Lifting trajectory tracking method and system fusing digital hydraulic and intelligent control

Abstract

The application provides a lifting motion trajectory tracking method and system combining digital hydraulic pressure and intelligent control, and relates to the technical field of industrial control. First, a two-way coupling linkage feature set of the lifting actuator motion posture and the digital hydraulic drive unit flow pressure is constructed, which carries two-way adaptive association information. Then, the two-way coupling linkage feature set is associated with the lifting operation dynamic load feature for cross-domain association empowerment processing to generate dynamic linkage empowerment parameters. Based on the dynamic linkage empowerment parameters, two-way collaborative driving control of intelligent control and digital hydraulic drive is realized to generate a trajectory tracking two-way driving control signal set. The trajectory tracking two-way driving control signal set is subjected to trajectory evolution deviation reverse source tracing correction to generate a target two-way driving control signal set. Finally, the target two-way driving signal set is used to regulate the lifting actuator posture evolution time sequence and the hydraulic drive dynamic output state to realize high-precision tracking of the lifting motion trajectory.

Description

Technical Field

[0001] This invention relates to the field of industrial control technology, and more specifically, to a lifting motion trajectory tracking method and system that integrates digital hydraulics and intelligent control. Background Technology

[0002] In industrial production and engineering operations, lifting actuators are widely used in various scenarios, such as construction lifting, logistics handling, and machinery manufacturing. Accurate tracking of their motion trajectory is crucial for ensuring operational safety and improving efficiency and quality. Traditional lifting motion trajectory tracking methods often employ a single control strategy, making it difficult to fully consider the complex motion posture changes of the lifting actuator and the dynamic flow and pressure characteristics of the digital hydraulic drive unit. For example, some methods based on simple open-loop control cannot adjust control parameters in real time according to load changes during actual operation, leading to deviations in the lifting trajectory and affecting operational accuracy. While some closed-loop control methods can provide a certain degree of feedback adjustment, they lack in-depth understanding of the bidirectional coupling relationship between the lifting actuator's motion posture and the digital hydraulic drive unit, making it difficult to achieve precise coordinated control.

[0003] Meanwhile, existing technologies often fail to cross-domain correlate load characteristics with the motion posture of the lifting actuator and the characteristics of digital hydraulic drive when dealing with dynamic load changes in lifting operations. This makes it impossible to dynamically adjust control parameters according to the continuous evolution of the load, further limiting the accuracy and stability of lifting motion trajectory tracking. Summary of the Invention

[0004] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide a lifting motion trajectory tracking method integrating digital hydraulics and intelligent control, the method comprising: A two-way coupling linkage feature set is constructed based on the continuous evolution characteristics of the motion posture of the lifting actuator and the dynamic coupling characteristics of the flow and pressure of the digital hydraulic drive unit. The two-way coupling linkage feature set carries the two-way adaptation and correlation information between the posture evolution of the lifting trajectory and the dynamic coupling of the hydraulic drive. The bidirectional coupling linkage feature set is cross-domain correlated with the dynamic load continuous evolution feature of the lifting operation to generate dynamic linkage empowerment parameters for digital hydraulics and intelligent control. Based on the aforementioned dynamic linkage empowerment parameters, a two-way coordinated drive control of intelligent control and digital hydraulic drive is realized, generating a set of two-way drive control signals for trajectory tracking of the lifting actuator. The trajectory tracking bidirectional drive control signal set is subjected to reverse source correction processing for trajectory evolution deviation to generate the target bidirectional drive control signal set for the lifting actuator. The attitude evolution timing and dynamic output state of the hydraulic drive are controlled based on the target bidirectional drive control signal set.

[0005] In another aspect, embodiments of the present invention also provide a lifting motion trajectory tracking system integrating digital hydraulics and intelligent control, including a processor and a machine-readable storage medium connected to the processor. The machine-readable storage medium is used to store programs, instructions, or code, and the processor is used to run the programs, instructions, or code in the machine-readable storage medium to implement the above-described method.

[0006] Based on the above, this embodiment of the invention constructs a bidirectional coupling linkage feature set based on the continuous evolution characteristics of the lifting actuator's motion posture and the dynamic coupling characteristics of the flow and pressure of the digital hydraulic drive unit. It deeply explores the bidirectional adaptation and correlation information between the lifting trajectory posture evolution and the dynamic coupling of the hydraulic drive. This bidirectional coupling linkage feature set is then cross-domain correlated with the continuous evolution characteristics of the dynamic load of the lifting operation, generating dynamic linkage empowerment parameters for digital hydraulics and intelligent control. This allows the control parameters to be adjusted in real time according to the dynamic changes in the load. Based on these dynamic linkage empowerment parameters, intelligent control and digital hydraulic drive are simultaneously achieved. By generating a set of bidirectional drive control signals for trajectory tracking of the lifting actuator through coordinated drive control, the problem of single control strategy and poor coordination in traditional methods is effectively solved, thus improving the accuracy of trajectory tracking. The bidirectional drive control signal set for trajectory tracking is then subjected to reverse tracing correction of trajectory evolution deviation to generate a target bidirectional drive control signal set, which further optimizes the control signals and reduces trajectory tracking errors. Finally, based on the target bidirectional drive control signal set, the attitude evolution timing of the lifting actuator and the dynamic output state of the hydraulic drive are controlled, achieving high-precision and stable tracking of the lifting motion trajectory and significantly improving the efficiency and quality of lifting operations. Attached Figure Description

[0007] Figure 1 This is a schematic diagram of the execution flow of the lifting motion trajectory tracking method that integrates digital hydraulics and intelligent control provided in an embodiment of the present invention.

[0008] Figure 2 This is a schematic diagram of the hardware architecture of the lifting motion trajectory tracking system that integrates digital hydraulics and intelligent control, provided in an embodiment of the present invention. Detailed Implementation

[0009] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a flowchart illustrating a lifting motion trajectory tracking method integrating digital hydraulics and intelligent control, provided in one embodiment of the present invention. The following is a detailed description of this lifting motion trajectory tracking method integrating digital hydraulics and intelligent control.

[0010] Step S110: Construct a bidirectional coupling linkage feature set based on the continuous evolution characteristics of the motion posture of the lifting actuator and the dynamic coupling characteristics of the flow and pressure of the digital hydraulic drive unit. The bidirectional coupling linkage feature set carries the bidirectional adaptation association information between the posture evolution of the lifting trajectory and the dynamic coupling of the hydraulic drive.

[0011] In this embodiment, the lifting actuator of a certain type of hydraulic crane is used as the application scenario. This lifting actuator mainly consists of components such as a telescopic boom, a luffing cylinder, and a hydraulic motor. The digital hydraulic drive unit includes a proportional multi-way valve, a hydraulic pump, and an accumulator. When constructing the bidirectional coupled linkage feature set, it is first necessary to obtain the continuous evolution characteristics of the lifting actuator's motion posture and the dynamic coupling characteristics of the flow and pressure of the digital hydraulic drive unit. Then, a series of feature processing methods are used to fuse the two, forming a feature set that reflects the bidirectional adaptive correlation information between posture evolution and hydraulic drive dynamic coupling.

[0012] Step S111: Collect motion attitude data of the lifting actuator during the trajectory stages of lifting start, trajectory advancement, and lifting stop. Extract continuous evolution features of attitude angle, continuous evolution features of attitude stability, and continuous evolution features of attitude adjustment response from the collected motion attitude data to generate a continuous evolution feature set of the lifting actuator's motion attitude. The continuous evolution feature set of the motion attitude includes feature sequences characterizing attitude angle changes in different lifting stages, feature sequences characterizing attitude stability, and feature sequences characterizing the time delay and adjustment amplitude from the input of the drive control signal to the start of attitude adjustment.

[0013] In the aforementioned application scenarios of hydraulic cranes, the lifting start-up phase refers to the period from when the crane receives the lifting command to when the telescopic boom begins to leave its initial position; the trajectory advancement phase refers to the period when the telescopic boom moves from its initial position to its target position according to a preset trajectory; and the lifting stop phase refers to the period when the telescopic boom approaches its target position and finally stops moving. To collect motion attitude data, attitude sensors, such as gyroscopes and tilt sensors, are installed at key locations on the telescopic boom, with a sampling frequency set to 100 Hz to ensure that subtle changes in attitude can be captured.

[0014] Step S1111: Control the lifting actuator to complete lifting actions with different loads and different trajectory plans in a simulated lifting scenario, synchronously collect motion attitude data of the trajectory stages of lifting start, trajectory advancement and lifting stop, record the attitude change data of each trajectory stage, and generate the original motion attitude data set of the lifting actuator.

[0015] In this embodiment, the simulated lifting scenarios include different load conditions, such as goods with different weights of 5 tons, 10 tons, 15 tons, etc., and different trajectory plans, such as straight trajectories, arc trajectories, etc. The lifting actuator is controlled by the crane control system to perform lifting actions under these simulated scenarios, and at the same time, the motion attitude data is collected in real time by the attitude sensor. For example, in the lifting scenario with a 5-ton load and a straight trajectory, starting from the start of lifting, the attitude data such as the pitch angle and azimuth angle of the telescopic boom are recorded every 0.01 seconds until the lifting stops. The above data is arranged in chronological order to form a subset of the original motion attitude data in this scenario. By integrating the subsets of the original motion attitude data under all simulated scenarios, a set of the original motion attitude data of the lifting actuator is generated.

[0016] Step S1112: Perform invalid data elimination and missing data supplementation processing on the set of the original motion attitude data of the lifting actuator, remove abnormal interference data, and supplement the missing data using the evolution trend of adjacent valid data to generate preprocessed motion attitude data.

[0017] Due to reasons such as sensor failures and electromagnetic interference, there may be invalid data in the set of the original motion attitude data collected. For example, the pitch angle suddenly appears with a value exceeding the normal range, or the data is missing at a certain time point. For invalid data, it is eliminated by setting a threshold. For example, when the change rate of the pitch angle exceeds 10 degrees per second, it is determined as abnormal interference data and removed. For missing data, linear interpolation is used for supplementation, that is, based on the adjacent valid data before and after the missing data, the attitude data at the missing moment is calculated. For example, if the pitch angles at times t1 and t3 are α1 and α3 respectively, and the data is missing at time t2 (t1 < t2 < t3), then the pitch angle α2 at time t2 = α1 + (α3 - α1) × (t2 - t1) / (t3 - t1). After the above processing, the preprocessed motion attitude data is obtained.

[0018] Step S1113: Extract the attitude angle change data in different trajectory stages from the preprocessed motion attitude data, and generate an attitude angle continuous evolution feature based on the attitude angle change data. The attitude angle continuous evolution feature carries the continuous change information of the attitude angle over time in different trajectory stages, and completely presents the change process of the attitude angle from the initial value to the target value.

[0019] The preprocessed motion attitude data includes attitude angle data for three trajectory stages: lift start, trajectory advancement, and lift stop. For each trajectory stage, the attitude angle variation data over time is extracted. For example, in the lift start stage, from time t0 to t1, the pitch angle gradually changes from the initial value α0 to α1. The pitch angle data during this time are arranged in chronological order to form the attitude angle variation sequence for this stage. Similarly, the attitude angle variation sequences for the trajectory advancement and lift stop stages are extracted separately. Then, the attitude angle variation sequences of these three stages are combined to form a continuous attitude angle evolution feature. This continuous attitude angle evolution feature can be represented as a three-dimensional array, where the first dimension represents the trajectory stage (lift start, trajectory advancement, lift stop), the second dimension represents the time point, and the third dimension represents the specific numerical value of the attitude angle (such as pitch angle, azimuth angle, etc.).

[0020] Step S1114: Extract attitude fluctuation data for different trajectory stages from the preprocessed motion attitude data, and generate attitude stability continuous evolution features based on the attitude fluctuation data. The attitude stability continuous evolution features carry the continuous change information of attitude fluctuations in different trajectory stages and reflect the change in the stability of the attitude during the evolution process.

[0021] Attitude fluctuation data can be obtained by calculating the deviation between the attitude angle at each time point and the average attitude angle within that trajectory phase. For example, in the trajectory advancement phase, the average pitch angle α_avg is calculated for all time points within that phase, and the pitch angle fluctuation value at each time point is Δα = α_i - α_avg, where α_i is the pitch angle at that time point. Arranging these fluctuation values ​​in chronological order forms the attitude fluctuation sequence for the trajectory advancement phase. Similarly, the attitude data for the lifting start-up and lifting stop phases are processed to obtain their respective attitude fluctuation sequences. Combining these three attitude fluctuation sequences generates a continuous evolution feature for attitude stability. This continuous evolution feature for attitude stability is also a three-dimensional array, with the first dimension representing the trajectory phase, the second dimension representing the time point, and the third dimension representing the attitude fluctuation value.

[0022] Step S1115: Extract the time difference data between the input time of the drive control signal and the time of the attitude adjustment response, as well as the attitude adjustment amplitude data, from the preprocessed motion attitude data. Generate a continuous evolution feature of the attitude adjustment response based on the time difference data between the input time of the drive control signal and the time of the attitude adjustment response, and the attitude adjustment amplitude data. The continuous evolution feature of the attitude adjustment response carries the continuous change information of the attitude response speed and response amplitude to the drive control signal at different trajectory stages, reflecting the response efficiency and effect of the attitude to the drive control signal.

[0023] The input time of the drive control signal can be obtained from the crane's control system. The attitude adjustment response time is determined by analyzing data collected by the attitude sensors; the attitude adjustment response time is when the attitude angle begins to change. The time difference data is the attitude adjustment response time minus the drive control signal input time. The attitude adjustment amplitude data is the difference in attitude angle before and after attitude adjustment. For example, in the lifting start-up phase, the lifting drive control signal is input at time t0, and the attitude angle begins to change at time t1. The time difference Δt = t1 - t0, and the attitude adjustment amplitude Δα = α1 - α0 (α0 is the attitude angle at time t0, and α1 is the attitude angle after attitude stabilization). The time difference data and attitude adjustment amplitude data are also acquired in the trajectory advancement phase and the lifting stop phase. Arranging the above data according to the trajectory phase and time sequence forms a continuous evolution characteristic of the attitude adjustment response. This continuous evolution characteristic of the attitude adjustment response is a two-dimensional array, where each row represents a trajectory phase, and each column represents the time difference and attitude adjustment amplitude corresponding to a time point.

