Method and device for real-time inversion of digital shafting attitude of hydraulic unit

Abstract

The invention relates to the field of hydroelectric generating set monitoring, in particular to a real-time inversion method and device for a digital shafting attitude of a hydroelectric generating set. According to the method, the sensors are arranged on the shafting key section corresponding to the digital model, and the data are synchronously acquired, so that the limitation of traditional single-point or local monitoring is broken through, the multi-position dynamic characteristics of the shafting can be covered, and the problem of position analysis limitation in the prior art is solved. And secondly, the original throw data is subjected to targeted filtering and noise suppression processing, so that null drift and power frequency interference are effectively filtered out, the data precision is improved, and the defect of low precision in the prior art is solved. And finally, on the basis of the purified throw data, the axis and cross section attitude is solved through a fitting or interpolation method, full-height shafting attitude parameters can be directly obtained, complex operation is not needed, the real-time performance of inversion is guaranteed, meanwhile, the comprehensive inversion capability of complex dynamic attitudes is also enhanced through combined solving of multi-section data, and the method is suitable for large-scale popularization and application. The problems of insufficient real-time performance and limited inversion capability in the prior art are solved.

Description

Technical Field

[0001] This invention relates to the field of hydropower unit monitoring, specifically to a method and device for real-time inversion of the digital shaft system attitude of a hydropower unit. Background Technology

[0002] In the field of hydraulic turbine shaft system monitoring, the operating status of the shaft system is directly related to the safe and efficient operation and maintenance of the unit. Existing technologies mostly analyze the shaft system status by collecting swing data through sensors, but most methods suffer from problems such as insufficient real-time performance, limited analysis location, and low accuracy. They also have limited ability to comprehensively invert the complex attitude and dynamic changes of the shaft system, making it difficult to meet the needs of efficient operation and maintenance as well as digitalization and intelligence. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide a method and apparatus for real-time inversion of the digital shaft system attitude of a hydraulic turbine unit, in order to solve the problems of insufficient real-time performance, limited analysis location, low accuracy, and limited ability to comprehensively invert the complex dynamic attitude of the shaft system in the existing technology.

[0004] In a first aspect, embodiments of the present invention provide a method for real-time inversion of the digital shaft system attitude of a hydraulic turbine unit, the method comprising:

[0005] Based on a pre-built digital model, corresponding sensors are set at key sections, and raw swing data is obtained by collecting swing data through the sensors. The key section is the monitoring position on the axis system represented by the digital model for collecting swing data. The original sway data is subjected to targeted filtering and noise suppression to obtain purified sway data; Based on the purified sway data, the axis attitude and cross-sectional attitude are solved by fitting or interpolation methods to obtain the axis attitude parameters.

[0006] Furthermore, before setting up corresponding sensors at key cross-sections based on a pre-built digital model, the method further includes: Obtain the original design drawings of the unit and extract key information about the shafting system; Based on the key information of the shaft system, a basic shaft system geometry composed of several concentric cross sections was constructed on a 3D CAD platform. Based on the basic axis geometry, preset axis parameters, and key point coordinate adjustment interface, an initial model that can be parameterized and controlled is obtained. The initial model is validated, and the validated initial model is used as the digital model.

[0007] Furthermore, the targeted filtering and noise suppression processing of the original slew rate data to obtain purified slew rate data includes: Extract key parameters from the original swing data, including the unit's rated rotational frequency and power frequency filtering; The original swing data is high-pass filtered using the rated frequency of the unit to obtain the first swing data with zero drift eliminated. Based on the first swing data, a power frequency low-pass filter is applied to obtain preliminary purification data to shield against power frequency electromagnetic interference; Based on the preset unit analysis requirements, the preliminary purification data is filtered to obtain the purified swing data.

[0008] Furthermore, the step of selectively filtering the preliminary purification data according to preset unit analysis requirements to obtain purified swing data includes: Obtain the preset unit analysis requirements corresponding to the preliminary purification data; Based on the preset unit analysis requirements, determine whether it is an ultra-low frequency analysis requirement or a high-order harmonic analysis requirement, and obtain the judgment result. Based on the judgment result, a corresponding filtering method is selected, and the filtering method is used to filter the preliminary purified data to obtain purified swing data.

[0009] Furthermore, based on the purified sway data, the axis attitude and cross-sectional attitude are solved by fitting or interpolation to obtain the axis system attitude parameters, including: After obtaining the purified swing data, the centroid displacements of each section are sorted and organized according to the height of the measured section to obtain the centroid displacement sequence arranged by height; The centroid displacement sequence is calculated and solved based on the number of measured sections to obtain the axis attitude at the full height position; Based on the axis orientation, calculate the normal vector of each cross section and solve the corresponding spatial rotation matrix to obtain the orientation transformation parameters of each cross section. The coordinate transformation of the non-centroidal geometric points of the cross-section is performed using the cross-sectional attitude transformation parameters. The axis and cross-sectional attitude results are then integrated to obtain the axis system attitude parameters.

[0010] Furthermore, the method also includes: The process data for solving the shaft system attitude is obtained, and the process data is used to check for any disorder issues where the high section falls below the low section, and the check results are obtained. Based on the investigation results, the source of the disorder was determined. Based on the source of the problem, a corresponding solution is matched, and the solution process is modified according to the solution to obtain the modified shaft attitude parameters.

[0011] Furthermore, the method also includes: The shaft rotation angle is calculated based on the shaft attitude parameters and unit speed, and the shaft bending and torsional shape is updated using the shaft rotation angle to obtain the updated shaft visualization geometric data. Generate an initial 3D shaft system animation based on the updated shaft system visualization geometry data; Analyze the matching degree between preset animation-related parameters, and determine the harmony principle based on the matching degree to obtain a parameter harmony scheme; The animation-related parameters of the initial three-dimensional axis system animation are adjusted according to the parameter harmonization scheme to obtain a real-time visualization of the three-dimensional axis system with the best display effect.

