such complex parameter correlation changes, resulting in a failure to timely detect abnormalities in device operation. Therefore, when implementing control of the bonding copper wire device through data abnormality identification in the prior art, there are many deficiencies in timely detecting abnormalities in device operation, and there is an urgent need to develop more effective monitoring and control methods to improve the stability and reliability of the bonding copper wire device operation. SUMMARY

[0003] The main purpose of the embodiments of the present application is to provide a method for realizing control of a bonding copper wire device based on an Internet of Things and a 5G network, aiming to solve the problem that the prior art cannot timely detect abnormalities in device operation when implementing control of the bonding copper wire device through data abnormality identification, thereby reducing the stability and reliability of the bonding copper wire device operation.

[0004] In a first aspect, the embodiments of the present application provide a method for realizing control of a bonding copper wire device based on an Internet of Things and a 5G network, comprising:

[0005] obtaining a target control result corresponding to a target device, and obtaining a similar control result corresponding to the target control result through the Internet of Things;

[0006] obtaining first sensing data corresponding to the target device under a target sensor under the similar control result, and performing abnormality identification on the first sensing data to obtain target abnormal data corresponding to the target sensor;

[0007] performing data elimination on the first sensing data according to the target abnormal data to obtain second sensing data;

[0008] performing correlation analysis according to the second sensing data to obtain a target relationship corresponding to the target sensors;

[0009] obtaining current sensing data corresponding to the target sensors of the target device under the target control result according to the 5G network;

[0010] determine a target control strategy corresponding to the target device according to the current sensor data and the target relationship, and perform device control on the target device according to the target control strategy to obtain a target control result.

[0011] In a second aspect, an embodiment of the present application provides a system for implementing bonding copper wire device control based on an Internet of Things and a 5G network, comprising:

[0012] a data acquisition module configured to obtain a target control result corresponding to a target device, and obtain a similar control result corresponding to the target control result through the Internet of Things;

[0013] an abnormality identification module configured to obtain first sensor data corresponding to the target device under a target sensor in a similar control result, and perform abnormality identification on the first sensor data to obtain target abnormality data corresponding to the target sensor;

[0014] a data elimination module configured to perform data elimination on the first sensor data according to the target abnormality data to obtain second sensor data;

[0015] a correlation analysis module configured to perform correlation analysis on the second sensor data to obtain a target relationship corresponding to the target sensors;

[0016] a data collection module configured to obtain current sensor data corresponding to the target sensors of the target device under the target control result through a 5G network;

[0017] a data control module configured to determine a target control strategy corresponding to the target device according to the current sensor data and the target relationship, and perform device control on the target device according to the target control strategy to obtain a target control result.

[0018] In a third aspect, an embodiment of the present application further provides a terminal device, comprising a processor, a memory, a computer program stored in the memory and executable by the processor, and a data bus for realizing connection communication between the processor and the memory, wherein the computer program is executable by the processor to realize the steps of any one of the methods for implementing bonding copper wire device control based on an Internet of Things and a 5G network provided in the specification of the present application.

[0019] In a fourth aspect, an embodiment of the present application further provides a storage medium for computer readable storage, characterized in that the storage medium stores one or more programs, and the one or more programs are executable by one or more processors to realize the steps of any one of the methods for implementing bonding copper wire device control based on an Internet of Things and a 5G network provided in the specification of the present application.

[0020] The embodiment of the present application provides a method for realizing control of a bonding copper wire device based on an Internet of Things and a 5G network, which comprises the following steps: obtaining a target control result corresponding to a target device, and obtaining a similar control result corresponding to the target control result through the Internet of Things; obtaining first sensing data corresponding to the target device under a target sensor under the similar control result, and obtaining target abnormal data corresponding to the target sensor by performing abnormality identification on the first sensing data; obtaining second sensing data by performing data elimination on the first sensing data according to the target abnormal data; obtaining a target relationship corresponding to the target sensors by performing correlation analysis according to the second sensing data, so as to help to deeply understand the operation mechanism of the target device, and to reveal the physical and chemical processes inside the device by mining the relationship, thereby providing theoretical support for the control of the device, so as to obtain current sensing data corresponding to the target sensor of the target device under the target control result according to the 5G network; determining a target control strategy corresponding to the target device according to the current sensing data and the target relationship, and performing device control on the target device according to the target control strategy to obtain the target control result, so as to cope with various changes in the operation process of the device in time by adjusting the control strategy in real time, thereby guaranteeing the stability and reliability of the device, reducing the device failure and downtime, and improving the accuracy and stability of the device control, reducing the operation cost, improving the production efficiency and product quality, and solving the problem that the bonding copper wire device control cannot discover the abnormality in the operation of the device in time by using data abnormality identification in the related art, thereby reducing the stability and reliability of the operation of the bonding copper wire device. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without any creative labor on the basis of these drawings.

[0022] Figure 1 A flowchart of a method for realizing control of a bonding copper wire device based on an Internet of Things and a 5G network provided by the embodiment of the present application is shown in the figure.

[0023] Figure 2 A module structure diagram of a system for realizing control of a bonding copper wire device based on an Internet of Things and a 5G network provided by the embodiment of the present application is shown in the figure.

[0024] Figure 3 A structure schematic diagram of a terminal device provided by the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0025] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described, obviously, the described embodiments are a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all the other embodiments obtained by those skilled in the art without creative work belong to the protection scope of the present application.

[0026] The flow chart shown in the drawings is only an example, not necessarily including all the contents and operations / steps, and not necessarily executed in the described order. For example, some operations / steps can be decomposed, combined or partially merged, so that the actual execution order can be changed according to the actual situation.

[0027] It should be understood that the terms used in the present application specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the present application specification and the appended claims, unless otherwise clearly indicated by the context, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0028] The embodiments of the present application provide a method for realizing bonding copper wire equipment control based on Internet of Things and 5G network. The method for realizing bonding copper wire equipment control based on Internet of Things and 5G network can be applied to a terminal device, which can be an electronic device such as a tablet computer, a notebook computer, a desktop computer, a personal digital assistant and a wearable device. The terminal device can be a server or a server cluster.

[0029] Some embodiments of the present application will be described in detail below with reference to the accompanying drawings. In the case of no conflict, the following embodiments and features in the embodiments can be combined with each other.

[0030] Please refer to Figure 1 , Figure 1 A flowchart of a method for realizing bonding copper wire equipment control based on Internet of Things and 5G network provided by the embodiments of the present application is provided.

