A generator verification method, system, device and storage medium

Abstract

This disclosure relates to the field of generator inspection technology, and in particular to a generator inspection method, system, device, and storage medium. The method includes: acquiring the total number of stator slots, core length, core axial width, and sensor type of the generator to be inspected; dividing the inspection area into multiple segments to obtain the number of segments; constructing an inspection matrix based on the total number of stator slots, the number of segments, and the sensor type; acquiring multiple data items of the generator to be inspected, mapping each data item to the index coordinates of the inspection matrix to obtain the actual acquisition matrix; determining the missing matrix based on the inspection matrix and the actual acquisition matrix, traversing the missing matrix, selecting elements that meet preset missed inspection conditions, and obtaining missed inspection points; aggregating all missed inspection points to obtain a supplementary inspection path, and performing supplementary inspection based on the supplementary inspection path, thereby accurately locating the missed inspection area and pinpointing its specific location after inspection, without the need for manual investigation.

Description

Technical Field

[0001] This disclosure relates to the field of generator verification technology, and in particular to a generator verification method, system, equipment and storage medium. Background Technology

[0002] As a key piece of equipment in the power system, the operational reliability of large generators directly affects the safety and stability of the power grid. To ensure the long-term stable operation of the equipment, regular internal maintenance and condition monitoring are necessary.

[0003] In existing technologies, manual labor or inspection robots equipped with multiple sensors are typically used to acquire images or perform non-destructive testing on areas such as the stator slots, core, and windings. However, due to the complex internal structure of generators, confined space, and long inspection paths, the inspection process often relies on preset trajectories or human experience for control.

[0004] The generator testing solutions provided by related technologies require manual troubleshooting of generator faults, which consumes a lot of manpower and time. Summary of the Invention

[0005] To address the aforementioned technical problems, this disclosure provides a generator verification method, system, device, and storage medium.

[0006] This disclosure provides a generator verification method, including: The total number of stator slots, core length, core axial width, and sensor type of the generator under test are obtained. Based on the core length and core axial width, the detection area of ​​the generator under test is divided into multiple segments to obtain the number of segments. A test matrix is ​​constructed based on the total number of stator slots, the number of segments, and the sensor type, and a first preset value is set for each element in the test matrix; Multiple data items of the generator under test are acquired, each data item is mapped to the index coordinate of the test matrix to obtain an actual acquisition matrix with the same dimension as the test matrix, and a second preset value is set for each element in the acquisition matrix. The missing matrix is ​​determined based on the first preset value in the detection matrix and the second preset value in the actual collection matrix. The missing matrix is ​​then traversed to select elements that meet the preset missed detection conditions to obtain the missed detection points. All the missed detection points are aggregated to obtain a supplementary detection path, and supplementary detection is performed based on the supplementary detection path.

[0007] Furthermore, the step of acquiring multiple data items of the generator under test, mapping each data item to the index coordinates of the test matrix to obtain an actual acquisition matrix with dimensions consistent with the test matrix, and setting a second preset value for each element in the acquisition matrix, further includes: Retrieve multiple data items and validate them. When each of the data items meets the preset quality conditions, it is set to the second preset value.

[0008] Further, the step of determining the missing matrix based on the first preset value in the expected detection matrix and the second preset value in the actual collection matrix, and traversing the missing matrix to select elements that meet the preset missed detection conditions to obtain missed detection points, includes: The difference between the first preset value in the test matrix and the second preset value in the actual collection matrix is ​​calculated to obtain the accuracy matrix; Traverse each element in the missing matrix, select the element with the first preset value, and obtain the missed detection points.

[0009] Further, the step of determining the missing matrix based on the first preset value in the expected detection matrix and the second preset value in the actual acquisition matrix, and traversing the missing matrix to select elements that meet the preset missed detection conditions to obtain the missed detection points, also includes: The overall coverage rate is determined by the required detection matrix and the actual data collection matrix. A histogram of missed slots is obtained based on the total number of stator slots and the overall coverage rate. Locate the missed detection points based on the slot histogram.

[0010] Furthermore, it also includes: Generate a structured entry set for the detection coordinates and corresponding detection data items of each generator under test; The structured entry set is checked to obtain the check results; Based on the verification results, a maintenance report is generated.

[0011] Furthermore, the step of verifying the structured entry set to obtain verification results includes: Obtain the coordinates of each structured entry, determine whether there is valid data in the data item within the current coordinates, and obtain the detection result; If the data item meets the preset quality conditions, the detection is considered valid; otherwise, there is no supporting evidence.

