A WJ999 type track correction system

Abstract

The present application relates to the technical field of intelligent control of track deviation correction, in particular to a WJ999 type track correction system, which comprises a multi-source data acquisition preprocessing module capable of regularizing and preprocessing multi-class heterogeneous raw data, a working condition stratification and deviation identification module capable of distinguishing working conditions and identifying effective deviation states, an adaptive correction coefficient operation module capable of matching and generating compensation correction coefficients, a closed-loop control instruction output module capable of realizing closed-loop control, and a data storage and human-computer interaction module capable of storing data and realizing human-computer interaction. The present application completes heterogeneous data time sequence alignment, parameter calibration and abnormality elimination through the multi-source data acquisition preprocessing module, regularizes complex device operation and load working condition raw data into standardized data sets, and solves the industry pain points of complex track device operation data types, disordered data, and many invalid interference data.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for track correction, and in particular to a WJ999 track correction system. Background Technology

[0002] The field of intelligent track deviation control technology encompasses core technologies such as tracked equipment status monitoring, deviation identification, speed regulation, operating condition control, data processing, and human-machine interaction control. Focusing on tracked equipment deviation correction scenarios, it primarily covers multi-source operating condition data acquisition and organization, operating condition classification and labeling, identification of valid track deviation signals, correction parameter matching and calculation, closed-loop control, and data storage and display technologies. It is a comprehensive control technology integrating status perception, data processing, operating condition identification, and control, and is widely applicable to stable operation control scenarios for various tracked work equipment.

[0003] The WJ999 track correction system is an automated correction control system applied in the field of intelligent track correction control to solve the problem of tracked equipment running off track. Addressing the issues of complex operating conditions, numerous offset interference signals, weak adaptability of fixed correction parameters, limited control methods, incomplete data archiving, and lack of manual correction channels in tracked equipment, the system collects and organizes raw data on equipment operation and load. It classifies and labels operating conditions based on equipment load, drive wheel operation, machine posture, and track tension. It uses a time-series filtering window to filter effective offset data, constructs a correspondence between operating conditions and correction coefficients based on historical correction data, and iteratively optimizes parameters. It achieves offset correction by adjusting the speed of the tracks differently on both sides, and completes multiple fine-tuning corrections and anomaly warning protection through a layered progressive judgment logic. Simultaneously, it enables the classification and archiving of operating condition data, visualization display, and manual parameter correction operations.

[0004] Traditional track deviation control relies solely on conventional data processing and basic control logic. The data organization method is rather general, failing to standardize heterogeneous raw data. The classification of working conditions and deviation screening lack refined data support. The correction parameters do not have the ability to dynamically iterate and update. Fixed parameters are difficult to adapt to dynamically changing complex working conditions. The correction and control process is relatively simple, failing to form a complete closed-loop judgment and anomaly protection mechanism. Data storage and classification are rather disorganized. Human-machine interaction correction functions are lacking. Under complex load and attitude conditions, deviation identification is prone to inaccurate. The correction adaptability is poor, deviation problems cannot be completely eliminated, and equipment operation data is difficult to trace and organize effectively. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the existing technology and propose a WJ999 track correction system.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a WJ999 track correction system, the system comprising a multi-source data acquisition and preprocessing module, used to acquire equipment operation data and load condition raw data under track running conditions, and to perform unified regularization and preprocessing on the acquired heterogeneous raw data. The working condition stratification and deviation identification module is used to extract features from the preprocessed standardized data, distinguish different operating conditions of the equipment according to the data features and bind corresponding working condition labels, filter the track offset data based on the labeled working condition labels, and identify the effective track deviation state. The adaptive correction coefficient calculation module is used to store historical correction data of the equipment, establish a corresponding mapping relationship between operating conditions and correction coefficients, match the corresponding correction coefficients according to the real-time operating conditions of the equipment, and generate compensation correction coefficients for track speed adjustment. The closed-loop control command output module is used to output differentiated speed adjustment commands to the drive units on both sides of the track according to the generated compensation correction coefficient, continuously collect the corrected equipment operation data to complete the closed-loop state determination, and execute corresponding control operations according to the abnormal operation state of the equipment. The data storage and human-computer interaction module is used to classify and store equipment operating condition data, track alignment process data, and full data of equipment abnormal events. It is also configured with data visualization display function and manual parameter correction interaction function.