[0031] As shown in Figure 1 , the method for realizing bonding copper wire equipment control based on Internet of Things and 5G network includes steps S101 to S106.

[0032] Step S101, obtaining a target control result corresponding to a target device, and obtaining a similar control result corresponding to the target control result through Internet of Things.

[0033] For example, the target equipment is a copper wire bonding device. Key parameters during the operation of the copper wire bonding device may include the bonding speed, temperature, pressure, and welding time. These parameters constitute the basic control requirements for the operation of the device. For a device control system with an open interface, the target control results corresponding to the target device are read through the interface protocol. The target control results are used to characterize the parameter requirements of the target device during operation or the expected operating results of the target device.

[0034] For example, ensure that the target device and other similar devices (which may be geographically distributed) are connected to the Internet of Things (IoT). Devices can be connected to the IoT platform via wireless communication technologies (such as Wi-Fi, Bluetooth, LoRa, etc.) or wired networks (such as Ethernet), allowing data from the target device and other similar devices to be uploaded to the platform for centralized management and analysis.

[0035] For example, the range of values ​​for parameters such as speed, temperature, and pressure in copper wire bonding is obtained based on the target control result, and then the similar control result corresponding to the target control result is defined based on the range of values. For example, control results with speed error within ±5%, temperature error within ±3℃, and pressure error within ±2% can be considered as similar control results.

[0036] In some implementations, obtaining the similar control result corresponding to the target control result via the Internet of Things includes: decomposing the target control result into data to obtain a first control parameter corresponding to the target control result; obtaining a first sub-parameter corresponding to the first control parameter under a first parameter type, and obtaining a first nearest neighbor parameter corresponding to the first sub-parameter in a database; obtaining a second nearest neighbor parameter corresponding to the first nearest neighbor parameter, and determining mutual nearest neighbor parameters between the first sub-parameter and the first nearest neighbor parameter based on the first nearest neighbor parameter and the second nearest neighbor parameter; determining a first weight corresponding to the first sub-parameter based on the mutual nearest neighbor parameters, and determining a first density information corresponding to the first sub-parameter based on the first weight combined with the first nearest neighbor parameter and the mutual nearest neighbor parameter; obtaining the first nearest neighbor parameter. The system obtains the corresponding second density information, and determines the first similarity between the first sub-parameter and the first nearest neighbor parameter based on the first density information and the second density information combined with the first nearest neighbor parameter and the mutual nearest neighbor parameter; obtains the second control parameter corresponding to the first nearest neighbor parameter from the database, and calculates the second similarity between the second sub-parameter corresponding to the first control parameter under the second parameter type and the third sub-parameter corresponding to the second control parameter under the second parameter type; fuses the first similarity and the second similarity to determine the target similarity between the first control parameter and the second control parameter; determines the third control parameter associated with the first control parameter based on the target similarity, and determines the similar control result corresponding to the target control result based on the third control parameter.

[0037] For example, the target equipment is a copper wire bonding machine, and its business scenario may be the packaging and production of electronic chips. The target control result may be the overall performance indicator of completing a copper wire bonding operation. For copper wire bonding equipment, the target control result may cover multiple aspects such as the quality of copper wire bonding (e.g., weld strength, connection stability) and production efficiency (e.g., bonding time). Based on the target control result, the specific requirements of the copper wire bonding equipment are determined, thereby obtaining the data range corresponding to various types of parameters during the operation of the target equipment, including but not limited to wire feeding speed, heating temperature, and pressure.

[0038] For example, the comprehensive information in the target control results can be mapped to specific first control parameters. For example, by analyzing the quality inspection data and production records of the copper wire bonding equipment, the quality of bonding can be mapped to the specific value range of first control parameters such as heating temperature and pressure.

[0039] For example, the first parameter type in the first control parameter includes, but is not limited to, the speed, temperature, pressure, welding time, etc. of copper wire bonding. Then, based on the first parameter type, the first sub-parameter corresponding to the first control parameter under the first parameter type is obtained from the first control parameter. In the database, the first sub-parameter is used as the query condition to search for the closest parameter according to a certain distance metric (such as Euclidean distance, Manhattan distance, etc.). These parameters are the first nearest neighbor parameters.

[0040] For example, based on the first nearest neighbor parameter, the database is searched again for the parameter that is closest to the first nearest neighbor parameter to obtain the second nearest neighbor parameter. When the second nearest neighbor parameter corresponding to the first nearest neighbor parameter contains the first sub-parameter, that is, the first sub-parameter is the nearest neighbor of the first nearest neighbor parameter, and the first nearest neighbor parameter is also the nearest neighbor of the first sub-parameter, it is determined that there is a mutual nearest neighbor relationship between them, and the corresponding parameter is the mutual nearest neighbor parameter.

[0041] For example, the third distance between each sub-nearest neighbor parameter in the first sub-parameter and the mutual nearest neighbor parameters is obtained, and then all third distances are summed to obtain the first sum value. Then, the fourth distance between the first sub-parameter and the data in the database that are not equal to the first sub-parameter under the first parameter type is calculated, and then all fourth distances are summed to obtain the second sum value. The ratio between the first sum value and the second sum value is calculated and then summed with a constant 1 to determine the first weight. Then, the first weight, the first nearest neighbor parameter, and the mutual nearest neighbor parameter are combined, and the first density information corresponding to the first sub-parameter is determined by a specific density calculation method, such as a method based on kernel density estimation.

[0042] For example, using the same method as calculating the first density information, the second density information corresponding to the first nearest neighbor parameter is calculated. The first density information reflects the local density of the first sub-parameter in the data space, while the second density information reflects the density of the first nearest neighbor parameter's distribution among its surrounding data points. Therefore, these two density information sets can be considered important indicators describing the positional characteristics of the first sub-parameter and the first nearest neighbor parameter in the data space. Thus, by applying an appropriate distance metric, such as Euclidean distance, Manhattan distance, or Mahalanobis distance, and combining the first and second density information, the distance between the first sub-parameter and the first nearest neighbor parameter can be calculated. Different distance metric methods are suitable for different data characteristics and analytical needs.