[0024] Step S1116: Perform trajectory stage association processing on the attitude angle continuous evolution feature, attitude stability continuous evolution feature and attitude adjustment response continuous evolution feature, bind each feature to the corresponding trajectory stage, and generate the comprehensive motion attitude evolution feature corresponding to each trajectory stage.

[0025] Data belonging to the same trajectory stage from the continuous evolution features of attitude angle, attitude stability, and attitude adjustment response are extracted and combined. For example, for the lifting start-up stage, the attitude angle change sequence, attitude fluctuation sequence, time difference, and attitude adjustment amplitude data of this stage are combined to form the comprehensive motion attitude evolution feature of the lifting start-up stage. Similarly, comprehensive motion attitude evolution features for the trajectory advancement stage and the lifting stop stage are generated. Each comprehensive motion attitude evolution feature contains information on various aspects such as attitude angle changes, stability, and response within that trajectory stage.

[0026] Step S1117: Perform time dimension expansion processing on the comprehensive motion posture evolution features corresponding to each trajectory stage, associate the comprehensive motion posture evolution features of adjacent trajectory stages, and generate extended motion posture features containing posture evolution association information of adjacent trajectory stages.

[0027] Taking the lifting start-up phase and the trajectory propulsion phase as examples, the attitude angle at the end of the lifting start-up phase is the same as the attitude angle at the beginning of the trajectory propulsion phase. By associating the attitude data at the end of the lifting start-up phase with the attitude data at the beginning of the trajectory propulsion phase, a transition relationship between the two phases is established. For example, in the time dimension, connecting the last time point of the lifting start-up phase with the first time point of the trajectory propulsion phase allows the extended motion attitude features to reflect the continuous evolution process from lifting start-up to trajectory propulsion. Similarly, by associating the comprehensive motion attitude evolution features of the trajectory propulsion phase and the lifting stop phase, extended motion attitude features are ultimately generated.

[0028] Step S1118: Perform weight allocation processing on the extended motion posture features. Based on the degree of influence of different posture evolution features in reflecting the overall state of motion posture, assign corresponding weight coefficients to generate weight coefficients corresponding to different posture evolution features.

[0029] Experts in the field were invited to evaluate the importance of continuous evolution characteristics of attitude angle, attitude stability, and attitude adjustment response in reflecting the overall state of motion posture, based on their experience and practical application needs. For example, experts considered the continuous evolution characteristics of attitude angle to be the most important, assigning it a weighting coefficient of 0.5; the continuous evolution characteristics of attitude stability were second most important, assigning them a weighting coefficient of 0.3; and the continuous evolution characteristics of attitude adjustment response had a weighting coefficient of 0.2. The sum of these weighting coefficients needed to be equal to 1 to ensure the rationality of subsequent weighting processing.

[0030] Step S1119: The extended motion posture features are weighted by the weight coefficients. The extended motion posture features of different dimensions are weighted according to the weight coefficients to generate weighted motion posture evolution feature components, and the weighted motion posture evolution feature components are combined into a weighted motion posture evolution feature vector.

[0031] The attitude angle continuous evolution feature component in the extended motion attitude features is multiplied by a weighting factor of 0.5, the attitude stability continuous evolution feature component by a weighting factor of 0.3, and the attitude adjustment response continuous evolution feature component by a weighting factor of 0.2 to obtain their respective weighted components. Then, these weighted components are arranged and combined according to time order and trajectory stage order to form a high-dimensional weighted motion attitude evolution feature vector. Each element in this weighted motion attitude evolution feature vector corresponds to the attitude feature data at a certain moment and a certain trajectory stage after weighting processing.

[0032] Step S11110: Generate a continuous evolution feature group of motion attitude of the lifting actuator based on the weighted motion attitude evolution feature vector.

[0033] The weighted motion attitude evolution feature vector generated above is used as the motion attitude continuous evolution feature group of the lifting actuator. This motion attitude continuous evolution feature group contains comprehensive information such as the attitude angle change, stability and response of the lifting actuator at different trajectory stages, and the importance of different features is highlighted through weighted processing.

[0034] Step S112: Collect dynamic data of flow and pressure of the digital hydraulic drive unit under different drive and control commands. Extract the flow and pressure co-evolution characteristics, the hysteresis evolution characteristics of pressure with flow, and the flow and pressure stability evolution characteristics from the collected dynamic data of flow and pressure to generate a dynamic coupling feature group of flow and pressure of the digital hydraulic drive unit. The dynamic coupling feature group of flow and pressure includes a feature sequence characterizing the co-change of flow and pressure, a feature sequence characterizing the hysteresis of pressure with respect to flow, and a feature sequence characterizing the flow and pressure entering a stable state.

[0035] In the digital hydraulic drive unit of a hydraulic crane, drive commands include commands to control the hydraulic pump displacement and commands to control the opening degree of the proportional multi-way valve. Flow sensors and pressure sensors are installed at the hydraulic pump outlet and the inlet and outlet of the proportional multi-way valve to collect flow and pressure data under different drive commands. The sampling frequency is also set to 100 Hz to ensure the timeliness and accuracy of the data.

[0036] Step S1121: Control the digital hydraulic drive unit to operate under different combinations of drive control commands, synchronously collect flow and pressure data, record the dynamic changes in flow and pressure under different drive control commands, and generate a set of dynamic raw data of flow and pressure of the digital hydraulic drive unit.

[0037] The drive control command combinations include different pump displacement commands (e.g., 20%, 50%, 80% of rated displacement) and different multi-way valve opening commands (e.g., 10%, 30%, 50%, 70%, 90% of maximum opening). Under each drive control command combination, the digital hydraulic drive unit operates, simultaneously acquiring flow and pressure data via flow and pressure sensors. For example, when the pump displacement command is 50% of rated displacement and the multi-way valve opening command is 50% of maximum opening, flow and pressure data are recorded every 0.01 seconds from the start of the command until the flow and pressure stabilize. This data is then arranged chronologically to form a subset of dynamic raw flow and pressure data for that drive control command combination. Integrating these subsets of raw data from all drive control command combinations generates a complete set of dynamic raw flow and pressure data.

[0038] Step S1122: Perform data cleaning processing on the raw flow and pressure dynamic data set of the digital hydraulic drive unit to remove noise data and abnormal peak data, and generate cleaned flow and pressure dynamic data.

[0039] Noise data is mainly caused by small fluctuations due to sensor inaccuracies or external electromagnetic interference, which can be removed using a moving average filtering method. For example, a moving average with a window size of 5 can be used for flow data, where the value for each data point is the average of that point and the two data points before and after it. Abnormal peak data refers to sudden maximum or minimum values ​​in flow or pressure. These are identified and removed by setting reasonable threshold ranges; for example, when the flow value exceeds 150% of the normal operating range, it is identified as abnormal peak data and removed. After processing, the cleaned dynamic flow and pressure data is obtained.

[0040] Step S1123: Extract the coordinated change data of flow and pressure over time from the cleaned flow and pressure dynamic data, and generate flow and pressure coordinated evolution characteristics based on the coordinated change data of flow and pressure over time. The flow and pressure coordinated evolution characteristics carry the synchronous change information of flow and pressure over time and reflect the dynamic correlation between the two.

[0041] For each combination of drive and control commands, the dynamic data of flow and pressure are used to extract the time-varying sequences of flow and pressure. For example, under a certain combination of drive and control commands, the flow sequence is Q(t0), Q(t1), ..., Q(tn), and the pressure sequence is P(t0), P(t1), ..., P(tn), where t0 to tn are time points. The flow and pressure sequences are combined to form a two-dimensional array, where each row represents a time point, the first column is the flow value, and the second column is the pressure value. This two-dimensional array represents the flow-pressure co-evolution feature components under that combination of drive and control commands. The flow-pressure co-evolution feature components under all combinations of drive and control commands are integrated to generate the flow-pressure co-evolution feature.

[0042] Step S1124: Extract the time delay data of pressure change relative to flow rate change from the dynamic data of flow rate and pressure after cleaning. Generate the hysteresis evolution feature of pressure change with flow rate based on the time delay data of pressure change relative to flow rate change. The hysteresis evolution feature of pressure change with flow rate carries the time interval information of pressure change lagging behind flow rate change, reflecting the dynamic response characteristics of the hydraulic system.

[0043] By analyzing the flow and pressure change curves, the moments when flow and pressure begin to change are determined, and the time difference between these moments is the time delay of pressure change relative to flow change. For example, under a certain combination of drive and control commands, flow begins to increase at time tq, and pressure begins to increase at time tp, with a time delay Δt = tp - tq. For the dynamic flow and pressure data under each combination of drive and control commands, the time delay data at multiple time points are calculated to form a time delay sequence. By integrating the time delay sequences under all combinations of drive and control commands, the hysteresis evolution characteristics of pressure changing with flow are generated.

[0044] Step S1125: Extract the time and stable value data of the flow and pressure entering a stable state from the dynamic data of the flow and pressure after cleaning. Generate flow and pressure stability evolution characteristics based on the time and stable value data of the flow and pressure entering a stable state. The flow and pressure stability evolution characteristics carry information on the time and stable value of the flow and pressure reaching a stable state, reflecting the steady-state performance of the hydraulic system.

[0045] The criteria for determining when flow and pressure reach a stable state are that, within a certain period (e.g., 2 seconds), the rate of change of both flow and pressure is less than a set threshold (e.g., flow rate change rate less than 0.5% / second, pressure rate change rate less than 1% / second). For the dynamic data of flow and pressure under each combination of drive and control commands, the time ts and stable values ​​Qs and Ps for flow and pressure to reach a stable state are determined. The above data are arranged in the order of drive and control command combinations to form a flow and pressure stability evolution characteristic. This flow and pressure stability evolution characteristic is a two-dimensional array, where each row represents a drive and control command combination, and the columns are flow stabilization time, flow stabilization value, pressure stabilization time, and pressure stabilization value, respectively.

[0046] Step S1126: Perform drive control command association processing on the flow-pressure co-evolution characteristics, the hysteresis evolution characteristics of pressure with flow rate change, and the flow-pressure stable evolution characteristics, bind each characteristic with the corresponding drive control command combination, and generate the comprehensive flow-pressure dynamic coupling characteristics corresponding to each drive control command combination.

[0047] Data belonging to the same drive-control command combination from the flow-pressure co-evolution characteristics, the hysteresis evolution characteristics of pressure changing with flow rate, and the flow-pressure stable evolution characteristics are extracted and combined. For example, for a drive-control command combination where the pump displacement command is 50% of the rated displacement and the multi-way valve opening command is 50% of the maximum opening, the flow-pressure co-evolution characteristic component, the hysteresis evolution characteristic component, and the stable evolution characteristic component under this combination are combined together to form the comprehensive flow-pressure dynamic coupling characteristics of this drive-control command combination.

[0048] Step S1127: Perform instruction dimension expansion processing on the comprehensive flow pressure dynamic coupling feature corresponding to each drive control command combination, associate the comprehensive flow pressure dynamic coupling features of adjacent drive control command combinations, and generate extended flow pressure dynamic coupling features containing the flow pressure dynamic coupling association information of adjacent drive control command combinations.

[0049] When the combination of drive control commands changes, such as from 50% pump displacement and 50% opening to 60% pump displacement and 50% opening, the flow and pressure states at the end of the previous command combination will affect the flow and pressure states at the beginning of the next command combination. By associating the combined flow and pressure dynamic coupling characteristics of adjacent drive control command combinations, for example, by using the stable flow and pressure values ​​of the previous command combination as the initial conditions of the next command combination, extended flow and pressure dynamic coupling characteristics are generated to reflect the dynamic coupling process of flow and pressure when drive control commands change continuously.

[0050] Step S1128: Perform weight allocation processing on the extended flow-pressure dynamic coupling features. Based on the degree of influence of different flow-pressure dynamic coupling features in reflecting the overall state of the hydraulic drive unit, assign corresponding weight coefficients to generate weight coefficients corresponding to different flow-pressure dynamic coupling features.

[0051] Similarly, domain experts were invited to assess the importance of the flow-pressure co-evolution characteristics, the lag evolution characteristics of pressure with flow rate changes, and the stable evolution characteristics of flow-pressure. It is assumed that the expert assessment results are: a weighting coefficient of 0.4 for the flow-pressure co-evolution characteristics, 0.3 for the lag evolution characteristics of pressure with flow rate changes, and 0.3 for the stable evolution characteristics of flow-pressure, totaling 1.

[0052] Step S1129: The extended flow pressure dynamic coupling feature is weighted by the weighting coefficient. The extended flow pressure dynamic coupling feature of different dimensions is weighted according to the weighting coefficient to generate weighted flow pressure dynamic coupling feature components, and the weighted flow pressure dynamic coupling feature components are combined into a weighted flow pressure dynamic coupling feature vector.

[0053] The flow-pressure co-evolution feature component in the extended flow-pressure dynamic coupling feature is multiplied by 0.4, the hysteresis evolution feature component by 0.3, and the stable evolution feature component by 0.3 to obtain their respective weighted components. These weighted components are then arranged and combined according to the order of drive and control commands and the time order to form a weighted flow-pressure dynamic coupling feature vector.

[0054] Step S11210: Generate the flow and pressure dynamic coupling feature group of the digital hydraulic drive unit based on the weighted flow and pressure dynamic coupling feature vector.

[0055] The weighted flow-pressure dynamic coupling feature vector generated above is used as the flow-pressure dynamic coupling feature group of the digital hydraulic drive unit. This flow-pressure dynamic coupling feature group comprehensively reflects the flow-pressure dynamic coupling characteristics of the digital hydraulic drive unit under different drive and control commands.

[0056] Step S113: Perform feature dimension collaborative mapping processing on the motion attitude continuous evolution feature group of the lifting actuator and the flow pressure dynamic coupling feature group of the digital hydraulic drive unit. By matching the feature dimension attributes of the motion attitude continuous evolution feature group and the flow pressure dynamic coupling feature group, establish a cross-dimensional correlation mapping relationship and generate a motion hydraulic bidirectional coupling feature vector with cross-domain correlation dimension.

