Multi-mode switching control method for variable displacement piston pump driven by servo motor

Abstract

The invention relates to the technical field of industrial automation, and discloses a multi-mode switching control method for a variable displacement piston pump driven by a servo motor, which comprises the following steps of: acquiring multi-dimensional operation state parameters and load power of a variable displacement piston pump system; performing dynamic mode decomposition on the load power signal, and executing preset state switching compensation mapping to complete mode switching on the basis of the extracted mode characteristics and the current state parameter under the condition of judging that the switching condition is satisfied; executing a corresponding control strategy in the new mode, and dynamically adjusting the pump displacement and the rotating speed of the servo motor through feedback control; and detecting the response and energy efficiency performance of the regulation and control system according to the performance index in the system operation process. By adopting a dynamic mode intelligent switching mechanism, the system can automatically match the optimal operation mode in different working environments, the energy utilization efficiency is greatly improved, and the effects of effectively reducing energy waste and reducing the total energy consumption of the system during stable load or low load operation are achieved.

Description

Technical Field

[0001] This invention relates to the field of industrial automation technology, specifically to a multi-mode switching control method for a servo motor-driven variable displacement piston pump. Background Technology

[0002] In modern industrial production systems, hydraulic systems, as key power transmission and control components, are widely used in a variety of complex working conditions, with variable displacement piston pumps being the core power components. Because they can flexibly adjust output flow and pressure according to load conditions, variable displacement piston pumps play an irreplaceable role in many fields such as construction machinery, heavy equipment manufacturing, and shipbuilding.

[0003] However, variable displacement piston pumps in current hydraulic systems face significant challenges in energy utilization. On one hand, traditional variable displacement piston pump control technologies are mostly based on mechanical, hydraulic, or simple electrical control logic, which struggles to accurately perceive changes in the real-time operating environment. Under stable load conditions, the system maintains high output power, resulting in the unnecessary consumption of large amounts of electrical energy. Conversely, during low-load operation, the pump's output parameters fail to adapt promptly, leading to severe energy waste. For example, in the hydraulic propulsion system of a large ship, during relatively stable load phases such as cruising, the variable displacement piston pump cannot automatically switch to energy-saving mode, resulting in an additional monthly energy consumption of several thousand kilowatt-hours, significantly increasing operating costs. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a multi-mode switching control method for a servo motor-driven variable displacement piston pump. This method solves the problem that existing hydraulic systems often use constant proportional or empirical parameter control methods, resulting in a fixed matching relationship between the servo motor speed and the pump displacement. This lack of dynamic response capability to actual operating conditions leads to the system maintaining high power output even when the load fluctuates little or is in a low-load operating phase, resulting in a large amount of energy waste and low overall operating efficiency.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a multi-mode switching control method for a servo motor-driven variable displacement piston pump, comprising the following steps: Collect multidimensional operating status parameters and load power of the variable displacement piston pump system; Dynamic mode decomposition is performed on the load power signal to extract multi-scale modal features representing the system's operating state; Based on the extracted modal features and current state parameters, the current operating mode of the system is determined. If the switching conditions are met, the preset state switching compensation mapping is executed to complete the mode switching; In the new mode, the corresponding control strategy is executed, and the pump displacement and servo motor speed are dynamically adjusted through feedback control. The controller parameters are adjusted in real time according to the performance indicators during system operation to detect and regulate the system response and energy efficiency.

[0006] Preferably, the state parameters include pressure, speed, oil temperature, and pump displacement. The load power signal is acquired and calculated through the current sensor and speed sensor of the servo motor, and the load power signal represents the actual power output of the servo motor load.

[0007] Preferably, the load power signal is calculated using the following formula: W = k × I × N; Where: k is the power constant; I is the current signal; N is the rotational speed signal.

[0008] Preferably, the modal feature extraction step includes performing empirical mode decomposition on the load power signal, decomposing the original power signal into multiple eigenmode functions with local time-varying characteristics and a trend term.

[0009] Preferably, the intrinsic mode function is subjected to wavelet packet transform to obtain the energy distribution of each mode component, and the current system operation mode is determined based on the changing trend of the energy distribution.