[0012] Secondly, embodiments of the present invention provide a real-time inversion device for the digital shaft system attitude of a hydraulic turbine unit, the device comprising: A construction module is used to set up corresponding sensors at key sections based on a pre-built digital model, and to collect swing data through the sensors to obtain raw swing data. The key section is the monitoring position on the axis system represented by the digital model for collecting swing data. The processing module is used to perform targeted filtering and noise suppression on the original slew data to obtain purified slew data; The calculation module is used to solve the axis attitude and cross-sectional attitude based on the cleaned swing data, and obtain the axis attitude parameters by fitting or interpolation.

[0013] Thirdly, embodiments of the present invention provide a computer device, including: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the method described in the first aspect or any corresponding embodiment thereof.

[0014] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing computer instructions that cause a computer to perform the method described in the first aspect or any of its corresponding embodiments.

[0015] This application overcomes the limitations of traditional single-point or local monitoring by arranging sensors on key cross-sections of the shaft system corresponding to the digital model and simultaneously collecting data. This allows for coverage of dynamic characteristics at multiple locations within the shaft system, solving the problem of limited positional analysis in existing technologies. Secondly, targeted filtering and noise suppression processing are applied to the raw sway data, effectively removing zero drift and power frequency interference, improving data accuracy, and addressing the low accuracy deficiency of existing technologies. Finally, based on the purified sway data, the attitude of the axis and cross-sections is solved using fitting or interpolation methods, directly obtaining the full-height shaft system attitude parameters without complex calculations, ensuring real-time inversion. Simultaneously, the joint solution of multi-section data enhances the comprehensive inversion capability for complex dynamic attitudes, overcoming the problems of insufficient real-time performance and limited inversion capabilities in existing technologies. Attached Figure Description

[0016] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating a method for real-time inversion of the digital shaft system attitude of a hydraulic turbine unit according to some embodiments of the present invention. Figure 2 This is a flowchart illustrating another method for real-time inversion of the digital shaft system attitude of a hydraulic turbine unit according to some embodiments of the present invention; Figure 3 This is a schematic diagram of a cross-sectional intersection problem according to some embodiments of the present invention; Figure 4 This is a flowchart illustrating another method for real-time inversion of the digital shaft system attitude of a hydraulic turbine unit according to some embodiments of the present invention; Figure 5 This is a schematic diagram illustrating the solution of the cross-sectional tilt attitude according to some embodiments of the present invention; Figure 6 This is a flowchart of real-time inversion of a shaft system digital model according to some embodiments of the present invention; Figure 7 This is a structural block diagram of a real-time inversion device for the digital shaft system attitude of a hydraulic turbine unit according to an embodiment of the present invention; Figure 8 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] According to an embodiment of the present invention, a method and apparatus for real-time inversion of the digital shaft system attitude of a hydraulic turbine unit are provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0020] This embodiment provides a method for real-time inversion of the digital shaft system attitude of a hydraulic turbine unit. Figure 1 This is a flowchart of a real-time inversion method for the digital shaft system attitude of a hydraulic turbine unit according to an embodiment of the present invention, as follows: Figure 1 As shown, the process includes the following steps: Step S101: Based on the pre-built digital model, corresponding sensors are set at the key sections, and the original swing data is obtained by collecting swing data through the sensors. The key section is the monitoring position on the axis system represented by the digital model for collecting swing data.

[0021] In this embodiment of the application, before setting corresponding sensors at key sections based on a pre-constructed digital model, the method further includes: acquiring the original design drawing of the unit and extracting key information of the shaft system; constructing a basic shaft system geometry composed of several concentric cross sections on a 3D CAD platform based on the key information of the shaft system; obtaining a parameterizable initial model based on the basic shaft system geometry, preset shaft parameters, and key point coordinate adjustment interface; verifying the initial model and using the verified initial model as the digital model.

[0022] Obtain the original design drawings of the unit and extract key information of the shaft system: Collect the original design drawings related to the shaft system of the hydraulic turbine unit, and accurately extract core information such as the overall structural dimensions of the shaft system, parameters of each concentric cross section, and installation positions of key components from the drawings. This provides complete basic data support for the subsequent construction of the shaft system geometry, ensuring that the constructed model is consistent with the actual shaft system structure of the unit.

[0023] Based on the key information of the shaft system, a basic shaft system geometry composed of several concentric cross sections is constructed on a 3D CAD platform: relying on the extracted key information of the shaft system, in the 3D CAD design platform, according to the structural characteristics of the shaft system, each concentric cross section is connected in series according to the actual installation height, and a basic shaft system geometry that fits the actual structure of the unit is accurately constructed, ensuring the structural integrity and dimensional accuracy of the geometry.

[0024] Based on the basic axis geometry, preset axis parameters, and key point coordinate adjustment interface, a parameterizable initial model is obtained: using the completed basic axis geometry as the carrier, preset axis parameter adjustment items and key point coordinate adjustment interface are used. By linking the key geometric parameters of the axis through the interface, the parameters and the model are linked, thereby forming an initial model that can be quickly adjusted and adapted to different axis postures.

[0025] The initial model is validated, and the validated initial model is used as the digital model: the parameterizable initial model is functionally validated, with a focus on validating the ability to quickly generate a new shaft system attitude model after parameter adjustment, ensuring that the model can be updated synchronously with the test data. After the initial model is verified to be error-free and meets the shaft system monitoring requirements, it is determined as the official digital model for subsequent monitoring and analysis.

[0026] In this embodiment, based on a pre-built digital model that matches the actual shaft system structure of the hydraulic turbine, radial eddy current or laser displacement sensors with an accuracy better than 1µm are arranged on key sections for sway monitoring, such as above the thrust disk, upper guide bearing, generator air gap, lower guide bearing, lower end shaft middle section, main shaft middle section, and water guide bearing, which are represented by the model. All sensor channels are synchronously triggered at a sampling frequency of not less than 16 times the runaway speed of the unit. The sensors capture the minute offset of the centroid of each bearing section in real time, and collect multi-channel original shaft sway data.

[0027] Step S102: Perform targeted filtering and noise suppression on the original swing data to obtain purified swing data.

[0028] In this embodiment of the application, the original slew rate data is subjected to targeted filtering and noise suppression processing to obtain purified slew rate data, including: Step A1: Extract key parameters from the raw swing data. These key parameters include the unit's rated rotational frequency and power frequency filtering.