[0007] As a further aspect of the present invention, the multi-source data acquisition and preprocessing module has built-in fixed preprocessing logic, which sequentially performs time-series alignment, parameter calibration, and abnormal data removal on the various types of heterogeneous raw data collected, thereby unifying and standardizing the multi-source heterogeneous data into a standardized working condition dataset.

[0008] As a further aspect of the present invention, the working condition stratification and deviation identification module classifies working condition types based on multi-dimensional equipment operation data characteristics, distinguishes different operating conditions according to the characteristic differences of equipment load status, drive wheel operation status, body attitude status, and track tension status, and matches and binds corresponding working condition tags to the real-time collected operating data.

[0009] As a further aspect of the present invention, the working condition stratification and deviation identification module is configured with a self-configurable time-series filtering window. Based on the effective offset sampling data within the filtering window, invalid offset signals corresponding to instantaneous interference generated during equipment operation are filtered out to determine the actual track misalignment event.

[0010] As a further aspect of the present invention, the adaptive correction coefficient calculation module has a built-in iteratively updatable working condition-correction coefficient mapping database, continuously collects effective correction events under each working condition, and iteratively updates the benchmark correction coefficients corresponding to each working condition based on the collected historical correction event data.

[0011] As a further aspect of the present invention, the adaptive correction coefficient calculation module is configured with fixed coefficient constraint logic to constrain the output range of the real-time generated speed compensation correction coefficient and standardize the output parameter state of the correction coefficient.

[0012] As a further aspect of the present invention, the closed-loop control command output module is configured with a hierarchical and progressive correction judgment process. After a single correction operation is completed, the off-track running status is verified. If the off-track status is not eliminated, a secondary fine-tuning correction process is initiated. If the off-track status cannot be eliminated after multiple fine-tuning operations, an abnormal working condition warning signal is output.

[0013] As a further aspect of the present invention, the closed-loop control command output module is associated with the overall speed control logic of the equipment. When the track is continuously detected to be misaligned and multiple correction operations fail to eliminate the misalignment, a protective control operation to reduce the speed of the entire machine is triggered.

[0014] As a further aspect of the present invention, the data storage and human-computer interaction module is configured with dedicated event classification and marking logic, which sets independent marks for events such as correction failure, abnormal operation offset, and abnormal equipment parameters, and binds the event marks with the full amount of operating condition data of the corresponding time period for classification and archiving.

[0015] As a further aspect of the present invention, the data storage and human-computer interaction module can call up the historical calibration data archived by the system, visually display the deviation events, calibration process and coefficient iteration records corresponding to each working condition, and support manual intervention to correct and update the calibration parameters built into the system.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, a multi-source data acquisition and preprocessing module completes heterogeneous data time-series alignment, parameter calibration, and anomaly removal, transforming complex raw data on equipment operation and load conditions into a standardized dataset. This addresses the industry pain points of complex data types, chaotic data, and a large amount of invalid interference data in tracked equipment operation. Simultaneously, relying on the working condition layering and deviation identification module, the operating conditions are divided into multiple dimensions such as load, drive wheel, machine posture, and track tension. Combined with a configurable time-series filtering window to filter out instantaneous interference offset signals, the true track deviation state can be identified. This completely avoids the problems of traditional correction methods being unable to distinguish working condition differences and being susceptible to instantaneous working condition interference, which lead to miscorrection. It significantly improves the accuracy of track deviation identification and the adaptability to different complex working conditions, providing reliable data support for subsequent correction. Attached Figure Description

[0017] Figure 1 This is a system flowchart of the present invention. Detailed Implementation

[0018] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0019] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0020] The following detailed description of the solution is based on specific embodiments. The detailed implementation methods described herein are only for explaining the working principle and operation process of the solution and are not intended to limit the scope of protection of the solution.

