Electromagnetic interference modeling method, device, and medium

Abstract

The application provides an electromagnetic interference modeling method, device and medium, the method comprising: obtaining a welding electromagnetic interference test data set, a preset power threshold and a preset pulse duration, the test data set comprising test data of multiple weldings in each round of welding; determining an average background noise power according to the test data set; determining model parameters of a welding electromagnetic interference model according to the test data set, the preset power threshold, the preset pulse duration and the average background noise power, the welding electromagnetic interference model being a time series model, and the model parameters comprising a pulse time interval, a pulse amplitude and a pulse duration; and constructing the welding electromagnetic interference model according to the model parameters and the average background noise power, so as to improve the accuracy of the welding electromagnetic interference model.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electromagnetic interference, in particular to an electromagnetic interference modeling method, device and medium. BACKGROUND

[0002] In the wireless communication network of the industrial Internet of Things (IIoT), electromagnetic interference (EMI) will affect the quality of IIoT wireless communication, and the EMI interference source comes from various industrial production equipment such as welding machines, motors and voltage regulators. In order to effectively solve the influence of industrial EMI, it is urgent to accurately model the EMI.

[0003] At present, the EMI modeling method includes two categories of physical modeling and statistical modeling. In the physical modeling, the ray tracing technology calculates different signals passing through different paths to reach the receiving point, which will be attenuated by various propagation phenomena in the transmission medium in the process, and a large amount of electromagnetic theory calculation is needed, which consumes too many resources and time, and is not suitable for modeling of industrial EMI. The equivalent circuit is easily disturbed by the outside world, and the diversity of industrial equipment makes it impossible to model the equivalent circuit of each device separately. The statistical modeling method does not measure and analyze a single type of equipment, which makes it impossible to obtain and predict industrial EMI at a certain time.

[0004] Therefore, the modeling method of industrial EMI in the prior art has certain limitations. SUMMARY

[0005] The purpose of the present application is to provide an electromagnetic interference modeling method, device and medium to solve the problem of the limitation of the modeling method of industrial EMI in the prior art.

[0006] To achieve the above purpose, the technical solutions adopted by the embodiments of the present application are as follows:

[0007] In a first aspect, the present application provides an electromagnetic interference modeling method, which comprises:

[0008] Obtaining a welding electromagnetic interference test data set, a preset power threshold and a preset pulse duration, the test data set comprising test data of multiple weldings in each welding cycle;

[0009] Determining the average background noise power according to the test data set;

[0010] Based on the test dataset, the preset power threshold, the preset pulse duration, and the average background noise power, the model parameters of the welding electromagnetic interference model are determined. The welding electromagnetic interference model is a time series model, and the model parameters include the pulse time interval, pulse amplitude, and pulse duration.

[0011] The welding electromagnetic interference model is constructed based on the model parameters and the average background noise power.

[0012] Secondly, embodiments of this application provide an electromagnetic interference modeling apparatus, the apparatus comprising:

[0013] The acquisition module is used to acquire the test dataset, preset power threshold and preset pulse duration of welding electromagnetic interference. The test dataset includes test data from multiple welding operations in each round of welding.

[0014] The determination module is used to determine the average background noise power based on the test dataset;

[0015] The determining module is further configured to determine the model parameters of the welding electromagnetic interference model based on the test dataset, the preset power threshold, the preset pulse duration, and the average background noise power. The welding electromagnetic interference model is a time series model, and the model parameters include the pulse time interval, pulse amplitude, and pulse duration.

[0016] A construction module is used to construct the welding electromagnetic interference model based on the model parameters and the average background noise power.

[0017] Thirdly, embodiments of this application provide a computer device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the steps of the electromagnetic interference modeling method as described in the first aspect above.

[0018] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the electromagnetic interference modeling method described in the first aspect above.

[0019] The beneficial effects of this application are:

[0020] This application provides an electromagnetic interference (EMI) modeling method, device, and apparatus. By performing frequency and time domain measurements on welding EMI, a test dataset of welding EMI is obtained, along with a preset power threshold and preset pulse duration. Based on the test dataset, it is determined that the welding EMI exhibits a specific pulse pattern in the time domain for each welding cycle. Based on the test data from the non-welding period during the entire welding EMI measurement process, the average background noise power is determined. Then, based on the test dataset, preset power threshold, preset pulse duration, and average background noise power, the pulse time interval, pulse amplitude, and pulse duration of the welding EMI are determined. These pulse time interval, pulse amplitude, and pulse duration are used as model parameters for the welding EMI model. Based on the model parameters and the average background noise power, and since the welding EMI exhibits a specific pulse pattern in the time domain for each welding cycle, the welding EMI is modeled according to a time series, accurately constructing the welding EMI model. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 A flowchart illustrating the electromagnetic interference modeling method provided in this application embodiment;

[0023] Figure 2 This is a schematic diagram of the structure of the welding electromagnetic interference measuring device provided in the embodiments of this application;

[0024] Figure 3 This is a schematic diagram of the environment for measuring welding electromagnetic interference provided in an embodiment of this application; Figure 4 Time-domain measurement diagram of welding electromagnetic interference for single-wheel stud welding provided in an embodiment of this application;

[0025] Figure 5 Time-domain measurement diagram of welding electromagnetic interference for five-wheel stud welding provided in the embodiments of this application;

[0026] Figure 6 A time-domain measurement diagram of welding electromagnetic interference for single-wheel arc welding provided in an embodiment of this application;

[0027] Figure 7 Time-domain measurement diagram of welding electromagnetic interference of five-round arc welding provided in the embodiments of this application;

[0028] Figure 8 A schematic diagram of a welding electromagnetic interference model provided in an embodiment of this application;

[0029] Figure 9 A flowchart illustrating the process of determining the model parameters of a welding electromagnetic interference model using the electromagnetic interference modeling method provided in this application embodiment;

[0030] Figure 10 A flowchart illustrating the process of determining the electromagnetic interference pulse sequence of each welding round in the electromagnetic interference modeling method provided in the embodiments of this application;

[0031] Figure 11 A schematic flowchart illustrating the process of determining the power of each welding electromagnetic interference pulse in each round of welding using the electromagnetic interference modeling method provided in this application embodiment;

[0032] Figure 12 A schematic diagram illustrating the process of determining the target welding electromagnetic interference pulse sequence, target peak power, and target delay time in the electromagnetic interference modeling method provided in this application embodiment;

[0033] Figure 13 A schematic diagram illustrating the process of determining the target peak power and target delay time of each welding electromagnetic interference pulse using the electromagnetic interference modeling method provided in this application embodiment.

[0034] Figure 14 Another flowchart illustrating the process of determining the model parameters of the welding electromagnetic interference model using the electromagnetic interference modeling method provided in this application embodiment;

[0035] Figure 15 A flowchart illustrating the verification of the welding electromagnetic interference model for the electromagnetic interference modeling method provided in this application embodiment;

[0036] Figure 16 A graph showing the cumulative distribution function of the absolute error of the welding electromagnetic interference model provided in the embodiments of this application;

[0037] Figure 17 A modular structure diagram of the electromagnetic interference modeling device provided in the embodiments of this application;

[0038] Figure 18 A schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the accompanying drawings in this application are for illustrative and descriptive purposes only and are not intended to limit the scope of protection of this application. Furthermore, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed or implemented simultaneously. In addition, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.

[0040] Furthermore, the described embodiments are merely some, not all, of the embodiments of this application. The components of the embodiments of this application described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0041] It should be noted that the term "comprising" will be used in the embodiments of this application to indicate the presence of the features declared thereafter, but does not exclude the addition of other features.