[0029] First, the core parameters of the unit's rated rotational frequency are accurately obtained from the original design data and test calibration documents. These parameters are the direct basis for setting the cutoff frequency of the subsequent high-pass filter. Simultaneously, the fixed filter parameter of 50Hz is defined as the core benchmark for low-pass filtering. During the extraction process, the acquisition and correlation information of the original swing data must be verified simultaneously to ensure that the rated rotational frequency is consistent with the actual operating calibration value of the unit, and that the power frequency parameter matches the grid frequency standard of the test environment, avoiding parameter deviations that could lead to subsequent filter failures. Furthermore, the extracted key parameters, along with the channel information and sampling frequency of the swing data, must be stored in association to provide parameter support for subsequent precise filtering by channel, ensuring accurate matching between the filter parameters and the swing data of each acquisition channel.

[0030] Step A2: Use the unit's rated frequency to perform high-pass filtering on the original swing data to obtain the first swing data with zero drift eliminated.

[0031] First, based on the extracted rated operating frequency of the unit, a high-pass filter cutoff frequency of 0.1 times the rated operating frequency is calculated. This cutoff frequency is used as the core threshold for filtering. A professional digital filtering algorithm is then used to uniformly process the raw swing data of each channel. During the filtering process, the synchronization of the data in each channel is maintained, ensuring consistency with the sampling frequency and time axis of the original data, thus avoiding time series misalignment caused by the filtering operation.

[0032] To address the issue of prolonged zero drift in the raw sway data, this high-pass filter precisely removes ultra-low frequency zero-drift signals below the cutoff frequency, retaining the valid sway vibration signal to obtain the first sway data without zero drift. After filtering, the validity of the first sway data needs to be verified to confirm that the zero-drift signal has been effectively filtered out without distorting the valid sway signal, ensuring the integrity and accuracy of the data.

[0033] Step A3: Apply a power frequency low-pass filter based on the first swing data to obtain preliminary purification data that shields against power frequency electromagnetic interference.

[0034] When applying a power frequency low-pass filter to the first swing angle data, the extracted 50Hz power frequency is used as the core cutoff frequency. A corresponding low-pass filter threshold is set, and an adapted digital filtering technique is employed to perform a secondary filtering process on the first swing angle data after zero drift has been eliminated. During processing, the principle of synchronization between channels is maintained, and filtering is performed strictly according to the original channel correspondence to ensure that the swing angle data from different sections and positions maintain consistency in the time dimension after filtering, meeting the synchronization requirements for subsequent axis attitude calculation. The core objective of this low-pass filter is to shield the 50Hz power frequency electromagnetic interference signal present in the test environment. This type of interference signal can affect the accuracy of the swing angle data. Filtering effectively removes this interference, yielding preliminary purified data shielded from power frequency electromagnetic interference. After filtering, the preliminary purified data needs to be verified for signal characteristics to confirm that the power frequency interference has been effectively shielded and the effective characteristics of the swing angle data have been fully preserved.

[0035] Step A4: Based on the preset unit analysis requirements, perform targeted filtering on the preliminary purification data to obtain the purified swing data.

[0036] Based on the preset unit analysis requirements, the preliminary purification data is filtered in a targeted manner to obtain the purified swing amplitude data. This includes: obtaining the preset unit analysis requirements corresponding to the preliminary purification data; determining whether the preset unit analysis requirements are for ultra-low frequency analysis or high-order harmonic analysis, and obtaining the determination result; selecting the appropriate filtering method based on the determination result, and using the filtering method to filter the preliminary purification data to obtain the purified swing amplitude data.

[0037] Specifically, to obtain the preset unit analysis requirements corresponding to the preliminary purification data, it is necessary to first extract the analysis objectives and directions that match the preliminary purification swing data of this batch from preset documents such as the unit test plan and data analysis task book. It is necessary to clarify which type of frequency characteristic swing signal of the shaft system is to be analyzed in this data processing, and at the same time, associate the unit operating conditions, measurement sections and other information corresponding to the data to ensure that the extracted analysis requirements accurately correspond to the preliminary purification data and there is no mismatch between requirements and data, so as to provide a clear core basis for subsequent filtering judgment.

[0038] Based on the preset unit analysis requirements, determine whether it is an ultra-low frequency analysis requirement or a high-order harmonic analysis requirement. The core of the determination is to identify the type based on the frequency analysis range specified in the requirements. If the requirement is to analyze the swing signal below 0.1 times the rated rotation frequency, such as studying the slow displacement change during the static swing phase of the unit, it is an ultra-low frequency analysis requirement. If the requirement is to analyze the high-frequency components of the swing above the 50Hz power frequency, such as studying the harmonic characteristics of the higher-order vibration of the shaft system, it is a high-order harmonic analysis requirement. There are no other ambiguous judgment situations, and a clear choice between the two is obtained directly.

[0039] Based on the judgment result, the appropriate filtering method is selected, and the preliminary purified data is filtered using the filtering method to obtain the purified swing data. If it is determined that the analysis requirement is for ultra-low frequency, the high-pass filter of 0.1 times the rated frequency is skipped, and the preliminary purified data that has only been filtered by the power frequency low-pass filter is directly retained. If it is determined that the analysis requirement is for high-order harmonics, an additional power frequency high-pass filter is applied to the preliminary purified data to filter out the power frequency and lower frequency signals, and only retain the high frequency components. During the filtering, the data synchronization of each channel is maintained, and finally the purified swing data that meets the analysis requirements is obtained.

[0040] Step S103: Based on the cleaned swing data, the axis attitude and cross-sectional attitude are solved by fitting or interpolation to obtain the axis attitude parameters.

[0041] In this embodiment of the application, based on the purified sway data, the axis attitude and cross-sectional attitude are solved by fitting or interpolation methods to obtain the axis system attitude parameters, including: Step B1: Obtain the purified swing data, sort and organize the centroid displacements of each section according to the height of the measured section, and obtain the centroid displacement sequence arranged by height.