[0012] Furthermore, the step of verifying the structured entry set and obtaining the verification result also includes: A defect set is obtained by manually screening multiple data items of the generator under test in advance. Iterate through each data item in the defect set and check whether each data item in the defect set has a corresponding detection coordinate. If the detection coordinate exists, mark it as abnormal evidence.

[0013] This disclosure also provides a generator verification system, including: The acquisition module is used to acquire the total number of stator slots, core length, core axial width and sensor type of the generator under test, and divide the detection area of ​​the generator under test into multiple segments based on the core length and core axial width to obtain the number of segments; The construction module is used to construct an inspection matrix based on the total number of stator slots, the number of segments, and the sensor type, and to set a first preset value for each element in the inspection matrix; The mapping module is used to acquire multiple data items of the generator under test, map each data item to the index coordinates of the test matrix, obtain an actual acquisition matrix with the same dimension as the test matrix, and set a second preset value for each element in the acquisition matrix. The selection module is used to determine the missing matrix based on the first preset value in the detection matrix and the second preset value in the actual collection matrix, and to traverse the missing matrix to select elements that meet the preset missed detection conditions to obtain the missed detection points. The supplementary inspection module is used to aggregate all the missed inspection points, obtain supplementary inspection paths, and perform supplementary inspections based on the supplementary inspection paths.

[0014] This disclosure also provides a computer device, including a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the generator verification method.

[0015] This disclosure also provides a computer-readable storage medium storing computer-readable instructions, which, when executed by a processor, implement the steps of the generator verification method.

[0016] The technical solution provided in this disclosure has the following advantages compared with the prior art: By acquiring the total number of stator slots, core length, core axial width, and sensor type of the generator under test, and dividing the detection area into multiple segments based on the core length and core axial width, the number of segments is obtained. A test matrix is ​​constructed based on the total number of stator slots, the number of segments, and the sensor type, and a first preset value is set for each element in the test matrix. Multiple data items of the generator under test are acquired, and each data item is mapped to the index coordinates of the test matrix to obtain an actual acquisition matrix with the same dimensions as the test matrix. A second preset value is set for each element in the acquisition matrix. A missing matrix is ​​determined based on the first preset value in the test matrix and the second preset value in the actual acquisition matrix. The missing matrix is ​​traversed, and elements that meet the preset missed detection conditions are selected to obtain missed detection points. All missed detection points are aggregated to obtain a supplementary inspection path, and supplementary inspection is performed based on the supplementary inspection path. This allows for accurate location of the missed detection area and its specific position after the initial detection, eliminating the need for manual investigation. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0018] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A schematic diagram of the generator verification method provided in this embodiment of the disclosure; Figure 2 A schematic diagram of a method for obtaining missed detection points provided in an embodiment of this disclosure; Figure 3 This is a schematic diagram of a generator verification system provided in an embodiment of this disclosure. Detailed Implementation

[0020] To better understand the above-mentioned objectives, features, and advantages of this disclosure, the solutions disclosed herein will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0021] Numerous specific details are set forth in the following description in order to provide a full understanding of this disclosure, but this disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some, and not all, of the embodiments of this disclosure.

[0022] Figure 1 This is a schematic diagram of the generator verification method provided in the embodiments of this disclosure; as shown Figure 1 As shown, a generator verification method includes: Step S1: Obtain the total number of stator slots, core length, core axial width, and sensor type of the generator under test. Based on the core length and core axial width, divide the detection area of ​​the generator under test into multiple segments to obtain the number of segments. In this embodiment, the physical structure of the generator under test is parametrically modeled, and the continuous space is transformed into a computable discrete structure. By obtaining key structural parameters such as the total number of stator slots, core length, and core axial width, the basic geometric framework of the equipment in both the circumferential and axial directions can be established. The total number of stator slots determines the number of circumferential discrete units, while the core length and core axial width are used to divide the axial space into multiple equally or approximately equally spaced segments, thus obtaining the number of segments. This division method discretizes the originally continuous detection path into a finite number of enumerable spatial units, enabling subsequent detection data to be expressed and managed within a unified spatial coordinate system. Simultaneously, the sensor type is introduced as a third dimension, allowing the coverage of different detection methods (such as visual inspection, eddy current inspection, and ultrasonic inspection) at the same spatial location to be independently characterized, thereby forming a multi-dimensional detection model. This transforms the complex three-dimensional detection space into a clearly structured discrete mesh, laying the foundation for subsequent data mapping, coverage calculation, and missed detection identification. It achieves a standardized expression of the detection area, providing a unified reference framework for detection by different devices or batches; at the same time, it provides the necessary premise for the quantitative assessment of detection integrity, transforming the question of whether the detection is in place from an empirical judgment into a calculable problem.