[0042] Currently, EMI modeling methods can be broadly categorized into physical modeling and statistical modeling. Physical modeling methods require extensive electromagnetic theory calculations, consuming excessive resources and time, making them unsuitable for industrial EMI modeling. Statistical modeling methods cannot perform measurement and analysis on a single type of industrial equipment, resulting in the inability to acquire and predict industrial EMI at a fixed time, thus exhibiting certain limitations.

[0043] This application proposes an electromagnetic interference (EMI) modeling method to address the aforementioned problems. It comprehensively measures welding EMI in the time domain to obtain a test dataset and determines the model parameters, including pulse time interval, pulse amplitude, and pulse duration. Based on these parameters, a welding EMI model is created, which is a time-series model. The model is then validated using the absolute error between the test dataset and the model, thereby improving its accuracy.

[0044] Figure 1This is a flowchart illustrating the electromagnetic interference modeling method provided in an embodiment of this application. The execution subject of this method can be any computer device with computing power. Figure 1 As shown, the method includes:

[0045] S101. Obtain the test dataset of welding electromagnetic interference, the preset power threshold and the preset pulse duration. The test dataset includes test data from multiple welding operations in each round of welding.

[0046] Optionally, during the welding process, the welding equipment is programmed by a Programmable Logic Controller (PLC) to automate the welding task. The PLC sends instructions to the welding equipment based on preset welding paths and parameters to precisely control the equipment's movement trajectory, including the welding start point, number of welds, welding speed, and welding direction. The welding equipment generates welding EMI during the welding process; this welding equipment can be a welding robot. Welding technologies include stud welding (SW) and arc welding (AW). SW EMI is the welding EMI generated during stud welding, while AW EMI is the welding EMI generated during arc welding.

[0047] Figure 2 This is a schematic diagram of the structure of the welding electromagnetic interference measuring device provided in the embodiments of this application, with reference to... Figure 2 The equipment for measuring electromagnetic interference in welding can include a directional log-periodic antenna, a signal analyzer, and a computer, wherein the signal analyzer is connected to both the directional log-periodic antenna and the computer. Figure 3 This is a schematic diagram of the environment for measuring welding electromagnetic interference provided in an embodiment of this application, with reference to... Figure 3 The test environment was a welding workshop in an automobile manufacturing plant, and the measurements were... Figure 3 The welding equipment used in production line stations L100-L110 is for welding the left outer side panel of a car. L100 uses arc welding, and L110 uses stud welding.

[0048] Continue to refer to Figure 2 and Figure 3 To ensure that the welding EMI generated by the welding equipment is not affected by EMI generated by other welding equipment, the received interference mode is a single type of welding EMI, and the test dataset L of SWEMI is obtained. s (t) and AW EMI test dataset L A(t), where the welding EMI test dataset includes test data from multiple welding operations in each round of welding. Specifically, during the measurement process, a directional log-periodic antenna is aligned with the welding nozzle worksheet of the welding equipment. The parameters of the signal analyzer are adjusted to accurately collect the electromagnetic waves radiated during the welding process, obtaining the welding EMI test dataset and storing it in a computer. The number of scanning points of the signal analyzer is configured to the maximum value of 32001 points, achieving the highest measurement fineness of the signal analyzer. Frequency domain measurement of the welding EMI is performed, and parameters such as the scanning bandwidth, resolution bandwidth, and video bandwidth of the signal analyzer are adjusted to accurately measure the frequency domain information of the welding EMI. Then, time domain measurement of the welding EMI is performed, and the measurement duration is controlled according to the preset measurement accuracy to obtain the morphology of the welding EMI. For example, when the preset measurement accuracy is a scanning time interval of 1ms, the measurement time is 32 seconds.

[0049] Stud welding is a welding technique that uses an electric arc to firmly attach a metal stud to a metal workpiece. During the welding process, the arc discharge can cause SWEMI. Figure 4 The time-domain measurement diagram of the welding electromagnetic interference of single-wheel stud welding provided in the embodiments of this application is as follows: Figure 4 As shown, the single-wheel stud welding process includes multiple welding operations. For example, the total duration of the single-wheel SW EMI is approximately 27 seconds, divided into 6 welding operations. Each stud welding operation consists of a very short pulse, lasting only about 0.11 seconds. Figure 5 The time-domain measurement diagram of the welding electromagnetic interference of the five-wheel stud welding provided in the embodiments of this application is as follows: Figure 5 As shown, the test dataset L of SW EMI s (t) includes test data from 5 rounds of welding, with 6 welding operations in each round. The pulse time interval, shape, and power of the same pulse are consistent in the time-domain measurement graphs of each round, indicating that SWEMI has deterministic characteristics, that is, SWEMI has a specific pulse shape in the time domain of each round of welding.

[0050] Arc welding is a welding process that uses an electric arc to form a fusion between two metal workpieces. When current passes through the electrode, a strong electric arc is generated between the electrode and the workpiece, thereby generating heat and AWEMI. Figure 6 The time-domain measurement diagram of welding electromagnetic interference for single-wheel arc welding provided in the embodiments of this application is as follows: Figure 6 As shown, the single-round arc welding process also includes multiple welding operations. For example, the total duration of a single-round AW EMI is approximately 18 seconds, divided into 6 welding operations, each consisting of a pulse lasting 1.5 seconds. Figure 7 The time-domain measurement diagram of the welding electromagnetic interference of five-round arc welding provided in the embodiments of this application is as follows: Figure 7 As shown, the test dataset L of AW EMIA (t) includes test data from 8 rounds of welding, with 6 welding operations in each round. Since arc welding is a sliding welding mode, the pulse shape of the word AWEMI showed small fluctuations. However, the pulse time interval, shape, and power of the same pulse remained consistent in the time domain measurement diagrams of each round, indicating that AWEMI also has deterministic characteristics, that is, AWEMI also has a specific pulse shape in the time domain of each round of welding.

[0051] Computer equipment acquires the test dataset L of measured SWEMI. s (t) and AW EMI test dataset L A (t), and receives a preset power threshold T and a preset pulse duration D input by the user. The preset power threshold T may include a preset power threshold T for stud welding. S The preset power threshold T for arc welding A For example, the preset power threshold T for stud welding S The preset power threshold T for arc welding can be -111dBm. A It can be -113dBm, and the preset pulse duration D can be 50 milliseconds.

[0052] S102. Determine the average background noise power based on the test dataset.

[0053] Optionally, continue to refer to Figure 5 According to the SW EMI test dataset L s (t), based on the test data during the non-soldering period of the entire SWEMI measurement process, the average background noise power σ of the stud weld is calculated. s 2 Accordingly, continue to refer to Figure 7 According to AW EMI's test dataset L A (t), based on the test data during the non-welding period of the entire AW EMI measurement process, the average background noise power σ of arc welding is calculated. A 2 .

[0054] S103. Based on the test dataset, preset power threshold, preset pulse duration, and average background noise power, determine the model parameters of the welding electromagnetic interference model. The welding electromagnetic interference model is a time series model, and the model parameters include pulse time interval, pulse amplitude, and pulse duration.

[0055] Optionally, since SW EMI exhibits specific pulse patterns in the time domain during each welding round, each round of SW EMI can be modeled as a time series model, resulting in individual SW EMI models. Before constructing each SW EMI model, the SW EMI test dataset L is used as a reference.s (t) Preset power threshold T for stud welding S The preset pulse duration D and the average background noise power σ of stud welding s 2 The model parameters for each SWEMI model are determined. These parameters include the pulse time interval, pulse amplitude, and pulse duration. The pulse time interval is characterized by the delay time of each individual SWEMI pulse relative to the first SWEMI pulse in a welding cycle; the pulse amplitude is characterized by the peak power of each individual SWEMI pulse in a welding cycle; and the duration is characterized by the sequence of individual SWEMI pulses in a welding cycle.