[0042] After obtaining the purified swing data, the precise matching of each channel's swing data with its corresponding measurement cross-section is first completed. This involves identifying the unique identifier and corresponding centroidal displacement data for each measurement cross-section, including those above the thrust disk, the upper guide bearing, the generator air gap, and the lower guide bearing. Simultaneously, the actual installation height parameters of each measurement cross-section within the shaft system are retrieved to establish a one-to-one correspondence between height and cross-section. Then, using the shaft system height as the sorting benchmark, the centroidal displacement data of all measurement cross-sections are arranged in an orderly manner according to a unified rule of ascending or descending height. During the sorting process, the original data characteristics of each cross-section's centroidal displacement are strictly preserved, including key information such as displacement amplitude and direction, to prevent data misalignment or feature loss. After sorting, a structured centroidal displacement sequence arranged by height is formed. This sequence must clearly label the cross-section height, cross-section position, and original channel information corresponding to each data node. This provides a regular, orderly, and complete set of foundational data for subsequent shaft attitude calculations, ensuring accurate correspondence to the centroidal displacement state at different heights during subsequent calculations.

[0043] Step B2: Calculate the centroid displacement sequence based on the number of measured sections to obtain the axis attitude at the full height position.

[0044] First, the number of measurement sections in the sequence is counted to serve as the core basis for selecting the solution algorithm, ensuring algorithm adaptability and calculation accuracy. If the number of measurement sections is ≤4, the centroid displacement sequence is calculated using a fitting method. Combining the overall structural characteristics of the shaft system, a suitable fitting model is selected. The centroid displacement data at each height are substituted into the model for parameter solving. Through fitting operations, a centroid displacement fitting curve covering the entire height of the shaft system is obtained, thereby solving for the radial runout at all height positions of the shaft system and completely restoring the overall spatial shape of the axis. If the number of measurement sections is ≥5, Newton's interpolation method is used for calculation. The centroid displacement at each height is used as the interpolation node. An interpolation polynomial is constructed according to the operation rules of Newton's interpolation. This polynomial is used to perform centroid displacement interpolation calculations at unmeasured positions within the entire height range of the shaft system, accurately obtaining the centroid displacement value at each height position.

[0045] Throughout the solution process, the height sorting rules of the centroid displacement sequence must be strictly followed to ensure that the correspondence between height and displacement is accurate. At the same time, the calculation results are initially verified to confirm that the trend of the centroid displacement at all heights is consistent with the characteristics of the shaft system structure. Finally, the axis attitude that can accurately reflect the spatial position of the shaft system is obtained.

[0046] Step B3: Based on the axis attitude, calculate the normal vector of each cross section and solve the corresponding spatial rotation matrix to obtain the attitude transformation parameters of each cross section.

[0047] First, the spatial position information of each cross-section of the axis system within the axis's attitude is extracted. Combined with the equivalent digital model of the axis system constructed in a 3D CAD platform, the intersection point of the geometric center of each cross-section and the axis is determined. Using this as a reference, the normal vector of each cross-section is calculated. The normal vector calculation must strictly follow spatial geometric operation rules to accurately reflect the orientation of the cross-section in space, ensuring a high degree of matching between the normal vector and the spatial characteristics of the axis's attitude. After obtaining the normal vectors of each cross-section, using the normal vectors and the spatial position of the cross-section as core parameters, the spatial rotation matrix corresponding to each cross-section is solved according to the solution method for the spatial rotation matrix, combined with the 3D spatial coordinate system of the axis system. The solution of the rotation matrix must ensure that the dimension is consistent with the coordinate system, accurately representing the spatial rotation state of the cross-section. The calculation of the normal vectors and the solution of the rotation matrices for all cross-sections must correspond one-to-one with the full height data of the axis's attitude, forming a complete set of parameters for the attitude transformation of each cross-section. This parameter set must clearly label the height and position of each cross-section corresponding to each parameter, ensuring the uniqueness and correlation of the parameters, providing accurate and effective calculation basis for the subsequent coordinate transformation of non-centroid geometric points of the cross-sections.

[0048] Step B4: Use the cross-sectional attitude transformation parameters to perform coordinate transformation on the non-centroidal geometric points of the cross-section, integrate the axis and cross-sectional attitude results, and obtain the axis system attitude parameters.

[0049] First, the original coordinate data of all non-centroidal geometric points of each cross-section in the equivalent digital model of the axis system in the 3D CAD platform are retrieved. A precise correlation is established between the original coordinates and the corresponding cross-sections, ensuring that each geometric point can be matched with the rotation matrix, normal vector, and other attitude transformation parameters of its respective cross-section. Then, according to the rules of spatial coordinate transformation, the original coordinates of the non-centroidal geometric points of each cross-section are substituted into the spatial rotation matrix of the corresponding cross-section for coordinate transformation calculation. During the calculation, the value standards of the attitude transformation parameters are strictly followed to ensure the accuracy of the coordinate transformation, resulting in the new spatial coordinates of each non-centroidal geometric point under the current axis attitude, thus reflecting the actual spatial attitude of the cross-section.

[0050] After completing the coordinate transformation of all non-centroidal geometric points of the cross-sections, the centroidal displacement data of the axis attitude at all heights and the spatial attitude data of each cross-section are integrated and uniformly incorporated into the three-dimensional spatial coordinate system of the axis system. The integrated data is then structured to clearly mark the axis position at each height of the axis system, the spatial orientation of each cross-section and the coordinates of the geometric points. Invalid and redundant data are eliminated, and finally, axis attitude parameters that can comprehensively and accurately characterize the overall spatial state of the axis system are formed. These parameters can be directly fed back to the equivalent digital model of the axis system to achieve accurate matching between the digital model and the actual axis attitude.

[0051] In the embodiments of this application, such as Figure 2 As shown, the method also includes: Step S201: Obtain the process data for solving the shaft system attitude, and check for any disorder issues where the high section falls below the low section based on the process data, and obtain the check results.