[0010] Preferably, the mode switching determination criteria include: the rate of change of modal energy, the rate of change of load power, the rate of change of pump outlet pressure, and the change of oil temperature. A threshold is set to determine whether the system needs to switch modes.

[0011] Preferably, the switching modes include energy-saving mode, high-response mode and protection mode, wherein energy-saving mode corresponds to stable load conditions, high-response mode corresponds to drastic load changes, and protection mode corresponds to abnormal conditions.

[0012] Preferably, the state switching compensation mapping step uses a smoothing algorithm to perform transition processing on the current system state, and the transition processing is used to generate system oscillations or discontinuities during the switching process.

[0013] Preferably, the smoothing algorithm includes linear interpolation, quadratic interpolation, or higher-order interpolation methods for smoothing transitions.

[0014] Preferably, the control strategy employs the optimal control algorithm in each operating mode, which minimizes energy consumption and maintains system stability by adjusting the ratio of servo motor speed to pump displacement in real time.

[0015] This invention provides a multi-mode switching control method for a servo motor-driven variable displacement piston pump. It offers the following advantages: 1. This invention employs a dynamic mode intelligent switching mechanism, enabling the system to automatically match the optimal operating mode under different working environments, significantly improving energy utilization efficiency and effectively reducing energy waste and lowering the total energy consumption of the system when the load is stable or low.

[0016] 2. This invention employs an efficient feedback control mechanism to quickly respond to load changes and system disturbances, enabling precise adjustment of the motor-pump ratio under high dynamic load conditions, resulting in a fast and stable system response, and significantly improving the system's dynamic tracking capability and emergency response speed.

[0017] 3. This invention combines a smoothing algorithm with an optimal control strategy to achieve a smooth transition between different control modes, thereby effectively suppressing system oscillations, reducing discontinuities in the control process, and improving the overall stability and reliability of the system operation.

[0018] 4. This invention achieves automatic switching to protection mode under abnormal operating conditions through intelligent mode switching and real-time monitoring mechanism, thereby avoiding overload, overheating and other faults, significantly improving system safety and extending the service life of key components. Attached Figure Description

[0019] Figure 1 This is a perspective view of the present invention. Detailed Implementation

[0020] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Please see the appendix Figure 1 This invention provides a multi-mode switching control method for a servo motor-driven variable displacement piston pump, comprising the following steps: Collect multidimensional operating status parameters and load power of the variable displacement piston pump system; Dynamic mode decomposition is performed on the load power signal to extract multi-scale modal features representing the system's operating state; Based on the extracted modal features and current state parameters, the current operating mode of the system is determined. If the switching conditions are met, the preset state switching compensation mapping is executed to complete the mode switching; In the new mode, the corresponding control strategy is executed, and the pump displacement and servo motor speed are dynamically adjusted through feedback control. The controller parameters are adjusted in real time according to the performance indicators during system operation to detect and regulate the system response and energy efficiency.

[0022] The status parameters include pressure, speed, oil temperature and pump displacement. The load power signal is acquired and calculated through the current sensor and speed sensor of the servo motor. The load power signal represents the actual power output of the servo motor load.