[0043] For example, the first nearest neighbor parameter is the parameter in the database that is closest to the first sub-parameter, while the second nearest neighbor parameter is the parameter that is closest to the first nearest neighbor parameter. The intersection of the first and second nearest neighbor parameters is then calculated to find parameters that belong to both the nearest neighbor range of the first nearest neighbor parameter and are related to the first sub-parameter. These parameters are the shared parameters between the first nearest neighbor parameter and the first sub-parameter. After obtaining the shared parameters, their quantity is counted and recorded as the first quantity. This quantity reflects the number of common data points between the first sub-parameter and the first nearest neighbor parameter within their nearest neighbor range; a higher quantity indicates a greater degree of overlap in their local data distribution. Simultaneously, the mutual nearest neighbor parameters are counted to obtain the second quantity. The mutual nearest neighbor parameters reflect the strength of the mutual neighbor relationship between the first sub-parameter and the first nearest neighbor parameter, and the magnitude of the second quantity reflects the tightness of this mutual neighbor relationship.

[0044] For example, the distance information between the previously obtained first sub-parameter and the first nearest neighbor parameter is squared to obtain the squared distance. Squaring amplifies distance differences, making larger distances have a more significant impact in subsequent calculations. Next, an exponential operation is performed with the natural number e as the base and the squared distance as the exponent, resulting in a new data point. Taking the reciprocal of this data point yields the first value. The purpose of this series of operations is to transform the distance information into a similarity-related value; the greater the distance, the smaller the first value, thus reflecting the inverse effect of distance on similarity. The first number of shared parameters and the second number of mutual nearest neighbor parameters are added together, and then averaged to obtain the average number. This average number comprehensively considers the degree of overlap and the closeness of the mutual neighbor relationship between the first sub-parameter and the first nearest neighbor parameter in the local data distribution. Again, an exponential operation is performed with the natural number e as the base and the average number as the exponent, resulting in the second value. The exponential operation causes changes in the average number to have an exponential effect on the second value, highlighting the important role of data distribution density in similarity. Finally, the first and second values ​​are multiplied to obtain the first similarity between the first sub-parameter and the first nearest neighbor parameter. This similarity value takes into account the distance between the first sub-parameter and the first nearest neighbor parameter, the degree of overlap in the local data distribution, and the closeness of their neighbor relationships. The closer the similarity value is to 1, the more similar the first sub-parameter and the first nearest neighbor parameter are; the closer the similarity value is to 0, the greater the difference between them.

[0045] For example, a second control parameter corresponding to the first nearest neighbor parameter is extracted from the database, and a third sub-parameter corresponding to the second parameter type is obtained from the second control parameter according to the second parameter type. Then, a second similarity between the second sub-parameter and the third sub-parameter under the second parameter type of the first control parameter is calculated. The calculation method of the second similarity is the same as the calculation method of the first similarity.

[0046] For example, a weighted fusion of the first and second similarities is performed to determine the target similarity between the first and second control parameters. Based on this target similarity, a third control parameter with a high similarity to the first control parameter is selected from the database. For instance, a similarity threshold is set, and control parameters with similarities higher than this threshold are used as the third control parameter. Since the third control parameter has a high similarity to the first control parameter, its corresponding control result can be considered as a similar control result to the target control result.

[0047] In some implementations, determining the first density information corresponding to the first sub-parameter based on the first weight combined with the first nearest neighbor parameter and the mutual nearest neighbor parameter includes: calculating a first distance between the first sub-parameter and any fourth sub-parameter among the mutual nearest neighbor parameters, and determining the first sub-density corresponding to the first sub-parameter based on the first distance and the first weight; performing a difference calculation on the first nearest neighbor parameter and the mutual nearest neighbor parameter to obtain a target set, wherein the data in the target set belongs to the first nearest neighbor parameter but not to the mutual nearest neighbor parameter; calculating a second distance between the first sub-parameter and any fifth sub-parameter in the target set, and determining the second sub-density corresponding to the first sub-parameter based on the second distance; obtaining all sub-parameters corresponding to the first parameter type from the database, and removing the first nearest neighbor parameter from the all sub-parameters to obtain the remaining sub-parameters; calculating a third distance between the first sub-parameter and any sixth sub-parameter among the remaining sub-parameters, and determining the third sub-density corresponding to the first sub-parameter based on the third distance; and fusing the first sub-density, the second sub-density, and the third sub-density to determine the first density information corresponding to the first sub-parameter.

[0048] For example, the first distance between the first sub-parameter and any fourth sub-parameter among the nearest neighbor parameters is calculated using Euclidean distance, Manhattan distance, or Mahalanobis distance. Then, an exponential operation is performed with the natural number e as the base and the reciprocal of the first distance as the exponent, resulting in new data. The purpose of this step is to transform the distance information into a density-related value; the closer the distance, the larger the result after the exponential operation. This new data is then multiplied by a pre-set first weight to obtain the first product. The first weight can be adjusted according to the actual situation to control the degree of influence of different fourth sub-parameters on the final result. The first products between the first sub-parameter and each fourth sub-parameter among the nearest neighbor parameters are summed to obtain the first target sum. This sum integrates the relationship between the first sub-parameter and all elements in the nearest neighbor parameters. The number of first parameters corresponding to the first nearest neighbor parameters and the number of second parameters corresponding to the nearest neighbor parameters are obtained. The number of first parameters and the number of second parameters corresponding to the nearest neighbor parameters are divided, and then multiplied by the first target sum to finally obtain the first sub-density. The first sub-density reflects the local density relationship between the first sub-parameter and the nearest neighbor parameters.

[0049] For example, by comparing each element in the first nearest neighbor parameter and the mutual nearest neighbor parameter, elements that belong to the first nearest neighbor parameter but are not in the mutual nearest neighbor parameter set are selected. These elements together constitute the target set. For instance, if the first nearest neighbor parameter is {A,B,C,D} and the mutual nearest neighbor parameter is {B,D}, then the target set obtained by difference set calculation is {A,C}.

[0050] For example, for each fifth sub-parameter in the target set, the distance between it and the first sub-parameter is calculated using either Euclidean distance or Manhattan distance; this distance is called the second distance. These two distance metrics are simple and intuitive, and can effectively reflect the relative positional relationships between data points. The first distance exponent is obtained by exponentially calculating the second distance with the natural number e as the base and the reciprocal of the second distance as the exponent. Similarly, the closer the distance, the larger the first distance exponent. The first distance exponents corresponding to all fifth sub-parameters in the target set are summed to obtain the second target sum. The number of third parameters corresponding to the target set is obtained, divided by the number of first parameters, and then multiplied by the second target sum to obtain the second sub-density. The second sub-density reflects the local density relationship between the first sub-parameter and the elements in the target set.