[0057] The dimensions of the continuous evolution feature group of motion attitude include trajectory stage, time point, and attitude feature type (attitude angle, stability, response), etc.; the dimensions of the dynamic coupling feature group of flow and pressure include drive and control command combination, time point, and flow and pressure feature type (co-evolution, hysteresis evolution, stable evolution), etc. By analyzing the intrinsic relationship between the motion attitude of the lifting actuator and the flow and pressure of the digital hydraulic drive unit, for example, a specific drive and control command combination will lead to a specific flow and pressure change, which in turn causes a specific attitude evolution. Based on the above relationship, the time-corresponding feature dimensions in the continuous evolution feature group of motion attitude and the dynamic coupling feature group of flow and pressure are matched to establish a correlation mapping relationship. For example, at a certain time point in the lifting start-up phase, the flow and pressure feature under the corresponding drive and control command combination is mapped to the attitude feature at that time point. Through the above mapping, the dimensions of the two feature groups are fused to generate a motion-hydraulic bidirectional coupling feature vector, the dimensions of which simultaneously contain relevant information about motion attitude and flow and pressure.

[0058] Step S114: Perform stage division and feature matching processing on the motion hydraulic bidirectional coupling feature vector of the lifting trajectory. Divide the trajectory stages according to the stroke nodes of the preset lifting trajectory, match the attitude evolution threshold feature and hydraulic drive coupling threshold feature corresponding to each trajectory stage, and generate matching association information of attitude evolution threshold feature and hydraulic drive coupling threshold feature corresponding to each stage of the preset lifting trajectory.

[0059] The stroke nodes of the pre-set lifting trajectory can be set according to actual operational needs. For example, the trajectory from 0 meters to 10 meters can be divided into three stages: 0-3 meters is the lifting start-up stage, 3-7 meters is the trajectory advancement stage, and 7-10 meters is the lifting stop stage. For each trajectory stage, attitude evolution threshold features are preset, such as the range of pitch angle changes, the maximum allowable value of attitude fluctuation, and the maximum allowable delay of attitude adjustment response. At the same time, hydraulic drive coupling threshold features are set, such as the upper and lower limits of flow rate, the upper and lower limits of pressure, and the maximum allowable value of pressure lag time. Then, the features belonging to each trajectory stage in the motion-hydraulic bidirectional coupling feature vector are matched with the corresponding threshold features to determine whether the actual features are within the threshold range, and the matching results are recorded to generate matching association information. For example, in the lifting start-up stage, if the actual pitch angle change range is within the preset threshold and the corresponding flow rate and pressure features are also within the preset threshold, the matching is successful; otherwise, the matching fails, and the deviation information is recorded.

[0060] Step S115: Based on the matching and association information, mine the bidirectional linkage relationship between motion posture and hydraulic drive, analyze the changes in the demand for hydraulic drive parameters caused by the posture evolution of the lifting actuator and the influence of changes in hydraulic drive parameters on posture evolution, and generate the characteristics of the evolution of the demand for hydraulic drive parameters caused by the posture evolution of the lifting actuator and the characteristics of the influence of changes in hydraulic drive parameters on posture evolution.

[0061] Based on the matching and correlation information, when the attitude evolution characteristics approach or exceed a threshold, the changes in hydraulic drive parameters are analyzed to determine the attitude evolution's demand for hydraulic drive parameters. For example, when the attitude adjustment response delay is too long, it may be necessary to increase the hydraulic flow rate to improve the response speed, thus obtaining the attitude evolution's flow rate demand evolution characteristics. Conversely, when the hydraulic drive parameters change, such as a sudden increase in flow rate, the changes in attitude evolution characteristics are analyzed, such as whether the attitude adjustment amplitude increases or the adjustment speed increases, thus obtaining the influence characteristics of hydraulic drive parameter changes on attitude evolution. These characteristics can be obtained by statistically analyzing the parameter change patterns under different matching conditions, for example, generating a correlation curve between the attitude angle change rate and the flow rate change rate as part of the influence characteristics.

[0062] Step S116: Deeply fuse the bidirectional linkage relationship with the contour evolution features of the lifting preset trajectory, integrate the trajectory contour evolution information, attitude evolution information and hydraulic drive coupling evolution information, and generate bidirectional fused features that fuse trajectory contour evolution, attitude evolution and hydraulic drive coupling evolution.

[0063] The contour evolution features of the pre-set lifting trajectory include the trajectory's shape (e.g., straight line, curve), the coordinates of points on the trajectory, and changes in the trajectory's slope. These contour evolution features are combined with a two-way linkage relationship (the mutual influence between attitude and hydraulic drive). For example, specific attitude adjustments and hydraulic drive parameters are required in the curved sections of the trajectory. By associating each point on the trajectory contour with its corresponding attitude evolution feature and hydraulic drive coupling evolution feature, a three-dimensional fused feature is formed, where one dimension represents the trajectory contour position, one dimension represents the attitude evolution feature, and one dimension represents the hydraulic drive coupling evolution feature. In this way, the two-way fused feature can comprehensively reflect the combined situation of trajectory, attitude, and hydraulic drive during the lifting process.

[0064] Step S117: Perform spatiotemporal continuous extension processing on the bidirectional fusion features, extend the feature temporal span in the time dimension to cover the complete lifting process, expand the feature association range in the spatial dimension to cover the entire trajectory position, and generate a spatiotemporal continuous extended coupled feature that includes continuous temporal evolution features in the time dimension and trajectory position association evolution features in the spatial dimension.

[0065] In the temporal dimension, the time span of the bidirectional fusion features is extended from the currently acquired time period to the entire lifting process, including the preparation stage before lifting and the closing stage after lifting. This is achieved by extrapolating trends from existing data and making reasonable assumptions to supplement the extended feature data. In the spatial dimension, the feature association range is expanded from the currently focused local trajectory location to all locations along the entire lifting trajectory, ensuring that each trajectory location has corresponding attitude and hydraulic drive features. For example, for locations on the lifting trajectory where data was not acquired, interpolation is performed based on feature data from adjacent locations. After this spatiotemporally continuous extension processing, the resulting spatiotemporally continuous extended coupled features can fully cover the temporal and spatial range of the lifting process.

[0066] Step S118: Perform feature association weight allocation processing on the spatiotemporal continuous extended coupling features. Based on the importance of different dimensional evolution features in bidirectional coupling modeling, assign corresponding weight coefficients to generate the association weight sequence of different dimensional evolution features in bidirectional coupling modeling.

[0067] Experts were invited to evaluate the importance of continuous temporal evolution features in the time dimension, trajectory position correlation evolution features in the spatial dimension, attitude evolution features, and hydraulic drive coupling evolution features in bidirectional coupling modeling. It is assumed that the evaluation results are: time dimension feature weight 0.2, spatial dimension feature weight 0.2, attitude evolution feature weight 0.3, hydraulic drive coupling evolution feature weight 0.3, and the sum is 1. These weight coefficients are arranged in order to form a correlation weight sequence.

[0068] Step S119: Perform weighted aggregation modeling on the spatiotemporal continuous extended coupling features through the associated weight sequence, aggregate the features of different dimensions according to the weight coefficients, generate a weighted aggregated bidirectional coupling feature vector, and generate a bidirectional coupling linkage feature set based on the weighted aggregated bidirectional coupling feature vector.

[0069] The temporal dimension feature component, spatial dimension feature component, attitude evolution feature component, and hydraulic drive coupling evolution feature component in the spatiotemporal continuous extended coupling feature are multiplied by 0.2, 0.2, 0.3, and 0.3 respectively, to obtain weighted feature components for each dimension. These weighted feature components are then aggregated according to temporal and spatial order to form a high-dimensional weighted aggregated bidirectional coupling feature vector. This weighted aggregated bidirectional coupling feature vector contains weighted spatiotemporal, attitude, and hydraulic drive coupling information, and is used as a set of bidirectional coupling linkage features.

[0070] Step S120: Perform cross-domain correlation empowerment processing on the bidirectional coupling linkage feature set and the dynamic load continuous evolution feature of the lifting operation to generate dynamic linkage empowerment parameters of digital hydraulics and intelligent control.

[0071] In the lifting operations of hydraulic cranes, dynamic load is a crucial factor affecting lifting trajectory tracking. This paper cross-domain correlates the bidirectional coupled linkage feature set (reflecting the coupling relationship between posture and hydraulic drive) with the continuous evolution characteristics of dynamic load (reflecting load changes). By analyzing the impact of load changes on posture and hydraulic drive, and how posture and hydraulic drive adapt to load changes, dynamic linkage enabling parameters that can guide the coordinated operation of intelligent control and digital hydraulic drive are generated.

[0072] Step S121: Collect real-time load dynamic data during the lifting operation. Combine the historical load evolution data of the lifting operation with the collected load dynamic data to extract the continuous load increase evolution characteristics, load mutation evolution characteristics, load stability evolution characteristics, and load evolution impact characteristics on lifting attitude. Generate a dynamic load continuous evolution characteristic group for the lifting operation. The dynamic load continuous evolution characteristic group includes a feature sequence representing the load value increasing over time, features representing the time and magnitude of load mutation, features representing the duration and average of the load stability phase, and correlation features representing the relationship between load value and attitude angle.

[0073] A tension sensor is installed on the lifting device of the hydraulic crane to collect load data in real time during the lifting operation, with a sampling frequency of 50 Hz. Simultaneously, historical load data of the crane's lifting operations under different operating scenarios over the past year is collected to form a historical load evolution database.

[0074] Step S1211: Deploy load detection equipment in the actual lifting operation scenario to continuously collect real-time load dynamic data during the lifting operation process. At the same time, collect historical load evolution data of lifting operations under different operation objects and different operation environments, integrate real-time data and historical data, and jointly generate a set of original load dynamic data.

[0075] Real-world lifting operations include equipment hoisting in factory workshops and material handling on construction sites. Different objects handled include equipment, steel, and concrete blocks, and different operating environments include sunny days, rainy days, high temperatures, and low temperatures. Real-time load dynamic data is collected by tension sensors on the lifting equipment. For example, when hoisting a large piece of equipment, the load value is recorded every 0.02 seconds from the moment the equipment leaves the ground. Historical load evolution data is obtained from the crane's operation logs and database, including load changes over time for each past lifting operation. The real-time load data and historical load data are integrated, and duplicate and obviously erroneous data are removed to generate a raw dynamic load data set.

[0076] Step S1212: Perform data cleaning processing on the original load dynamic data set, remove abnormal load data and invalid load data, remove data that exceeds the normal load range and data that has no actual load significance, and generate cleaned load dynamic data.

[0077] The normal load range is determined based on the crane's rated lifting capacity. For example, if the rated lifting capacity of this hydraulic crane is 20 tons, then load data exceeding 25 tons (considering a safety factor of 1.25) is considered abnormal load data. Invalid load data includes zero or fixed values ​​collected when sensors are not connected. By setting these rules, the raw load dynamic data set is filtered, retaining valid data and generating cleaned load dynamic data.

[0078] Step S1213: Extract the load change data with lifting time from the cleaned load dynamic data, and generate a continuous load evolution feature based on the load change data with lifting time. The continuous load evolution feature carries the information of continuous load change with the lifting process, and fully presents the load increase process from the initial value to the peak value.

[0079] During lifting operations, as the spreader gradually raises the cargo, the load gradually increases with the lifting height (considering factors such as cargo swaying which may cause load fluctuations, but the overall trend is upward). From the dynamic load data after cleaning, segments showing a continuous upward trend in load over time are extracted, such as the period from the start of lifting to the point where the load reaches its maximum value. The load data for this period are arranged chronologically to form a continuous upward load evolution characteristic, which is a one-dimensional array where each element represents the load value at a different time point.

[0080] Step S1214: Extract the data of mutation nodes whose load change rate exceeds the preset range and the load data before and after the mutation from the cleaned load dynamic data. Generate load mutation evolution features based on the data of mutation nodes whose load change rate exceeds the preset range and the load data before and after the mutation. The load mutation evolution features carry information on the time and magnitude of the load mutation and clarify the key information of the load mutation.

[0081] The preset load change rate range is no more than 5% of the rated lifting capacity per second. When the load change rate exceeds this range, it is determined to be a load mutation. For example, during lifting, if the cargo suddenly swings, causing the load to increase by 2 tons within 0.1 seconds, exceeding the preset range, the time of this mutation node, the load value before the mutation, and the load value after the mutation are recorded. All the above mutation node data are arranged in chronological order to form a load mutation evolution feature. This load mutation evolution feature is a two-dimensional array, with each row containing the time of the mutation, the load before the mutation, and the load after the mutation.

[0082] Step S1215: Extract stable phase data with load change rate within a preset range from the cleaned load dynamic data, and generate load stability evolution features based on the stable phase data with load change rate within a preset range. The load stability evolution features carry information on the duration of the load stability phase and the load stability value, reflecting the core parameters of the load stability state.

[0083] The stage where the load change rate is within a preset range (e.g., the change per second does not exceed 1% of the rated lifting capacity) is the load stabilization stage. These stabilization stages are identified from the cleaned load dynamic data, and the start time, end time (duration), and average load value (load stability value) of each stabilization stage are recorded. The above data are arranged in chronological order to form a load stabilization evolution feature, which is a two-dimensional array, with each row containing the start time, end time, and load stability value of the stabilization stage.

[0084] Step S1216: Extract lifting posture change data under different load conditions from the cleaned load dynamic data, and generate load evolution impact characteristics on lifting posture based on the lifting posture change data under different load conditions. The load evolution impact characteristics on lifting posture carry the correlation information between load change and posture change, and reflect the relationship between load change and posture.

[0085] Different load conditions include light load (below 30% of rated lifting capacity), medium load (30%-70% of rated lifting capacity), and heavy load (above 70% of rated lifting capacity). For each load condition, corresponding lifting attitude data (such as pitch angle and azimuth angle) are extracted, and the relationship between load value and attitude angle is analyzed. For example, under heavy load conditions, the rate of change of pitch angle may be slower, while under light load conditions, the rate of change of pitch angle may be faster. By statistically analyzing the attitude angle changes corresponding to different load values, the influence characteristics of load evolution on lifting attitude are generated. This influence characteristic can be a matrix, where rows represent load conditions, columns represent attitude angle parameters, and elements represent the characteristics of attitude angle changes (such as rate of change, range of change, etc.) under that load condition.

[0086] Step S1217: Perform work scenario association processing on the load continuous increase evolution feature, load sudden change evolution feature, load stable evolution feature and load evolution impact feature on lifting attitude, bind each feature with the corresponding work scenario, and generate comprehensive load evolution feature corresponding to each work scenario.

[0087] The operational scenarios include different operational objects and environments, such as factory workshops for hoisting equipment and construction sites for handling steel. Data belonging to the same operational scenario are extracted from the characteristics of continuous load increase evolution, sudden load change evolution, stable load evolution, and the impact of load evolution on lifting posture, and then combined. For example, in the scenario of hoisting equipment in a factory workshop, the characteristics of continuous load increase, sudden load change, stable load, and the impact on posture are combined to form the comprehensive load evolution characteristics for that operational scenario.