[0052] The process data for shaft system attitude calculation is acquired, and based on this data, an investigation is conducted to identify any issues where higher cross-sections fall below lower cross-sections. The investigation results are then obtained. First, complete process data is extracted from the entire shaft system attitude calculation process. This includes the original centroid displacement values ​​of each measurement cross-section sorted by height, the interpolated / fitted full-height centroid displacement sequence, the normal vectors and spatial rotation matrices of each cross-section, and the 3D coordinate data of non-centroid geometric points after coordinate transformation. Simultaneously, the corresponding cross-section height identifiers, algorithm parameters, and data acquisition channels are retained to ensure traceability and completeness of the process data. Then, using the height axis of the shaft system's 3D coordinate system as the core benchmark, the spatial position data of each cross-section is checked one by one in ascending order of height. Special attention is paid to verifying whether the centroid and key geometric point height axis coordinates of higher-height cross-sections are consistently higher than those of lower-height cross-sections. For example, the Z-axis coordinates of the high cross-section of the generator air gap and the low cross-section of the lower guide bearing are checked. Simultaneously, the spatial arrangement of each cross-section of the shaft system is reconstructed through 3D visualization to visually check for any instances of cross-section spatial position overlap or higher cross-sections sinking below lower cross-sections.

[0053] After a comprehensive check of all cross sections, the findings are summarized to form a clear investigation result, clearly indicating whether there are any errors or confusions. If so, the specific cross section location where the error occurred, the numerical deviation of the error, and the frequency of occurrence are clearly identified, providing an accurate basis for subsequent problem determination.

[0054] Step S202: Based on the investigation results, determine the source of the disorder problem.

[0055] If the investigation results show no disorder where the high section falls below the low section, then it is directly determined that there is no abnormality and no further tracing is needed. If the investigation results confirm the existence of the disorder, then the source of the problem is determined one by one based on the characteristics of the disorder, analyzing from two dimensions: signal interference and algorithm interpolation error. If the disorder occurs periodically at a specific section position, and the deviation pattern is consistent with the trend of sensor zero drift, for example, if a fixed disorder deviation occurs at the section above the thrust disk after each stable operation of the unit, then it is determined that the problem originates from the low-frequency zero drift of the swing signal, and the zero drift signal that has not been effectively filtered out causes the distortion of the calculated data.

[0056] If the errors occur randomly in different cross-sections without a fixed location, and the deviation values ​​are irregular and sudden (e.g., errors appear in the lower shaft mid-section in one calculation and in the water guide bearing section in the next), then the problem is determined to originate from signal glitches and noise, specifically high-frequency interference causing abnormal calculation data. If the errors are concentrated at the cross-sections where rigid and elastic segments connect, such as the connection between the rigid segment of the motor rotor and the elastic segment of the upper shaft, and the deviation stems from insufficient adaptability of the interpolation algorithm to shaft segments with different characteristics, then the problem is determined to be a rigid-elastic segment mixed interpolation error, caused by the calculation algorithm's failure to segment and process shafts with different structural characteristics.

[0057] Step S203: Match the corresponding solution based on the source of the problem, and correct the solution process according to the solution to obtain the corrected shaft attitude parameters.

[0058] If the problem originates from low-frequency zero drift in the sway signal, the solution is to increase the high-pass cutoff frequency. Based on the original 0.1 times the rated rotation frequency, the cutoff threshold of the high-pass filter is appropriately increased, and the original sway data is re-filtered to eliminate residual low-frequency zero drift signals. Then, the shaft attitude interpolation / fitting, coordinate transformation, and other solution steps are performed again. If the problem originates from signal glitches or noise, the low-pass cutoff frequency is lowered, and the threshold of the 50Hz power frequency low-pass filter is tightened to enhance the shielding effect against high-frequency glitches. The sway data is then re-purified before performing shaft attitude calculations.

[0059] If the problem originates from a mixed interpolation error between rigid body and elastic segment, the interpolation mode is switched and the centroid trajectory solution algorithm is refined by segmentation. For example, the motor rotor is equivalent to a rigid body, ensuring that its upper and lower cross sections are completely parallel. The centroid position is interpolated linearly, while the upper and lower shafts use Newton interpolation with boundary constraints to ensure continuity with the motor segment axis C¹. The centroid displacement sequence is re-interpolated according to the segmented refined algorithm, and then the cross section normals, rotation matrices are solved sequentially, and the coordinate transformations of non-centroid geometric points are performed. Finally, the corrected shaft system attitude parameters are obtained, ensuring that the spatial arrangement of each cross section of the shaft system conforms to the actual structural characteristics and there are no errors.

[0060] like Figure 3 As shown, section 1 and section 2 are cross-sections at different heights on the axis, but they intersect in space (i.e., the higher section falls below the lower section). Based on the technical solution, if this misalignment occurs periodically at a specific location, it is likely due to low-frequency zero drift of the sway signal, requiring an increase in the high-pass cutoff frequency. If it occurs randomly, it is caused by signal glitches and noise, requiring a decrease in the low-pass cutoff frequency. If it is caused by rigid-elastic segment interpolation errors, the interpolation mode needs to be switched, for example, by treating the motor rotor as a rigid body and using a segmented, refined algorithm to correct it, thus avoiding cross-section intersections.

[0061] In the embodiments of this application, such as Figure 4 As shown, the method also includes: Step S301: Calculate the shaft rotation angle based on the shaft attitude parameters and unit speed, and update the shaft bending and torsional shape using the shaft rotation angle to obtain the updated shaft visualization geometric data.

[0062] First, retrieve the calculated shaft system attitude parameters, which include core data such as the centroid displacement at full height, normal vectors of each cross section, spatial rotation matrix, and coordinates of non-centroid geometric points. At the same time, obtain the real-time operating speed or preset speed parameters of the unit, convert the speed into rotation angles per unit time, and establish a dynamic calculation model of shaft system rotation angles in combination with the time dimension. Accurately calculate the rotation angle values ​​of the entire shaft system and each cross section at different times according to the time step.

[0063] Subsequently, based on the equivalent digital model of the shaft system constructed on the 3D CAD platform, the calculated rotation angle and shaft system attitude parameters are integrated. According to the spatial rotation matrix of each cross section, the geometric shape of each component of the shaft system is dynamically adjusted. The overall bending and twisting shape of the shaft system is updated according to the bending and twisting characteristics of the shaft attitude, accurately restoring the actual spatial shape of the shaft system in the rotation state. For example, the elastic deflection of the shaft system and the yaw rotation of each cross section when the unit rotates at high speed are reflected through the update of geometric data.

