PLC electric signal relative coordinate calibration method and device and freezer power panel thereof

Abstract

The invention relates to the technical field of power panel detection, and particularly discloses a PLC electric signal relative coordinate calibration method and device and a freezer power panel thereof. The method comprises the following steps: judging whether the power panel is located at a detection station or not through the position detection unit; if yes, executing a PLC electric signal discontinuous test process, identifying the target electronic element based on the first probe contact, and obtaining charge information of the target electronic element; mapping the element and the charge information thereof to a preset relative coordinate model; performing convolution kernel operation and loss entropy judgment on the charge information through a Transform neural network layer in the model to obtain a virtual circuit loop containing the current charge quantity and a direction vector; and finally, performing compliance judgment on the loop parameters and outputting a calibration result. According to the method, through combination of accurate positioning, intelligent signal acquisition, space coordinate mapping and deep learning analysis, dynamic and deep detection of the circuit state of the power panel is realized, hidden faults such as pseudo soldering and parameter drift can be effectively identified, and the automation degree, precision and reliability of detection are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of power supply board technology, and in particular to a method and apparatus for calibrating the relative coordinates of PLC electrical signals and a power supply board for a freezer. Background Technology

[0002] In the production and testing of power boards, especially uninterruptible power supplies (UPS) boards used in medical and scientific refrigerators, ensuring the correctness and stability of their internal circuit connections is crucial. Traditional testing methods mainly rely on manual point-to-point continuity testing with multimeters or static parameter measurements using simple online testing instruments. However, these methods are inefficient and heavily dependent on operator experience, making it difficult to detect potential cold solder joints, internal cracks, and latent faults that only manifest under high-frequency switching signals. Furthermore, while traditional flying probe testing achieves automation, its testing logic is based on preset, fixed physical connections, failing to model and verify the dynamic signal integrity and virtual current paths of the power board under complex operating conditions. Especially for power boards containing complex power topologies such as inverters and automatic transfer switches, current technology lacks an efficient and intelligent testing solution that can simulate real operating conditions and deeply uncover the hidden electrical relationships between components. This leads to some power boards with latent defects entering the market, potentially causing power outages in refrigerator systems during critical periods and resulting in incalculable losses.

[0003] Therefore, there is an urgent need in the field for an advanced detection method that can perform dynamic, in-depth signal analysis on power boards and automatically calibrate their circuit logic health status. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the existing defects and provide a PLC electrical signal relative coordinate calibration method, device and its freezer power board, which can effectively solve the problems in the background art.

[0005] To achieve the above objectives, the present invention will be further described below with reference to embodiments. A method for calibrating the relative coordinates of PLC electrical signals includes the following steps: The position detection unit determines whether the power board is located at the detection station. If the power supply board is located at the testing station, a PLC electrical signal interruption test process is performed, which includes: The target electronic component on the power board is identified based on the first probe contact, and the charge information on the target electronic component is obtained. Each target electronic component and its corresponding charge information are uploaded and mapped into a preset relative coordinate model; Through the Transformer neural network layer in the relative coordinate model, convolution kernel operations and loss entropy judgments are performed on the charge information of each element to obtain virtual circuit loops between each element at the data signal level. The virtual circuit loops include current charge and its corresponding direction vector. The compliance of the current charge and its corresponding direction vector of the virtual circuit loop is judged, and the calibration result is output.

[0006] Furthermore, the step of determining whether the power board is located at the detection station through the position detection unit includes: The positioning marks on the edge of the power board are detected by photoelectric sensors installed at the inspection station; When the photoelectric sensor receives the light signal reflected by the positioning mark, it determines the precise positioning status of the power board.

[0007] Furthermore, the step of identifying the target electronic component on the power board based on the first probe contact and obtaining the charge information on the target electronic component includes: Drive the probe array controlled by a linear stepper motor to move, so that the first probe contact makes physical contact with the predetermined test point on the power board; Apply a PLC test signal of a specific frequency and amplitude to the predetermined test point; By monitoring the response of the test signal in the path through the sampling circuit, the components are identified and their voltage, current and phase information are obtained as charge information.

[0008] Furthermore, the step of uploading and mapping each target electronic component and its corresponding charge information to a preset relative coordinate model includes: A two-dimensional coordinate system with the specific physical characteristics of the power board as the origin is constructed as the relative coordinate model; The identified component locations are mapped to coordinates and bound to their real-time charge information to form timestamped data elements.