[0021] This solution is a WJ999 track correction system, applied to the automatic correction of track misalignment during the travel and operation of tracked engineering equipment. The WJ999 tracked equipment has a total weight of 32 tons, a rated working load of 18 tons, a track pitch of 203 mm, 21 teeth on the drive wheel, and a rated drive wheel speed of 130 rpm. The equipment is equipped with two independent drive units on the left and right sides, each of which can independently receive speed adjustment commands.

[0022] The system collects equipment operation data and load condition raw data under track running conditions through a multi-source data acquisition and preprocessing module. It performs unified and standardized preprocessing on the collected heterogeneous raw data. The system extracts features from the preprocessed standardized data through a condition stratification and deviation identification module. Based on the data features, it distinguishes different operating conditions of the equipment and binds corresponding condition labels. Based on the labeled condition labels, it filters the track deviation data to identify the effective track deviation state. The system stores historical correction data of the equipment through an adaptive correction coefficient calculation module, establishes a corresponding mapping relationship between operating conditions and correction coefficients, matches the corresponding correction coefficients according to the real-time operating conditions of the equipment, and generates compensation correction coefficients for track speed adjustment. The system outputs differentiated speed adjustment commands to the drive units on both sides of the track according to the generated compensation correction coefficients through a closed-loop control command output module. It continuously collects the corrected equipment operation data to complete the closed-loop state determination and executes corresponding control operations according to the abnormal operating state of the equipment. The system stores equipment operating condition data, track correction process data, and full data of abnormal equipment events through a data storage and human-machine interaction module. It also configures data visualization display function and manual parameter correction interaction function.

[0023] The above five modules are connected sequentially according to the process of data acquisition, working condition identification, coefficient calculation, instruction output, and storage interaction, together forming the complete calibration link of this solution.

[0024] Example 1, please refer to Figure 1 This embodiment is mainly used for the acquisition and unified preprocessing of multi-source heterogeneous operating data in the WJ999 track correction system, providing standardized data input for subsequent working condition identification and correction calculations. Specifically, it includes a multi-source data acquisition and preprocessing module, which is used to acquire equipment operating data and load condition raw data under track operating conditions, and to perform unified preprocessing on the acquired multi-type heterogeneous raw data. This multi-source data acquisition and preprocessing module has built-in fixed preprocessing logic, which sequentially performs time sequence alignment, parameter calibration, and abnormal data removal on the acquired heterogeneous raw data, unifying the multi-source heterogeneous data into a standardized working condition dataset.

[0025] In the specific implementation process, the WJ999 tracked equipment is equipped with a total of ten sensing units, including a left track speed sensor, a right track speed sensor, a left drive wheel speed sensor, a right drive wheel speed sensor, a fuselage yaw angle gyroscope, a fuselage pitch angle sensor, a fuselage roll angle sensor, a left track tension sensor, a right track tension sensor, and a load weight sensor. The sampling frequency of each sensing unit is uniformly set to 100 Hz, that is, data sampling is completed every 10 milliseconds. During the actual implementation, the equipment was in a medium-load paving operation state. The load weight sensor collected paving material load production data of 13 tons. After adding the equipment's own weight of 32 tons, the total weight production data of the whole machine was 45 tons. The original speed production data collected by the left track speed sensor was 125.5 rpm, the original speed production data collected by the right track speed sensor was 124.8 rpm, the original speed production data collected by the left drive wheel speed sensor was 126.1 rpm, the original speed production data collected by the right drive wheel speed sensor was 125.5 rpm, the original yaw angle production data collected by the fuselage yaw angle gyroscope was 2.3 degrees, the original pitch angle production data collected by the fuselage pitch angle sensor was 1.2 degrees, the original roll angle production data collected by the fuselage roll angle sensor was 0.8 degrees, the original tension production data collected by the left track tension sensor was 8500 N, and the original tension production data collected by the right track tension sensor was 8600 N.

[0026] Because the working principles and output formats of various sensors are different, some sensors use pulse output, some use analog voltage output, and some use digital bus output. The time when the data from each sensor arrives at the main control unit varies by milliseconds, and each sensor has individual errors at the factory. Therefore, it is necessary to use fixed preprocessing logic to perform three processes on the raw data in sequence: timing alignment, parameter calibration, and abnormal data removal.