[0051] For example, all sub-parameters belonging to the first parameter type are filtered from the database; these sub-parameters form a complete set. Then, the first nearest neighbor parameter is removed from all sub-parameters, and the remaining sub-parameters constitute the residual sub-parameters. For each sixth sub-parameter in the residual sub-parameters, the distance between it and the first sub-parameter, i.e., the third distance, is calculated using a suitable distance metric (such as Euclidean distance, Manhattan distance, etc.). The second distance exponent is obtained by exponentiation with the natural number e as the base and the reciprocal of the third distance as the exponent. The third target sum is obtained by summing the second distance exponents corresponding to all sixth sub-parameters in the residual sub-parameter set.

[0052] For example, the total number of all sub-parameters is obtained, and the total number of sub-parameters corresponding to the first nearest neighbor parameter is subtracted from the total number of sub-parameters, and then the reciprocal is calculated. The reciprocal is then multiplied by the third objective sum to obtain the third sub-density. The third sub-density reflects the local density relationship between the first sub-parameter and other sub-parameters besides the first nearest neighbor parameter.

[0053] For example, the first sub-density, the second sub-density, and the third sub-density are summed to obtain the first density information corresponding to the first sub-parameter. This first density information comprehensively considers the local density relationship between the first sub-parameter and different types of sub-parameters, and can more comprehensively and accurately reflect the distribution characteristics of the first sub-parameter in the dataset.

[0054] Specifically, by calculating and fusing the density relationships with different types of sub-parameters separately, the local characteristics of the first sub-parameter in the data space can be characterized more meticulously, avoiding the information loss caused by considering only certain parameters, and making the understanding of the data more in-depth.

[0055] It should be noted that the second density information corresponding to the first nearest neighbor parameter can also be obtained in the same way as the first density information corresponding to the first sub-parameter, which will not be described in detail here.

[0056] Step S102: Obtain the first sensing data of the target device under the target sensor under the similar control result, and perform anomaly identification on the first sensing data to obtain the target anomaly data corresponding to the target sensor.

[0057] For example, the first sensing data of the target device under the target sensor under similar control results is obtained from the database. The target sensor can be various sensors installed on the bonding copper wire device, such as temperature sensors, pressure sensors, displacement sensors, etc.

[0058] For example, the target sensor includes at least a first sensor and a second sensor. First data corresponding to the first sensor and second data corresponding to the second sensor are obtained from the first sensing data. Anomaly identification algorithms, such as statistical analysis-based methods or machine learning-based methods, are then used to identify anomalies in the first data to obtain first anomalous data and a corresponding first anomalous time. Similarly, anomaly identification algorithms are used to identify anomalies in the second data to obtain second anomalous data and a second anomalous time. The intersection of the first and second anomalous times is then calculated to obtain the target anomalous time. The purpose of this intersection calculation is to find the time point when two sensors simultaneously detect anomalies. Because multiple sensors detecting anomalies simultaneously likely indicates a genuine problem with the target object, rather than a misjudgment by a single sensor. For example, when both a temperature sensor and a vibration sensor detect anomalies at the same time, this time point is identified as the target anomalous time. Finally, data at that time point is extracted from the first sensing data based on the target anomalous time and identified as the target anomalous data. This target anomalous data is obtained through multiple screening and analysis processes, resulting in higher reliability and analytical value.

[0059] In some embodiments, the target sensor includes at least a first sensor and a second sensor, and the first sensing data includes at least first data corresponding to the first sensor and second data corresponding to the second sensor. The step of obtaining target abnormal data corresponding to the target sensor by anomaly identification of the first sensing data includes: obtaining first sub-data corresponding to the first sensor at a target time from the first data, and obtaining second sub-data corresponding to the second sensor at the target time from the second data; obtaining a first frequency corresponding to the first sub-data from the first data and a second frequency corresponding to the second sub-data from the second data; and combining the first sub-data and the second sub-data with the first frequency and the second frequency... The frequency determines the target entropy value corresponding to the target sensor at the target time; the target entropy value determines the first anomaly representation value corresponding to the target sensor at the target time; a first discrete representation value corresponding to the first sub-data is obtained from the first data, and a second discrete representation value corresponding to the second sub-data is obtained from the second data; a second anomaly representation value corresponding to the target sensor at the target time is determined based on the first discrete representation value and the second discrete representation value; a target anomaly representation value corresponding to the target sensor at the target time is determined based on the first anomaly representation value and the second anomaly representation value; and anomaly identification is performed on the first sensing data based on the target anomaly representation value to obtain the target anomaly data corresponding to the target sensor.

[0060] For example, the target time is the time information specified when analyzing the first data and the second data simultaneously. After determining the target time, the first sub-data corresponding to the first sensor at that time is extracted from the first data, and the second sub-data corresponding to the second sensor at that time is extracted from the second data. Then, for the first sub-data, the number of times it appears in the first data is counted to calculate the first frequency. Similarly, for the second sub-data, the number of times it appears in the second data is counted to calculate the second frequency.

[0061] For example, the target entropy value corresponding to the target sensor at the target time is calculated using the entropy calculation formula based on the first frequency and the second frequency. The larger the target entropy value, the higher the uncertainty of the data and the more possible anomalies there may be. Then, the target entropy value is normalized, and the normalization result is determined as the first anomaly characterization value. The first anomaly characterization value can intuitively reflect the degree of anomaly of the target time data based on the entropy value.

[0062] For example, a first discrete representation value is obtained by calculating the discrete representation value of the first sub-data in the first data based on the variance or standard deviation; and a second discrete representation value is obtained by calculating the discrete representation value of the second sub-data in the second data. The greater the degree of dispersion, the more dispersed the data is, and the greater the possibility of anomalies.

[0063] For example, the first discrete characterization value and the second discrete characterization value are added together to obtain the target discrete characterization value at the target time, and then the target discrete characterization value is normalized to obtain the second abnormal characterization value.

[0064] For example, the first anomaly representation value and the second anomaly representation value are weighted and summed, with different weights assigned to each. These weights are then added together to obtain the target anomaly representation value. The weights can be adjusted according to specific application scenarios and requirements to highlight the importance of different anomaly judgment factors. A threshold for the target anomaly representation value is then set. When the target anomaly representation value exceeds this threshold, the first sensing data at the target time is considered anomaly data and is identified as the target anomaly data corresponding to the target sensor.