[0009] Furthermore, the step of performing convolution kernel operations and loss entropy determination on the charge information of each element through the Transformer neural network layer in the relative coordinate model includes: The serialized charge information is input into the Transformer encoder, and the feature correlation is calculated using the self-attention mechanism. Local feature extraction is performed using a one-dimensional convolutional kernel to generate a higher-order feature map. Calculate the cross-entropy loss value between the higher-order feature map and the pre-stored qualified sample features, and determine the connectivity of the virtual circuit loop based on the loss value.

[0010] Furthermore, the step of performing compliance judgment on the current charge and its corresponding direction vector of the virtual circuit loop, and outputting the calibration result, includes: The current charge is compared with a preset qualified threshold range; Calculate the spatial angle between the stated direction vector and the preset standard direction vector; When the current charge is within the acceptable threshold range and the spatial angle is less than or equal to the permissible deviation angle, a "qualified" calibration result is output; otherwise, a "unqualified" calibration result is output and an alarm is triggered.

[0011] Furthermore, a method for calibrating the relative coordinates of PLC electrical signals includes: The precise positioning and detection unit is used to determine whether the power board is located at the detection station through the position detection unit; The intelligent signal testing and acquisition unit is used to perform a PLC electrical signal intermittent testing process when the power supply board is located at the testing station. The unit includes: The component identification unit is used to identify target electronic components on the power board based on the first probe contact and obtain their charge information; The coordinate mapping unit is used to upload the information of each component and its charge to the preset relative coordinate model; The signal relationship analysis unit has a built-in Transformer neural network layer, which is used to calculate the charge information to obtain the virtual circuit loop; The compliance judgment and output unit is used to judge the parameters of the virtual circuit loop and output the calibration results.

[0012] Furthermore, the memory of the power board being tested stores a computer program, which, when executed by a processor, implements the steps of the PLC electrical signal relative coordinate calibration method described above.

[0013] This invention provides a PLC electrical signal relative coordinate calibration method, device, and refrigerator power board, which has the following beneficial effects: Through the coordinated work of photoelectric sensors and probe arrays, the positioning, contact, and signal acquisition of the power board are automatically completed, replacing the traditional inefficient and error-prone manual multimeter measurement, significantly improving detection efficiency and consistency. Furthermore, by introducing a relative coordinate model and Transformer neural network, it can extract the implicit electrical connections between components from dynamic test data and construct virtual circuit loops. This method can not only detect hard faults such as continuity and short circuits, but also accurately identify hidden defects that are difficult to detect by traditional methods, such as poor soldering, parameter drift, and poor high-frequency characteristics. Through convolution kernel operations and cross-entropy loss judgment, the unit has self-learning and adaptive capabilities, and can perform reasonable comparisons based on a qualified sample library, resulting in objective and accurate judgment results. This greatly reduces the dependence on operator experience and ensures the consistency and high reliability of the power board's factory quality. Attached Figure Description

[0014] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a flowchart illustrating a method for relative coordinate calibration of PLC electrical signals according to an embodiment of the present invention. Figure 2 This is a structural block diagram of a PLC electrical signal relative coordinate calibration device according to an embodiment of the present invention. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0016] Reference Figure 1 The diagram below illustrates a method for calibrating the relative coordinates of PLC electrical signals proposed in this invention, comprising the following steps: S100 uses a position detection unit to determine whether the power board is located at the detection station; S200, if the power supply board is located at the testing station, then a PLC electrical signal interruption test process is performed, the test process including: S210, Identify the target electronic component on the power board based on the first probe contact, and obtain the charge information on the target electronic component; S220, upload and map each target electronic component and its corresponding charge information to the preset relative coordinate model; S300, through the Transformer neural network layer in the relative coordinate model, the charge information of each element is subjected to convolution kernel operation and loss entropy judgment to obtain the virtual circuit loop between each element at the data signal level. The virtual circuit loop includes the current charge and its corresponding direction vector. S400, performs compliance judgment on the current charge and its corresponding direction vector of the virtual circuit loop, and outputs the calibration result.