[0056] Accordingly, since AW EMI also exhibits specific pulse patterns in the time domain of each welding round, each round of AW EMI can be modeled as a time series model, resulting in individual AW EMI models. Before constructing each AW EMI model, the AW EMI test dataset L is used... A (t) Preset power threshold T for arc welding A The preset pulse duration D and the average background noise power σ of arc welding A 2 The model parameters for each AW EMI model are determined. These parameters include the pulse time interval, pulse amplitude, and pulse duration. The pulse time interval is characterized by the delay time of each individual AW EMI pulse relative to the first AW EMI pulse in a welding cycle; the pulse amplitude is characterized by the peak power of each individual AW EMI pulse in a welding cycle; and the duration is characterized by the sequence of individual AW EMI pulses in a welding cycle.

[0057] S104. Based on the model parameters and average background noise power, construct a welding electromagnetic interference model.

[0058] Alternatively, based on the average background noise power σ of the stud weld s 2 In addition to the pulse time interval, pulse amplitude, and pulse duration parameters in the SW EMI model, a single-round SW EMI model S(t) is constructed based on the following expression:

[0059]

[0060] Where S(t) is a single-round SWEMI model, N S P represents the number of welds in a single-wheel stud welding process. i S This represents the peak power of each individual SWEMI pulse during a welding cycle. This refers to the time-series pulses of each individual SWEMI event during a round of welding, i.e., the sequence of individual SWEMI pulses during a round of welding. For each individual SWEMI pulse during a round of welding, t i σ represents the delay time of each individual SWEMI pulse relative to the first SWEMI pulse during a welding cycle. S 2 This represents the average background noise power of the stud weld.

[0061] The single-wheel SWEMI model S(t) is extended to construct an arbitrary-wheel SWEMI model I based on the following expression. S (t):

[0062]

[0063] Among them, I S S(t) represents an arbitrary-wheel SWEMI model, and S(t) represents a single-wheel SWEMI model. j Let F be the delay time for the j-th round of stud welding, and F be the number of welding rounds of stud welding. S 2 This represents the average background noise power of the stud weld.

[0064] Accordingly, based on the average background noise power σ of arc welding A 2 In addition to the pulse time interval, pulse amplitude, and pulse duration parameters in the AWEMI model, a single-round AWEMI model A(t) is constructed based on the following expression:

[0065]

[0066] Where A(t) is a single-wheel AW EMI model, N A This refers to the number of welds in a single-round arc welding process. This represents the peak power of each individual AW EMI pulse during a welding cycle. This refers to the time-series pulses of each individual AW EMI during a round of welding, i.e., the pulse sequence of each individual AW EMI during a round of welding. For each single AW EMI pulse during a round of welding, t i σ represents the delay time of each individual AW EMI pulse relative to the first AW EMI pulse during a welding cycle. A 2 This represents the average background noise power of the stud weld.

[0067] The single-wheel AWEMI model A(t) is extended to construct an arbitrary-wheel AWEMI model I based on the following expression. A (t):

[0068]

[0069] Among them, I A A(t) represents the AW EMI model of any wheel, and A(t) represents the AW EMI model of a single wheel. j Let G be the delay time for the j-th round of arc welding, and G be the number of arc welding rounds. A 2 This represents the average background noise power of arc welding.

[0070] Figure 8 A schematic diagram of the welding electromagnetic interference model provided in the embodiments of this application is shown below. Figure 8 The image shows schematic diagrams of two welding EMI models after removing the average background noise power. Figure 8 (a) The left side is a schematic diagram of the single-round SWEMI model S(t) after removing the average background noise power. Figure 8 (a) The right side shows the second single-pulse sequence in the single-round SWEMI model S(t) after removing the average background noise power. The diagram is shown below. Figure 4 The single-round SWEMI model S(t) and the SWEMI test dataset L S The interference characteristics of the single-wheel SWEMI in (t) are basically consistent, indicating that the constructed single-wheel SWEMI model S(t) is relatively accurate.

[0071] Figure 8 (b) The left side is a schematic diagram of the single-round AW EMI model A(t) after removing the average background noise power. Figure 8 (b) The right side shows the second single-pulse sequence in the single-round AW EMI model A(t) after removing the average background noise power. The diagram is shown below. Figure 6 The single-round AWEMI model A(t) and the AWEMI test dataset L A The interference characteristics of the single-wheel AW EMI in (t) are basically consistent, indicating that the constructed single-wheel AW EMI model A(t) is relatively accurate.

[0072] In this embodiment, a test dataset of welding electromagnetic interference (EMI) is obtained by performing frequency and time domain measurements. A preset power threshold and preset pulse duration are also acquired. Based on the test dataset, it is determined that the EMI exhibits a specific pulse pattern in the time domain for each welding round. The average background noise power is determined based on test data from non-welding periods throughout the EMI measurement process. Furthermore, the pulse time interval, pulse amplitude, and pulse duration of the EMI are determined using the test dataset, preset power threshold, preset pulse duration, and average background noise power. These parameters are then used as model parameters for the EMI model. Based on the model parameters and the average background noise power, and considering the specific pulse pattern of the EMI in the time domain for each welding round, the EMI is modeled according to a time series, thus accurately constructing the EMI model.

[0073] The following is a detailed explanation of the process of determining the model parameters of the welding electromagnetic interference model based on the test dataset, preset power threshold, preset pulse duration, and average background noise power.

[0074] Figure 9 A flowchart illustrating the process of determining the model parameters of a welding electromagnetic interference model using the electromagnetic interference modeling method provided in this application embodiment is shown below. Figure 9 As shown, step S103 above, which involves determining the model parameters of the welding electromagnetic interference model based on the test dataset, preset power threshold, preset pulse duration, and average background noise power, includes:

[0075] S901. Based on the test dataset, preset power threshold, and preset pulse duration, determine the electromagnetic interference pulse sequence for each round of welding.

[0076] Optionally, continue to refer to Figure 4 and Figure 5 SW EMI's test dataset L s (t) includes not only test data from multiple welding operations during each stud welding process, but also background noise. To avoid the influence of some high-power background noise on the determination of SWEMI model parameters, the test dataset L of SWEMI is used. s (t) Preset power threshold T for stud welding S And the preset pulse duration D, for the SWEMI test dataset L s (t) is used to screen and determine the SWEMI pulse sequence in each round of stud welding.

[0077] Among them, the SWEMI pulse sequence in each stud welding process is represented as follows: N sThis refers to the number of welds in a single-wheel stud welding process.

[0078] Accordingly, continue to refer to Figure 6 and Figure 7 AW EMI's test dataset L A (t) includes not only test data from multiple welds in each arc welding process, but also background noise. To avoid the influence of some high-power background noise on the determination of AW EMI model parameters, it is necessary to use the AW EMI test dataset L. A (t) Preset power threshold T for arc welding A And the preset pulse duration D, for the AW EMI test dataset L A (t) is used to screen and determine the AW EMI pulse sequence in each round arc welding process.

[0079] The AW EMI pulse sequence in each round arc welding process is represented as follows: N A This refers to the number of welding passes in a single-round arc welding process.

[0080] S902. Based on the test dataset, welding electromagnetic interference pulse sequence, and average background noise power, determine the power of each welding electromagnetic interference pulse in each round of welding.