[0065] In some implementations, obtaining the first discrete representation value corresponding to the first sub-data from the first data includes: obtaining the first nearest neighbor data corresponding to the first sub-data from the first data, and obtaining the relevant data corresponding to the first sub-data as the nearest neighbor data; determining the first neighborhood data corresponding to the first sub-data based on the relevant data and the first nearest neighbor data; determining a target nearest neighbor range, and obtaining a first distance between the first sub-data and the first nearest neighbor data based on the target nearest neighbor range; obtaining a first expected distance corresponding to the target nearest neighbor range, and determining a first distance difference corresponding to the first sub-data under the target nearest neighbor range based on the first distance and the first expected distance. The process involves: determining a second distance difference for the first sub-data within the entire target nearest neighbor range based on the first distance difference; determining a first representation parameter for the first sub-data based on the first nearest neighbor data, and obtaining a second representation parameter for each third sub-data in the first neighborhood data; determining a bandwidth value between the first sub-data and the third sub-data based on the first representation parameter and the second representation parameter; determining a kernel density estimate for the first sub-data based on the bandwidth value and the first neighborhood data; and determining a first discrete representation value for the first sub-data based on the second distance difference and the kernel density estimate. The first discrete representation value is obtained according to the following formula:

[0066]

[0067] Among them, disc i sum_diff represents the first discrete representation value corresponding to the i-th first sub-data point. i This represents the second distance difference corresponding to the i-th first sub-data item, count(all) i ) represents the first neighborhood data all corresponding to the i-th first sub-data. i The number of data, m i Let m represent the first representation parameter corresponding to the i-th first sub-data point. j Let pi represent the second characterization parameter corresponding to the j-th third sub-data in the first neighborhood data corresponding to the i-th first sub-data, where pi represents an irrational number, d represents a positive integer, exp represents an exponential function, and dis represents a distance function.

[0068] For example, the first nearest neighbor data that is closest to the first sub-data is found from the first data. Here, "near" can be defined according to a specific distance metric (such as Euclidean distance, Manhattan distance, etc.). Simultaneously, the relevant data corresponding to the first sub-data as the nearest neighbor data of other data is found. The first neighborhood data corresponding to the first sub-data is obtained by performing a union calculation based on the previously obtained relevant data and the first nearest neighbor data.

[0069] For example, a target nearest neighbor range is determined. The target nearest neighbor range can be a first nearest neighbor range, a second nearest neighbor range, or a third nearest neighbor range. That is, when the target nearest neighbor range is k, the value of k is defined. After the target nearest neighbor range is determined, the distance between the first sub-data and the corresponding first nearest neighbor data under the target nearest neighbor range is calculated to obtain the first distance.

[0070] For example, the relevant neighbor data corresponding to each sub-data in the first data within the target nearest neighbor range is obtained, and the distance between the sub-data and the relevant neighbor data is calculated. The sum of all distances is then taken as the average to obtain the first expected distance. The first expected distance is obtained by subtracting the first distance from the first expected distance based on the average distance between the data distances of the first data within the target nearest neighbor range, thus obtaining the first distance difference corresponding to the first sub-data within the target nearest neighbor range.

[0071] For example, for the entire target nearest neighbor range, repeat the steps of calculating the first distance and the first distance difference, and finally sum the first distance differences of the first sub-data in the entire target nearest neighbor range to obtain the corresponding second distance difference.

[0072] For example, the average data distance between the first nearest neighbor data and the first sub-data is determined, and this average value is used as the first representation parameter corresponding to the first sub-data. Simultaneously, a second representation parameter corresponding to each third sub-data in the first neighborhood data is calculated.

[0073] For example, the first and second characterization parameters are multiplied and then square-rooted; the result is then determined as the bandwidth value between the first and third sub-data points. The bandwidth value plays a crucial role in kernel density estimation, controlling the smoothness of the kernel function.

[0074] For example, the kernel density estimate corresponding to the first sub-data is obtained based on the bandwidth value and the first neighborhood data according to the following formula:

[0075]

[0076] Among them, kel_v i Represents the kernel density estimate corresponding to the i-th first sub-data point, count(all) i) represents the first neighborhood data all corresponding to the i-th first sub-data. i The number of data, m i Let m represent the first representation parameter corresponding to the i-th first sub-data point. j Let pi represent the second characterization parameter corresponding to the j-th third sub-data in the first neighborhood data corresponding to the i-th first sub-data, where pi represents an irrational number, d represents a positive integer, exp represents an exponential function, and dis represents a distance function.

[0077] For example, based on the previously obtained second distance difference and kernel density estimate, the first discrete representation value corresponding to the first sub-data is calculated using the following formula. This formula comprehensively considers the distance information and probability density information of the data, and can more comprehensively reflect the degree of dispersion of the first sub-data:

[0078]

[0079] Among them, disc i sum_diff represents the first discrete representation value corresponding to the i-th first sub-data point. i kel_v represents the second distance difference corresponding to the i-th first sub-data point. i This represents the kernel density estimate corresponding to the i-th first sub-data point.

[0080] Specifically, by comprehensively considering multiple factors such as nearest neighbor data, distance information, expected distance, and kernel density estimation, the discrete characteristics of the first sub-data can be more accurately characterized, making it easier to identify data points with abnormal dispersion, thereby improving the accuracy and reliability of anomaly detection.

[0081] It should be noted that the above steps can also be used to obtain the second discrete representation value corresponding to the second sub-data from the second data, and this application will not repeat them here.

[0082] In some implementations, determining the first characterization parameter corresponding to the first sub-data based on the first nearest neighbor data includes: obtaining the nearest neighbor distance between the first sub-data and the first nearest neighbor data; counting the number of targets corresponding to the first nearest neighbor data; and determining the first characterization parameter corresponding to the first sub-data based on the nearest neighbor distance and the number of targets.

[0083] For example, the nearest distance between the first sub-data and each first nearest neighbor data is calculated based on Euclidean distance or Manhattan distance, and the number of first nearest neighbor data is counted to obtain the target number.

[0084] For example, the first representation parameter corresponding to the first sub-data is obtained by summing all nearest neighbor distances and then dividing by the target number. The first representation parameter comprehensively considers the distance between the first sub-data and its nearest neighbor data, as well as the number of nearest neighbor data, and can well reflect the local characteristics of the data.

[0085] For example, by combining nearest neighbor distance and the number of targets to determine the first characterization parameter, the limitations of single-index analysis are avoided. Nearest neighbor distance reflects the similarity between data points, while the number of targets reflects the density information of the data. Combining the two allows for a more comprehensive and accurate description of the data characteristics, thereby improving the accuracy of data analysis.

[0086] Step S103: Based on the target abnormal data, perform data removal on the first sensing data to obtain the second sensing data.

[0087] For example, target anomalous data is removed from the first sensing data to obtain the second sensing data.