[0017] As described in step S100 above, the position detection unit determines whether the power board is located at the detection station. The technical purpose of this detection is to ensure the precise positioning of the power board, providing a stable reference platform for subsequent testing. By employing a reflective photoelectric sensor to detect the dedicated positioning marks on the edge of the power board, and combining this with a laser displacement sensor to verify flatness, a multi-layered positioning verification mechanism is formed to solve the problems of poor contact of test probes and large fluctuations in test data caused by inaccurate positioning. The precise positioning of the photoelectric sensor, combined with the mechanical limiting device, achieves a positioning accuracy of ±0.05mm, laying the foundation for subsequent precision testing.

[0018] As described in step S200 above, the PLC electrical signal intermittent test process is performed. The technical purpose of the intermittent test is to obtain accurate electrical parameters while protecting the circuit under test. By using a periodically switching test signal, data is collected during the power-on period and the circuit is allowed to recover during the power-off period, avoiding overheating damage to components that may be caused by continuous testing. The test cycle is set to 100ms, with a power-on duration of 20ms and an intermittent duration of 80ms, ensuring both the integrity of data acquisition and the safety of the circuit.

[0019] As described in step S210 above, components on the power board are identified based on the first probe contact, and charge information is acquired. The technical purpose of component identification and information acquisition is to accurately identify the component type and obtain its electrical parameters. A high-precision probe array forms reliable contact with the test point, applies a PLC test signal of a specific frequency, and simultaneously acquires voltage, current, and phase information. By using beryllium copper alloy probes with precision pressure control, the contact resistance is less than 10mΩ, solving the measurement error problem caused by contact resistance in traditional testing, and achieving a measurement accuracy of 0.1%.

[0020] As described in step S220 above, each target electronic component and its corresponding charge information are uploaded and mapped to a preset relative coordinate model. The purpose of coordinate mapping is to establish the correlation between electrical parameters and spatial location. By constructing a two-dimensional coordinate system with the physical characteristics of the power board as the origin, the component positions are coordinateized and bound to real-time electrical parameters, forming timestamped data elements. This mapping relationship enables spatial visualization of test data, providing an intuitive reference framework for subsequent analysis.

[0021] As described in step S300 above, analysis is performed using a Transformer neural network layer in the relative coordinate model. The technical purpose of neural network analysis is to deeply analyze electrical signals and identify potential circuit connection relationships. The self-attention mechanism of the Transformer neural network is used to analyze signal correlation, features are extracted through convolutional kernels, and loss entropy is calculated to determine loop connectivity. This method can discover hidden faults that are difficult to detect using traditional testing methods, such as cold solder joints and parameter drift, thus improving fault detection coverage.

[0022] As described in step S400 above, the virtual circuit loop undergoes compliance judgment and calibration results are output. The technical purpose of compliance judgment is to objectively assess the circuit status and provide clear quality inspection conclusions. The results are output by comparing measured parameters with preset thresholds and combining multi-dimensional judgment logic. A dynamic threshold mechanism is adopted to automatically adjust the judgment criteria based on ambient temperature, achieving automated and standardized judgment of test results, greatly improving detection efficiency and consistency.

[0023] In one embodiment, step S100, which involves determining whether the power board is located at the detection station using the position detection unit, includes: S110 uses a photoelectric sensor installed at the testing station to detect the positioning marks on the edge of the power board; S120, when the photoelectric sensor receives the light signal reflected by the positioning mark, it determines the precise positioning status of the power board.

[0024] In practical implementation, on the automated production line for power boards, a reflective photoelectric sensor is used at the testing station, mounted on a precision-machined positioning base. The sensor's optical axis forms a 45° angle with the horizontal plane, and the detection distance is precisely set to 15mm. A reflective film with a reflectivity of 1000cd / lx / m² is attached to the edge of the power board. When the power board enters the testing station via the conveyor belt, the 880nm infrared light emitted by the sensor is reflected by the reflective film. When the receiver detects that the light intensity exceeds a set threshold, the output level jumps from low to high. This signal is then input to the PLC's X0 input point after optical isolation. The system simultaneously activates three auxiliary sensors: two laser displacement sensors detect the flatness of the power board, and a proximity switch verifies complete positioning. Only when all sensors simultaneously output ready signals, and the flatness error is less than 0.1mm, does the system determine that the power board is accurately positioned. This embodiment, through a multi-sensor verification mechanism, solves the problem of misjudgment caused by dust or oil contamination that can occur with traditional single sensors, resulting in a high positioning success rate and providing a reliable benchmark platform for subsequent precision testing.