[0064] During the update process, the correlation between geometric points of each cross section and the integrity of the shaft system structure are maintained. All feature information of rotation angle and bending / torsion shape is integrated into the geometric data. Finally, the shaft system visualization geometric data containing dynamic rotation and bending / torsion features can be directly used for visualization rendering. This data needs to be stored in a time-series structured manner to ensure the continuity of subsequent animation generation.

[0065] Step S302: Generate an initial 3D shaft system animation based on the updated shaft system visualization geometry data.

[0066] First, import the time-series structured visualization geometry data of the axis system into a professional 3D graphics rendering platform. Select a suitable graphics API (such as OpenGL) as the rendering core. Alternatively, a video pre-rendering mode can be used according to actual needs. First, complete the basic configuration of rendering parameters, including rendering resolution, viewpoint selection, material and texture mapping of axis system components, lighting effects, etc., to ensure that the rendered 3D axis system can clearly present its bending shape and rotation state.

[0067] Subsequently, according to the time step of the geometric data, the spatial geometric information of the axis system is extracted frame by frame. Using the real-time rendering function of the graphics API, the geometric shape of the axis system at each time node is rendered into an independent three-dimensional frame, or the entire time series of frames is generated at once through the video pre-rendering mode. During the rendering process, the continuity of the frame images is ensured to avoid problems such as frame skipping and abrupt changes in geometric shape.

[0068] After rendering all frames, the frames are stitched together in time sequence, and basic animation parameters such as playback rate are added to generate an initial 3D axis system animation that can intuitively show the dynamic process of axis rotation and bending. This animation fully restores the dynamic changes of the axis system based on attitude parameters, but does not adjust parameters such as rotation frequency, sampling rate, and refresh rate, and is only a basic visualization result.

[0069] Step S303: Analyze the matching degree between preset animation-related parameters, determine the harmony principle based on the matching degree, and obtain the parameter harmony scheme.

[0070] First, we sorted out all the preset parameters related to the animation display, including the unit rotation frequency, swing signal sampling rate, data processing cycle, and screen refresh rate. At the same time, we identified the key indicators such as the important vibration frequency of the shaft system and the minimum frequency fmin that we were interested in, and converted all parameters into values ​​in the frequency dimension for easy comparison and analysis.

[0071] Subsequently, the matching relationship between each parameter was checked one by one. The focus was on checking whether the physical resolution frame rate reached 2.56 times the switching frequency and important frequency, whether the screen refresh rate was not lower than the physical resolution frame rate, whether the data processing speed was faster than the passage of physical time, and whether the data processing cycle could accommodate the whole cycle waveform of the lowest concerned frequency. For example, if the unit switching frequency is 50Hz, the physical resolution frame rate should be at least 128Hz. If the actual screen refresh rate is only 60Hz, it is determined that the refresh rate and resolution frame rate do not match. If the number of data points in the data processing cycle cannot meet the requirement of fs / (sfmin), it is determined that the data processing cycle does not match the lowest concerned frequency.

[0072] Based on the matching and verification results of each parameter, parameter combinations with mismatch issues are identified. Then, according to the four preset harmonization principles, a corresponding harmonization direction is determined for each type of mismatch. For example, if the physical resolution frame rate is insufficient, the principle is to increase the rendering frame rate; if the screen refresh rate is lower than the resolution frame rate, the principle is to set slow motion; and if the data processing speed is insufficient, the principle is to discard some test data. Finally, all harmonization directions are integrated, clarifying the adjustment goals, adjustment ranges, and linkage adjustment rules of each parameter, forming a specific and executable parameter harmonization plan. The plan must indicate the mismatch issues and core basis corresponding to each adjustment measure to ensure the targeted nature of the harmonization measures.

[0073] Step S304: Adjust the animation-related parameters of the initial three-dimensional axis system animation according to the parameter harmonization scheme to obtain a real-time visualization of the three-dimensional axis system with the best display effect.

[0074] First, according to the requirements of the harmonization scheme, the core parameters are adjusted step by step. If the scheme requires increasing the physical resolution frame rate, the resolution frame rate is increased to more than 2.56 times the turning frequency and important frequencies by optimizing the graphics API rendering algorithm and improving hardware computing efficiency, so as to avoid substantial aliasing of the axis attitude. If the scheme requires setting slow motion, a reasonable playback multiplier is calculated so that the actual playback frame rate = playback multiplier × physical resolution frame rate. At the same time, the maximum playback speed on the user end is limited to less than 0.39 times the turning frequency and important frequencies to avoid display aliasing.

[0075] To address the issue of insufficient data processing speed, based on the current playback speed, in each data processing cycle T... d Select only the first sf s T d Rendering is performed on each data point, and redundant data is discarded to improve processing speed, ensuring that a signal with a physical duration of t seconds can be rendered within t seconds; to address the issue of data processing cycle, the time interval between data acquisition and processing is adjusted to ensure that each data processing can select no less than fs / (sfmin) data points to fully present the whole-cycle waveform of the lowest frequency of interest.

[0076] During parameter adjustment, emphasis is placed on the interrelationship of various parameters. For example, adjusting the playback magnification should be synchronized with the data processing cycle and parsing frame rate to avoid new mismatch issues caused by adjusting a single parameter. After all parameter adjustments are completed, the adjusted parameters are re-imported into the 3D rendering platform to re-render and configure the playback of the initial 3D axis animation. The display effect of the visualization is previewed in real time to check whether the rotation and bending of the axis are continuous and clear, and whether there are problems such as aliasing, stuttering, and distortion. Fine-tuning is made for the details in the preview to finally obtain a real-time 3D axis visualization with matched parameters, no display defects, and accurate reproduction of the dynamic posture of the axis.

[0077] As an example, taking a common vertical pumped-storage unit with a rotational speed of 428.6 rpm and a power of 300 MW as an example, the specific implementation process of the method of the present invention is explained.