[0023] Step S2: Construct an inspection matrix based on the total number of stator slots, the number of segments, and the sensor type, and set a first preset value for each element in the inspection matrix; In this embodiment, a complete set of detection targets, i.e., a detection matrix, is constructed based on the established discrete space. The matrix forms a three-dimensional structure with dimensions including the total number of stator slots, the number of axial segments, and the sensor type. Each matrix element corresponds to a minimum detection unit. By setting a first preset value for each element in the matrix (e.g., uniformly setting it to represent the state that needs to be detected), the entire detection space is essentially initialized with full coverage, meaning that all spatial units should be covered by detection data by default. This transforms the detection task from fuzzy path execution to a clear unit coverage task, giving each minimum spatial unit a clear detection responsibility. Simultaneously, some non-detection areas (such as non-effective areas at the ends of the iron core) can be removed or marked based on the equipment's structural characteristics, further improving the model's accuracy. A complete and comprehensive theoretical detection benchmark is constructed, allowing all subsequent detection results to be compared with this benchmark, thus achieving an objective evaluation of detection completeness. Furthermore, the matrix provides a reference template for subsequent differential calculations, serving as a crucial foundation for achieving missed detection identification.

[0024] Step S3: Acquire multiple data items of the generator to be inspected, map each data item to the index coordinates of the inspection matrix, obtain the actual acquisition matrix with the same dimension as the inspection matrix, and set a second preset value for each element in the acquisition matrix; In this embodiment, multiple data items (such as image frames, waveform data, etc.) generated by the robot during the detection process are acquired. Encoder data, visual positioning results, or other positioning information are used to extract the slot number position and axial position corresponding to each data item, thereby mapping them to specific index coordinates in the detection matrix. Simultaneously, the position of each data item in the third dimension is determined based on its sensor type, ultimately uniquely assigning each data point to a specific spatial unit. Based on this, an actual acquisition matrix consistent with the dimensions of the detection matrix is ​​constructed, and units with valid data coverage are marked with a second preset value (e.g., indicating "detected"). Multiple data points may exist in the same spatial unit; as long as one data point meets the quality requirements, the unit is considered covered. A data quality evaluation mechanism can be introduced to filter out low-quality or invalid data, ensuring the reliability of the mapping results. This achieves precise binding between detection data and physical spatial location, transforming originally discrete and disordered data into a structured and indexable dataset. Simultaneously, it provides a real data foundation for subsequent coverage calculation and missed detection identification, significantly improving the manageability and analyzability of the detection data.

[0025] Further, step S3, acquiring multiple data items of the generator under test, mapping each data item to the index coordinates of the test matrix to obtain an actual acquisition matrix with the same dimension as the test matrix, and setting a second preset value for each element in the acquisition matrix, also includes: acquiring multiple data items and verifying the multiple data items; and setting the second preset value when each data item meets the preset quality conditions.

[0026] Specifically, a data quality verification mechanism is introduced on top of the basic data mapping process to ensure the reliability of the data entering the actual acquisition matrix. After acquiring multiple data items, they are not directly mapped to the corresponding positions in the detection matrix. Instead, each data item is first verified for quality. For example, it can be evaluated based on indicators such as image clarity, contrast, signal-to-noise ratio, waveform integrity, or sensor stability to form a unified data quality judgment standard. Only when a data item meets the preset quality conditions is its corresponding matrix element set to the second preset value (i.e., the "effectively detected" state). Otherwise, even if data is acquired in a spatial location, it will not be considered as effective coverage, thus avoiding misjudgment problems caused by blurred images, noisy data, or abnormal acquisition results. This significantly improves the accuracy of detection coverage judgment, making the coverage results more realistic and reliable; at the same time, it avoids the problem of low-quality data covering missed areas, thereby improving the robustness and engineering practicality of the overall detection system.