[0088] Step S104: Perform correlation analysis based on the second sensing data to obtain the target relationships between the target sensors.

[0089] For example, the correlation coefficients between different target sensors are calculated using Pearson correlation coefficient, Spearman correlation coefficient, etc., based on the second sensor data, thereby determining the target relationship between the target sensors. If the correlation coefficient is close to 1 or -1, it indicates that there is a strong linear positive or negative correlation between the target sensors.

[0090] In some implementations, obtaining the target relationship between the target sensors by performing correlation analysis based on the second sensing data includes: obtaining a first sensor and a second sensor from the target sensors, and obtaining third data corresponding to the first sensor and fourth data corresponding to the second sensor from the second sensing data; combining the third data and the fourth data according to the acquisition time to obtain initial combined data, and determining the probability distribution corresponding to each sub-combined data based on the initial combined data; performing feature selection from the initial combined data based on the probability distribution to obtain the target feature corresponding to the initial combined data; determining the target entropy information corresponding to the initial combined data based on the target feature, and determining the target function corresponding to the initial combined data based on the target entropy information; obtaining the optimal solution corresponding to the target function according to the Lagrange multiplier method, and obtaining the correlation function corresponding to the first sensor and the second sensor based on the optimal solution; and determining the target relationship between the first sensor and the second sensor based on the correlation function.

[0091] For example, a first sensor and a second sensor are determined from the target sensors, and third data corresponding to the first sensor and fourth data corresponding to the second sensor are obtained from the second sensing data.

[0092] For example, the third and fourth data points are paired and combined according to their acquisition time to form initial combined data. For instance, the temperature value collected by the first sensor and the humidity value collected by the second sensor at the same time are combined to form a data pair. The initial combined data is then grouped to obtain multiple sub-combined data points. The probability distribution of each sub-combined data point is estimated by statistically analyzing its frequency. If the data volume is large, methods such as histograms and kernel density estimation can be used to determine the probability distribution more accurately.

[0093] For example, based on the probability distribution, feature evaluation methods such as information gain and chi-square test are used to evaluate each feature of the initial combined data, measuring the importance of each feature in distinguishing different data patterns. Based on the evaluation results, important features are selected as target features from the initial combined data. A threshold can be set to retain only features with evaluation scores higher than this threshold.

[0094] For example, the target entropy information of the initial combined data is calculated based on the target features, and then the target entropy information is used as the basis to construct the objective function corresponding to the initial combined data. The objective function is usually designed to achieve a certain optimization objective, such as maximizing information gain or minimizing error.

[0095] For example, the Lagrange multiplier method is used to solve for the optimal solution of the objective function. This method transforms the constrained optimization problem into an unconstrained optimization problem by introducing Lagrange multipliers. Based on the obtained optimal solution, the correlation function between the first and second sensors is determined.

[0096] For example, the correlation function describes the mathematical relationship between two sensor data points. Based on the form and parameters of the correlation function, the target relationship between the first and second sensors can be analyzed. If the correlation function exhibits a linear relationship, it indicates a linear correlation between the two sensor data points; if it exhibits a non-linear relationship, it indicates a more complex correlation, and thus the target relationship between the first and second sensors can be characterized through the correlation function.

[0097] Step S105: Obtain the current sensing data of the target sensor corresponding to the target sensor of the target device under the target control result based on the 5G network.

[0098] For example, ensure the target device has 5G communication capabilities, install a suitable 5G communication module, and configure it accordingly to allow the target device to attempt to connect to the 5G network. Identify the target sensor associated with the target device. The target sensor may be installed on the target device or in its surrounding environment to monitor the device's operating status, environmental parameters, etc. Establish an association between the target device and the target sensor, binding the target sensor's identifier to the target device. After the target device establishes a connection with the 5G network and is associated with the target sensor, it sends a data request to the target sensor via the 5G network. The data request may include information such as the target sensor's identifier and the data acquisition time range to ensure accurate current sensor data is obtained.

[0099] Step S106: Determine the target control strategy corresponding to the target device based on the current sensing data and the target relationship, and perform device control on the target device according to the target control strategy to obtain the target control result.

[0100] For example, machine learning algorithms, such as decision trees and neural networks, are used to perform relationship analysis on the current sensor data to obtain the current data relationships. These current data relationships are then compared with the target relationship. The target relationship can be a pre-defined functional relationship. Appropriate evaluation metrics are determined to measure the degree of difference between the current data relationship and the target relationship, such as deviation rate or similarity. If the evaluation result exceeds a pre-defined acceptable range, the current data relationship is determined not to meet the target relationship. If the current data relationship does not meet the target relationship, anomaly identification methods, such as machine learning methods, are applied to the data corresponding to each sensor in the current sensor data. After anomaly identification for each sensor's data, the number of anomaly data corresponding to each sensor is counted. Sensors with a smaller number of anomaly data are designated as invariant sensors. These sensors are chosen because their data is relatively stable and reliable, providing a more accurate basis for subsequently determining the target control strategy.

[0101] For example, the specific association between a constant sensor and another sensor that needs to be controlled is determined based on the target relationship. Then, based on the target relationship and the current data of the constant sensor, the target value corresponding to the other sensor is calculated. Based on the calculated target value, a control strategy for the other sensor is formulated. The control strategy should specify concrete control actions and parameters, such as adjusting the operating power associated with the device under that sensor, or turning the device's components on or off.

[0102] For example, during the execution of the target control strategy by the target device, the current sensing data of the target sensors are continuously monitored to obtain the target control result. The obtained target control result is compared and evaluated with the expected target. If the result does not meet the expectation, the cause is analyzed and the target control strategy is adjusted, and the control process is executed again until a satisfactory target control result is obtained.

[0103] Please see Figure 2 , Figure 2 This application provides a system 200 for controlling a copper wire bonding device based on the Internet of Things (IoT) and a 5G network. The system 200 includes: a data acquisition module 201, an anomaly identification module 202, a data rejection module 203, a correlation analysis module 204, a data acquisition module 205, and a data control module 206. The data acquisition module 201 is used to obtain a target control result corresponding to the target device and to obtain a similar control result corresponding to the target control result through the IoT. The anomaly identification module 202 is used to obtain first sensing data of the target device under the target sensor under the similar control result and to perform anomaly detection on the first sensing data. The system identifies and obtains target anomaly data corresponding to the target sensor; a data removal module 203 is used to remove data from the first sensor data based on the target anomaly data to obtain second sensor data; a correlation analysis module 204 is used to perform correlation analysis based on the second sensor data to obtain the target relationship between the target sensors; a data acquisition module 205 is used to obtain the current sensor data corresponding to the target sensor of the target device under the target control result based on the 5G network; and a data control module 206 is used to determine the target control strategy corresponding to the target device based on the current sensor data and the target relationship, and to perform device control on the target device based on the target control strategy to obtain the target control result.