[0027] Timing alignment is based on the main control unit's system clock, aligning each sensor data stream to a 10-millisecond time slice, ensuring that all ten data streams within the same time slice correspond to the same device operating time. In practice, if the arrival timestamp of a frame from the left track speed sensor lags behind the system reference time by 3 milliseconds, and the arrival timestamp of a frame from the right track speed sensor precedes the system reference time by 2 milliseconds, the timing alignment logic aligns both frames to the same 10-millisecond time slice. This ensures that the left and right track speeds used in subsequent calculations are taken from the same moment, preventing distortion in speed difference calculations due to timing misalignment.

[0028] The parameter calibration process uses the calibration coefficients and zero-point offsets recorded in the factory calibration certificates of each sensor to linearly correct the original sampled values. The correction formula is: the calibrated value equals the original sampled value multiplied by the calibration coefficient plus the zero-point offset. In specific implementation, the calibration coefficient for the left track speed sensor is 1.000, and the zero-point offset is 0 rpm. Therefore, the calibrated value of the left track speed is 125.5 multiplied by 1.000 plus 0, resulting in 125.5 rpm. The calibration coefficient for the right track speed sensor is 1.000, and the zero-point offset is 0 rpm. Therefore, the calibrated value of the right track speed is 124.8 multiplied by 1.000 plus 0, resulting in 124.8 rpm. The calibration coefficient of the left track tension sensor is 1.002, and the zero-point offset is -15 N. Therefore, the calibrated left track tension value is 8500 multiplied by 1.002 plus -15, resulting in 8517 minus 15, which equals 8502 N. The calibration coefficient of the right track tension sensor is 1.001, and the zero-point offset is -10 N. Therefore, the calibrated right track tension value is 8600 multiplied by 1.001 plus -10, resulting in 8608.6 minus 10, which equals 8598.6 N, rounded down to 8599 N. The calibration coefficient of the fuselage yaw angle gyroscope is 1.000, and the zero-point offset is 0.1 degrees. Therefore, the calibrated yaw angle value is 2.3 multiplied by 1.000 plus 0.1, resulting in 2.4 degrees. The calibration coefficients for the other drive wheel speed, pitch angle, roll angle, and load weight sensors are all 1.000 and the zero-point offset is 0. The calibrated values ​​are consistent with the original values, and they will not be listed one by one here.

[0029] The abnormal data removal process adopts the three-standard-deviation criterion. The mean and standard deviation of each channel data are calculated within a sliding window consisting of 50 consecutive sampling points. When the absolute value of the difference between the value of a certain sampling point and the mean of the window exceeds three standard deviations, the sampling point is determined to be abnormal data caused by electromagnetic interference or sensor jitter and is removed. After removal, the gap is filled by linear interpolation of two adjacent valid sampling points within the window.

[0030] In the specific implementation process, the mean production data of the left track speed channel within a certain 50-point sliding window is 125.2 revolutions per minute, the standard deviation production data is 0.15 revolutions per minute, and the result of three times the standard deviation is 0.45 revolutions per minute. The original production data of a certain sampling point within this window is 130.2 revolutions per minute. The difference between this value and the mean is 130.2 minus 125.2, which equals 5.0 revolutions per minute. Since 5.0 is greater than 0.45, this sampling point is determined to be abnormal data and is removed. Linear interpolation is performed using the effective sampling values ​​of 125.4 revolutions per minute and 125.3 revolutions per minute before and after this point, and the filler value is 125.35 revolutions per minute, which is rounded to one decimal place as 125.4 revolutions per minute.

[0031] After time alignment, parameter calibration, and outlier data removal, the data from the ten heterogeneous sensors are standardized into a standardized working condition dataset containing eleven fields: timestamp, left track speed, right track speed, left drive wheel speed, right drive wheel speed, yaw angle, pitch angle, roll angle, left track tension, right track tension, and load weight. Each data record corresponds to a 10-millisecond time slice. This standardized working condition dataset is directly output to the working condition stratification and deviation identification module.