[0025] In one embodiment, step S210, which identifies components on the power board based on the first probe contact and obtains charge information, includes: S211, drive the probe array controlled by the linear stepper motor to move, so that the first probe contact forms physical contact with the predetermined test point on the power board; S212, apply a PLC test signal of a specific frequency and amplitude to the predetermined test point; S213, by monitoring the response of the test signal in the path through the sampling circuit, the component is identified and its voltage, current and phase information are obtained as charge information.

[0026] In the specific implementation, a probe array is used in the communication power board testing workshop. Sixteen probes are arranged in a 4×4 matrix with a probe spacing of 25mm. The probe contact resistance is less than 10mΩ. The drive mechanism uses a stepper motor in conjunction with a THK precision ball screw to improve positioning accuracy. At the start of the test, the motor descends rapidly at 50mm / s, switching to a precise positioning speed of 1mm / s when 1mm from the test point. When the pressure sensor detects a contact pressure of 0.5N±0.1N, the motor stops and maintains its position. The signal source outputs a 1kHz, 5Vpp sine wave, coupled to the probe through an isolation transformer. The sampling circuit synchronously samples the ADC, acquiring 16 channel signals at a sampling rate of 100kSPS. Phase measurement is achieved through an FPGA, realizing precise phase difference measurement with a resolution of 0.1°. This embodiment, through a precise mechanical positioning and signal acquisition system, solves the measurement error problem caused by poor contact in traditional testing. The voltage measurement accuracy reaches ±1mV, and the current measurement accuracy is ±0.1mA, providing a high-quality data foundation for subsequent analysis.

[0027] In one embodiment, step S220, which involves uploading and mapping the charge information of each target electronic component and its corresponding component to a preset relative coordinate model, includes: S221, Construct a two-dimensional coordinate system with the specific physical characteristics of the power board as the origin as the relative coordinate model; S222 converts the identified component position coordinates and binds them with its real-time charge information to form a data element with a timestamp.

[0028] In practical implementation, a two-dimensional Cartesian coordinate system is established at the industrial control power supply board test station, with the center of the mounting hole at the lower left corner of the power supply board as the origin (0,0). Using a pre-imported Gerber file, the system automatically identifies the theoretical coordinates of key test points. When the probe contacts a test point, the actual coordinates are obtained through the encoder feedback of the motion control card and matched with the theoretical coordinates. Each test point's data packet includes: component ID, coordinate value, effective voltage value, effective current value, phase angle, and timestamp. The data is transmitted to the analysis computer via Gigabit Ethernet and stored using a MySQL database. The system creates an independent data file for each power supply board, with the filename containing information such as product serial number, test time, and operator ID. This embodiment establishes a complete association between electrical parameters and spatial location through precise coordinate mapping and data binding, realizing structured management and visual display of test data, and providing an accurate spatial reference framework for subsequent intelligent analysis.

[0029] In one embodiment, step S300, which involves analysis via a Transformer neural network layer in a relative coordinate model, includes: S301, input the serialized charge information into the Transformer encoder, and use the self-attention mechanism to calculate the feature correlation degree; S302 uses a one-dimensional convolutional kernel to extract local features and generate a high-order feature map. S303, calculate the cross-entropy loss value between the higher-order feature map and the pre-stored qualified sample features, and determine the connectivity of the virtual circuit loop based on the loss value.

[0030] In practical implementation, at the power board quality inspection center, a deep learning framework based on TensorFlow 2.4 was adopted. The Transformer network contains 8 encoder layers, each with 12 attention heads and a hidden layer dimension of 512. The input sequence is current time-series data of 16 test points, with a sequence length of 1000. The network first extracts features through a one-dimensional convolutional layer, and then inputs them into the Transformer encoder. A self-attention mechanism calculates the correlation weights of current signals at different test points to identify potential circuit connection relationships. The feedforward network uses two fully connected layers with a hidden layer dimension of 2048 and uses the GELU activation function. The training process uses 50,000 sets of labeled data, employs the Adam optimizer, and the learning rate decays by the inverse square root after a 10,000-step warm-up. When the cross-entropy loss value is below 0.1, the virtual loop is determined to be connected. This embodiment successfully identified several hidden faults that are difficult to detect by traditional testing methods through deep learning, including poor solder joints and increased capacitor ESR in high-frequency circuits, greatly improving the reliability of product quality.