[0078]

[0079] Step 1: The unit is in a suspended arrangement. Eddy current displacement probes are installed at seven sections: the middle section of the upper shaft, the upper guide bearing, the generator air gap, the lower guide bearing, the middle section of the lower shaft, the middle section of the main shaft, and the water guide bearing. One probe is arranged along the X and Y directions at each section, for a total of 14 channels. The height of all measuring points is measured with the center of the rotor as the zero point.

[0080] Step 2: The sampling frequency is set to 1000Hz. The online algorithm first performs a 0.7Hz high-pass filter to remove zero drift, and then performs a 50Hz low-pass filter to suppress power frequency interference. Near the upper rack, the 100Hz higher harmonics are quite noticeable; therefore, the original waveform should be copied separately and band-pass filtered and analyzed separately.

[0081] Step 3: After filtering, construct the difference quotient table and solve for the coefficients of the sixth-order Newton polynomial. Taking the X-direction as an example, construct the following difference quotient table:

[0082] The method for calculating the first-order difference quotient is as follows:

[0083] Other methods for calculating step quotients are as follows:

[0084] in Let the polynomial coefficients be:

[0085] An interpolating polynomial in the X direction can be constructed:

[0086] Substituting any height value yields the distance of the local centroid from the axis in the X direction. The Y direction is constructed in the same way. .

[0087] When the shaft system is not bent, the height is at any noncentroid point Calculate the new position after bending using the following steps. .

[0088] (1) Differentiate the centroid offset and substitute it into any height value. The normalized normal vector of the cross section at this height can be obtained as:

[0089] (2) Use the cross product to calculate the rotation axis vector of the normal vector. and rotation angle :

[0090]

[0091] (3) Calculate the Rodrigues matrix :

[0092] (4) such as Figure 5 As shown, matrix multiplication yields the coordinate changes of non-centroid points relative to the centroid, and adding the offset of the centroid point gives the new positions of all non-centroid points.

[0093]

[0094] For any given Since all points on the cross section have the same Rodrigues matrix, the coordinates of non-centroid points should be updated in batches according to the cross section height.

[0095] Assuming a sampling rate of 1000Hz, a screen refresh rate of 30Hz, a maximum frequency of interest of 50Hz (mains frequency), a minimum frequency of interest of 7.14Hz (rotation frequency), and a combined rendering speed of 20fps, the following processing method can be selected based on the aforementioned harmonization principle: a physical resolution frame rate of 128Hz, a playback speed of 0.2x (actual playback frame rate ≈ 25fps), processing 700 data points every 0.7s, but only rendering the first 140 data points, and looping the axis posture of these 140 data points during playback to ensure that the data processing time is consistently less than the physical time.

[0096] refer to Figure 6On the left is the physical structure of the unit's shaft system, marked with seven key measurement sections (1-7), including locations above the thrust disk, the upper guide bearing, and the generator air gap, used to deploy high-precision sensors to collect swing data. The middle section is the parametric construction stage of the equivalent digital model, transforming the physical shaft system into a 3D CAD model composed of concentric cross-sections connected in series, based on the original design drawings. This supports rapid reconstruction by adjusting axis parameters and key point coordinates, enabling synchronous updates of the model and test data. On the right is the algorithm processing stage, which filters and denoises the collected multi-section swing data, solves for axis and cross-sectional attitudes, checks and corrects for misalignment, and then obtains the dynamic attitude parameters of the shaft system through inversion calculation. Finally, based on the updated shaft system parameters, real-time visualization is performed through sampling and rendering collaboration. By calculating rotation angles and updating bending and twisting shapes, a 3D shaft system animation is generated using a graphics API or pre-rendering.

[0097] This embodiment also provides a real-time inversion device for the digital shaft system attitude of a hydraulic turbine unit. This device is used to implement the above embodiments and preferred embodiments, and details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0098] This embodiment provides a real-time inversion device for the digital shaft system attitude of a hydraulic turbine unit, such as... Figure 7 As shown, it includes: The construction module 501 is used to set up corresponding sensors at key sections based on a pre-built digital model, and to collect swing data through the sensors to obtain raw swing data. The key section is the monitoring position on the axis system represented by the digital model for collecting swing data. Processing module 502 is used to perform targeted filtering and noise suppression on the original slew data to obtain purified slew data; The calculation module 503 is used to solve the axis attitude and cross-sectional attitude based on the cleaned swing data by fitting or interpolation methods to obtain the axis attitude parameters.

[0099] In this embodiment of the application, the device further includes: a construction module, used to acquire the original design drawing of the unit and extract key information of the shaft system; build a basic shaft system geometry composed of several concentric cross sections on a three-dimensional CAD platform based on the key information of the shaft system; obtain an initial model that can be parameterized and controlled based on the basic shaft system geometry, preset shaft parameters and key point coordinate adjustment interface; verify the initial model and use the verified initial model as a digital model.

[0100] In this embodiment, the processing module 502 is specifically used to extract key parameters from the original swing data, including the unit's rated rotation frequency and power frequency filtering; to perform high-pass filtering on the original swing data using the unit's rated rotation frequency to obtain first swing data with zero drift eliminated; to apply power frequency low-pass filtering based on the first swing data to obtain preliminary purified data shielded from power frequency electromagnetic interference; and to perform targeted filtering on the preliminary purified data according to preset unit analysis requirements to obtain purified swing data.

[0101] In this embodiment of the application, the processing module 502 is specifically used to obtain the preset unit analysis requirements corresponding to the preliminary purification data; determine whether it is an ultra-low frequency analysis requirement or a high-order harmonic analysis requirement based on the preset unit analysis requirements, and obtain the judgment result; select the corresponding filtering method according to the judgment result, and use the filtering method to filter the preliminary purification data to obtain the purified swing data.