[0027] Step S4: Determine the missing matrix based on the first preset value in the detection matrix and the second preset value in the actual collection matrix, and traverse the missing matrix to select elements that meet the preset missed detection conditions to obtain the missed detection points. In this embodiment, the detection coverage and missed detection points are calculated by comparing the detection matrix and the actual acquisition matrix. By performing logical operations on the two matrices, a missing matrix is ​​obtained, where elements marked as missing indicate that the corresponding spatial units are not covered by valid data. By traversing the missing matrix and combining it with preset missing detection criteria (e.g., whether it belongs to a valid detection area, whether it is continuously missing, etc.), all missed detection points are filtered out. The problem of whether there are missed detections is transformed into a standard matrix difference problem, giving the missed detection identification process a clear mathematical definition and computational logic. Compared with traditional methods that rely on manual experience or trajectory records to determine missed detections, this method can accurately locate specific slot numbers, axial segments, and sensor types, significantly improving positioning accuracy. Simultaneously, overall coverage and local coverage can be calculated based on matrix statistical results. This achieves automatic identification and precise positioning of missed detection areas, making the detection process verifiable and traceable; furthermore, the quantitative indicator output provides an objective basis for detection quality assessment.

[0028] Furthermore, Figure 2This is a schematic diagram of a method for obtaining missed detection points provided in an embodiment of this disclosure; as shown. Figure 2 As shown, step S4, determining the missing matrix based on the first preset value in the test matrix and the second preset value in the actual acquisition matrix, and traversing the missing matrix to select elements that meet the preset missed detection conditions to obtain the missed detection points, includes: step S41, performing a difference operation between the first preset value in the test matrix and the second preset value in the actual acquisition matrix to obtain the missing matrix; step S42, traversing each element in the missing matrix and selecting the element with the first preset value to obtain the missed detection points.

[0029] Specifically, this can be understood as comparing each spatial unit: if a position in the detection matrix is ​​a detection-prone state, but is not marked as a detected state in the actual acquisition matrix, then a first preset value is retained in the missing matrix, indicating that this position is a missed detection point. Subsequently, by traversing each element in the missing matrix and filtering out the set of elements that still retain the first preset value, all missed detection points can be obtained. This transforms missed detection identification from empirical judgment to rigorous matrix operations, giving it clear logical boundaries and calculation rules. It achieves automated and batch identification of missed detection points, significantly reducing the cost of manual judgment; and through a unified mathematical expression, the missed detection identification process has repeatability and consistency, facilitating its application in different devices or different detection tasks.

[0030] Step S4, based on the first preset value in the inspection matrix and the second preset value in the actual acquisition matrix, determines the missing matrix, and traverses the missing matrix to select elements that meet the preset missed detection conditions to obtain the missed detection points. It also includes: determining the overall coverage rate through the inspection matrix and the actual acquisition matrix; obtaining the missed detection slot histogram based on the total number of stator slots and the overall coverage rate; and locating the missed detection points according to the slot histogram.

[0031] Specifically, based on the calculation of the missing matrix, coverage statistics and visualization analysis methods are further introduced. First, by comparing the required inspection matrix with the actual acquisition matrix, the overall coverage rate can be calculated, which is the proportion of all inspected units to the required inspection units, reflecting the completion status of the inspection task from a macro perspective. Then, combined with the total number of stator slots, the inspection coverage of each slot is statistically analyzed, generating a histogram distribution of "slot-coverage rate". By analyzing the histogram, slot areas with significantly lower-than-average coverage can be quickly identified. These areas often correspond to problems such as obstructed robotic arm movement, sensor malfunctions, or unreasonable path planning. Furthermore, by combining the specific index information in the missing matrix, these low-coverage slots can be precisely located, thereby determining the spatial distribution of missed inspection points. This not only provides quantitative overall evaluation indicators but also enhances the intuitiveness of problem localization through visualization, enabling maintenance personnel to quickly identify systemic problems. Simultaneously, it helps optimize the inspection path and equipment structure design, improving overall inspection efficiency and quality.

[0032] Step S5: Aggregate all missed detection points to obtain supplementary detection paths, and perform supplementary detection based on the supplementary detection paths.