[0031] In one embodiment, step S400 of performing compliance judgment on the virtual circuit loop and outputting the result includes: S401, compare the current charge with a preset qualified threshold range; S402, Calculate the spatial angle between the direction vector and the preset standard direction vector; S403, when the current charge is within the qualified threshold range and the spatial angle is less than or equal to the permissible deviation angle, output a "qualified" calibration result; otherwise, output a "unqualified" calibration result and trigger an alarm.

[0032] In practical implementation, the compliance judgment system on the electronic power module testing line adopts a multi-level threshold mechanism. The acceptable range of current charge is dynamically adjusted according to the power board model, with a basic range of [9.5mA, 10.5mA] and a temperature compensation coefficient of 0.1% / ℃. The permissible deviation angle of the direction vector is set to 5°, which can be relaxed to 10° for high-frequency circuits. The judgment logic adopts a three-level verification: first, data validity is checked to eliminate abnormal sampling points; then, the average current and direction angle of the loop are calculated; finally, it is compared with the dynamic threshold. The qualified result is displayed on the industrial touch screen, including a green "PASS" logo, detailed test data, and a QR code. The unqualified result triggers a red LED flashing and a buzzer alarm, while the specific parameters exceeding the standard are displayed on the screen. All results are automatically saved to the database and support the export of PDF test reports. This embodiment, through automated compliance judgment, increases the testing efficiency from 150 tests per person per day to 450 tests per day in traditional manual testing, and reduces the false judgment rate from 5% to less than 0.5%, significantly improving production efficiency and product quality.

[0033] Reference Figure 2 Here is a structural block diagram of a PLC electrical signal relative coordinate calibration device according to an embodiment of the present invention, comprising: The precise positioning and detection unit is used to determine whether the power board is located at the detection station through the position detection unit; The intelligent signal testing and acquisition unit is used to perform a PLC electrical signal intermittent testing process when the power supply board is located at the testing station. The unit includes: The component identification unit is used to identify target electronic components on the power board based on the first probe contact and obtain their charge information; The coordinate mapping unit is used to upload the information of each component and its charge to the preset relative coordinate model; The signal relationship analysis unit has a built-in Transformer neural network layer, which is used to calculate the charge information to obtain the virtual circuit loop; The compliance judgment and output unit is used to judge the parameters of the virtual circuit loop and output the calibration results.

[0034] In this embodiment, the specific implementation of each unit in the above device embodiment is described in the above method embodiment, and will not be repeated here.

[0035] This invention also provides a freezer power board, which includes a processor, memory, display screen, input device, network interface, and database connected via a system bus. The processor in this freezer power board provides computing and control capabilities. The memory of the freezer power board includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and the freezer power board programs stored in the non-volatile storage medium. The database of the freezer power board stores the data corresponding to this embodiment. The network interface of the freezer power board is used for communication with external terminals via a network connection. When executed by the processor, the freezer power board implements the above-described method.

[0036] Those skilled in the art will understand that the block diagram is merely a partial structural representation of the present invention and does not constitute a limitation on the freezer power board to which the present invention is applied.

[0037] In summary, this invention achieves precise positioning of the power board through photoelectric sensors and a multi-positioning mechanism; it drives a high-precision probe array to form reliable contact with the test points, and obtains multi-dimensional charge information such as voltage, current, and phase of the target electronic components by applying specific PLC test signals and synchronously acquiring responses; the system coordinates the physical position of the components and binds it with real-time electrical parameters, mapping it to a preset relative coordinate model to construct a data-driven spatial association; based on the Transformer neural network layer, it uses a self-attention mechanism to analyze signal correlation, and through convolution kernel operations and cross-entropy loss judgment, it mines the implicit electrical relationships between components from dynamic data, constructing a virtual circuit loop at the data signal level; by performing multi-dimensional compliance judgment on the current charge and its direction vector in the loop, it automatically outputs a calibration result of "qualified" or "unqualified". This method realizes full-process automation from precise positioning, intelligent testing, deep analysis to automatic judgment, improving detection efficiency and reducing the false judgment rate to less than 0.5%, significantly improving the accuracy, efficiency, and reliability of industrial quality inspection.