[0104] In some implementations, the system 200 for controlling the bonding copper wire equipment based on the Internet of Things and 5G networks can be applied to terminal devices.

[0105] It should be noted that those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system 200 for controlling the bonding copper wire equipment based on the Internet of Things and 5G network described above can be referred to the corresponding process in the aforementioned method embodiment for controlling the bonding copper wire equipment based on the Internet of Things and 5G network, and will not be repeated here.

[0106] Please see Figure 3 , Figure 3 This is a schematic block diagram of the structure of a terminal device provided in an embodiment of the present invention.

[0107] likeFigure 3 As shown, the terminal device 300 includes a processor 301 and a memory 302, which are connected by a bus 303, such as an I2C (Inter-integrated Circuit) bus.

[0108] Specifically, processor 301 provides computing and control capabilities to support the operation of the entire terminal device. Processor 301 can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0109] Specifically, the memory 302 can be a Flash chip, a read-only memory (ROM) disk, an optical disk, a USB flash drive, or a portable hard drive, etc.

[0110] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the embodiments of the present invention, and does not constitute a limitation on the terminal device to which the embodiments of the present invention are applied. A specific server may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0111] The processor is used to run a computer program stored in a memory, and when executing the computer program, implements any of the methods for controlling a copper wire bonding device based on the Internet of Things and 5G network provided in the embodiments of the present invention.

[0112] In one embodiment, the processor is configured to run a computer program stored in memory, and when executing the computer program, perform the following steps:

[0113] Obtain the target control result corresponding to the target device, and obtain the similar control result corresponding to the target control result through the Internet of Things;

[0114] Obtain the first sensing data of the target device under the target sensor under the similar control result, and perform anomaly identification on the first sensing data to obtain the target anomaly data corresponding to the target sensor;

[0115] The first sensing data is removed based on the target abnormal data to obtain the second sensing data;

[0116] Based on the second sensing data, correlation analysis is performed to obtain the target relationships between the target sensors.

[0117] The target device obtains the current sensing data corresponding to the target sensor under the target control result based on the 5G network.

[0118] Based on the current sensing data and the target relationship, a target control strategy corresponding to the target device is determined, and the target device is controlled according to the target control strategy to obtain the target control result.

[0119] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the terminal device described above can be referred to the corresponding process in the aforementioned method embodiment for controlling the bonding copper wire device based on the Internet of Things and 5G network, and will not be repeated here.

[0120] This invention also provides a storage medium for computer-readable storage, wherein the storage medium stores one or more programs that can be executed by one or more processors to implement the steps of any of the methods for controlling a copper wire bonding device based on the Internet of Things and 5G networks as provided in the specification of this invention.

[0121] The storage medium can be an internal storage unit of the terminal device described in the foregoing embodiments, such as the hard drive or memory of the terminal device. Alternatively, the storage medium can be an external storage device of the terminal device, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the terminal device.

[0122] It will be understood by those skilled in the art that all or some of the steps, systems, or apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware embodiments, the division between functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0123] It should be understood that the term "and / or" as used in this specification and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations. It should be noted that, herein, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0124] The sequence numbers of the above embodiments of the present invention are merely for descriptive purposes and do not represent the superiority or inferiority of the embodiments. The above descriptions are only specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for controlling a copper wire bonding device based on the Internet of Things and 5G networks, characterized in that, The method includes: Obtain the target control result corresponding to the target device, and obtain the similar control result corresponding to the target control result through the Internet of Things; Obtain the first sensing data of the target device under the target sensor under the similar control result, and perform anomaly identification on the first sensing data to obtain the target anomaly data corresponding to the target sensor; The first sensing data is removed based on the target abnormal data to obtain the second sensing data; Based on the second sensing data, correlation analysis is performed to obtain the target relationships between the target sensors. The target device obtains the current sensing data corresponding to the target sensor under the target control result based on the 5G network. Based on the current sensing data and the target relationship, a target control strategy corresponding to the target device is determined, and the target device is controlled according to the target control strategy to obtain the target control result.

2. The method according to claim 1, characterized in that, The step of obtaining a similar control result corresponding to the target control result through the Internet of Things includes: The target control result is decomposed into data to obtain the first control parameter corresponding to the target control result; Obtain the first sub-parameter corresponding to the first control parameter under the first parameter type, and obtain the first nearest neighbor parameter corresponding to the first sub-parameter in the database; Obtain the second nearest neighbor parameter corresponding to the first nearest neighbor parameter, and determine the mutual nearest neighbor parameter between the first sub-parameter and the first nearest neighbor parameter based on the first nearest neighbor parameter and the second nearest neighbor parameter; The first weight corresponding to the first sub-parameter is determined based on the mutual nearest neighbor parameter, and the first density information corresponding to the first sub-parameter is determined based on the first weight combined with the first nearest neighbor parameter and the mutual nearest neighbor parameter. Obtain the second density information corresponding to the first nearest neighbor parameter, and determine the first similarity between the first sub-parameter and the first nearest neighbor parameter based on the first density information and the second density information combined with the first nearest neighbor parameter and the mutual nearest neighbor parameter; Obtain the second control parameter corresponding to the first nearest neighbor parameter from the database, and calculate the second similarity between the second sub-parameter corresponding to the first control parameter under the second parameter type and the third sub-parameter corresponding to the second control parameter under the second parameter type. The target similarity between the first control parameter and the second control parameter is determined by fusing the first similarity and the second similarity. Based on the target similarity, a third control parameter associated with the first control parameter is determined, and based on the third control parameter, the similar control result corresponding to the target control result is determined.

3. The method according to claim 2, characterized in that, The step of determining the first density information corresponding to the first sub-parameter based on the first weight, combined with the first nearest neighbor parameter and the mutual nearest neighbor parameter, includes: Calculate the first distance between the first sub-parameter and any fourth sub-parameter among the mutual nearest neighbor parameters, and determine the first sub-density corresponding to the first sub-parameter based on the first distance and the first weight; The target set is obtained by performing a difference calculation on the first nearest neighbor parameter and the mutual nearest neighbor parameter, wherein the data in the target set belongs to the first nearest neighbor parameter but does not belong to the mutual nearest neighbor parameter; Calculate the second distance between the first sub-parameter and any fifth sub-parameter in the target set, and determine the second sub-density corresponding to the first sub-parameter based on the second distance; Obtain all sub-parameters corresponding to the first parameter type from the database, and remove the first nearest neighbor parameter from all sub-parameters to obtain the remaining sub-parameters; Calculate the third distance between the first sub-parameter and any sixth sub-parameter among the remaining sub-parameters, and determine the third sub-density corresponding to the first sub-parameter based on the third distance; The first density information corresponding to the first sub-parameter is determined by fusing the first sub-density, the second sub-density, and the third sub-density.