[0023] Specifically, the entire control process is built upon a continuous state observation and discrimination logic. The key lies in the high-fidelity acquisition of the load power signal and the accurate extraction of state variables. Changes in state parameters, as explicit characteristics of the system's internal operating state, provide a basis for modal discrimination. Dynamic feedback of the load power signal plays a crucial role in response-driven and control closed-loop correction. Real-time sensing and dynamic management of the plunger pump's operating state are typically required, especially in situations with frequent load fluctuations and high response speed requirements. Traditional fixed control strategies suffer from strong lag and high control rigidity. The multi-mode switching control scheme proposed in this invention can adaptively switch between energy-saving, high-response, or protection control modes according to the system's operating characteristics, thereby improving overall operating performance. The pump displacement signal is indirectly obtained by measuring the angular displacement of the servo motor drive through the encoder, and then solved by the conversion model between the pump swashplate angle and the displacement. The load power signal is calculated by the input current and real-time speed of the servo motor, and is used to reflect the power demand of the system under the current operating conditions. This power signal is approximately calculated using the following expression: P = k·I·n. in: P represents the real-time load power, measured in watts (W). I represents the servo motor input current, measured in amperes (A), which is detected by a Hall current sensor; n represents the real-time speed of the servo motor, in revolutions per minute (rpm); k is a power conversion constant, which depends on factors such as motor efficiency, load coupling coefficient, and motor rated power factor, and is expressed in volts per second. The system also introduces a dynamic correction term ΔP to compensate for the back electromotive force of the servo motor and the hysteresis effect of the nonlinear load, making the final load power estimate closer to the actual output state. P load =k·I·n+ΔP; in: ΔP is dynamically compensated by an empirical model or neural network prediction module, and can be obtained by calibration through training data; P load This represents the real-time load power after compensation. The acquisition period of the current signal needs to be synchronized with the PWM control period of the servo driver to ensure that the response speed of the load power feedback is not affected by the sampling delay. The current signal is obtained by processing the three-phase current through Clarke Transform and Park Transform to extract the d-axis current I. d The component is used for subsequent load calculations. In light-load areas, the system can set a load threshold P. th To avoid triggering the wrong mode; The load power signal not only provides a criterion for mode switching but also serves as a real-time basis for controller parameter adjustment. When the load power fluctuation exceeds the preset range, the system will trigger a dynamic modal recognition mechanism to perform empirical mode decomposition (EMD) on the current state signal and update the system operating mode based on the decomposition results. The rapid fluctuation characteristics of the load power can also be used to predict the impending shock wave or back pressure abnormality event in the hydraulic system, thereby switching to the protection mode in advance and extending the service life of the equipment. The state parameter acquisition mechanism and the load power signal extraction method have all achieved closed-loop closure through the standard sensor system and controller data channel, providing a complete physical foundation and data support for subsequent modal recognition and multi-mode switching control.

[0024] The load power signal is calculated using the following formula: W = k × I × N; Where: k is the power constant; I is the current signal; N is the rotational speed signal.

[0025] Specifically, after obtaining the load power signal, the sampling periods of the current signal and the speed signal must be strictly consistent, typically between 1ms and 10ms, to ensure that the dynamic response of the load power matches the system control commands. The value of the power constant k will be related to factors such as the rated parameters of the servo motor used, the pump load type, and the drive control method. A preliminary estimate can generally be made using the following empirical model: Where P rated I rated N rated These are the motor's rated power, current, and speed, respectively. A compensation correction term ΔW can also be introduced to correct power distortion errors caused by high-frequency load fluctuations or temperature changes. The final output power expression is: W actual =k·I·N+ΔW; ΔW can be dynamically given by a temperature-related compensation model or a data-driven method (such as a BP neural network). The compensation module is automatically activated when the oil temperature exceeds a certain threshold (such as 60℃). The proposed formula simplifies the problem of simultaneously acquiring torque and angular velocity in traditional power measurement. Only the current and speed parameters need to be acquired to complete the estimation of servo load power, which is convenient for engineering deployment and conducive to rapid deployment in industrial control systems. Based on the above formulas and parameter definitions, those skilled in the art can achieve real-time load power estimation and controller feedback correction in a conventional hydraulic control platform (such as a PLC or embedded controller) by serially calling sampling, calculation and judgment modules. This has strong engineering reproducibility and system integration convenience.

[0026] The modal feature extraction step includes performing empirical mode decomposition on the load power signal, decomposing the original power signal into multiple eigenmode functions with local time-varying characteristics and a trend term.

[0027] Specifically, the modal feature extraction step includes performing empirical mode decomposition on the real-time acquired load power signal to identify local variation features of system operation and use them for subsequent operation mode identification and state mapping. Specifically, EMD decomposition decomposes the original power signal W(t) into a finite number of intrinsic mode functions (IMFs) and a residual trend term r(t), which is expressed as follows: in: c i (t) represents the i-th order intrinsic mode function (IMF), which reflects the local characteristic components of the system at different time scales; r(t) is the final residual term, which usually represents the overall trend of the signal or low-frequency drift behavior; n is the total number of mode functions obtained from EMD decomposition, which is dynamically determined based on signal complexity.