[0088] Step S1218: Perform time dimension expansion processing on the comprehensive load evolution characteristics corresponding to each operation scenario, associate the comprehensive load evolution characteristics of different lifting periods, and generate extended load evolution characteristics containing load evolution association information of different lifting periods.

[0089] Different lifting times, including morning, noon, afternoon, and evening, can be affected by environmental factors (such as temperature and humidity). This study correlates the comprehensive load evolution characteristics across different lifting times within the same work scenario to analyze the temporal patterns of load evolution. For example, comparing the stable load evolution characteristics in the morning and noon under the same work scenario reveals differences. Through correlation processing, extended load evolution characteristics are generated to reflect the impact of time factors on load evolution.

[0090] Step S1219: Perform weight allocation processing on the extended load evolution features. Based on the degree of influence of different load evolution features in reflecting the overall load evolution state, assign corresponding weight coefficients to generate weight coefficients corresponding to different load evolution features.

[0091] Experts were invited to evaluate the importance of the characteristics of continuous load increase evolution, abrupt load change evolution, stable load evolution, and the impact of load evolution on lifting attitude. The evaluation results are assumed to be: a weight of 0.25 for continuous load increase evolution, 0.3 for abrupt load change evolution, 0.25 for stable load evolution, and 0.2 for the impact of load evolution on lifting attitude, totaling 1.

[0092] Step S12110: The extended load evolution features are weighted by the weight coefficients. The extended load evolution features of different dimensions are weighted according to the weight coefficients to generate weighted load evolution feature components, and the weighted load evolution feature components are combined into a weighted load evolution feature vector.

[0093] The continuously increasing load evolution feature component in the extended load evolution feature is multiplied by 0.25, the sudden load evolution feature component by 0.3, the stable load evolution feature component by 0.25, and the influence of load evolution on lifting attitude feature component by 0.2, to obtain their respective weighted components. These weighted components are then arranged and combined according to the operation scenario and time sequence to form a weighted load evolution feature vector.

[0094] Step S12111: Generate a dynamic load continuous evolution feature group for the lifting operation based on the weighted load evolution feature vector.

[0095] The weighted load evolution feature vector generated above is used as the dynamic load continuous evolution feature group for lifting operations. This dynamic load continuous evolution feature group comprehensively reflects the continuous evolution of lifting load under different operating scenarios and different times and its impact on attitude.

[0096] Step S122: Perform feature association deep mining on the bidirectional coupling linkage feature set and the dynamic load continuous evolution feature set of the lifting operation to explore the potential correlation between features in the two feature sets and generate cross-domain correlation feature vectors of bidirectional coupling features and dynamic load evolution features.

[0097] Data mining algorithms, such as association rule mining and cluster analysis, are used to analyze features in the bidirectional coupling and linkage feature set and the dynamic load continuous evolution feature group. For example, association rule mining reveals that when the load undergoes a sudden change, the corresponding hydraulic drive pressure also undergoes a corresponding change, and the attitude stability decreases. Through cluster analysis, samples with similar load evolution characteristics and similar bidirectional coupling characteristics are grouped into one class, identifying feature association patterns between different categories. These potential associations are represented as feature vectors, generating cross-domain association feature vectors. The elements of these cross-domain association feature vectors represent the association strength or association pattern between different features.

[0098] Step S123: Perform hydraulic drive adaptation requirement mining processing on the cross-domain correlation feature vector under load evolution, analyze the flow and pressure adaptation requirements of the digital hydraulic drive unit under different load evolution stages, and generate the flow dynamic adaptation requirement characteristics and pressure dynamic adaptation requirement characteristics of the digital hydraulic drive unit corresponding to different load evolution stages.

[0099] Different load evolution stages include the continuous load increase stage, the sudden load change stage, and the stable load stage. For each load evolution stage, based on the correlation between load and hydraulic drive in the cross-domain correlation feature vector, the required flow and pressure range of the digital hydraulic drive unit is analyzed. For example, in the continuous load increase stage, to ensure the lifting speed, the flow rate needs to increase appropriately with the increase of load; in the sudden load change stage, to avoid shock to the hydraulic system, the pressure needs to be quickly adjusted. The above flow and pressure adaptation requirements are organized according to the load evolution stage to generate dynamic flow adaptation requirement features and dynamic pressure adaptation requirement features. Each feature is a sequence containing requirement parameters for different stages.

[0100] Step S124: Based on the dynamic adaptation requirements of flow rate and pressure, mine the dynamic adaptation relationship of intelligent control strategy, analyze the change mode of intelligent control strategy that matches the adaptation requirements of hydraulic drive, and generate the evolution characteristics of intelligent control strategy that match the adaptation requirements of hydraulic drive.

[0101] Intelligent control strategies include PID control parameters and fuzzy control rules. Based on the dynamic adaptation requirements of flow and pressure, the intelligent control strategy is analyzed to determine how to adjust to meet these requirements. For example, when flow demand increases rapidly, the proportional gain of the PID controller should be appropriately increased to accelerate the response speed; when pressure needs to be stabilized at a certain value, the integral gain should be adjusted to reduce steady-state error. These adjustments of the intelligent control strategy are recorded according to time and load evolution stages to generate intelligent control strategy evolution characteristics. These characteristics reflect the process of the intelligent control strategy changing with the adaptation requirements of hydraulic drive.

[0102] Step S125: Perform bidirectional empowerment and fusion processing on the dynamic adaptation requirement characteristics of the digital hydraulic drive and the evolution characteristics of the intelligent control strategy to establish a mutual empowerment relationship between the hydraulic drive parameters and the intelligent control strategy, and generate a linkage fusion feature in which the hydraulic drive parameters and the intelligent control strategy mutually empower each other.

[0103] The dynamic adaptation requirements of hydraulic drive parameters influence the adjustment of intelligent control strategies, which in turn affect the output of hydraulic drive parameters. By establishing a feedback relationship between the two, bidirectional empowerment and fusion are achieved. For example, when a sudden load change leads to an increase in the dynamic pressure adaptation requirement, the intelligent control strategy adjusts the PID parameters to increase the hydraulic pump's output pressure. This increase in pressure, in turn, affects the load change, thus influencing the next round of adaptation requirements and control strategy adjustments. Representing these interactions as features generates a linked fusion feature, which contains evolutionary information about the hydraulic drive parameters and the intelligent control strategy during their interaction.

[0104] Step S126: Perform dynamic empowerment parameter evolution trend prediction processing on the linkage fusion feature, analyze the change trend of dynamic empowerment parameters in different stages of lifting motion, and generate continuous evolution trend features of dynamic empowerment parameters in different stages of lifting motion.

[0105] Based on the historical evolution data of hydraulic drive parameters and intelligent control strategies in the linkage and fusion characteristics, time series prediction algorithms, such as ARIMA models and LSTM neural networks, are used to predict the changing trends of dynamic empowerment parameters at different stages of the lifting motion (such as lifting start-up, trajectory advancement, and lifting stop). For example, it predicts the rate at which flow demand will increase and how the proportional coefficient of the PID controller will be adjusted during the trajectory advancement stage. The above prediction results are organized by stage to generate continuous evolution trend characteristics.

[0106] Step S127: Based on the evolution trend characteristics, continuously optimize the dynamic empowerment parameters, adjust the dynamic empowerment parameters to adapt to the needs of different load evolution stages and trajectory evolution stages, and generate optimized dynamic empowerment parameter characteristics that meet the needs of different load evolution stages and trajectory evolution stages.

[0107] Based on the predicted evolutionary trend characteristics, the dynamic empowerment parameters are adjusted and optimized in advance. For example, if a sudden load change is predicted, the proportional coefficient of the PID controller for pressure control is increased in advance to cope with the sudden load change. In this way, the dynamic empowerment parameters can better adapt to changes in load and trajectory, generating optimized dynamic empowerment parameter characteristics.

[0108] Step S128: Perform dimensional standardization and temporal alignment processing on the optimized dynamic empowerment parameter features, unify the dimensional attributes of parameters of different dimensions, align the temporal nodes of the parameters to match the lifting process, and generate standardized temporal empowerment parameter features that meet the requirements of cross-domain association empowerment.

[0109] Dynamic empowerment parameters of different dimensions may have different units of measurement, such as flow rate in liters per minute, pressure in megapascals, and PID parameters being dimensionless. Standardization transforms these parameters to the same unit of measurement, for example, normalizing all parameters to the range of 0-1. Simultaneously, the time-series nodes of different parameters are aligned to ensure that parameters at the same point in time correspond to each other and remain consistent with the timeline of the lifting process. After processing, standardized time-series empowerment parameter characteristics are generated.

[0110] Step S129: Perform feature enhancement processing on the standardized time-series empowerment parameter features to enhance the recognizability and correlation of the parameter features and generate enhanced cross-domain correlation dynamic empowerment features.

[0111] Feature enhancement techniques, such as Principal Component Analysis (PCA) and wavelet transform, are employed to process the standardized time-series empowered parameter features. PCA extracts the main feature components, reducing data dimensionality while retaining key information; wavelet transform performs time-frequency analysis on the parameter features, highlighting feature variations across different frequency ranges. These processes enhance the discriminative power of the parameter features, making the correlations between different features more apparent, and generating enhanced cross-domain correlated dynamic empowered features.

[0112] Step S1210: Generate dynamic linkage empowerment parameters for digital hydraulics and intelligent control based on the enhanced cross-domain correlation dynamic empowerment features.

[0113] Key parameters are extracted from the enhanced cross-domain dynamic empowerment features, such as optimized and standardized flow adaptation requirements, pressure adaptation requirements, and PID control parameters. These parameters are then organized according to time sequence and functional category to form dynamic linkage empowerment parameters for digital hydraulics and intelligent control. This set of dynamic linkage empowerment parameters guides the intelligent control and digital hydraulic drive units to work collaboratively during the lifting process.

[0114] Step S130: Based on the dynamic linkage empowerment parameters, realize the bidirectional coordinated drive control of intelligent control and digital hydraulic drive, and generate a set of bidirectional drive control signals for trajectory tracking of the lifting actuator.

[0115] Dynamic linkage enabling parameters provide specific parameter guidance for the coordinated operation of intelligent control and digital hydraulic drive. The intelligent control system adjusts its control strategy based on these parameters, while the digital hydraulic drive unit adjusts its flow and pressure output accordingly. Together, they work to drive and control the lifting actuator to track a preset trajectory.

[0116] Step S131: Extract intelligent control strategy empowerment parameter features, digital hydraulic drive empowerment parameter features, and time sequence collaborative empowerment parameter features from the dynamic linkage empowerment parameters to generate a dynamic linkage empowerment parameter feature group. The dynamic linkage empowerment parameter feature group includes an intelligent control parameter sequence, a digital hydraulic drive parameter sequence, and a time synchronization mark sequence representing the collaborative timing of the two.

[0117] The characteristics of intelligent control strategy-enabled parameters include the time-varying sequences of parameters such as the proportional coefficient, integral coefficient, and derivative coefficient of PID control; the characteristics of digital hydraulic drive-enabled parameters include the time-varying sequences of parameters such as flow commands and pressure commands; and the characteristics of time-coordinated enabled parameters include the synchronization time markers of intelligent control parameters and hydraulic drive parameters, such as a flow command and its corresponding PID parameter adjustment taking effect at the same time. These characteristics are extracted and arranged in chronological order to form a dynamic linkage enabled parameter characteristic group.

[0118] Step S132: Based on the dynamic linkage empowerment parameter feature group, generate intelligent control dynamic drive strategy features that are matched in real time with the dynamic adaptation requirements of hydraulic drive, and adjust the intelligent control drive strategy according to the changes in the dynamic adaptation requirements of hydraulic drive.

[0119] By analyzing the digital hydraulic drive parameter sequence and timing coordination markers in the dynamic linkage enabling parameter feature group, the dynamic adaptation requirements of the hydraulic drive at different times are determined. For example, when the flow command in the hydraulic drive parameter sequence increases, the intelligent control strategy needs to adjust the PID parameters accordingly to accelerate the response. Based on these requirements, intelligent control dynamic drive strategy features are generated. These features include the specific content of the intelligent control strategy at different time points, such as the values ​​of PID parameters and the selection of control modes.

[0120] Step S133: Based on the characteristics of the intelligent control dynamic drive strategy, the intelligent control parameters are continuously and dynamically adjusted. The intelligent control parameters are continuously adjusted according to the changes in the hydraulic drive state and the load evolution state, generating real-time dynamic control parameter characteristics of intelligent control. The real-time dynamic control parameter characteristics evolve continuously with the changes in the hydraulic drive state and the load evolution state.

[0121] Step S1331: Convert the intelligent control dynamic drive strategy features into corresponding intelligent control parameter adjustment instructions. The intelligent control parameter adjustment instructions carry the direction and magnitude of parameter adjustment information, and clarify the adjustment method of each intelligent control parameter.

[0122] The intelligent control dynamic drive strategy includes requirements for adjusting intelligent control parameters such as PID parameters, such as "increasing the proportional coefficient from Kp1 to Kp2". These requirements are translated into specific adjustment instructions, such as "the proportional coefficient is adjusted by increasing, and the adjustment range is Kp2 - Kp1". Each intelligent control parameter has a corresponding adjustment instruction, clearly defining its adjustment method.

[0123] Step S1332: Based on the intelligent control parameter adjustment command, perform preliminary adjustment on the intelligent control parameters, change the initial value of the intelligent control parameters according to the direction and magnitude of the adjustment command, and generate the initially adjusted intelligent control parameters.

[0124] The initial values ​​of the intelligent control parameters can be system default values ​​or parameter values ​​from the previous moment. Based on the adjustment command, the initial values ​​are modified. For example, if the initial proportional coefficient Kp0 = 5 and the adjustment command is to increase it by 2, then the initially adjusted proportional coefficient Kp1 = 5 + 2 = 7. This same process is performed on all intelligent control parameters to generate a set of initially adjusted intelligent control parameters.

[0125] Step S1333: Collect the intelligent control output status data and corresponding digital hydraulic drive response data under the action of the initially adjusted intelligent control parameters, record the output information of intelligent control and the feedback information of digital hydraulic drive, and generate parameter adjustment response data.