[0081] Optionally, based on the SW EMI test dataset L s (t), each SWEMI pulse sequence in each round of stud welding and the average background noise power σ of stud welding. S 2 Test dataset L with SWEMI removed s The influence of background noise in (t) on the power of each SWEMI pulse in each SWEMI pulse sequence during stud welding is obtained, and the power of each SWEMI pulse after removing the background noise of stud welding is obtained. Accordingly, based on the AW EMI test dataset L A (t), each AW EMI pulse sequence and the average background noise power σ of arc welding in each arc welding process. A 2 Test dataset L with AW EMI removed A The influence of background noise in (t) on the power of each AWEMI pulse in each AWEMI pulse sequence during each round of arc welding is obtained, and the power of each AWEMI pulse after removing the background noise of arc welding is obtained.

[0082] S903. Based on the welding electromagnetic interference pulse sequence and the power of each welding electromagnetic interference pulse, perform power normalization processing on each welding electromagnetic interference pulse sequence to determine the target welding electromagnetic interference pulse sequence, the target peak power of each welding electromagnetic interference pulse in each welding round, and the target delay time of each welding electromagnetic interference pulse in each welding round.

[0083] Optionally, based on the SW EMI pulse sequences in each stud welding operation and the power of each SW EMI pulse after removing the background noise of the stud welding, power normalization is performed on each SW EMI pulse sequence to align the SW EMI pulse sequences in each stud welding operation, thereby obtaining the target SW EMI pulse sequence, the target peak power, and the target delay time of each SW EMI pulse in each stud welding operation. Correspondingly, based on the AW EMI pulse sequences in each arc welding operation and the power of each AW EMI pulse after removing the background noise of the arc welding, power normalization is performed on each AW EMI pulse sequence to align the AW EMI pulse sequences in each arc welding operation, thereby obtaining the target AW EMI pulse sequence, the target peak power, and the target delay time of each AW EMI pulse in each arc welding operation.

[0084] S904. Based on the target welding electromagnetic interference pulse sequence, the target peak power of each welding electromagnetic interference pulse in each welding round, and the target delay time of each welding electromagnetic interference pulse in each welding round, determine the model parameters of the welding electromagnetic interference model.

[0085] Optionally, based on the target SWEMI pulse sequence of each target in each round of stud welding, the target peak power of each target SWEMI pulse in each round of stud welding, and the target delay time of each target SWEMI pulse in each round of stud welding, the mean pulse duration, pulse amplitude, and pulse time interval in the model parameters of the SWEMI model are determined respectively, so as to construct the mean time series model of SWEMI based on the mean model parameters.

[0086] Accordingly, based on the target AWEMI pulse sequence of each target in each round of arc welding, the target peak power of each target AWEMI pulse in each round of arc welding, and the target delay time of each target AWEMI pulse in each round of arc welding, the mean pulse duration, pulse amplitude, and pulse time interval in the model parameters of the AWEMI model are determined respectively, so as to construct the mean time series model of AWEMI based on the mean model parameters.

[0087] In this embodiment, based on the test dataset, preset power threshold, and preset pulse duration, the welding electromagnetic interference (EMI) pulse sequences for each round of welding are selected. Then, based on the test dataset, the EMI pulse sequences, and the average background noise power, the influence of background noise on the power of each EMI pulse in the test dataset is removed, thus determining the power of each EMI pulse in each round of welding. Based on each EMI pulse sequence and its power after background noise removal, power normalization is performed on each EMI pulse sequence to determine the target EMI pulse sequence, the target peak power, and the target delay time for each EMI pulse in each round of welding. This, in turn, determines the mean pulse duration, pulse amplitude, and pulse time interval in the model parameters of the welding EMI model. This allows for the construction of a mean time series model of welding EMI based on the mean model parameters, improving model accuracy.

[0088] The following is a detailed explanation of the process of determining the electromagnetic interference pulse sequence for each round of welding based on the test dataset, preset power threshold, and preset pulse duration.

[0089] Figure 10 A flowchart illustrating the determination of the electromagnetic interference pulse sequence for each welding round in the electromagnetic interference modeling method provided in this application embodiment is shown below. Figure 10 As shown, step S901 above, which determines the electromagnetic interference pulse sequence for each round of welding based on the test dataset, preset power threshold, and preset pulse duration, includes:

[0090] S1001. Based on the test dataset and the preset power threshold, determine each target pulse in each round of welding, where the power of the target pulse is greater than the preset power threshold.

[0091] Optionally, continue to refer to Figure 4 and Figure 5 The power of background noise during stud welding is much lower than the power of SWEMI pulses. According to the SWEMI test dataset L... s (t) and the preset power threshold T for stud welding S By power screening, the target SWEMI pulses in each round of stud welding are identified. Specifically, pulses with power greater than the preset power threshold T for stud welding are selected. S The pulse is used as the target SWEMI pulse in stud welding, that is, the pulse with power greater than -111dBm is used as the target SWEMI pulse in stud welding.

[0092] Accordingly, continue to refer to Figure 6 and Figure 7The power of background noise during arc welding is also much lower than the power of AW EMI pulses, according to the AW EMI test dataset L. A (t) and the preset power threshold T for arc welding A By power screening, the target AW EMI pulses in each round of arc welding are determined. Specifically, the pulse power is set to be greater than the preset power threshold T for arc welding. A The pulse is used as the target AW EMI pulse in arc welding, that is, the pulse with power greater than -113dBm is used as the target AW EMI pulse in arc welding.

[0093] S1002. Based on the test dataset, the preset pulse duration, and the target pulses in each round of welding, determine the welding electromagnetic interference pulse sequence in each round of welding. The pulse duration of each welding electromagnetic interference pulse in the welding electromagnetic interference pulse sequence is greater than the preset pulse duration.

[0094] Optionally, continue to refer to Figure 4 and Figure 5 The duration of background noise during stud welding is much shorter than the pulse duration of SWEMI pulses. According to the SWEMI test dataset L... s (t) and a preset pulse duration D are used to filter the pulse durations and determine the SWEMI pulses in each round of stud welding from the target SWEMI pulses in each round of stud welding, thereby determining the SWEMI pulse sequence in each round of stud welding. Specifically, pulses with a pulse duration greater than the preset pulse duration D are taken as SWEMI pulses in stud welding, that is, pulses with a pulse duration greater than 50 milliseconds are taken as SWEMI pulses in stud welding, thus obtaining the SWEMI pulse sequence in each round of stud welding.

[0095] Accordingly, continue to refer to Figure 6 and Figure 7 The duration of background noise during arc welding is much shorter than the pulse duration of AWEMI pulses. According to the AWEMI test dataset L... A (t) and a preset pulse duration D are used to filter the pulse durations to determine the AWEMI pulses in each round of arc welding from the target AWEMI pulses in each round of arc welding, thereby determining the AWEMI pulse sequence in each round of arc welding. Specifically, pulses with a pulse duration greater than the preset pulse duration D are taken as AWEMI pulses in arc welding, that is, pulses with a pulse duration greater than 50 milliseconds are taken as AWEMI pulses in arc welding, thereby obtaining the AWEMI pulse sequence in each round of arc welding.

[0096] In this embodiment, firstly, based on the test dataset and a preset power threshold, target welding electromagnetic interference pulses in each round of welding are determined through power filtering. Pulses with power exceeding the preset power threshold are used as target pulses in each round of welding. Then, based on the test dataset and a preset pulse duration, welding electromagnetic interference pulses in each round of welding are determined from the target pulses in each round of welding through pulse duration filtering, thereby determining the sequence of welding electromagnetic interference pulses in each round of welding. Through power filtering and pulse duration filtering, background noise in the test dataset is eliminated, improving the accuracy of the welding electromagnetic interference pulse sequence in each round of welding.