[0032] Example 2: Building upon the multi-source data acquisition and standardized preprocessing completed in Example 1, this example further realizes the hierarchical identification of equipment operating conditions and the determination of effective track deviation status. Specifically, it includes a working condition hierarchical and deviation identification module, which is used to extract features from the preprocessed standardized data, distinguish different operating conditions of the equipment according to the data features and bind corresponding working condition labels, filter the track deviation data based on the labeled working condition labels, and identify the effective track deviation status.

[0033] This working condition stratification and deviation identification module classifies working condition types based on multi-dimensional equipment operation data characteristics. It distinguishes different operating conditions according to the differences in equipment load status, drive wheel operating status, machine attitude status, and track tension status, and matches and binds corresponding working condition tags to the real-time collected operating data. This module is equipped with a customizable temporal filtering window. Based on the effective offset sampling data within the filtering window, it filters out invalid offset signals corresponding to instantaneous interference generated during equipment operation, thus identifying the actual track misalignment events.

[0034] The working condition stratification process extracts features from four dimensions—load status, drive wheel operating status, machine attitude status, and track tension status—and combines them to determine the working condition type. Load status is divided into three levels based on load weight: light load, medium load, and heavy load. A load weight less than 30 tons is light load; a load weight greater than or equal to 30 tons and less than or equal to 50 tons is medium load; and a load weight greater than 50 tons is heavy load. Drive wheel operating status is divided into three states based on the difference in speed between the left and right drive wheels and the steering command signal: straight ahead, left turn, and right turn. A difference in speed between the left and right drive wheels less than 1 revolution per minute and no steering command indicates straight ahead; a left turn indicates a left turn if a left turn command is in effect or the left drive wheel speed is significantly lower than the right drive wheel speed; and a right turn indicates a right turn if a right turn command is in effect or the right drive wheel speed is significantly lower than the left drive wheel speed. The fuselage attitude is categorized into five states based on pitch and roll angles: flat ground, uphill, downhill, left roll, and right roll. A pitch angle with an absolute value less than 2 degrees and a roll angle with an absolute value less than 2 degrees indicates flat ground; a pitch angle greater than or equal to 2 degrees indicates uphill; a pitch angle less than or equal to -2 degrees indicates downhill; a roll angle greater than or equal to 2 degrees indicates left roll; and a roll angle less than or equal to -2 degrees indicates right roll. Track tension is categorized into three states based on the tension difference between the left and right tracks and its absolute value: normal tension, low tension, and high tension. A left-right tension difference of less than 500 N and a single track tension between 7000 N and 12000 N indicates normal tension; a single track tension below 7000 N indicates low tension; and a single track tension above 12000 N indicates high tension.

[0035] During implementation, the production data for the load weight was 45 tons, falling within the 30-50 ton range, thus classifying the load as medium load. The left and right drive wheel speeds were 126.1 rpm and 125.5 rpm respectively, with a speed difference of 0.6 rpm (126.1 - 125.5). Since 0.6 is less than 1, and the main control unit did not output a steering command, the drive wheel operation was determined to be straight-line. The production data for the pitch angle was 1.2 degrees, and the production data for the roll angle was 0.8 degrees. Both absolute values ​​were less than 2 degrees, thus classifying the fuselage attitude as flat ground.

[0036] The left track tension production data is 8502 N, and the right track tension production data is 8599 N. The tension difference between the left and right tracks is 8599 - 8502 = 97 N. Since 97 is less than 500, and both track tensions are within the range of 7000 N to 12000 N, the track tension is determined to be normal. Combining the results from these four dimensions, the production data for the current operating data is labeled as "normal tension for medium-load, straight-line, flat terrain."