[0102] In this embodiment, the calculation module 503 is specifically used to acquire the purified swing data, sort and organize the centroid displacements of each cross-section according to the height of the measured cross-section, and obtain a centroid displacement sequence arranged by height; calculate and solve the centroid displacement sequence according to the number of measured cross-sections to obtain the axis attitude at the full height position; calculate the normal vector of each cross-section and solve the corresponding spatial rotation matrix based on the axis attitude to obtain the attitude transformation parameters of each cross-section; use the cross-section attitude transformation parameters to perform coordinate transformation on the non-centroid geometric points of the cross-section, integrate the axis and cross-section attitude results, and obtain the axis system attitude parameters.

[0103] In this embodiment of the application, the device further includes: a correction module, used to acquire process data of shaft system attitude solution, and based on the process data, check whether there is a disorder problem where the high section falls below the low section, and obtain the investigation result; based on the investigation result, determine the source of the disorder problem; match the corresponding solution according to the source of the problem, and correct the solution process according to the solution to obtain the corrected shaft system attitude parameters.

[0104] In this embodiment, the device further includes: a display module, configured to calculate the shaft rotation angle based on shaft attitude parameters and unit speed, and update the shaft bending and twisting shape using the shaft rotation angle to obtain updated shaft visualization geometric data; generate an initial three-dimensional shaft animation based on the updated shaft visualization geometric data; analyze the matching degree between preset animation-related parameters, determine the harmonization principle based on the matching degree, and obtain a parameter harmonization scheme; and adjust the animation-related parameters of the initial three-dimensional shaft animation according to the parameter harmonization scheme to obtain a three-dimensional shaft real-time visualization screen with the best display effect.

[0105] Please see Figure 8 , Figure 8This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 8 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system).

[0106] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0107] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0108] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device as shown by a landing page for an app. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, which can be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0109] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include combinations of the above types of memory. The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0110] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0111] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for real-time inversion of the attitude of a digital shaft system of a hydraulic turbine unit, characterized in that, The method includes: Based on a pre-built digital model, corresponding sensors are set at key sections, and raw swing data is obtained by collecting swing data through the sensors. The key section is the monitoring position on the axis system represented by the digital model for collecting swing data. The original sway data is subjected to targeted filtering and noise suppression to obtain purified sway data; Based on the purified sway data, the axis attitude and cross-sectional attitude are solved by fitting or interpolation methods to obtain the axis attitude parameters.

2. The method according to claim 1, characterized in that, Before setting up corresponding sensors at key cross-sections based on a pre-built digital model, the method further includes: Obtain the original design drawings of the unit and extract key information about the shafting system; Based on the key information of the shaft system, a basic shaft system geometry composed of several concentric cross sections was constructed on a 3D CAD platform. Based on the basic axis geometry, preset axis parameters, and key point coordinate adjustment interface, an initial model that can be parameterized and controlled is obtained. The initial model is validated, and the validated initial model is used as the digital model.

3. The method according to claim 1, characterized in that, The process of performing targeted filtering and noise suppression on the original sway data to obtain purified sway data includes: Extract key parameters from the original swing data, including the unit's rated rotational frequency and power frequency filtering; The original swing data is high-pass filtered using the rated frequency of the unit to obtain the first swing data with zero drift eliminated. Based on the first swing data, a power frequency low-pass filter is applied to obtain preliminary purification data to shield against power frequency electromagnetic interference; Based on the preset unit analysis requirements, the preliminary purification data is filtered to obtain the purified swing data.

4. The method according to claim 3, characterized in that, The step of filtering the preliminary purification data according to preset unit analysis requirements to obtain purified swing data includes: Obtain the preset unit analysis requirements corresponding to the preliminary purification data; Based on the preset unit analysis requirements, determine whether it is an ultra-low frequency analysis requirement or a high-order harmonic analysis requirement, and obtain the judgment result. Based on the judgment result, a corresponding filtering method is selected, and the filtering method is used to filter the preliminary purified data to obtain purified swing data.

5. The method according to claim 1, characterized in that, The axis attitude parameters are obtained by solving the axis attitude and cross-sectional attitude based on the purified sway data through fitting or interpolation methods, including: After obtaining the purified swing data, the centroid displacements of each section are sorted and organized according to the height of the measured section to obtain the centroid displacement sequence arranged by height; The centroid displacement sequence is calculated and solved based on the number of measured sections to obtain the axis attitude at the full height position; Based on the axis orientation, calculate the normal vector of each cross section and solve the corresponding spatial rotation matrix to obtain the orientation transformation parameters of each cross section. The coordinate transformation of the non-centroidal geometric points of the cross-section is performed using the cross-sectional attitude transformation parameters. The axis and cross-sectional attitude results are then integrated to obtain the axis system attitude parameters.

6. The method according to claim 1, characterized in that, The method further includes: The process data for solving the shaft system attitude is obtained, and the process data is used to check for any disorder issues where the high section falls below the low section, and the check results are obtained. Based on the investigation results, the source of the disorder was determined. Based on the source of the problem, a corresponding solution is matched, and the solution process is modified according to the solution to obtain the modified shaft attitude parameters.

7. The method according to claim 1, characterized in that, The method further includes: The shaft rotation angle is calculated based on the shaft attitude parameters and unit speed, and the shaft bending and torsional shape is updated using the shaft rotation angle to obtain the updated shaft visualization geometric data. Generate an initial 3D shaft system animation based on the updated shaft system visualization geometry data; Analyze the matching degree between preset animation-related parameters, and determine the harmony principle based on the matching degree to obtain a parameter harmony scheme; The animation-related parameters of the initial three-dimensional axis system animation are adjusted according to the parameter harmonization scheme to obtain a real-time visualization of the three-dimensional axis system with the best display effect.

8. A real-time inversion device for the attitude of a hydraulic turbine digital shaft system, characterized in that, The device includes: A construction module is used to set up corresponding sensors at key sections based on a pre-built digital model, and to collect swing data through the sensors to obtain raw swing data. The key section is the monitoring position on the axis system represented by the digital model for collecting swing data. The processing module is used to perform targeted filtering and noise suppression on the original slew data to obtain purified slew data; The calculation module is used to solve the axis attitude and cross-sectional attitude based on the cleaned swing data, and obtain the axis attitude parameters by fitting or interpolation.

9. A computer device, characterized in that, include: A memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, the processor executing the computer instructions to perform the method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the method of any one of claims 1 to 7.

Images