[0097] The following is a detailed explanation of the process of determining the power of each welding electromagnetic interference pulse in each round of welding based on the test dataset, the welding electromagnetic interference pulse sequence, and the average background noise power.

[0098] Figure 11 A flowchart illustrating the determination of the power of each welding electromagnetic interference pulse in each round of welding using the electromagnetic interference modeling method provided in this application embodiment is shown below. Figure 11 As shown, step S902 above, which involves determining the power of each welding electromagnetic interference pulse in each round of welding based on the test dataset, the welding electromagnetic interference pulse sequence, and the average background noise power, includes:

[0099] S1101. Based on the test dataset, determine the initial power of each welding electromagnetic interference pulse in the welding electromagnetic interference pulse sequence.

[0100] Optionally, based on the SW EMI test dataset L s (t) and the SW EMI pulse sequences during each stud welding process, the initial power of each SW EMI pulse is determined. The initial power value of the SW EMI pulse is affected by the background noise of the stud welding. The unit of the initial power of the SW EMI pulse is dBm. Accordingly, based on the AW EMI test dataset L... A (t) and each AW EMI pulse sequence in each round of arc welding, determine the initial power of each AW EMI pulse. The initial power value of the AW EMI pulse will be affected by the background noise of arc welding. The unit of the initial power of the AW EMI pulse is dBm.

[0101] S1102. Convert the initial power of the welding electromagnetic interference pulse into the initial actual power of the welding electromagnetic interference pulse.

[0102] Optionally, the initial power of the SW EMI pulse is converted to the initial actual power of the SW EMI pulse through power value conversion, so that the unit of the initial actual power of the SW EMI pulse is dB; the initial power of the AW EMI pulse is converted to the initial actual power of the AW EMI pulse, so that the unit of the initial actual power of the AW EMI pulse is dB.

[0103] S1103. The difference between the initial actual power of the welding electromagnetic interference pulse and the average background noise power is taken as the actual power of the welding electromagnetic interference pulse.

[0104] Optionally, to remove the influence of stud solder background noise on the power of the SW EMI pulse, the average background noise power σ of the stud solder is subtracted from the initial actual power of the SW EMI pulse. S 2 The initial actual power of the SW EMI pulse and the average background noise power σ of the stud weld were obtained. S 2 The difference is taken as the actual power of the SW EMI pulse after removing the background noise from stud welding. Correspondingly, to remove the influence of arc welding background noise on the power of the AW EMI pulse, the average background noise power σ of arc welding is subtracted from the initial actual power of the AW EMI pulse. A 2 The initial actual power of the AW EMI pulse and the average background noise power σ of arc welding were obtained. A 2 The difference is used as the actual power of the AW EMI pulse after removing the background noise of arc welding.

[0105] S1104. Determine the power of the welding electromagnetic interference pulse based on its actual power.

[0106] Optionally, by power value conversion, the actual power of the SW EMI pulse is converted to the power of the SW EMI pulse, so that the unit of the SW EMI pulse power is dBm. The power of the SW EMI pulse is the power of the SW EMI pulse after removing the background noise of stud welding, thus eliminating the influence of stud welding background noise on the power of the SW EMI pulse. Correspondingly, the actual power of the AW EMI pulse is converted to the power of the AW EMI pulse, so that the unit of the AW EMI pulse power is dBm, thus eliminating the influence of arc welding background noise on the power of the AW EMI pulse.

[0107] In this embodiment, based on the test dataset and the welding electromagnetic interference (EMI) pulse sequences in each round of welding, the initial power of each EMI pulse is determined. Then, through power value conversion, the initial power of the EMI pulse is converted into its initial actual power. The average background noise power is subtracted from the initial actual power of the EMI pulse, and the difference between the initial actual power and the average background noise power is taken as the actual power of the EMI pulse. This actual power is then converted back into the actual power of the EMI pulse through power value conversion again. This eliminates the influence of background noise on the power of the EMI pulse.

[0108] The following is a detailed explanation of the process of determining the target welding electromagnetic interference pulse sequence, the target peak power of each welding electromagnetic interference pulse, and the target delay time of each welding electromagnetic interference pulse in each round of welding by performing power normalization on each welding electromagnetic interference pulse sequence and the power of each welding electromagnetic interference pulse in each round of welding.

[0109] Figure 12 A flowchart illustrating the process of determining the target welding electromagnetic interference pulse sequence, target peak power, and target delay time in the electromagnetic interference modeling method provided in this application embodiment is shown below. Figure 12 As shown, step S903 above, which involves normalizing the power of each welding electromagnetic interference pulse sequence based on the sequence and power of each pulse, and determining the target welding electromagnetic interference pulse sequence, the target peak power, and the target delay time of each pulse in each welding round, includes:

[0110] S1201. Determine the peak power of each welding electromagnetic interference pulse based on the sequence of each welding electromagnetic interference pulse and the power of each welding electromagnetic interference pulse.

[0111] Optionally, based on the SW EMI pulse sequence and power of each SW EMI pulse during stud welding, the maximum power of each SW EMI pulse is obtained through iteration, and the maximum power of each SW EMI pulse is taken as the peak power of each SW EMI pulse. Correspondingly, based on the AW EMI pulse sequence and power of each AW EMI pulse during arc welding, the maximum power of each AW EMI pulse is obtained through iteration, and the maximum power of each AW EMI pulse is taken as the peak power of each AW EMI pulse.

[0112] S1202. The quotient of the power peak value of the welding electromagnetic interference pulse and the welding electromagnetic interference pulse is used as the normalized welding electromagnetic interference pulse, thus obtaining the normalized welding electromagnetic interference pulse sequence.

[0113] Optionally, the SW EMI pulse is divided by its peak power to obtain the quotient of the two SW EMI pulses. This quotient is used as the normalized SW EMI pulse, and a normalized SW EMI pulse sequence is obtained based on each normalized SW EMI pulse. Similarly, the AW EMI pulse is divided by its peak power to obtain the quotient of the two AW EMI pulses. This quotient is used as the normalized AW EMI pulse, and a normalized AW EMI pulse sequence is obtained based on each normalized AW EMI pulse.

[0114] S1203. Based on the normalized welding electromagnetic interference pulses, determine the target peak power and target delay time of each welding electromagnetic interference pulse.

[0115] Optionally, based on each normalized SW EMI pulse, the target peak power of each SW EMI pulse is obtained by averaging its peak power; and the target delay time of each SW EMI pulse is obtained by averaging its delay time. Correspondingly, based on each normalized AW EMI pulse, the target peak power of each AW EMI pulse is obtained by averaging its peak power; and the target delay time of each AW EMI pulse is obtained by averaging its delay time.

[0116] S1204. Based on the normalized welding electromagnetic interference pulse sequence, determine the welding electromagnetic interference pulse sequence of each target in each round of welding.

[0117] Optionally, the average value of each normalized SW EMI pulse sequence in each round of stud welding is calculated to determine the average value of each normalized SW EMI pulse sequence in each round of stud welding, and this average value is used as the target SW EMI pulse sequence in each round of stud welding. Correspondingly, the average value of each normalized AW EMI pulse sequence in each round of arc welding is calculated to determine the average value of each normalized AW EMI pulse sequence in each round of arc welding, and this average value is used as the target AW EMI pulse sequence in each round of arc welding. Both the target SW EMI pulse sequence and the target AW EMI pulse sequence are mean pulse sequences.