[0126] The initially adjusted intelligent control parameters are input into the intelligent control system, which outputs drive signals to the digital hydraulic drive unit. Simultaneously, the system collects output status data from the intelligent control system (such as the amplitude and frequency of the control signal) and response data from the digital hydraulic drive unit (such as actual flow rate and actual pressure). This data is recorded in chronological order to generate parameter adjustment response data.

[0127] Step S1334: Extract the deviation data between the intelligent control output state and the expected output state from the parameter adjustment response data, compare the difference between the actual output state and the expected output state, and perform secondary adjustment on the intelligent control parameters based on the deviation data between the intelligent control output state and the expected output state to correct the parameters to reduce the deviation and generate the secondary adjusted intelligent control parameters.

[0128] The expected output state can be determined based on the dynamic linkage enabling parameters. For example, if the expected flow rate is Q0 and the actual flow rate is Q1, then the deviation data is ΔQ = Q1 - Q0. Based on the magnitude and direction of the deviation data, the intelligent control parameters are adjusted a second time. For example, if the actual flow rate is less than the expected flow rate and the deviation is large, it may be necessary to further increase the proportional coefficient. The adjusted intelligent control parameters are generated using the above method.

[0129] Step S1335: Repeat the process of parameter adjustment, response data acquisition, deviation analysis and secondary parameter adjustment until the deviation between the intelligent control output state and the expected output state is within an acceptable range. Optimize the parameters through multiple rounds of iteration to generate the iteratively optimized intelligent control parameters.

[0130] Set an acceptable deviation range, such as flow rate deviation not exceeding 5% of the expected value and pressure deviation not exceeding 3% of the expected value. If the parameters after the second adjustment still do not bring the deviation within the acceptable range, continue parameter adjustment, data acquisition, and deviation analysis, repeating this process until the deviation meets the requirements. After multiple iterations, the iteratively optimized intelligent control parameters are obtained.

[0131] Step S1336: Based on the iteratively optimized intelligent control parameters, combined with the load evolution characteristics and hydraulic drive state characteristics, analyze the pattern of parameter changes with load and hydraulic state, and generate the continuous adjustment trend of intelligent control parameters.

[0132] This study analyzes the values ​​of the iteratively optimized intelligent control parameters under different load conditions (light, medium, and heavy load) and different hydraulic drive conditions (flow rate and pressure) to identify patterns in parameter variation. For example, the larger the load, the larger the proportional coefficient; the higher the flow rate demand, the smaller the integral coefficient. Based on these patterns, the study predicts the adjustment trends of the intelligent control parameters under future changes in load and hydraulic conditions, generating a continuous adjustment trend.

[0133] Step S1337: Adjust the intelligent control parameters in real time according to the continuous adjustment trend of the intelligent control parameters, and continuously change the parameter values ​​as the load evolves and the hydraulic drive state changes, thereby generating continuous change data of the intelligent control parameters.

[0134] Based on continuous adjustment trends, the system monitors load evolution and hydraulic drive status characteristics in real time during the lifting process. When these characteristics change, the intelligent control parameters are adjusted according to the trends. For example, when the load is detected to increase from 5 tons to 10 tons, the proportional coefficient is adjusted from 7 to 9 based on the continuous adjustment trend. The parameter adjustment process is recorded in chronological order to generate continuous change data of the intelligent control parameters.

[0135] Step S1338: Extract parameter values ​​and parameter change rates at different time points from the continuous change data of the intelligent control parameters, record the parameter status and change rate at each time point, and generate real-time dynamic control parameter characteristics of the intelligent control.

[0136] In the continuously changing data of intelligent control parameters, parameter values ​​are extracted at regular time intervals (e.g., 0.1 seconds), and the rate of change of parameters between adjacent time points (parameter change divided by the time interval) is calculated. The above parameter values ​​and rates of change are arranged in chronological order to form real-time dynamic control parameter characteristics, which reflect the state and rate of change of intelligent control parameters at different times.

[0137] Step S1339: Smooth the real-time dynamic control parameter features of the intelligent control in the time dimension to generate smoothed real-time dynamic control parameter features, and generate the final real-time dynamic control parameter features of the intelligent control based on the smoothed real-time dynamic control parameter features.

[0138] Due to sensor noise or system disturbances, the real-time dynamic control parameter characteristics may fluctuate. A moving average or exponential smoothing method is used to smooth them; for example, a moving average with a window size of 3 is used, where the parameter value at each time point is the average of the parameter values ​​at that point and the one time point before and after it. After smoothing, the smoothed real-time dynamic control parameter characteristics are obtained, serving as the final intelligent control real-time dynamic control parameter characteristics.

[0139] Step S134: Generate digital hydraulic dynamic drive control parameter features that match the intelligent control strategy in real time based on the dynamic linkage empowerment parameter feature group, and adjust the digital hydraulic drive control parameters according to the changes in the intelligent control strategy.

[0140] By analyzing the intelligent control parameter sequence and time-series coordination markers in the dynamic linkage empowerment parameter feature group, the changes in the intelligent control strategy at different times are determined. For example, when the proportional coefficient in the intelligent control parameters increases, it indicates a need to accelerate the response speed; in this case, the flow command in the digital hydraulic drive control parameters should increase accordingly. Based on these changes, digital hydraulic dynamic drive control parameter features are generated, which include parameters such as flow command and pressure command of the digital hydraulic drive unit at different time points.

[0141] Step S135: Based on the characteristics of the digital hydraulic dynamic drive control parameters, the flow rate and pressure of the digital hydraulic drive unit are continuously and collaboratively adjusted. The flow rate and pressure are adjusted synchronously to match the changes in intelligent control parameters and load evolution state, thereby generating the real-time dynamic output characteristics of the digital hydraulic drive unit. The real-time dynamic output characteristics continuously evolve with the changes in intelligent control parameters and load evolution state.

[0142] The digital hydraulic drive unit adjusts flow and pressure based on flow and pressure commands from the dynamic drive control parameters, controlling the opening of the proportional multi-way valve and the displacement of the hydraulic pump. During adjustment, it ensures that changes in flow and pressure are coordinated; for example, while increasing flow, pressure is adjusted appropriately according to the load to avoid system overload or reduced efficiency. The actual output flow and pressure of the digital hydraulic drive unit are recorded chronologically to generate real-time dynamic output characteristics.

[0143] Step S136: Perform bidirectional collaborative correlation modeling processing on the real-time dynamic control parameter characteristics of the intelligent control and the real-time dynamic output characteristics of the digital hydraulic drive unit, establish a collaborative working correlation model between the two, and generate bidirectional collaborative drive and control characteristics of intelligent control and digital hydraulic drive.

[0144] System identification methods or machine learning algorithms, such as neural network models, are employed to model the real-time dynamic control parameter characteristics of intelligent control and the real-time dynamic output characteristics of the digital hydraulic drive unit. Using intelligent control parameters as input and digital hydraulic output as output, a mathematical relationship model is established between the two. This mathematical relationship model reflects how intelligent control parameters affect the digital hydraulic output, and also adjusts the intelligent control parameters based on feedback from the digital hydraulic output, thereby generating a bidirectional collaborative drive-control characteristic. This bidirectional collaborative drive-control characteristic embodies the dynamic relationship between intelligent control and digital hydraulic drive during collaborative operation.

[0145] Step S137: Perform continuous encoding processing on the bidirectional cooperative drive control features to convert the cooperative drive control features into a continuous signal encoding form, and generate a continuous timing drive control signal encoding sequence for the lifting actuator.

[0146] The drive control signals can be encoded using pulse width modulation (PWM) encoding or digital signal encoding. Based on the parameter information in the bidirectional cooperative drive control characteristics, such as flow commands, pressure commands, and intelligent control parameters, they are converted into corresponding electrical signal codes. For example, the flow command value is converted into the duty cycle of a PWM signal, and the pressure command value is converted into the binary code of a digital signal. These codes are then arranged in chronological order to form a continuous timing drive control signal encoding sequence.

[0147] Step S138: Based on the continuous timing drive control signal encoding sequence, perform dynamic adaptation and conversion processing of the signal format, adjust the signal encoding format to meet the requirements of the real-time control protocol of the lifting actuator, and generate dynamic format drive control signal features that meet the requirements of the real-time control protocol of the lifting actuator.

[0148] The real-time control protocol of the lifting actuator may specify the communication format, data bit length, and verification method of the signals. The continuously sequenced drive control signal encoding sequence is formatted according to these protocol requirements, such as adding start bits, stop bits, and parity bits, and adjusting the data bit length to the length specified in the protocol. After conversion, a dynamic format drive control signal characteristic is generated, ensuring that the drive control signal can be correctly identified and received by the lifting actuator.

[0149] Step S139: Perform signal enhancement and anti-interference processing on the dynamic format drive control signal features to strengthen the signal strength and reduce the influence of interference signals, thereby generating enhanced bidirectional cooperative drive control signal features.

[0150] A signal amplification circuit is used to enhance the strength of the dynamic format drive signal, ensuring that the signal does not attenuate excessively during transmission. Simultaneously, filtering techniques, such as low-pass and band-pass filters, are employed to remove high-frequency noise and interference signals. For digital signals, error control coding, such as CRC checksum, can also be used to improve the reliability of signal transmission. After processing, an enhanced bidirectional cooperative drive signal characteristic is generated.

[0151] Step S1310: Generate a set of bidirectional drive control signals for trajectory tracking of the lifting actuator based on the enhanced bidirectional collaborative drive control signal characteristics.

[0152] Different types of signals (such as signals controlling the telescopic boom and signals controlling the luffing cylinder) in the enhanced bidirectional collaborative drive and control signal characteristics are classified and organized to form a trajectory tracking bidirectional drive and control signal set. This trajectory tracking bidirectional drive and control signal set contains drive and control signals that control various components of the lifting actuator, enabling precise tracking of the lifting trajectory.

[0153] Step S140: Perform reverse tracing correction processing on the trajectory evolution deviation of the trajectory tracking bidirectional drive control signal set to generate the target bidirectional drive control signal set of the lifting actuator.

[0154] During the movement of the lifting actuator following the bidirectional drive control signal set along the trajectory, various factors (such as load changes, hydraulic system leakage, sensor errors, etc.) may cause deviations between the actual trajectory and the preset trajectory. By reverse tracing and correction processing, the cause of the deviation is identified, and the drive control signals are corrected to reduce the deviation and generate the target bidirectional drive control signal set.

[0155] Step S141: Extract signal amplitude evolution features, signal timing evolution features, and signal co-evolution features from the trajectory tracking bidirectional drive control signal set to generate a trajectory tracking bidirectional drive control signal feature group. The trajectory tracking bidirectional drive control signal feature group includes the amplitude sequence of the drive control signal, the time interval sequence of the signal transmission, and the time difference sequence between the intelligent control signal and the hydraulic drive signal.

[0156] Signal amplitude evolution characteristics refer to the sequence of changes in the voltage or current amplitude of the drive and control signal over time; signal timing evolution characteristics refer to the sequence of changes in the time interval between signal transmissions over time, such as a control signal being sent every 0.1 seconds, which may vary depending on the system load; signal co-evolution characteristics refer to the sequence of changes in the transmission time difference between the intelligent control signal and the hydraulic drive signal over time, reflecting their synergy. These characteristics are extracted to form a feature set for the bidirectional drive and control signal for trajectory tracking.

[0157] Step S142: Based on the trajectory tracking bidirectional drive control signal feature group, predict the trajectory evolution deviation characteristics that may occur within a preset time period in the future. At the same time, trace the root cause characteristics of the hydraulic drive parameter deviation and the intelligent control parameter deviation, and combine the motion posture evolution characteristics of the lifting actuator and the dynamic coupling characteristics of the digital hydraulic drive to analyze the deviation trend and the root cause.

[0158] For example, in step S1421: based on the trajectory tracking bidirectional drive control signal feature group, combined with the motion posture evolution characteristics of the lifting actuator and the dynamic coupling characteristics of the digital hydraulic drive, a trajectory evolution deviation prediction model is established, and relevant features are integrated to construct the logical relationship of deviation prediction.

[0159] Data on trajectory evolution deviations, trajectory tracking bidirectional drive control signal features, motion attitude evolution features, and dynamic coupling features of digital hydraulic drive from historical lifting operations are collected. Machine learning algorithms, such as Support Vector Machine (SVM) and Random Forest, are used to train a trajectory evolution deviation prediction model. The model's input consists of the trajectory tracking bidirectional drive control signal feature set, motion attitude evolution features, and dynamic coupling features of digital hydraulic drive; the output is the trajectory evolution deviation features within a preset future time period. Through training, the model learns the logical relationship between these features and trajectory deviations.

[0160] Step S1422: Simulate the motion trajectory of the lifting actuator within a preset time period using the trajectory evolution deviation prediction model, deduce the trajectory position at different time nodes, and generate predicted trajectory data.

[0161] The current trajectory tracking bidirectional drive control signal feature group, motion posture evolution feature and digital hydraulic drive dynamic coupling feature are input into the trajectory evolution deviation prediction model. The model outputs the position coordinates of the lifting actuator at different time nodes (e.g., every 0.1 seconds) within a future preset time period (e.g., 5 seconds), forming the predicted trajectory data.

[0162] Step S1423: Compare the predicted trajectory data with the preset trajectory data, extract the deviation data between the two, calculate the positional difference between the predicted trajectory and the preset trajectory at each time node, and generate trajectory evolution deviation features that may occur in the future preset time period based on the deviation data between the predicted trajectory data and the preset trajectory data. The trajectory evolution deviation features carry information on the direction, magnitude and occurrence time of the deviation.

[0163] The preset trajectory data is the ideal trajectory planned before the lifting operation, containing the position coordinates of different time nodes. The position coordinates of each time node in the predicted trajectory data are compared with the corresponding position coordinates of the preset trajectory data to calculate the position difference (such as the distance difference in a three-dimensional coordinate system). Based on the direction of the position difference (such as positive deviation in the X-axis direction and negative deviation in the Y-axis direction), the magnitude (size of the difference), and the time of occurrence (corresponding time node), trajectory evolution deviation characteristics are generated.

[0164] Step S1424: Collect the digital hydraulic drive parameter data and intelligent control parameter data at the current moment, record the parameter values ​​under the current working state, and generate the current parameter data set.

[0165] The system collects parameter data such as flow rate, pressure, pump displacement, and valve opening degree of the digital hydraulic drive unit at the current moment through sensors and control system, as well as parameter data such as PID parameters and control mode of intelligent control system. The above data is integrated to form the current parameter data set.