[0037] After the operating condition label is bound, deviation identification processing performs time-series filtering and effective deviation determination on the track offset data. The time-series filtering window can be configured independently. In the specific implementation, the filtering window duration is 500 milliseconds, corresponding to 50 10-millisecond sampling points. Within the filtering window, 50 frames of yaw angle data are continuously collected, and the arithmetic mean of the yaw angles within the window is calculated as the effective offset of the window. In the specific implementation, the arithmetic mean of the yaw angles generated within the 500-millisecond window is 2.1 degrees. This includes several frames of instantaneous yaw angle jumps caused by road bumps, with the highest value in a single frame reaching 4.6 degrees and the lowest value in a single frame reaching 0.3 degrees. After arithmetic mean filtering, the instantaneous jumps are diluted by the majority of stable data within the window, and the effective offset stabilizes at 2.1 degrees.

[0038] Effective track deviation determination is based on the deviation threshold corresponding to the operating condition label. Different operating condition labels correspond to different yaw angle deviation thresholds. For straight-ahead operating conditions, the deviation threshold is 1.5 degrees. For turning operating conditions, the deviation threshold is relaxed accordingly to avoid misjudging normal turning as deviation. In specific implementation, the current operating condition label is medium-load, straight-ahead, flat ground, normal tension, corresponding to a deviation threshold of 1.5 degrees. The effective window offset is 2.1 degrees. Since 2.1 is greater than 1.5, it is determined that there is an effective track deviation, and the deviation direction is to the left. The determination is based on the left track speed of 125.5 rpm being higher than the right track speed of 124.8 rpm, and the left track traveling a greater distance than the right, causing the fuselage to turn to the left. If the effective window offset is less than or equal to 1.5 degrees, it is determined to be a momentary disturbance or normal attitude fluctuation, and subsequent correction calculations are not triggered. After operating condition stratification and deviation identification processing, the operating condition label and the effective deviation determination result are output together to the adaptive correction coefficient calculation module.

[0039] Example 3, building upon the operational condition stratification and effective deviation identification achieved in Example 2, further implements adaptive correction coefficient calculation and output constraints based on operational condition mapping. Specifically, it includes an adaptive correction coefficient calculation module for storing historical equipment correction data, establishing a mapping relationship between operating conditions and correction coefficients, matching corresponding correction coefficients based on real-time equipment operating conditions, and generating compensation correction coefficients for track speed adjustment. This adaptive correction coefficient calculation module has a built-in iteratively updatable operational condition correction coefficient mapping database, continuously collecting effective correction events under various operating conditions, and iteratively updating the baseline correction coefficients corresponding to each operating condition based on the collected historical correction event data. This adaptive correction coefficient calculation module is configured with fixed coefficient constraint logic to constrain the output range of the real-time generated speed compensation correction coefficients, standardizing the output parameter state of the correction coefficients.

[0040] The working condition correction coefficient mapping database pre-stores the correspondence between each working condition label and the reference correction coefficient. The reference correction coefficient represents the compensation ratio required to eliminate a unit speed difference under that working condition; its physical meaning is the ratio of the compensation speed to the measured speed difference. The value of the reference correction coefficient is affected by factors such as ground adhesion coefficient, track tension, load distribution, and equipment wear condition. The optimal coefficient varies under different working conditions, therefore, it needs to be stored separately for each working condition and iteratively updated. In the specific implementation process, the working condition labels stored in the mapping database include twelve typical working conditions, such as light-load straight-line flat ground, medium-load straight-line flat ground, heavy-load straight-line flat ground, light-load uphill, and medium-load uphill. Among them, the reference correction coefficient production data for the medium-load straight-line flat ground normal tension working condition is 0.82.

[0041] The correction coefficient matching process uses the operating condition label output in Example 2 as the search key to accurately match the corresponding benchmark correction coefficient in the mapping database. If an accurate match fails, the coefficient of the nearest operating condition is used for interpolation estimation. In the specific implementation, the current operating condition label is medium load, straight, flat ground, normal tension. An accurate match is successfully found in the mapping database, and the benchmark correction coefficient is retrieved as 0.82. This coefficient indicates that under the current operating condition, when the compensation speed is 0.82 times the measured speed difference, the deviation can be effectively eliminated without over-correction.