[0118] In this embodiment, based on the welding electromagnetic interference (EMI) pulse sequence and its power in each welding round, the maximum power of each EMI pulse is obtained, and this maximum power is taken as the peak power of each EMI pulse. The quotient of the EMI pulse and its peak power is taken as the normalized EMI pulse, and a normalized EMI pulse sequence is obtained based on these normalized EMI pulses. The peak power and delay time of each normalized EMI pulse are averaged to obtain the target peak power and target delay time. The average value of each normalized EMI pulse sequence is calculated to determine the average value of each normalized EMI pulse sequence in each welding round, and this average value is taken as the target EMI pulse sequence in each welding round. The target peak power, target delay time, and target welding electromagnetic interference pulse sequence of the mean values ​​were determined so that a mean time series model of welding electromagnetic interference could be constructed based on the target peak power, target delay time, and target welding electromagnetic interference pulse sequence of the mean values.

[0119] The following is a detailed explanation of the process of determining the target peak power and target delay time of each welding electromagnetic interference pulse based on the normalized welding electromagnetic interference pulses.

[0120] Figure 13 A flowchart illustrating the process of determining the target peak power and target delay time of each welding electromagnetic interference pulse using the electromagnetic interference modeling method provided in this application embodiment is shown below. Figure 13As shown, step S1203 above, which involves determining the target peak power and target delay time of each welding electromagnetic interference pulse based on the normalized welding electromagnetic interference pulses, includes:

[0121] S1301. Based on each normalized welding electromagnetic interference pulse, determine the peak power of each normalized welding electromagnetic interference pulse and the delay time of each normalized welding electromagnetic interference pulse.

[0122] Optionally, based on each normalized SWEMI pulse sequence, the peak power and delay time of each normalized SWEMI pulse are extracted, wherein the peak power of the normalized SWEMI pulse is the maximum power of the normalized SWEMI pulse sequence. Correspondingly, based on each normalized AWEMI pulse sequence, the peak power and delay time of each normalized AWEMI pulse are extracted, wherein the peak power of the normalized AWEMI pulse is the maximum power of the normalized AWEMI pulse sequence.

[0123] S1302. The average value of the peak power of each normalized welding electromagnetic interference pulse in each round of welding is taken as the target peak power of each welding electromagnetic interference pulse in each round of welding.

[0124] Optionally, the average peak power of each normalized SW EMI pulse in each stud weld is calculated to determine the average peak power of each normalized SW EMI pulse in each stud weld, and this average value is used as the target peak power of each SW EMI pulse in each stud weld. Correspondingly, the average peak power of each normalized AW EMI pulse in each arc weld is calculated to determine the average peak power of each normalized AW EMI pulse in each arc weld, and this average value is used as the target peak power of each AW EMI pulse in each arc weld. Here, the target peak power is an average peak power.

[0125] S1303. The average delay time of each normalized welding electromagnetic interference pulse in each round of welding is taken as the target delay time of each welding electromagnetic interference pulse in each round of welding.

[0126] Optionally, the average delay time of each normalized SW EMI pulse in each stud weld is calculated to determine the average delay time of each normalized SW EMI pulse in each stud weld, and this average value is used as the target delay time of each SW EMI pulse in each stud weld. Correspondingly, the average delay time of each normalized AW EMI pulse in each arc weld is calculated to determine the average delay time of each normalized AW EMI pulse in each arc weld, and this average value is used as the target delay time of each AW EMI pulse in each arc weld. Here, the target delay time is an average delay time.

[0127] In this embodiment, the peak power and delay time of each normalized welding electromagnetic interference (EMI) pulse are extracted based on the normalized EMI pulses. The average peak power of each normalized EMI pulse in each welding round is calculated, and this average is used as the target peak power for each EMI pulse in each welding round. Similarly, the average delay time of each normalized EMI pulse in each welding round is calculated, and this average is used as the target delay time for each EMI pulse in each welding round. Through these average calculations, the mean peak power and mean delay time of each welding EMI pulse are determined.

[0128] The following section details the process of determining the model parameters of the welding electromagnetic interference model based on the target welding electromagnetic interference pulse sequence, the target peak power of each welding electromagnetic interference pulse in each welding round, and the target delay time of each welding electromagnetic interference pulse in each welding round.

[0129] Figure 14 Another flowchart illustrating the determination of model parameters for the welding electromagnetic interference model using the electromagnetic interference modeling method provided in this application embodiment is shown below. Figure 14 As shown, step S904 above, which involves determining the model parameters of the welding electromagnetic interference model based on the target welding electromagnetic interference pulse sequence, the target peak power of each welding electromagnetic interference pulse in each round of welding, and the target delay time of each welding electromagnetic interference pulse in each round of welding, includes:

[0130] S1401. Based on the electromagnetic interference pulse sequence of each target in each round of welding, determine the pulse duration of each target electromagnetic interference pulse in each round of welding.

[0131] Optionally, the duration is characterized by the pulse sequence of each single SW EMI pulse in a welding round, that is, the pulse duration of each target SW EMI pulse in each round of stud welding is determined based on the target SW EMI pulse sequence in each round of stud welding. Correspondingly, the duration is characterized by the pulse sequence of each single AW EMI pulse in a welding round, that is, the pulse duration of each target AW EMI pulse in each round of arc welding is determined based on the target AW EMI pulse sequence in each round of arc welding. Here, the pulse duration of each target SW EMI pulse and the pulse duration of each target AW EMI pulse are both average pulse durations.

[0132] S1402. Determine the pulse amplitude of each welding electromagnetic interference pulse in each welding round based on the target peak power of each welding electromagnetic interference pulse in each welding round.

[0133] Optionally, the pulse amplitude is characterized by the peak power of each individual SW EMI pulse in a welding round, i.e., the pulse amplitude of each SW EMI pulse in each round of stud welding is determined based on the target peak power of each SW EMI pulse in each round of stud welding. Correspondingly, the pulse amplitude is characterized by the peak power of each individual AW EMI pulse in a welding round, i.e., the pulse amplitude of each AW EMI pulse in each round of arc welding is determined based on the target peak power of each AW EMI pulse in each round of arc welding. Here, both the pulse amplitude of each target SW EMI pulse and the pulse amplitude of each target AW EMI pulse are average pulse amplitudes.

[0134] S1403. Determine the pulse time interval of each welding electromagnetic interference pulse in each welding round based on the target delay time of each welding electromagnetic interference pulse in each welding round.

[0135] Optionally, the pulse time interval is characterized by the delay time of each individual SW EMI pulse relative to the first SW EMI pulse in a welding round, i.e., the pulse time interval of each SW EMI pulse in each round of stud welding is determined based on the target delay time of each SW EMI pulse in each round of stud welding. Correspondingly, the pulse time interval is characterized by the delay time of each individual AW EMI pulse relative to the first AW EMI pulse in a welding round, i.e., the pulse time interval of each AW EMI pulse in each round of arc welding is determined based on the target delay time of each AW EMI pulse in each round of arc welding. Here, both the pulse time interval of each target SW EMI pulse and the pulse time interval of each target AW EMI pulse are average pulse time intervals.

[0136] In this embodiment, the pulse duration of each target welding electromagnetic interference pulse in each welding round is determined based on the target welding electromagnetic interference pulse sequence in each welding round. The pulse amplitude of each welding electromagnetic interference pulse in each welding round is determined based on the target peak power of each welding electromagnetic interference pulse in each welding round. The pulse time interval of each welding electromagnetic interference pulse in each welding round is determined based on the target delay time of each welding electromagnetic interference pulse in each welding round. The pulse duration, pulse amplitude, and pulse time interval of the mean value of each target welding electromagnetic interference pulse are determined respectively, so as to determine the model parameters of the mean value and then construct the mean value time series model of welding electromagnetic interference.