4. The method according to claim 1, characterized in that, The target sensor includes at least a first sensor and a second sensor, and the first sensing data includes at least first data corresponding to the first sensor and second data corresponding to the second sensor. The step of obtaining target abnormal data corresponding to the target sensor by performing anomaly identification on the first sensing data includes: The first sub-data corresponding to the first sensor at the target time is obtained from the first data, and the second sub-data corresponding to the second sensor at the target time is obtained from the second data; Obtain the first frequency corresponding to the first sub-data from the first data and obtain the second frequency corresponding to the second sub-data from the second data; The target entropy value corresponding to the target sensor at the target time is determined based on the first sub-data and the second sub-data combined with the first frequency and the second frequency. Determine the first anomaly characterization value corresponding to the target sensor at the target time based on the target entropy value; Obtain the first discrete representation value corresponding to the first sub-data from the first data, and obtain the second discrete representation value corresponding to the second sub-data from the second data; Determine the second anomaly characterization value corresponding to the target sensor at the target time based on the first discrete characterization value and the second discrete characterization value; The target anomaly characterization value corresponding to the target sensor at the target time is determined based on the first anomaly characterization value and the second anomaly characterization value. Based on the target anomaly characterization value, anomaly identification is performed on the first sensing data to obtain the target anomaly data corresponding to the target sensor.

5. The method according to claim 4, characterized in that, Obtaining the first discrete representation value corresponding to the first sub-data from the first data includes: Obtain the first nearest neighbor data corresponding to the first sub-data from the first data, and obtain the relevant data corresponding to the first sub-data as the nearest neighbor data; The first neighborhood data corresponding to the first sub-data is determined based on the relevant data and the first nearest neighbor data. Determine the target nearest neighbor range, and obtain the first distance between the first sub-data and the first nearest neighbor data based on the target nearest neighbor range; Obtain the first expected distance corresponding to the target nearest neighbor range, and determine the first distance difference corresponding to the first sub-data under the target nearest neighbor range based on the first distance and the first expected distance; Based on the first distance difference, determine the second distance difference corresponding to the first sub-data in the entire target nearest neighbor range; Based on the first nearest neighbor data, determine the first representation parameter corresponding to the first sub-data, and obtain the second representation parameter corresponding to each third sub-data in the first neighborhood data; The bandwidth value between the first sub-data and the third sub-data is determined based on the first characterization parameter and the second characterization parameter; The kernel density estimate corresponding to the first sub-data is determined based on the bandwidth value and the first neighborhood data. The first discrete representation value corresponding to the first sub-data is determined based on the second distance difference and the kernel density estimate. The first discrete representation value is obtained according to the following formula: Among them, disc i sum_diff represents the first discrete representation value corresponding to the i-th first sub-data point. i This represents the second distance difference corresponding to the i-th first sub-data item, count(all) i ) represents the first neighborhood data all corresponding to the i-th first sub-data. i The number of data, m i Let m represent the first representation parameter corresponding to the i-th first sub-data point. j Let pi represent the second characterization parameter corresponding to the j-th third sub-data in the first neighborhood data corresponding to the i-th first sub-data, where pi represents an irrational number, d represents a positive integer, exp represents an exponential function, and dis represents a distance function.

6. The method according to claim 5, characterized in that, The step of determining the first representation parameter corresponding to the first sub-data based on the first nearest neighbor data includes: Obtain the nearest neighbor distance between the first sub-data and the first nearest neighbor data; Count the number of targets corresponding to the first nearest neighbor data; The first characterization parameter corresponding to the first sub-data is determined based on the nearest neighbor distance and the number of targets.

7. The method according to claim 1, characterized in that, The step of obtaining the target relationship between the target sensors by performing correlation analysis based on the second sensing data includes: The first sensor and the second sensor are obtained from the target sensor, and the third data corresponding to the first sensor and the fourth data corresponding to the second sensor are obtained from the second sensor data; The third and fourth data are combined according to the collection time to obtain initial combined data, and the probability distribution corresponding to each sub-combined data is determined based on the initial combined data; Based on the probability distribution, feature selection is performed from the initial combined data to obtain the target features corresponding to the initial combined data; Based on the target features, the target entropy information corresponding to the initial combined data is determined, and the target entropy information is used to determine the target function corresponding to the initial combined data. The optimal solution corresponding to the objective function is obtained according to the Lagrange multiplier method, and the correlation function corresponding to the first sensor and the second sensor is obtained according to the optimal solution; The target relationship between the first sensor and the second sensor is determined based on the correlation function.

8. A system for controlling a copper wire bonding device based on the Internet of Things and 5G networks, characterized in that, include: The data acquisition module is used to obtain the target control result corresponding to the target device, and to obtain the similar control result corresponding to the target control result through the Internet of Things; An anomaly identification module is used to obtain the first sensing data of the target device under the target sensor under the similar control results, and to perform anomaly identification on the first sensing data to obtain the target anomaly data corresponding to the target sensor. The data removal module is used to remove data from the first sensing data based on the target abnormal data to obtain the second sensing data; The correlation analysis module is used to perform correlation analysis based on the second sensing data to obtain the target relationships between the target sensors. The data acquisition module is used to obtain the current sensing data of the target sensor of the target device under the target control result based on the 5G network; The data control module is used to determine the target control strategy corresponding to the target device based on the current sensing data and the target relationship, and to perform device control on the target device according to the target control strategy to obtain the target control result.

9. A terminal device, characterized in that, The terminal device includes a processor and a memory; The memory is used to store computer programs; The processor is used to execute the computer program and, in executing the computer program, implement the method for controlling a bonding copper wire device based on the Internet of Things and 5G network as described in any one of claims 1 to 7.

10. A computer storage medium for computer storage, characterized in that, The computer storage medium stores one or more programs, which can be executed by one or more processors to implement the steps of the method for controlling a copper wire bonding device based on the Internet of Things and 5G network as described in any one of claims 1 to 7.

Images