[0028] In this embodiment, the EMD algorithm is executed using the following iterative strategy: Initial settings: Let h0(t) = W(t), and set the allowable error threshold ∈; Extreme value identification: finding h k All local maxima and minima in (t); Envelope fitting: The upper envelope e is constructed by fitting the maxima and minima using cubic spline interpolation. max (t) and lower envelope e min (t); Mean calculation: Calculate the envelope mean. Signal Update: Obtain new candidate IMF components h k+1 (t)=h k (t)-m k (t); Judgment condition: when h k+1 If (t) satisfies both IMF criteria (the difference between the number of extreme values ​​and the number of zero crossings is at most 1; the local mean is close to 0), then it is denoted as c. i If the residual term r(t) is positive, then continue the sifting process. This process is repeated until the residual term r(t) becomes a monotonic function or there are no further decomposable features. To avoid endpoint effects and mode aliasing, the EMD algorithm can incorporate boundary mirror extension and **ensemble empirical mode decomposition (EEMD)** techniques to suppress high-frequency noise. These enhancements significantly improve the physical consistency and engineering interpretability of the mode decomposition. After the decomposition is completed, each c is obtained i (t) Each modal component corresponds to a specific frequency bandwidth and power density concentration region. The system identifies the modal characteristics of the operating state by analyzing characteristic parameters such as the instantaneous frequency, energy weight, and oscillation number of these IMF components. In one possible implementation, each c i The average envelope energy E of (t) i It can be calculated using the following formula: in: T is the sampling period; E i Let be the average energy intensity of the i-th order IMF component; Changes in energy spectrum distribution can serve as a sensitive indicator of a system entering a new operating mode, and can also be used as a basis for fault early warning. After completing EMD decomposition, short-time Fourier transform (STFT) or Hilbert transform can be performed on key IMF components to extract their instantaneous frequency components, further enriching the modal discrimination input. The Hilbert spectrum of the key components can be expressed as: in: A i (t) represents the instantaneous amplitude; ω i (t) is the instantaneous frequency, derived from the Hilbert transform; This type of transient spectrum structure provides important input for subsequent mode switching mapping and pattern recognition classifiers. Through EMD processing of the load power signal, this invention can separate complex nonlinear dynamic behavior into a series of modal features with physical meaning, which facilitates the system to identify energy concentration, impact disturbance or energy-saving sections under different operating states, thereby providing reliable data support for multi-mode switching control strategies. All processing flows can be implemented in real time through embedded controllers, and are compatible with conventional industrial platforms such as Cortex-M4 and TIDSP, with good industrial feasibility and algorithm versatility.

[0029] The intrinsic mode functions are subjected to wavelet packet transform to obtain the energy distribution of each mode component, and the current system operation mode is determined based on the changing trend of the energy distribution.

[0030] Specifically, wavelet packet transform is an extension of traditional wavelet transform. It not only decomposes the subband signal into low frequencies but also further decomposes the high-frequency components, resulting in higher frequency resolution. It is particularly suitable for full-band analysis of non-stationary signals. The specific implementation steps are as follows: Modality selection: Select several representative IMF components from {c1(t),c2(t),…,c(t)} obtained from EMD decomposition, denoted as c(t). These can be selected based on energy weights or frequency band focusing. Construct a WPT tree: Perform wavelet packet decomposition on each c(t) up to the set decomposition level J, constructing a complete wavelet packet binary tree structure. Node energy calculation: Calculate the energy value E,j,k of each leaf node signal w,j,k(t). The calculation formula is as follows: in: j represents the decomposition level; k represents the index of the current node; E jk This represents the local energy carried by the k-th node in the j-th layer; Constructing the energy distribution vector: Normalize the energy of all sub-band nodes in sequence to form the energy distribution vector E = [e1, e2, ..., e m ],in: This energy vector reflects the relative weight of different frequency band signals in the current IMF component. Different operating modes (such as light load, heavy load, impact condition, energy recovery mode, etc.) will show typical energy distribution differences in the frequency domain. Under normal steady-state operating mode, the system energy is mostly concentrated in the low and medium frequency bands. In the impact or sudden load mode, the energy of high frequency nodes increases significantly. In the no-load or light load range, low frequency energy dominates and the total energy intensity decreases significantly. By observing the dynamic change trend of each frequency band in the energy distribution vector, the current operating status of the system can be effectively determined. However, this invention proposes the following operating mode discrimination criteria: Define the characteristic frequency band intervals, such as [f1,f2],[f3,f4]; The total corresponding energy is denoted as E. L E H These represent low-frequency and high-frequency energy integration, respectively. Construction modal ratio index: Where ∈ is a small constant to prevent the denominator from being zero (e.g., 10). -6 ); Set a threshold interval η1, η2, when: η<η1→The system is in a stable / low-power mode; η1≤η<η2→The system is in a transition state; η≥η2→The system is in a high-frequency impact / load mutation mode.