[0137] Figure 15 A flowchart illustrating the verification of the welding electromagnetic interference model for the electromagnetic interference modeling method provided in this application embodiment is shown below. Figure 15 As shown, electromagnetic interference modeling methods also include:

[0138] S1501. Based on the test dataset and the welding electromagnetic interference model, determine the absolute error between the test dataset and the welding electromagnetic interference model.

[0139] Optionally, based on the SW EMI test dataset L s (t) and the extended arbitrary wheel SW EMI model I S (t), the test dataset L of SWEMI is calculated based on the following formula. s (t) and SWEMI model I S The absolute error (AE) of (t):

[0140]

[0141] Among them, AE S For the test dataset L of SW EMI s (t) and SWEMI model I S The absolute error of (t), L s (t) represents the test dataset for SWEMI, I S (t) represents an arbitrary SWEMI model, p s (t) represents the time-series pulses of SWEMI.

[0142] Accordingly, based on the AW EMI test dataset L A (t) and the extended arbitrary wheel AW EMI model I A (t), the test dataset L of AW EMI is calculated based on the following formula. A (t) and AW EMI model I A The absolute error of (t):

[0143]

[0144] Among them, AE A For the test dataset L of AW EMI A (t) and AW EMI model I A The absolute error of (t), L A (t) represents the test dataset for AWEMI, I A (t) represents an arbitrary wheel AW EMI model, p A (t) represents the time-series pulse of AW EMI.

[0145] S1502. Based on the absolute error, draw the cumulative distribution function graph of the absolute error.

[0146] Optionally, based on the SW EMI test dataset L s (t) and SWEMI model I S The absolute error AE of (t) S Plot the cumulative distribution function (CDF) of the absolute error of SW EMI, and use the test dataset L of AW EMI as a reference. A (t) and AW EMI model I A The absolute error AE of (t) A Draw the CDF diagram of AW EMI. Figure 16 The cumulative distribution function of the absolute error of the welding electromagnetic interference model provided in the embodiments of this application is shown in the figure. Figure 16 As shown, Figure 16 (a) is the CDF plot of SWEMI. Figure 16 (b) is the CDF diagram of AW EMI.

[0147] S1503. Verify the welding electromagnetic interference model based on the cumulative distribution function graph.

[0148] Optionally, continue to refer to Figure 16 (a) 90th, 95th, and 99th percentile AE of SW EMI S The SW EMI model was validated with AE values ​​of 0.71dB, 0.89dB, and 1.17dB, respectively. S The values ​​are mainly within 1dB, indicating that the constructed SWEMI model is relatively accurate and meets communication requirements.

[0149] Accordingly, continue to refer to Figure 16 (b) AE of AW EMI at the 90th, 95th, and 99th percentiles AThe AW EMI model was validated with values ​​of 2.47dB, 3dB, and 3.85dB respectively. The AE of the AW EMI model was also verified. A The values ​​are mainly concentrated within 3dB, indicating that the constructed AW EMI model is relatively accurate and meets communication requirements. Specifically, the AE of the AW EMI model... A Compared to SW EMI's AE S The main reason is that arc welding is a sliding welding process, and the friction with the metal surface can cause large fluctuations in AW EMI.

[0150] In this embodiment, based on the test dataset and the welding electromagnetic interference model, the absolute error between the test dataset and the welding electromagnetic interference model is calculated, and the cumulative distribution function of the absolute error is plotted. The accuracy of the welding electromagnetic interference model is verified based on the absolute error values ​​at the 90th, 95th, and 99th percentiles of the cumulative distribution function.

[0151] Based on the same inventive concept, this application also provides an electromagnetic interference modeling device corresponding to the electromagnetic interference modeling method. Since the principle of the device in this application is similar to the electromagnetic interference modeling method described above in this application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0152] Figure 17 A module structure diagram of the electromagnetic interference modeling device provided in the embodiments of this application is shown below. Figure 17 As shown, the device includes:

[0153] The acquisition module 1701 is used to acquire the test dataset, preset power threshold and preset pulse duration of welding electromagnetic interference. The test dataset includes test data of multiple welding operations in each round of welding.

[0154] The determination module 1702 is used to determine the average background noise power based on the test dataset.

[0155] The determination module 1702 is also used to determine the model parameters of the welding electromagnetic interference model based on the test dataset, preset power threshold, preset pulse duration and average background noise power. The welding electromagnetic interference model is a time series model, and the model parameters include pulse time interval, pulse amplitude and pulse duration.

[0156] Module 1703 is used to construct a welding electromagnetic interference model based on model parameters and average background noise power.

[0157] As an optional implementation, the determining module 1702 is specifically used for:

[0158] Based on the test dataset, preset power threshold, and preset pulse duration, determine the electromagnetic interference pulse sequence for each round of welding.

[0159] Based on the test dataset, welding electromagnetic interference pulse sequence, and average background noise power, the power of each welding electromagnetic interference pulse in each round of welding is determined.

[0160] Based on the welding electromagnetic interference pulse sequence and the power of each welding electromagnetic interference pulse, the power of each welding electromagnetic interference pulse sequence is normalized to determine the target welding electromagnetic interference pulse sequence, the target peak power of each welding electromagnetic interference pulse in each welding round, and the target delay time of each welding electromagnetic interference pulse in each welding round.

[0161] The model parameters of the welding electromagnetic interference model are determined based on the target welding electromagnetic interference pulse sequence, the target peak power of each welding electromagnetic interference pulse in each welding round, and the target delay time of each welding electromagnetic interference pulse in each welding round.

[0162] As an optional implementation, the determining module 1702 is specifically used for:

[0163] Based on the test dataset and the preset power threshold, the target pulses in each round of welding are determined, and the power of the target pulses is greater than the preset power threshold.

[0164] Based on the test dataset, the preset pulse duration, and the target pulses in each round of welding, the welding electromagnetic interference pulse sequence in each round of welding is determined, and the pulse duration of each welding electromagnetic interference pulse in the welding electromagnetic interference pulse sequence is greater than the preset pulse duration.

[0165] As an optional implementation, the determining module 1702 is specifically used for:

[0166] Based on the test dataset, determine the initial power of each welding electromagnetic interference pulse in the welding electromagnetic interference pulse sequence.

[0167] The initial power of the welding electromagnetic interference pulse is converted into the initial actual power of the welding electromagnetic interference pulse.

[0168] The difference between the initial actual power of the welding electromagnetic interference pulse and the average background noise power is taken as the actual power of the welding electromagnetic interference pulse.

[0169] The power of the welding electromagnetic interference pulse is determined based on its actual power.

[0170] As an optional implementation, the determining module 1702 is specifically used for:

[0171] The peak power of each welding electromagnetic interference pulse is determined based on the sequence and power of each pulse.

[0172] The quotient of the welding electromagnetic interference pulse and the peak power of the welding electromagnetic interference pulse are used as the normalized welding electromagnetic interference pulses, thus obtaining the normalized welding electromagnetic interference pulse sequences.

[0173] Based on the normalized welding electromagnetic interference pulses, the target peak power and target delay time of each welding electromagnetic interference pulse are determined.

[0174] Based on the normalized welding electromagnetic interference pulse sequences, the welding electromagnetic interference pulse sequences of each target in each round of welding are determined.

[0175] As an optional implementation, the determining module 1702 is specifically used for:

[0176] Based on each normalized welding electromagnetic interference pulse, determine the peak power and delay time of each normalized welding electromagnetic interference pulse.

[0177] The average peak power of each normalized welding electromagnetic interference pulse in each round of welding is taken as the target peak power of each welding electromagnetic interference pulse in each round of welding.

[0178] The average delay time of each normalized welding electromagnetic interference pulse in each round of welding is taken as the target delay time of each welding electromagnetic interference pulse in each round of welding.