[0166] Step S1425: Compare the current parameter data set with the historical optimal parameter data set, extract the parameter deviation data, and find the difference between the current parameters and the optimal parameters.

[0167] The historical best parameter dataset refers to the set of parameter data from past lifting operations where trajectory tracking accuracy was highest and deviation was lowest. Each parameter in the current parameter dataset is compared to its corresponding parameter in the historical best parameter dataset, and the parameter deviation (current parameter value minus the optimal parameter value) is calculated to generate parameter deviation data.

[0168] Step S1426: Based on the parameter deviation data and combined with the trajectory evolution deviation characteristics, analyze the correlation between hydraulic drive parameter deviation and trajectory evolution deviation, determine the degree of influence of hydraulic drive parameter deviation on trajectory deviation, and generate the root cause characteristics of hydraulic drive parameter deviation.

[0169] For example, the analysis examines whether the trajectory evolution deviation is also positive (actual position leads the preset position) when the flow parameter deviation is positive (actual flow is greater than the optimal flow), and the proportional relationship between the deviation magnitudes. Through statistical analysis, the influence weights of different hydraulic drive parameter deviations on the trajectory evolution deviation are determined, generating root cause features of hydraulic drive parameter deviations. These root cause features contain information on the degree of influence of each hydraulic drive parameter deviation on the trajectory deviation.

[0170] Step S1427: Based on the parameter deviation data and combined with the trajectory evolution deviation characteristics, analyze the correlation between intelligent control parameter deviation and trajectory evolution deviation, determine the degree of influence of intelligent control parameter deviation on trajectory deviation, and generate the root cause characteristics of intelligent control parameter deviation.

[0171] Similarly, we analyze the changes in trajectory evolution deviation when the PID proportional coefficient deviation is positive (the actual proportional coefficient is greater than the optimal proportional coefficient), determine the degree of influence of intelligent control parameter deviation on trajectory deviation, and generate the root cause characteristics of intelligent control parameter deviation.

[0172] Step S1428: Quantify the degree of deviation of the root cause features of the hydraulic drive parameter deviation and the root cause features of the intelligent control parameter deviation, and use specific feature values ​​to reflect the severity of the deviation to generate quantified root cause features of deviation.

[0173] For example, the impact of hydraulic drive parameter deviations on trajectory deviations is divided into levels 0-10, where level 0 indicates no impact and level 10 indicates the greatest impact, with corresponding numerical values ​​assigned based on the actual degree of impact. Similarly, the root cause characteristics of intelligent control parameter deviations are quantified to generate quantified deviation root cause characteristics.

[0174] Step S1429: Perform time-dimensional correlation processing on the quantified deviation root cause features, correlate deviation root cause features at different time points, and generate deviation root cause evolution features at different time points.

[0175] The quantitative deviation root cause characteristics at different time points are arranged in chronological order, and the changes of deviation root cause characteristics over time are analyzed to generate deviation root cause evolution characteristics. These deviation root cause evolution characteristics reflect the development trend of deviation root causes during the lifting process.

[0176] Step S14210: Based on the deviation root cause evolution characteristics and trajectory evolution deviation characteristics at different time points, generate corresponding correlation characteristics between deviation root causes and trajectory deviations, and fully present the root causes of trajectory evolution deviations.

[0177] By associating the root cause evolution characteristics of deviation at each time point with the corresponding trajectory evolution deviation characteristics, a one-to-one correspondence is established, generating corresponding associated features. These associated features can show when, and by which parameter deviations, the trajectory deviation was caused and by how much.

[0178] Step S143: Based on the trajectory evolution deviation characteristics, hydraulic drive parameter deviation root cause characteristics, and intelligent control parameter deviation root cause characteristics, generate reverse correction strategy characteristics for hydraulic drive parameter deviation and reverse correction strategy characteristics for intelligent control parameter deviation, and formulate the correction method and adjustment range for the corresponding parameters.

[0179] Based on the direction and magnitude of the trajectory evolution deviation, as well as the root causes of the hydraulic drive parameter deviation and the intelligent control parameter deviation, the correction direction and adjustment magnitude for the hydraulic drive parameter and the intelligent control parameter are determined. For example, if the trajectory evolution deviation is positive and large in magnitude, and the flow deviation in the hydraulic drive parameter is positive (actual flow is too large) is the main root cause, then the reverse correction strategy for the hydraulic drive parameter is to reduce the flow rate, and the adjustment magnitude is determined according to the degree of deviation. Similarly, if the proportional coefficient deviation in the intelligent control parameter is positive (proportional coefficient is too large) is also one of the root causes, then the reverse correction strategy for the intelligent control parameter is to reduce the proportional coefficient, and the adjustment magnitude is also determined according to the degree of deviation. These correction strategies are represented in feature form to generate reverse correction strategy features.

[0180] Step S144: Generate digital hydraulic drive correction parameter features based on the reverse correction strategy features for hydraulic drive parameter deviations, and adjust the digital hydraulic drive parameters according to the correction strategy.

[0181] Based on the characteristics of the reverse correction strategy for hydraulic drive parameter deviations, determine the digital hydraulic drive parameters (such as flow rate, pressure, pump displacement, etc.) that need adjustment and their specific values. For example, if the correction strategy is "reduce flow rate by 10%", then the correction value for the flow rate parameter in the digital hydraulic drive correction parameter characteristics is the current flow rate value multiplied by 0.9. Arrange the correction values ​​of all hydraulic drive parameters that need adjustment according to parameter type and time sequence to generate the digital hydraulic drive correction parameter characteristics.

[0182] Step S145: Generate intelligent control correction parameter features based on the reverse correction strategy features for intelligent control parameter deviations, and adjust the intelligent control parameters according to the correction strategy.

[0183] Similarly, based on the characteristics of the reverse correction strategy for the deviation of intelligent control parameters, the intelligent control parameters that need to be adjusted (such as PID proportional coefficient, integral coefficient, etc.) and the specific values ​​to be adjusted are determined, and the intelligent control correction parameter characteristics are generated.

[0184] Step S146: Perform collaborative fusion processing on the digital hydraulic drive correction parameter features and the intelligent control correction parameter features, integrate the two correction parameter features to achieve a collaborative correction effect, and generate bidirectional collaborative correction features.

[0185] This study analyzes the interaction between the correction parameter characteristics of digital hydraulic drive and intelligent control to ensure that their correction directions and magnitudes are matched. For example, when reducing the flow rate, it may be necessary to adjust the PID parameters of the intelligent control accordingly to maintain system stability. Through collaborative fusion processing, the two correction parameter characteristics are integrated into a unified bidirectional collaborative correction characteristic, ensuring that the corrected parameters can work together to effectively reduce trajectory deviation.

[0186] Step S147: Based on the bidirectional collaborative correction feature, perform correction encoding processing on the trajectory tracking bidirectional drive control signal, adjust the encoding of the drive control signal to incorporate correction parameter information, and generate a corrected drive control signal encoding sequence.

[0187] Based on the correction parameters in the bidirectional collaborative correction features, the encoding of the bidirectional drive control signal for trajectory tracking is adjusted. For example, if the digital hydraulic drive correction parameter features require a reduction in flow rate, the encoded value of the flow rate command is reduced accordingly; if the intelligent control correction parameter features require a reduction in the proportional coefficient, the corresponding control signal encoding value is adjusted. After correction encoding processing, a corrected drive control signal encoding sequence is generated.

[0188] Step S148: The modified drive control signal encoding sequence is fused with the original trajectory tracking bidirectional drive control signal feature group to integrate the original drive control signal information and the modified drive control signal information, and generate a signal feature vector that fuses the original drive control and reverse correction.

[0189] The corrected drive control signal encoding sequence is fused with the signal amplitude evolution characteristics, signal temporal evolution characteristics, and signal co-evolution characteristics from the feature group of the original trajectory tracking bidirectional drive control signal. For example, the corrected signal amplitude sequence and the original signal amplitude sequence are weighted and averaged, with the weights determined according to the reliability of the correction. Through fusion processing, a signal feature vector is generated that integrates the original drive control signal and the reverse correction. This signal feature vector contains information from both the original drive control signal and the correction information.

[0190] Step S149: Perform signal format adaptation processing on the signal feature vector of the fused original drive control and reverse correction, adjust the signal format to meet the requirements of the lifting actuator control protocol, and generate target signal format features that conform to the lifting actuator control protocol.

[0191] Similar to step S138, the fused signal feature vector is converted to the correct format according to the real-time control protocol of the lifting actuator to ensure that the signal format is correct and can be recognized by the lifting actuator.

[0192] Step S1410: Generate a set of target bidirectional drive control signals for the lifting actuator based on the target signal format characteristics that conform to the lifting actuator control protocol.

[0193] Different types of signals in the target signal format characteristics are classified and organized to form a target bidirectional drive control signal set. This target bidirectional drive control signal set is a drive control signal that has been corrected by reverse tracing, which can more accurately control the lifting actuator to track the preset trajectory.

[0194] Step S150: Based on the target bidirectional drive control signal set, adjust the attitude evolution timing of the lifting actuator and the dynamic output state of the hydraulic drive.

[0195] The target bidirectional drive control signal set contains optimized and corrected control signals. The control system of the lifting actuator controls the movement of mechanical components according to these signals, and the digital hydraulic drive unit adjusts the flow and pressure output according to these signals, thereby achieving precise control of the lifting actuator's attitude evolution timing and the dynamic output state of the hydraulic drive.

[0196] Step S151: Extract attitude control timing features and hydraulic drive output timing features from the target bidirectional drive control signal set, filter out timing information related to attitude control and hydraulic drive output, and generate target drive control core feature group.

[0197] The attitude control timing characteristics include information such as the timing of the telescopic boom pitch angle adjustment, the adjustment duration, and the adjustment interval; the hydraulic drive output timing characteristics include information such as the timing of flow output, the timing of pressure output, and the continuous output time of flow and pressure. These timing information are extracted from the target bidirectional drive control signal set to form the target drive control core feature group.

[0198] Step S152: Based on the attitude control timing features in the target drive control core feature group, control the attitude adjustment time and attitude adjustment range of the lifting actuator, determine the execution time and adjustment amount of each attitude adjustment action, and generate the attitude evolution timing control signal of the lifting actuator.

[0199] Based on the adjustment timing and amplitude information in the attitude control timing characteristics, determine when the telescopic boom starts adjusting the pitch angle, to what angle, and for how long the adjustment action lasts. Convert this information into control signals, such as electrical signals controlling the extension and retraction of the luffing cylinder, to generate attitude evolution timing control signals.

[0200] Step S153: Based on the attitude evolution timing control signal, drive the mechanical transmission components of the lifting actuator to perform coordinated actions, control the motion state of the mechanical transmission components to adjust the attitude of the lifting actuator, adjust the motion attitude of the lifting actuator, and make the attitude evolution of the lifting actuator conform to the attitude requirements of the preset trajectory.

[0201] The attitude evolution timing control signal is sent to the mechanical transmission components of the lifting actuator, such as the control valve of the luffing cylinder and the drive motor of the telescopic boom. The control valve adjusts its opening according to the signal, controlling the flow of hydraulic oil, thereby controlling the extension and retraction of the luffing cylinder and adjusting the pitch angle of the telescopic boom. The drive motor adjusts its speed and direction according to the signal, controlling the extension and retraction of the telescopic boom. Through the coordinated action of these mechanical transmission components, the attitude of the lifting actuator is adjusted to conform to the attitude requirements of the preset trajectory.

[0202] Step S154: Based on the hydraulic drive output timing characteristics in the target drive control core feature group, adjust the flow output time, flow output amplitude, pressure output time, and pressure output amplitude of the digital hydraulic drive unit to determine the time nodes and output quantities of flow and pressure output, and generate dynamic output control signals for the digital hydraulic drive.

[0203] Based on the flow and pressure output timing and amplitude information in the hydraulic drive output timing characteristics, it is determined when the hydraulic pump starts outputting flow, how much flow it outputs, and when the pressure reaches what value. This information is then converted into control signals, such as electrical signals controlling the hydraulic pump's displacement and controlling the opening degree of the proportional multi-way valve, generating dynamic output control signals for the digital hydraulic drive.

[0204] Step S155: Based on the dynamic output control signal of the digital hydraulic drive, drive the flow regulation component and pressure regulation component of the digital hydraulic drive unit to perform coordinated action, control the working state of the flow regulation component and pressure regulation component to adjust the hydraulic oil output, adjust the output state of the hydraulic oil, and make the output state of the digital hydraulic drive unit match the posture evolution requirements of the lifting actuator.

[0205] The dynamic output control signal of the digital hydraulic drive is sent to the flow regulation components (such as the displacement adjustment mechanism of the hydraulic pump) and pressure regulation components (such as the relief valve and proportional pressure reducing valve) of the digital hydraulic drive unit. The displacement adjustment mechanism adjusts the pump displacement according to the signal, thereby changing the flow output; the relief valve and proportional pressure reducing valve adjust the pressure setpoint according to the signal, thereby controlling the system pressure. Through the coordinated action of these components, the output state of the hydraulic oil is adjusted to ensure that it matches the posture evolution requirements of the lifting actuator (such as the need for a larger flow rate during posture adjustment and a stable pressure during posture stabilization).

[0206] Step S156: Collect the real-time attitude data of the lifting actuator and the real-time output data of the digital hydraulic drive unit after adjustment, record the actual working status data after adjustment, and generate real-time adjustment status data.

[0207] Real-time attitude data (such as pitch angle and azimuth angle) of the lifting actuator is collected by attitude sensors, and real-time output data (such as actual flow rate and actual pressure) of the digital hydraulic drive unit is collected by flow sensors and pressure sensors. The above data is recorded in chronological order to generate real-time control status data.

[0208] Step S157: Compare the real-time control state data with the preset attitude evolution standard data and hydraulic drive output standard data, extract the control deviation data, and find the difference between the actual state and the standard state.

[0209] The preset attitude evolution standard data consists of attitude data at each time point along the preset trajectory, while the hydraulic drive output standard data consists of flow and pressure standard values ​​at each time point determined based on the preset trajectory and load conditions. The attitude data and hydraulic output data in the real-time control status data are compared with their corresponding standard data, and the differences are calculated to generate control deviation data.

[0210] Step S158: Based on the control deviation data, fine-tune the target bidirectional drive control signal set, correct the drive control signal to reduce the control deviation, and generate the fine-tuned target bidirectional drive control signal set.