[0042] The compensation correction coefficient generation process calculates the compensation speed based on the matched baseline correction coefficient and the real-time speed difference. The calculation formula is: compensation speed equals the baseline correction coefficient multiplied by the left and right track speed difference, where the left and right track speed difference equals the left track speed minus the right track speed. A positive speed difference indicates that the left track is faster than the right track, resulting in a leftward deviation. The compensation direction is to decrease the left track speed while increasing the right track speed. In specific implementation, the production data for the left track speed is 125.5 rpm, and the production data for the right track speed is 124.8 rpm. The speed difference is calculated as 125.5 minus 124.8 equals 0.7 rpm. The baseline correction coefficient is 0.82, and the compensation speed is calculated as 0.82 multiplied by 0.7 equals 0.574 rpm, rounded to two decimal places as 0.57 rpm. After the compensation speed is generated, it is distributed to the left and right tracks according to the principle of equal distribution on both sides. The speed adjustment of the left track is half of the compensation speed, which is 0.574 divided by 2 equals 0.287 rpm, so we take 0.29 rpm and adjust it in the direction of deceleration. The speed adjustment of the right track is also 0.287 rpm, so we take 0.29 rpm and adjust it in the direction of acceleration. Therefore, the target speed on the left is 125.5 minus 0.29 equals 125.21 rpm, so we take 125.2 rpm. The target speed on the right is 124.8 plus 0.29 equals 125.09 rpm, so we take 125.1 rpm. After adjustment, the speed difference between the two sides is 125.2 minus 125.1 equals 0.1 rpm, and the tendency to deviate is effectively suppressed.

[0043] The coefficient constraint logic sets upper and lower limits for the reference correction coefficient and the compensation speed to prevent damage to the drive mechanism caused by excessive adjustment commands due to sensor malfunctions or sudden changes in operating conditions. The allowable range for the reference correction coefficient is 0.5 to 1.2 (based on production data). When the coefficient after matching or iteration is lower than 0.5, it is forcibly set to 0.5; when it is higher than 1.2, it is forcibly set to 1.2. The allowable upper limit for the compensation speed is 5% of the rated speed (130 rpm). 5% is calculated as 130 multiplied by 0.05, which equals 6.5 rpm. When the calculated compensation speed exceeds 6.5 rpm, it is forcibly set to 6.5 rpm. In actual implementation, the reference correction coefficient of 0.82 is within the range of 0.5 to 1.2, and the compensation speed of 0.574 rpm is much less than 6.5 rpm. Neither constraint is triggered, and the coefficients and compensation speed are output normally. If the speed difference reaches 10 revolutions per minute under a certain operating condition, the compensation speed is calculated to be 8.2 revolutions per minute with a coefficient of 0.82. Since 8.2 is greater than 6.5, the upper limit constraint is triggered, and the compensation speed is forced to be 6.5 revolutions per minute to avoid the drive unit receiving excessively large step commands.

[0044] After each effective correction event is completed, the iterative update process writes the working condition label, initial speed difference, residual speed difference after correction, and actual effective coefficient of the correction into the mapping database. The benchmark correction coefficient of the working condition is updated using an exponentially weighted moving average method. The update formula is that the new benchmark coefficient is equal to the old benchmark coefficient multiplied by 0.9 plus the actual effective coefficient of this time multiplied by 0.1. In the specific implementation process, the original reference coefficient for the medium-load straight-line flat road condition was 0.82. After this correction, the residual speed difference was 0.1 revolutions per minute. The correction efficiency was calculated as: initial difference 0.7 minus residual difference 0.1 divided by initial difference 0.7, which equals 0.857. Based on this, the actual effective coefficient for this operation was calculated to be 0.857. The updated reference coefficient was calculated as: 0.82 multiplied by 0.9 plus 0.857 multiplied by 0.1, which equals 0.738 plus 0.0857, which equals 0.8237. Retaining two decimal places, this is 0.82. This reference coefficient for the operating condition completed one micro-iteration. As correction events continue to accumulate, the reference coefficient will gradually converge to the optimal value for this operating condition. After coefficient calculation, constraint verification, and iterative update processing, the compensation correction coefficient and the target speeds on both sides are output to the closed-loop control command output module.