[0033] In this embodiment, by analyzing the spatial distribution of missed detection points in the missing matrix, adjacent or consecutive missed detection units are merged to form several continuous regions or path segments, thereby avoiding the inefficiency of point-by-point re-inspection. Path optimization algorithms (such as shortest path or continuous path priority strategies) can be combined to generate one or more efficient re-inspection paths, which are then sent to the robot control system for execution. This not only achieves automatic repair of missed detection problems but also dynamically adjusts the re-inspection strategy based on the actual detection situation, improving overall detection efficiency. Furthermore, by updating the actual acquisition matrix again after re-inspection, a closed-loop verification mechanism can be formed to ensure that all inspectable units are ultimately covered. The missed detection identification results are directly converted into executable re-inspection actions, achieving closed-loop control of the detection process. Simultaneously, through path optimization and region aggregation, re-inspection efficiency is improved, duplicate detection and invalid movement are reduced, significantly enhancing the automation and intelligence level of the overall maintenance operation.

[0034] In some possible implementations, the method further includes: generating a structured entry set from the detection coordinates and corresponding detection data items of each generator to be inspected; verifying the structured entry set to obtain the verification results; and generating a maintenance report based on the verification results.

[0035] In this embodiment, the coordinate information (such as slot number, axial position) of each detection unit is bound to the corresponding data items (images, waveforms, etc.) to generate structured entries in a unified format, which are then managed as a set. Verification operations, such as integrity checks and consistency verification, are performed on the set of structured entries to obtain the verification results. Finally, a maintenance report is automatically generated based on the verification results, achieving automated transformation from raw data to the final report. Establishing a structured relationship between "data—location—result" enables previously scattered data to be organized and accessed in a unified manner. This significantly improves data management efficiency, making the maintenance report generation process more standardized and automated; it also enhances report traceability, allowing each conclusion to be traced back to its specific data source, thus improving the report's credibility and auditability.

[0036] In some possible implementations, the structured entry set is checked to obtain the check results, including: obtaining the coordinates of each structured entry, determining whether there is valid data in the data item within the current coordinates, and obtaining the detection result; if the data item meets the preset quality conditions, then there is valid data, and the detection passes; otherwise, there is no evidence to support it.

[0037] In this embodiment, for each item in the structured item set, its corresponding spatial coordinates are first extracted. Then, a search is conducted in the original dataset to determine if there are any valid data items within that coordinate range. If data that meets the preset quality conditions exists, the item is deemed to have passed the inspection, meaning the conclusion is supported by data. Conversely, if no valid data is found or only data that does not meet the quality conditions exists, it is deemed to lack supporting evidence, ensuring that each maintenance conclusion has a reliable data basis. This effectively avoids situations where maintenance reports are based on experience or contain misjudgments, improving the rigor of the conclusions. Simultaneously, it provides auditors with clear verification criteria, enhancing the system's credibility and standardization.

[0038] In some possible implementation methods, checking the structured set of items to obtain the check results also includes: pre-screening multiple data items of the generator to be inspected by manual means to obtain a defect set; traversing each data item in the defect set, checking whether each data item in the defect set has a corresponding detection coordinate, and if a detection coordinate exists, marking it as abnormal evidence.

[0039] In this embodiment, all collected data is filtered manually or using auxiliary algorithms to extract data items with defect characteristics or abnormal signs, forming a defect set. Then, each data item in the defect set is traversed to check whether it has a corresponding detection coordinate in the structured item set or is referenced by a report item. If a correspondence exists, it indicates that the anomaly has been identified and included in the analysis; if no correspondence exists, the data item is marked as abnormal evidence, indicating that it may have been overlooked by the analysts. This effectively avoids the problem of missed defects and improves the completeness of the detection results. Simultaneously, through a two-way verification mechanism (combined with the previous implementation method), a closed-loop quality control system is formed, making the entire maintenance process more rigorous, comprehensive, and traceable.

[0040] In some possible implementations, define the device structure parameters and input the following parameters: N_{slot}: Total number of stator slots (e.g., 48, 54, 84, etc.); L_{core}: Effective length of stator core (mm); W_{zone}: Effective axial width of a single data point (set according to the field of view (FOV) or probe physical dimensions, e.g., 50mm); TYPE_{sensor}: Set of onboard sensors. .

[0041] The target matrix M_{target} is generated by dividing the detection region along the axial direction into K segments, where:

[0042] The initialization matrix is ​​set to 1 for all elements, meaning that by default, all slots, all axial segments, and all sensor types must be covered by data. For special non-detection areas (such as the non-core areas at both ends of the stator), the corresponding matrix elements are set to 0.

[0043] Spatial normalization mapping of collected data, parsing of raw data, and data packets transmitted back by the robot.