[0179] As an optional implementation, the determining module 1702 is specifically used for:

[0180] Based on the electromagnetic interference pulse sequence of each target in each round of welding, the pulse duration of each target electromagnetic interference pulse in each round of welding is determined.

[0181] The pulse amplitude of each welding electromagnetic interference pulse in each welding round is determined based on the target peak power of each welding electromagnetic interference pulse in each welding round.

[0182] The pulse time interval of each welding electromagnetic interference pulse in each welding round is determined based on the target delay time of each welding electromagnetic interference pulse in each welding round.

[0183] As an optional implementation, the determining module 1702 is further configured to:

[0184] Based on the test dataset and the welding electromagnetic interference model, determine the absolute error between the test dataset and the welding electromagnetic interference model.

[0185] Based on the absolute error, plot the cumulative distribution function of the absolute error.

[0186] The welding electromagnetic interference model was verified based on the cumulative distribution function plot.

[0187] This application also provides a computer device, such as... Figure 18 The diagram shown is a schematic representation of the structure of a computer device provided in an embodiment of this application, including: a processor 181, a memory 182, and a bus 183. The memory 182 stores machine-readable instructions executable by the processor 181 (e.g., ...). Figure 17 The device in the middle obtains the execution instructions corresponding to the module 1701, the determination module 1702 and the construction module 1703, etc. When the computer device is running, the processor 181 and the memory 182 communicate through the bus 183. When the machine-readable instructions are executed by the processor 181, the steps of the electromagnetic interference modeling method in the above embodiment are executed.

[0188] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the electromagnetic interference modeling method described in the above embodiments.

[0189] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and devices described above can be referred to the corresponding processes in the method embodiments, and will not be repeated here. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed mutual coupling or direct coupling or communication connection can be through some communication interfaces; the indirect coupling or communication connection of devices or modules can be electrical, mechanical, or other forms.

[0190] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. If the functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program code.

[0191] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. An electromagnetic interference modeling method, characterized in that, include: Acquire a test dataset of welding electromagnetic interference, a preset power threshold, and a preset pulse duration. The test dataset includes test data from multiple welding operations in each round of welding. Based on the test dataset, determine the average background noise power; Based on the test dataset, the preset power threshold, the preset pulse duration, and the average background noise power, the model parameters of the welding electromagnetic interference model are determined. The welding electromagnetic interference model is a time series model, and the model parameters include the pulse time interval, pulse amplitude, and pulse duration. The welding electromagnetic interference model is constructed based on the model parameters and the average background noise power. The step of determining the model parameters of the welding electromagnetic interference model based on the test dataset, the preset power threshold, the preset pulse duration, and the average background noise power includes: Based on the test dataset, the preset power threshold, and the preset pulse duration, determine the welding electromagnetic interference pulse sequence for each round of welding; Based on the test dataset, the welding electromagnetic interference pulse sequence, and the average background noise power, the power of each welding electromagnetic interference pulse in each round of welding is determined. Based on the welding electromagnetic interference pulse sequence and the power of the welding electromagnetic interference pulse, the power normalization process is performed on the welding electromagnetic interference pulse sequence to determine the target welding electromagnetic interference pulse sequence, the target peak power of the welding electromagnetic interference pulse in each round of welding, and the target delay time of the welding electromagnetic interference pulse in each round of welding. The model parameters of the welding electromagnetic interference model are determined based on the target welding electromagnetic interference pulse sequence, the target peak power of each welding electromagnetic interference pulse in each welding round, and the target delay time of each welding electromagnetic interference pulse in each welding round.

2. The method according to claim 1, characterized in that, The step of determining the electromagnetic interference pulse sequence for each round of welding based on the test dataset, the preset power threshold, and the preset pulse duration includes: Based on the test dataset and the preset power threshold, each target pulse in each round of welding is determined, wherein the power of the target pulse is greater than the preset power threshold; Based on the test dataset, the preset pulse duration, and the target pulses in each round of welding, a welding electromagnetic interference pulse sequence is determined for each round of welding, wherein the pulse duration of each welding electromagnetic interference pulse in the welding electromagnetic interference pulse sequence is greater than the preset pulse duration.

3. The method according to claim 1, characterized in that, The step of determining the power of each welding electromagnetic interference pulse in each round of welding based on the test dataset, the welding electromagnetic interference pulse sequence, and the average background noise power includes: Based on the test dataset, determine the initial power of each welding electromagnetic interference pulse in the welding electromagnetic interference pulse sequence; The initial power of the welding electromagnetic interference pulse is converted into the initial actual power of the welding electromagnetic interference pulse; The difference between the initial actual power of the welding electromagnetic interference pulse and the average background noise power is taken as the actual power of the welding electromagnetic interference pulse. The power of the welding electromagnetic interference pulse is determined based on its actual power.

4. The method according to claim 1, characterized in that, The step of performing power normalization processing on each welding electromagnetic interference pulse sequence based on the welding electromagnetic interference pulse sequence and the power of each welding electromagnetic interference pulse to determine the target welding electromagnetic interference pulse sequence, the target peak power of each welding electromagnetic interference pulse in each welding round, and the target delay time of each welding electromagnetic interference pulse in each welding round includes: The peak power of each welding electromagnetic interference pulse is determined based on the sequence of each welding electromagnetic interference pulse and the power of each welding electromagnetic interference pulse. The quotient of the welding electromagnetic interference pulse and the peak power of the welding electromagnetic interference pulse is used as the normalized welding electromagnetic interference pulse to obtain the normalized welding electromagnetic interference pulse sequence. Based on the normalized welding electromagnetic interference pulses, the target peak power and target delay time of each welding electromagnetic interference pulse are determined. Based on the normalized welding electromagnetic interference pulse sequences, the welding electromagnetic interference pulse sequences of each target in each round of welding are determined.

5. The method according to claim 4, characterized in that, The step of determining the target peak power and target delay time of each welding electromagnetic interference pulse based on the normalized welding electromagnetic interference pulses includes: Based on the normalized welding electromagnetic interference pulses, determine the peak power and delay time of each normalized welding electromagnetic interference pulse. The average value of the peak power of each normalized welding electromagnetic interference pulse in each round of welding is taken as the target peak power of each welding electromagnetic interference pulse in each round of welding. The average delay time of each normalized welding electromagnetic interference pulse in each round of welding is taken as the target delay time of each welding electromagnetic interference pulse in each round of welding.

6. The method according to claim 1, characterized in that, The step of determining the model parameters of the welding electromagnetic interference model based on the target welding electromagnetic interference pulse sequence in each round of welding, the target peak power of each welding electromagnetic interference pulse in each round of welding, and the target delay time of each welding electromagnetic interference pulse in each round of welding includes: Based on the electromagnetic interference pulse sequence of each target welding in each round of welding, the pulse duration of each target welding electromagnetic interference pulse in each round of welding is determined; The pulse amplitude of each welding electromagnetic interference pulse in each welding round is determined based on the target peak power of each welding electromagnetic interference pulse in each welding round. The pulse time interval of each welding electromagnetic interference pulse in each welding round is determined based on the target delay time of each welding electromagnetic interference pulse in each welding round.

7. The method according to claim 1, characterized in that, The method further includes: Based on the test dataset and the welding electromagnetic interference model, determine the absolute error between the test dataset and the welding electromagnetic interference model; Based on the absolute error, plot the cumulative distribution function of the absolute error; The welding electromagnetic interference model is verified based on the cumulative distribution function graph.

8. A computer device, characterized in that, include: The computer device includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the steps of the electromagnetic interference modeling method as described in any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, performs the steps of the electromagnetic interference modeling method as described in any one of claims 1 to 7.

Images