[0045] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A WJ999 type track alignment system, characterized in that, The system includes: The multi-source data acquisition and preprocessing module is used to collect equipment operation data and load condition raw data under track running conditions, and to perform unified regularization and preprocessing on the collected heterogeneous raw data. The working condition stratification and deviation identification module is used to extract features from the preprocessed standardized data, distinguish different operating conditions of the equipment according to the data features and bind corresponding working condition labels, filter the track offset data based on the labeled working condition labels, and identify the effective track deviation state. The adaptive correction coefficient calculation module is used to store historical correction data of the equipment, establish a corresponding mapping relationship between operating conditions and correction coefficients, match the corresponding correction coefficients according to the real-time operating conditions of the equipment, and generate compensation correction coefficients for track speed adjustment. The closed-loop control command output module is used to output differentiated speed adjustment commands to the drive units on both sides of the track according to the generated compensation correction coefficient, continuously collect the corrected equipment operation data to complete the closed-loop state determination, and execute corresponding control operations according to the abnormal operation state of the equipment. The data storage and human-computer interaction module is used to classify and store equipment operating condition data, track alignment process data, and full data of equipment abnormal events. It is also configured with data visualization display function and manual parameter correction interaction function.

2. The WJ999 track alignment system according to claim 1, characterized in that: The multi-source data acquisition and preprocessing module has built-in fixed preprocessing logic, which sequentially performs time-series alignment, parameter calibration, and abnormal data removal on the various types of heterogeneous raw data collected, and unifies the multi-source heterogeneous data into a standardized working condition dataset.

3. The WJ999 track alignment system according to claim 1, characterized in that: The working condition stratification and deviation identification module classifies working condition types based on multi-dimensional equipment operation data characteristics, distinguishes different operating conditions according to the characteristic differences of equipment load status, drive wheel operation status, body attitude status, and track tension status, and matches and binds corresponding working condition tags to the real-time collected operating data.

4. The WJ999 track alignment system according to claim 1, characterized in that: The working condition stratification and deviation identification module is configured with a customizable time-series filtering window. Based on the effective offset sampling data within the filtering window, it filters out invalid offset signals corresponding to instantaneous interference generated during equipment operation, thereby determining the actual track misalignment event.

5. The WJ999 track alignment system according to claim 1, characterized in that: The adaptive correction coefficient calculation module has a built-in iteratively updatable working condition-correction coefficient mapping database, which continuously collects effective correction events under each working condition and iteratively updates the benchmark correction coefficients corresponding to each working condition based on the collected historical correction event data.

6. The WJ999 track alignment system according to claim 1, characterized in that: The adaptive correction coefficient calculation module is configured with fixed coefficient constraint logic to constrain the output range of the real-time generated speed compensation correction coefficient and standardize the output parameter state of the correction coefficient.

7. The WJ999 track alignment system according to claim 1, characterized in that: The closed-loop control command output module is set up with a hierarchical and progressive correction judgment process. After a single correction operation is completed, the off-track running status is verified. If the off-track status is not eliminated, a secondary fine-tuning correction process is started. If the off-track status still cannot be eliminated after multiple fine-tuning operations, an abnormal working condition warning signal is output.

8. The WJ999 track alignment system according to claim 1, characterized in that, The closed-loop control command output module is associated with the overall speed control logic of the equipment. When the track is continuously detected to be misaligned and multiple correction operations fail to eliminate the misalignment, the system triggers a protective control operation to reduce the overall speed.

9. The WJ999 track alignment system according to claim 1, characterized in that, The data storage and human-computer interaction module is configured with dedicated event classification and marking logic. Independent marks are set for events such as correction failure, abnormal operation offset, and abnormal equipment parameters. The event marks are bound to the full amount of operating data for the corresponding time period and then classified and archived for storage.

10. The WJ999 track alignment system according to claim 1, characterized in that, The data storage and human-computer interaction module can call up the historical calibration data archived by the system, and visually display the deviation events, calibration process and coefficient iteration records corresponding to each working condition. It also supports manual intervention to correct and update the calibration parameters built into the system.

Images