[0044] ; in, For encoder The current Cao number calculated visually. Current axial depth; Coordinate discretization mapping, for each data item Calculate its index (s,k,t) in the grid: ; ; ;

[0045] Establish the actual data acquisition matrix Dimensions and Consistent, initial value is 0. If valid evidence exists... Falling into the traction Within the scope, and data quality assessment (e.g., image clarity), then Multiple frames of data may exist within the same grid; if at least one frame is valid, it is considered as coverage.

[0046] Coverage integrity calculation and missed detection identification, differential operation: Calculate the missing matrix : ; The elements in the matrix with a value of 1 are the missed detection points.

[0047] The coverage rate metric output calculates the overall coverage rate. : ; Simultaneously output local coverage rate according to slot number. It is used to generate a "slot number-coverage" histogram to quickly locate specific slots with low coverage (possibly due to the robotic arm's movement being obstructed in that slot).

[0048] Generate supplementary inspection instructions and iterate through... All traction values ​​of 1 Aggregate into a continuous list of supplementary inspection paths Send it to the control system for path planning. A two-way verification mechanism for reports and evidence, with structured report conclusions, parses the final maintenance report (Word / PDF) into a set of structured items. Each entry must contain attributes. .

[0049] Reverse verification (false / empty conclusion detection) for each report entry: Extract its coordinates (s, z), and search the original database for valid data files falling within that coordinate range. If the search results are empty, or the retrieved data is... If an entry is marked as "invalid / vague", the system will issue an alert: "The conclusion is not supported by evidence", indicating that the audit of this entry has failed.

[0050] Forward verification (omission analysis and detection) involves filtering out a set of suspected defects from the raw data through manual labeling and other methods. .

[0051] Iterate through each data item in D_{suspect} and check if it is referenced by any entry in the report set R. If a suspected defective data item is not referenced by any report entry, the system will issue an alert: "Unprocessed abnormal evidence," indicating that the analyst may have missed the defect.

[0052] Figure 3 This is a schematic diagram of a generator verification system provided in an embodiment of this disclosure; as shown below. Figure 3 As shown, this disclosure also provides a generator inspection system, including: an acquisition module 301, used to acquire the total number of stator slots, core length, core axial width, and sensor type of the generator to be inspected, and to divide the inspection area of ​​the generator to be inspected into multiple segments based on the core length and core axial width to obtain the number of segments; a construction module 302, used to construct an inspection matrix based on the total number of stator slots, the number of segments, and the sensor type, and to set a first preset value for each element in the inspection matrix; a mapping module 303, used to acquire multiple data items for inspecting the generator to be inspected, to map each data item to the index coordinates of the inspection matrix to obtain an actual acquisition matrix with dimensions consistent with the inspection matrix, and to set a second preset value for each element in the acquisition matrix; a selection module 304, used to determine a missing matrix based on the first preset value in the inspection matrix and the second preset value in the actual acquisition matrix, and to traverse the missing matrix to select elements that meet preset missed inspection conditions to obtain missed inspection points; and a supplementary inspection module 305, used to aggregate all missed inspection points to obtain a supplementary inspection path, and to perform supplementary inspection based on the supplementary inspection path.

[0053] In some possible implementations, the mapping module 303 is also used to: acquire multiple data items and verify the multiple data items; and set each data item as a second preset value when it meets a preset quality condition.

[0054] Among some possible implementations, module 304 is selected, specifically for: The difference between the first preset value in the detection matrix and the second preset value in the actual collection matrix is ​​calculated to obtain the true matrix; each element in the missing matrix is ​​traversed, and the element with the first preset value is selected to obtain the missed detection points.

[0055] In some possible implementations, module 304 is selected to further: determine the overall coverage rate through the inspection matrix and the actual acquisition matrix; obtain a histogram of missed slots based on the total number of stator slots and the overall coverage rate; and locate the missed points according to the slot histogram.

[0056] In some possible implementations, the system also includes a generation module (not shown in the figure), which is specifically used to: generate a set of structured entries from the detection coordinates and corresponding detection data items of each generator to be inspected; verify the set of structured entries to obtain the verification results; and generate a maintenance report based on the verification results.

[0057] This disclosure also provides a computer device, including a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the generator verification method.

[0058] This disclosure also provides a computer-readable storage medium storing computer-readable instructions, which, when executed by a processor, implement the steps of a generator verification method.