[0211] Based on the magnitude and direction of the control deviation data, the corresponding signals in the target bidirectional drive control signal set are fine-tuned. For example, if the actual pitch angle is less than the standard pitch angle, with a deviation of Δα, the amplitude of the signal controlling the extension of the luffing cylinder is appropriately increased or the signal duration is extended to increase the pitch angle and reduce the deviation. After fine-tuning, a fine-tuned target bidirectional drive control signal set is generated.

[0212] Step S159: Re-input the fine-tuned target bidirectional drive control signal set into the lifting actuator and digital hydraulic drive unit to perform a new round of control processing and control the equipment to work according to the corrected signal.

[0213] The fine-tuned target bidirectional drive control signal set is sent to the control system and digital hydraulic drive unit of the lifting actuator. The control process of steps S151 to S158 is repeated to continuously optimize and adjust the attitude evolution timing of the lifting actuator and the dynamic output state of the hydraulic drive until the lifting operation is completed or the control deviation is within an acceptable range.

[0214] Figure 2 This diagram illustrates the hardware structure of a lifting motion trajectory tracking system 100 that integrates digital hydraulics and intelligent control for implementing the aforementioned lifting motion trajectory tracking method, as provided in an embodiment of the present invention. Figure 2 As shown, the lifting motion trajectory tracking system 100, which integrates digital hydraulics and intelligent control, may include a processor 110, a machine-readable storage medium 120, a bus 130, and a communication unit 140.

[0215] Machine-readable storage medium 120 can store data and / or instructions. In some embodiments, machine-readable storage medium 120 can store data acquired from an external terminal. In some embodiments, machine-readable storage medium 120 can store data and / or instructions used by the lifting motion trajectory tracking system 100 integrating digital hydraulics and intelligent control to execute or use in order to complete the exemplary methods described in this invention. In a specific implementation, one or more processors 110 execute the computer-executable instructions stored in machine-readable storage medium 120, enabling processor 110 to execute the lifting motion trajectory tracking method integrating digital hydraulics and intelligent control as described in the above method embodiments. Processor 110, machine-readable storage medium 120, and communication unit 140 are connected via bus 130, and processor 110 can be used to control the transmission and reception actions of communication unit 140. The specific implementation process of processor 110 can be found in the various method embodiments executed by the lifting motion trajectory tracking system 100 integrating digital hydraulics and intelligent control described above, and their implementation principles and technical effects are similar, so they will not be repeated here.

[0216] Furthermore, this embodiment of the invention also provides a readable storage medium containing computer-executable instructions. When the processor executes the computer-executable instructions, the lifting motion trajectory tracking method integrating digital hydraulics and intelligent control, as described above, is implemented.

[0217] It should be noted that, in order to simplify the description of this invention and thus aid in the understanding of one or more embodiments, the foregoing description of the embodiments of this invention sometimes combines multiple features into a single embodiment, drawing, or description thereof. Similarly, it should be noted that, in order to simplify the description of this invention and thus aid in the understanding of one or more embodiments, the foregoing description of the embodiments of this invention sometimes combines multiple features into a single embodiment, drawing, or description thereof.

Claims

1. A method for tracking lifting motion trajectory that integrates digital hydraulics and intelligent control, characterized in that, The method includes: A two-way coupling linkage feature set is constructed based on the continuous evolution characteristics of the motion posture of the lifting actuator and the dynamic coupling characteristics of the flow and pressure of the digital hydraulic drive unit. The two-way coupling linkage feature set carries the two-way adaptation and correlation information between the posture evolution of the lifting trajectory and the dynamic coupling of the hydraulic drive. The bidirectional coupling linkage feature set is cross-domain correlated with the dynamic load continuous evolution feature of the lifting operation to generate dynamic linkage empowerment parameters for digital hydraulics and intelligent control. Based on the aforementioned dynamic linkage empowerment parameters, a two-way coordinated drive control of intelligent control and digital hydraulic drive is realized, generating a set of two-way drive control signals for trajectory tracking of the lifting actuator. The trajectory tracking bidirectional drive control signal set is subjected to reverse source correction processing for trajectory evolution deviation to generate the target bidirectional drive control signal set for the lifting actuator. The attitude evolution timing and dynamic output state of the hydraulic drive are controlled based on the target bidirectional drive control signal set.

2. The lifting motion trajectory tracking method integrating digital hydraulics and intelligent control according to claim 1, characterized in that, The set of bidirectional coupled linkage features, constructed based on the continuous evolution characteristics of the motion posture of the lifting actuator and the dynamic coupling characteristics of the flow and pressure of the digital hydraulic drive unit, includes: The motion attitude data of the lifting actuator during the trajectory stages of lifting start, trajectory advancement, and lifting stop are collected. From the collected motion attitude data, continuous evolution features of attitude angle, continuous evolution features of attitude stability, and continuous evolution features of attitude adjustment response are extracted to generate a continuous evolution feature set of the motion attitude of the lifting actuator. The continuous evolution feature set of the motion attitude includes feature sequences characterizing attitude angle changes in different lifting stages, feature sequences characterizing attitude stability, and feature sequences characterizing the time delay and adjustment amplitude from the input of the drive control signal to the start of attitude adjustment. Dynamic flow and pressure data of a digital hydraulic drive unit under different drive and control commands are collected. The flow and pressure co-evolution characteristics, the hysteresis evolution characteristics of pressure with flow, and the flow and pressure stability evolution characteristics are extracted from the collected dynamic flow and pressure data to generate a dynamic coupling feature group of flow and pressure for the digital hydraulic drive unit. The dynamic coupling feature group of flow and pressure includes a feature sequence characterizing the co-change of flow and pressure, a feature sequence characterizing the hysteresis of pressure with respect to flow, and a feature sequence characterizing the flow and pressure entering a stable state. The motion posture continuous evolution feature group of the lifting actuator and the flow pressure dynamic coupling feature group of the digital hydraulic drive unit are subjected to feature dimension collaborative mapping processing. By matching the feature dimension attributes of the motion posture continuous evolution feature group and the flow pressure dynamic coupling feature group, a cross-dimensional correlation mapping relationship is established to generate a motion hydraulic bidirectional coupling feature vector with cross-domain correlation dimension. The motion-hydraulic bidirectional coupling feature vector is divided into stages and features are matched for the lifting trajectory. The trajectory stages are divided according to the stroke nodes of the preset lifting trajectory. The attitude evolution threshold feature and hydraulic drive coupling threshold feature corresponding to each trajectory stage are matched to generate the matching association information of attitude evolution threshold feature and hydraulic drive coupling threshold feature corresponding to each stage of the preset lifting trajectory. Based on the matching and association information, the bidirectional linkage between motion posture and hydraulic drive is mined, the changes in the demand for hydraulic drive parameters due to the posture evolution of the lifting actuator and the influence of the changes in hydraulic drive parameters on posture evolution are analyzed, and the characteristics of the evolution of the demand for hydraulic drive parameters due to the posture evolution of the lifting actuator and the characteristics of the influence of the changes in hydraulic drive parameters on posture evolution are generated. The bidirectional linkage relationship is deeply fused with the contour evolution features of the lifting preset trajectory, integrating trajectory contour evolution information, attitude evolution information and hydraulic drive coupling evolution information to generate a bidirectional fused feature that integrates trajectory contour evolution, attitude evolution and hydraulic drive coupling evolution. The bidirectional fusion features are subjected to spatiotemporal continuous expansion processing, extending the feature temporal span in the time dimension to cover the complete lifting process, and expanding the feature association range in the spatial dimension to cover the entire trajectory position, generating a spatiotemporal continuous expansion coupled feature that includes continuous temporal evolution features in the time dimension and trajectory position association evolution features in the spatial dimension. The spatiotemporal continuous extended coupling features are processed by feature association weight allocation. Based on the importance of different dimensional evolution features in bidirectional coupling modeling, corresponding weight coefficients are assigned to generate the association weight sequence of different dimensional evolution features in bidirectional coupling modeling. The spatiotemporal continuous extended coupling features are modeled by weighted aggregation using the associated weight sequence. After weighted calculation of features of different dimensions according to the weight coefficients, they are aggregated to generate a weighted aggregated bidirectional coupling feature vector. A bidirectional coupling linkage feature set is generated based on the weighted aggregated bidirectional coupling feature vector.

3. The lifting motion trajectory tracking method integrating digital hydraulics and intelligent control according to claim 1, characterized in that, The process of cross-domain correlation and empowerment processing of the bidirectional coupling and linkage feature set with the dynamic load continuous evolution characteristics of the lifting operation to generate dynamic linkage empowerment parameters for digital hydraulics and intelligent control includes: Real-time load dynamic data during the lifting operation is collected and combined with historical load evolution data of the lifting operation. The continuous load increase evolution characteristics, load mutation evolution characteristics, load stability evolution characteristics, and load evolution impact characteristics on lifting attitude are extracted from the collected load dynamic data to generate a dynamic load continuous evolution characteristic group for the lifting operation. The dynamic load continuous evolution characteristic group includes a feature sequence representing the load value increasing over time, a feature representing the time and magnitude of load mutation, a feature representing the duration and average of the load stability phase, and a correlation feature representing the relationship between load value and attitude angle. The bidirectional coupling linkage feature set and the dynamic load continuous evolution feature set of the lifting operation are subjected to feature association deep mining to explore the potential correlation between features in the two feature sets and generate cross-domain correlation feature vectors between bidirectional coupling features and dynamic load evolution features. The cross-domain correlation feature vector is processed to mine the hydraulic drive adaptation requirements of the load evolution, and the flow and pressure adaptation requirements of the digital hydraulic drive unit under different load evolution stages are analyzed to generate the flow dynamic adaptation requirement characteristics and pressure dynamic adaptation requirement characteristics of the digital hydraulic drive unit corresponding to different load evolution stages. Based on the dynamic adaptation requirements of flow rate and pressure, the dynamic adaptation relationship of intelligent control strategies is mined, the change mode of intelligent control strategies that match the adaptation requirements of hydraulic drive is analyzed, and the evolution characteristics of intelligent control strategies that match the adaptation requirements of hydraulic drive are generated. The dynamic adaptation requirements of the digital hydraulic drive and the evolution characteristics of the intelligent control strategy are bidirectionally empowered and fused to establish a mutual empowerment relationship between hydraulic drive parameters and intelligent control strategy, and generate a linkage fusion feature in which hydraulic drive parameters and intelligent control strategy mutually empower each other. The dynamic empowerment parameters of the linkage and fusion feature are subjected to dynamic empowerment parameter evolution trend prediction processing, the change trend of dynamic empowerment parameters in different stages of lifting motion is analyzed, and continuous evolution trend features of dynamic empowerment parameters in different stages of lifting motion are generated. Based on the aforementioned evolutionary trend characteristics, the dynamic empowerment parameters are continuously optimized to adapt to the needs of different load evolution stages and trajectory evolution stages, thereby generating optimized dynamic empowerment parameter characteristics that meet the needs of different load evolution stages and trajectory evolution stages. The optimized dynamic empowerment parameter features are subjected to dimensional standardization and temporal alignment processing to unify the dimensional attributes of parameters of different dimensions, align the temporal nodes of the parameters to match the lifting process, and generate standardized temporal empowerment parameter features that meet the requirements of cross-domain association empowerment. The standardized time-series empowerment parameter features are subjected to feature enhancement processing to strengthen the discriminability and correlation of the parameter features, and to generate enhanced cross-domain correlation dynamic empowerment features. Based on the enhanced cross-domain correlation dynamic empowerment feature, dynamic linkage empowerment parameters for digital hydraulics and intelligent control are generated.

4. The lifting motion trajectory tracking method integrating digital hydraulics and intelligent control according to claim 1, characterized in that, The method of achieving bidirectional coordinated drive and control of intelligent control and digital hydraulic drive based on the dynamic linkage empowerment parameters, generating a set of bidirectional drive and control signals for trajectory tracking of the lifting actuator, includes: The intelligent control strategy empowerment parameter features, digital hydraulic drive empowerment parameter features, and time sequence coordination empowerment parameter features are extracted from the dynamic linkage empowerment parameters to generate a dynamic linkage empowerment parameter feature group. The dynamic linkage empowerment parameter feature group includes an intelligent control parameter sequence, a digital hydraulic drive parameter sequence, and a time synchronization mark sequence representing the coordination time sequence of the two. Based on the dynamic linkage empowerment parameter feature group, an intelligent control dynamic drive strategy feature is generated that matches the dynamic adaptation requirements of hydraulic drive in real time, and the intelligent control drive strategy is adjusted according to the changes in the dynamic adaptation requirements of hydraulic drive. Based on the characteristics of the intelligent control dynamic drive strategy, the intelligent control parameters are continuously and dynamically adjusted. The intelligent control parameters are continuously adjusted according to the changes in the hydraulic drive state and the load evolution state, generating real-time dynamic control parameter characteristics of intelligent control. The real-time dynamic control parameter characteristics evolve continuously with the changes in the hydraulic drive state and the load evolution state. Based on the dynamic linkage empowerment parameter feature group, digital hydraulic dynamic drive control parameter features are generated and matched with the intelligent control strategy in real time. The digital hydraulic drive control parameters are adjusted according to the changes in the intelligent control strategy. Based on the characteristics of the digital hydraulic dynamic drive control parameters, the flow rate and pressure of the digital hydraulic drive unit are continuously and collaboratively adjusted. The flow rate and pressure are adjusted synchronously to match the changes in intelligent control parameters and load evolution state, thereby generating the real-time dynamic output characteristics of the digital hydraulic drive unit. The real-time dynamic output characteristics evolve continuously with the changes in intelligent control parameters and load evolution state. The real-time dynamic control parameter characteristics of the intelligent control and the real-time dynamic output characteristics of the digital hydraulic drive unit are subjected to bidirectional collaborative correlation modeling processing to establish a collaborative working correlation model between the two and generate bidirectional collaborative drive and control characteristics of intelligent control and digital hydraulic drive. The bidirectional collaborative drive control features are subjected to continuous encoding processing of drive control signals, converting the collaborative drive control features into a continuous signal encoding form, and generating a continuous timing drive control signal encoding sequence for the lifting actuator; Based on the continuous timing drive control signal encoding sequence, dynamic adaptation and conversion processing of the signal format is performed to adjust the signal encoding format to meet the requirements of the real-time control protocol of the lifting actuator, and dynamic format drive control signal characteristics that meet the requirements of the real-time control protocol of the lifting actuator are generated. The dynamic format drive control signal features are subjected to signal enhancement and anti-interference processing to strengthen the signal strength and reduce the influence of interference signals, thereby generating enhanced bidirectional cooperative drive control signal features. Based on the enhanced bidirectional collaborative drive and control signal characteristics, a set of bidirectional drive and control signals for trajectory tracking of the lifting actuator is generated.