[0038] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0039] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for calibrating the relative coordinates of PLC electrical signals, characterized in that, Includes the following steps: The position detection unit determines whether the power board is located at the detection station. If so, then execute the PLC electrical signal discontinuity test process, which includes: identifying the target electronic component on the power board based on the first probe contact, obtaining the charge information on the target electronic component, uploading each target electronic component and its corresponding charge information and mapping it to a preset relative coordinate model; Through the Transformer neural network layer in the relative coordinate model, convolution kernel operations and loss entropy judgments are performed on the charge information of each element to obtain virtual circuit loops between each element at the data signal level. The virtual circuit loops include current charge and its corresponding direction vector. The compliance of the current charge and its corresponding direction vector of the virtual circuit loop is judged, and the calibration result is output.

2. The PLC electrical signal relative coordinate calibration method according to claim 1, characterized in that, The step of determining whether the power board is located at the detection station through the position detection unit includes: The positioning marks on the edge of the power board are detected by photoelectric sensors installed at the inspection station; When the photoelectric sensor receives the light signal reflected by the positioning mark, it determines the precise positioning status of the power board.

3. The PLC electrical signal relative coordinate calibration method according to claim 1, characterized in that, The step of identifying the target electronic component on the power board based on the first probe contact and obtaining the charge information on the target electronic component includes: Drive the probe array controlled by a linear stepper motor to move, so that the first probe contact makes physical contact with the predetermined test point on the power board; Apply a PLC test signal of a specific frequency and amplitude to the predetermined test point; By monitoring the response of the test signal in the path through the sampling circuit, the components are identified and their voltage, current and phase information are obtained as charge information.

4. The PLC electrical signal relative coordinate calibration method according to claim 1, characterized in that, The step of uploading and mapping each target electronic component and its corresponding charge information to a preset relative coordinate model includes: A two-dimensional coordinate system with the specific physical characteristics of the power board as the origin is constructed as the relative coordinate model; The identified component locations are mapped to coordinates and bound to their real-time charge information to form timestamped data elements.

5. The PLC electrical signal relative coordinate calibration method according to claim 1, characterized in that, The step of performing convolution kernel operations and loss entropy determination on the charge information of each element through the Transformer neural network layer in the relative coordinate model includes: The serialized charge information is input into the Transformer encoder, and the feature correlation is calculated using the self-attention mechanism. Local feature extraction is performed using a one-dimensional convolutional kernel to generate a higher-order feature map. Calculate the cross-entropy loss value between the higher-order feature map and the pre-stored qualified sample features, and determine the connectivity of the virtual circuit loop based on the loss value.

6. The PLC electrical signal relative coordinate calibration method according to claim 1, characterized in that, The step of performing compliance judgment on the current charge and its corresponding direction vector of the virtual circuit loop, and outputting the calibration result, includes: The current charge is compared with a preset qualified threshold range, and the spatial angle between the direction vector and the preset standard direction vector is calculated. When the current charge is within the acceptable threshold range and the spatial angle is less than or equal to the permissible deviation angle, a "qualified" calibration result is output; otherwise, a "unqualified" calibration result is output and an alarm is triggered.

7. A method for calibrating the relative coordinates of PLC electrical signals, used to implement the method according to any one of claims 1 to 6, characterized in that, include: The precise positioning and detection unit is used to determine whether the power board is located at the detection station through the position detection unit; The intelligent signal testing and acquisition unit is used to perform a PLC electrical signal intermittent testing process when the power board is located at the detection station. The testing process includes: identifying the target electronic components on the power board based on the first probe contact, acquiring the charge information on the target electronic components, uploading each target electronic component and its corresponding charge information and mapping it to a preset relative coordinate model. The signal relationship analysis unit has a built-in Transformer neural network layer, which is used to calculate the charge information to obtain the virtual circuit loop; The compliance judgment and output unit is used to judge the parameters of the virtual circuit loop and output the calibration results.

8. A freezer power supply board, characterized in that, The power board is the power board being tested as described in claim 1; The power board's memory stores a computer program, which, when executed by a processor, implements the steps of the PLC electrical signal relative coordinate calibration method as described in any one of claims 1 to 6.

Images