[0059] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0060] The above description is merely a specific embodiment of this disclosure, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not to be limited to the embodiments described herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A generator verification method, characterized in that, include: The total number of stator slots, core length, core axial width, and sensor type of the generator under test are obtained. Based on the core length and core axial width, the detection area of ​​the generator under test is divided into multiple segments to obtain the number of segments. A test matrix is ​​constructed based on the total number of stator slots, the number of segments, and the sensor type, and a first preset value is set for each element in the test matrix; Multiple data items of the generator under test are acquired, each data item is mapped to the index coordinate of the test matrix to obtain an actual acquisition matrix with the same dimension as the test matrix, and a second preset value is set for each element in the acquisition matrix. The missing matrix is ​​determined based on the first preset value in the detection matrix and the second preset value in the actual collection matrix. The missing matrix is ​​then traversed to select elements that meet the preset missed detection conditions to obtain the missed detection points. All the missed detection points are aggregated to obtain a supplementary detection path, and supplementary detection is performed based on the supplementary detection path.

2. The generator verification method according to claim 1, characterized in that, The process of acquiring multiple data items for detecting the generator under test, mapping each data item to the index coordinates of the test matrix to obtain an actual acquisition matrix with dimensions consistent with the test matrix, and setting a second preset value for each element in the acquisition matrix, further includes: Retrieve multiple data items and validate them. When each of the data items meets the preset quality conditions, it is set to the second preset value.

3. The generator verification method according to claim 1, characterized in that, The process of determining a missing matrix based on a first preset value in the expected detection matrix and a second preset value in the actual acquisition matrix, and then traversing the missing matrix to select elements that meet preset missed detection conditions to obtain missed detection points, includes: The difference between the first preset value in the test matrix and the second preset value in the actual collection matrix is ​​calculated to obtain the accuracy matrix; Traverse each element in the missing matrix, select the element with the first preset value, and obtain the missed detection points.

4. The generator verification method according to claim 1, characterized in that, The step of determining the missing matrix based on the first preset value in the expected detection matrix and the second preset value in the actual acquisition matrix, and traversing the missing matrix to select elements that meet the preset missed detection conditions to obtain the missed detection points, further includes: The overall coverage rate is determined by the required detection matrix and the actual data collection matrix. A histogram of missed slots is obtained based on the total number of stator slots and the overall coverage rate. Locate the missed detection points based on the slot histogram.

5. The generator verification method according to any one of claims 1 to 4, characterized in that, Also includes: Generate a structured entry set for the detection coordinates and corresponding detection data items of each generator under test; The structured entry set is checked to obtain the check results; Based on the verification results, a maintenance report is generated.

6. The generator verification method according to claim 5, characterized in that, The verification of the structured entry set to obtain the verification results includes: Obtain the coordinates of each structured entry, determine whether there is valid data in the data item within the current coordinates, and obtain the detection result; If the data item meets the preset quality conditions, the detection is considered valid; otherwise, there is no supporting evidence.

7. The generator verification method according to claim 5, characterized in that, The step of verifying the structured entry set and obtaining the verification result also includes: A defect set is obtained by manually screening multiple data items of the generator under test in advance. Iterate through each data item in the defect set and check whether each data item in the defect set has a corresponding detection coordinate. If the detection coordinate exists, mark it as abnormal evidence.

8. A generator verification system, characterized in that, include: The acquisition module is used to acquire the total number of stator slots, core length, core axial width and sensor type of the generator under test, and divide the detection area of ​​the generator under test into multiple segments based on the core length and core axial width to obtain the number of segments; The construction module is used to construct an inspection matrix based on the total number of stator slots, the number of segments, and the sensor type, and to set a first preset value for each element in the inspection matrix; The mapping module is used to acquire multiple data items of the generator under test, map each data item to the index coordinates of the test matrix, obtain an actual acquisition matrix with the same dimension as the test matrix, and set a second preset value for each element in the acquisition matrix. The selection module is used to determine the missing matrix based on the first preset value in the detection matrix and the second preset value in the actual collection matrix, and to traverse the missing matrix to select elements that meet the preset missed detection conditions to obtain the missed detection points. The supplementary inspection module is used to aggregate all the missed inspection points, obtain supplementary inspection paths, and perform supplementary inspections based on the supplementary inspection paths.

9. A computer device, characterized in that, The device includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the generator verification method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the generator verification method as described in any one of claims 1 to 7.

Images