5. The lifting motion trajectory tracking method integrating digital hydraulics and intelligent control according to claim 1, characterized in that, The step of performing reverse tracing correction processing on the trajectory evolution deviation of the trajectory tracking bidirectional drive control signal set to generate the target bidirectional drive control signal set for the lifting actuator includes: The trajectory tracking bidirectional drive control signal feature set is extracted from the set of trajectory tracking bidirectional drive control signals to generate a trajectory tracking bidirectional drive control signal feature set. The trajectory tracking bidirectional drive control signal feature set includes the amplitude sequence of the drive control signal, the time interval sequence of the signal transmission, and the time difference sequence between the intelligent control signal and the hydraulic drive signal. Based on the trajectory tracking bidirectional drive control signal feature group, the trajectory evolution deviation characteristics that may occur within a preset time period are predicted. At the same time, the root causes of the deviation in hydraulic drive parameters and intelligent control parameters are traced. The deviation trend and root causes are analyzed by combining the motion posture evolution characteristics of the lifting actuator and the dynamic coupling characteristics of digital hydraulic drive. Based on the trajectory evolution deviation characteristics, the root cause characteristics of hydraulic drive parameter deviation, and the root cause characteristics of intelligent control parameter deviation, reverse correction strategy characteristics for hydraulic drive parameter deviation and reverse correction strategy characteristics for intelligent control parameter deviation are generated, and the correction method and adjustment range of the corresponding parameters are formulated. Based on the reverse correction strategy features for the deviation of hydraulic drive parameters, digital hydraulic drive correction parameter features are generated, and the digital hydraulic drive parameters are adjusted according to the correction strategy. Based on the reverse correction strategy features for the deviation of intelligent control parameters, intelligent control correction parameter features are generated, and intelligent control parameters are adjusted according to the correction strategy. The digital hydraulic drive correction parameter features and the intelligent control correction parameter features are synergistically fused to integrate the two correction parameter features to achieve a synergistic correction effect and generate bidirectional synergistic correction features. Based on the aforementioned bidirectional collaborative correction feature, the trajectory tracking bidirectional drive control signal is corrected and encoded. The encoding of the drive control signal is adjusted to incorporate correction parameter information, thereby generating a corrected drive control signal encoding sequence. The modified drive control signal encoding sequence is fused with the original trajectory tracking bidirectional drive control signal feature group to integrate the original drive control signal information and the modified drive control signal information, and generate a signal feature vector that fuses the original drive control and reverse correction. The signal feature vectors of the fused original drive control and reverse correction are subjected to signal format adaptation processing to adjust the signal format to meet the requirements of the lifting actuator control protocol, and a target signal format feature that conforms to the lifting actuator control protocol is generated. A set of target bidirectional drive control signals for the lifting actuator is generated based on the target signal format characteristics that conform to the lifting actuator control protocol.

6. The lifting motion trajectory tracking method integrating digital hydraulics and intelligent control according to claim 2, characterized in that, The process involves collecting motion attitude data of the lifting actuator during the lifting start-up, trajectory advancement, and lifting stop phases. From this collected motion attitude data, continuous evolution features of attitude angles, attitude stability, and attitude adjustment response are extracted to generate a set of continuous motion attitude evolution features for the lifting actuator, including: The system controls the lifting actuator to complete lifting actions with different loads and different trajectory plans in a simulated lifting scenario. It simultaneously collects motion attitude data of the trajectory stages of lifting start, trajectory advancement, and lifting stop, records the attitude change data of each trajectory stage, and generates a raw set of motion attitude data of the lifting actuator. The original motion posture data set of the lifting actuator is processed by removing invalid data and supplementing missing data, removing abnormal interference data, and using the evolution trend of adjacent valid data to supplement missing data, thereby generating preprocessed motion posture data. The attitude angle change data of different trajectory stages are extracted from the preprocessed motion attitude data. Based on the attitude angle change data, attitude angle continuous evolution features are generated. The attitude angle continuous evolution features carry the information of the continuous change of attitude angle over time in different trajectory stages, and fully present the change process of attitude angle from initial value to target value. From the preprocessed motion posture data, attitude fluctuation data of different trajectory stages are extracted, and attitude stability continuous evolution features are generated based on the attitude fluctuation data. The attitude stability continuous evolution features carry the continuous change information of attitude fluctuations at different trajectory stages and reflect the change in the stability of the attitude during the evolution process. The time difference data between the input time of the drive control signal and the time of the attitude adjustment response and the attitude adjustment amplitude data of different trajectory stages are extracted from the preprocessed motion attitude data. Based on the time difference data between the input time of the drive control signal and the time of the attitude adjustment response and the attitude adjustment amplitude data, a continuous evolution feature of the attitude adjustment response is generated. The continuous evolution feature of the attitude adjustment response carries the continuous change information of the attitude response speed and response amplitude to the drive control signal at different trajectory stages, reflecting the response efficiency and effect of the attitude to the drive control signal. The attitude angle continuous evolution feature, attitude stability continuous evolution feature and attitude adjustment response continuous evolution feature are processed by trajectory stage association, and each feature is bound to the corresponding trajectory stage to generate the comprehensive motion attitude evolution feature corresponding to each trajectory stage. The comprehensive motion posture evolution features corresponding to each trajectory stage are extended in time dimension, and the comprehensive motion posture evolution features of adjacent trajectory stages are associated to generate extended motion posture features containing posture evolution association information of adjacent trajectory stages. The extended motion posture features are weighted and assigned according to their respective roles in reflecting the overall state of the motion posture. This generates weight coefficients for different posture evolution features. The extended motion posture features are weighted by the weighting coefficients. The extended motion posture features of different dimensions are weighted according to the weighting coefficients to generate weighted motion posture evolution feature components. The weighted motion posture evolution feature components are then combined into a weighted motion posture evolution feature vector. Based on the weighted motion posture evolution feature vector, a continuous evolution feature group of motion posture of the lifting actuator is generated.

7. The lifting motion trajectory tracking method integrating digital hydraulics and intelligent control according to claim 3, characterized in that, The real-time load dynamic data collected during the lifting operation is combined with historical load evolution data of the lifting operation. From the collected load dynamic data, continuous load increase evolution characteristics, sudden load evolution characteristics, stable load evolution characteristics, and the impact of load evolution on lifting attitude are extracted to generate a dynamic load continuous evolution feature set for the lifting operation, including: In actual lifting operation scenarios, load detection equipment is deployed to continuously collect real-time load dynamic data during the lifting operation process. At the same time, historical load evolution data of lifting operations under different work objects and different work environments are collected. Real-time data and historical data are integrated to jointly generate a set of original load dynamic data. The original load dynamic data set is cleaned to remove abnormal load data and invalid load data, remove data that exceeds the normal load range and data that has no actual load significance, and generate cleaned load dynamic data. Extract the load change data with lifting time from the cleaned load dynamic data, and generate a continuous load evolution feature based on the load change data with lifting time. The continuous load evolution feature carries the information of continuous load change with the lifting process, and fully presents the load increase process from the initial value to the peak value. Extract the data of mutation nodes whose load change rate exceeds a preset range and the load data before and after the mutation from the cleaned load dynamic data. Generate load mutation evolution features based on the data of mutation nodes whose load change rate exceeds a preset range and the load data before and after the mutation. The load mutation evolution features carry information on the time and magnitude of the load mutation and clarify the key information of the load mutation. Extract stable phase data where the load change rate is within a preset range from the cleaned load dynamic data. Generate load stability evolution features based on the stable phase data where the load change rate is within a preset range. The load stability evolution features carry information about the duration of the load stability phase and the load stability value, reflecting the core parameters of the load stability state. Extract lifting posture change data under different load conditions from the cleaned load dynamic data, and generate load evolution impact characteristics on lifting posture based on the lifting posture change data under different load conditions. The load evolution impact characteristics on lifting posture carry the correlation information between load change and posture change, and reflect the relationship between load change and posture. The continuous load evolution characteristics, sudden load evolution characteristics, stable load evolution characteristics, and the influence of load evolution on lifting attitude are processed by operation scenario association, and each feature is bound to the corresponding operation scenario to generate a comprehensive load evolution characteristic for each operation scenario. The comprehensive load evolution characteristics corresponding to each operation scenario are extended in time dimension, and the comprehensive load evolution characteristics of different lifting periods are associated to generate extended load evolution characteristics containing load evolution association information of different lifting periods. The extended load evolution features are weighted and assigned according to their respective roles in reflecting the overall load evolution state, thereby generating weight coefficients for different load evolution features. The extended load evolution features are weighted using the weighting coefficients. The extended load evolution features of different dimensions are weighted according to the weighting coefficients to generate weighted load evolution feature components. The weighted load evolution feature components are then combined into a weighted load evolution feature vector. Based on the weighted load evolution feature vector, a dynamic load continuous evolution feature group for lifting operations is generated.

8. The lifting motion trajectory tracking method integrating digital hydraulics and intelligent control according to claim 4, characterized in that, The process of continuously and dynamically adjusting the intelligent control parameters based on the characteristics of the intelligent control dynamic drive and control strategy to generate real-time dynamic control parameter characteristics for intelligent control includes: The intelligent control dynamic drive strategy features are converted into corresponding intelligent control parameter adjustment instructions. The intelligent control parameter adjustment instructions carry the direction and magnitude of parameter adjustment information, and clarify the adjustment method of each intelligent control parameter. Based on the intelligent control parameter adjustment command, the intelligent control parameters are initially adjusted, and the initial value of the intelligent control parameters is changed according to the direction and magnitude of the adjustment command to generate the initially adjusted intelligent control parameters. Collect intelligent control output status data and corresponding digital hydraulic drive response data under the action of the initially adjusted intelligent control parameters, record the output information of intelligent control and the feedback information of digital hydraulic drive, and generate parameter adjustment response data; The deviation data between the intelligent control output state and the expected output state is extracted from the parameter adjustment response data. The difference between the actual output state and the expected output state is compared. Based on the deviation data between the intelligent control output state and the expected output state, the intelligent control parameters are adjusted a second time to correct the parameters and reduce the deviation, thereby generating the intelligent control parameters after the second adjustment. Repeat the process of parameter adjustment, response data acquisition, deviation analysis and secondary parameter adjustment until the deviation between the intelligent control output state and the expected output state is within an acceptable range. Parameters are optimized through multiple rounds of iteration to generate iteratively optimized intelligent control parameters. Based on the iteratively optimized intelligent control parameters, combined with load evolution characteristics and hydraulic drive state characteristics, the law of parameter change with load and hydraulic state is analyzed, and the continuous adjustment trend of intelligent control parameters is generated. The intelligent control parameters are dynamically adjusted in real time according to the continuous adjustment trend of the intelligent control parameters, and the parameter values ​​are continuously changed in accordance with the load evolution and the change of hydraulic drive state, so as to generate continuous change data of intelligent control parameters. Extract parameter values ​​and parameter change rates at different time points from the continuous change data of the intelligent control parameters, record the parameter status and change rate at each time point, and generate real-time dynamic control parameter characteristics of the intelligent control. The real-time dynamic control parameter features of the intelligent control are smoothed in the time dimension to generate smoothed real-time dynamic control parameter features, and the final real-time dynamic control parameter features of the intelligent control are generated based on the smoothed real-time dynamic control parameter features.

9. The lifting motion trajectory tracking method integrating digital hydraulics and intelligent control according to claim 1, characterized in that, The control of the attitude evolution timing and dynamic output state of the hydraulic drive based on the target bidirectional drive control signal set includes: Extract attitude control timing features and hydraulic drive output timing features from the target bidirectional drive control signal set, filter out timing information related to attitude control and hydraulic drive output, and generate target drive control core feature group; Based on the attitude control timing features in the target drive control core feature group, the attitude adjustment time and attitude adjustment range of the lifting actuator are controlled, the execution time and adjustment amount of each attitude adjustment action are determined, and the attitude evolution timing control signal of the lifting actuator is generated. Based on the attitude evolution timing control signal, the mechanical transmission components of the lifting actuator are driven to perform coordinated actions, and the motion state of the mechanical transmission components is controlled to adjust the attitude of the lifting actuator, thereby adjusting the motion attitude of the lifting actuator so that the attitude evolution of the lifting actuator conforms to the attitude requirements of the preset trajectory. Based on the hydraulic drive output timing characteristics in the target drive control core feature group, the flow output time, flow output amplitude, pressure output time, and pressure output amplitude of the digital hydraulic drive unit are adjusted to determine the time nodes and output quantities of flow and pressure output, and to generate dynamic output control signals for the digital hydraulic drive. Based on the dynamic output control signal of the digital hydraulic drive, the flow regulation component and pressure regulation component of the digital hydraulic drive unit are driven to perform coordinated actions, control the working state of the flow regulation component and pressure regulation component to adjust the hydraulic oil output, and adjust the output state of the hydraulic oil so that the output state of the digital hydraulic drive unit matches the posture evolution requirements of the lifting actuator. Collect real-time attitude data of the lifting actuator and real-time output data of the digital hydraulic drive unit after adjustment, record the actual working status data after adjustment, and generate real-time adjustment status data. The real-time control state data is compared with the preset attitude evolution standard data and hydraulic drive output standard data to extract the control deviation data and find the difference between the actual state and the standard state. Based on the aforementioned control deviation data, the target bidirectional drive control signal set is fine-tuned to correct the drive control signal and reduce the control deviation, thereby generating a fine-tuned target bidirectional drive control signal set. The finely tuned target bidirectional drive control signal set is re-input into the lifting actuator and digital hydraulic drive unit to perform a new round of control processing, and the equipment is controlled to work according to the corrected signal.

10. A lifting motion trajectory tracking system integrating digital hydraulics and intelligent control, characterized in that, The lifting motion trajectory tracking system integrating digital hydraulics and intelligent control includes a processor and a memory, the memory and the processor being connected. The memory is used to store programs, instructions or code, and the processor is used to run the programs, instructions or code in the memory to implement the lifting motion trajectory tracking method integrating digital hydraulics and intelligent control as described in any one of claims 1-9.

Images