[0031] K-means clustering can be used to cluster energy distribution vectors within multiple time windows to construct modality class centers. During online runtime, optimal classification is performed based on the Euclidean distance between the current energy vector and the modality center, enabling fast pattern matching. The algorithm is as follows: in: E t The wavelet packet energy distribution for the current time slice; μ i Let i be the i-th cluster center; The output Mode(t) is the current running state number; By employing a two-level analysis structure of "EMD+WPT" and combining modal component selection, wavelet node energy extraction, energy spectrum construction, and operating mode recognition mechanisms, a complete modal feature extraction and recognition process is constructed. Its structure is clear, its implementation is flexible, and it possesses high reproducibility, making it suitable for various variable loads and highly disturbed hydraulic servo system operating conditions in the field.

[0032] The criteria for mode switching include: the rate of change of modal energy, the rate of change of load power, the rate of change of pump outlet pressure, and the change of oil temperature. Thresholds are set to determine whether the system needs to switch modes.

[0033] Specifically, the criteria for mode switching include not only the energy change trend of modal components, but also multiple physical state parameters such as load power change rate, pump outlet pressure change rate, and oil temperature change trend. This constructs a more robust and faster-responding intelligent switching judgment framework. Through the following multivariate switching index system, each index reflects dynamic characteristics in the form of time series derivatives or change slopes: Modal energy change rate: Let the current wavelet packet energy vector be E(t), and the energy at the previous time step be E(t-Δt). The rate of change is calculated as follows: in: ||·||2 represents the L2 norm; R E (t) represents the rate of energy change, which is the primary indicator for determining whether a system has entered the unsteady-state region; Load power change rate: If the current real-time load power is W(t), then its first-order difference is approximately: This indicator reflects the changes in system response caused by external load disturbances, and whether the corresponding control strategy needs to be accelerated or buffered; Pump outlet pressure change rate: that is, the pump outlet pressure P(t) changes at a certain rate: When the system experiences mechanical shock, control lag, or sudden load changes, R P The pressure will increase significantly, indicating that the system pressure is becoming unstable and it is necessary to switch to high-response control mode. Oil temperature change rate: The trend of oil temperature T(t) change has a significant impact on the long-term performance of the control system, and its change rate is expressed as: When R T If the threshold is exceeded, it indicates that the system may experience a decrease in efficiency or a change in oil viscosity due to long-term operation, and it should be considered to enter a protection or energy-saving operation state. By comparing the above four rates of change with preset thresholds, a mode switching logic criterion is constructed, assuming: θ E : Modal energy change rate threshold; θ W : Load power change rate threshold; θ P : Threshold for rate of change of pressure; θ T: Threshold for oil temperature change rate; The system enters the mode switching determination window when any two or more of the following conditions are met: The decision model can be designed as soft logic (such as fuzzy weight sum) or hard logic (such as Boolean expression) to adapt to the needs of different control strategies; By constructing a multi-dimensional dynamic rate-of-change index system including modal energy, power, pressure, and temperature, and combining threshold judgment with operational strategy matching logic, a multi-mode switching criterion mechanism with fast response, robust judgment, and sustainable scalability is realized. Its structure is complete, its logic is clear, and it possesses good engineering feasibility.

[0034] The switching modes include energy-saving mode, high-response mode, and protection mode. Energy-saving mode is for stable load conditions, high-response mode is for drastic load changes, and protection mode is for abnormal conditions.

[0035] Specifically, the system operates in three modes: energy-saving mode, high-response mode, and protection mode. Automatic switching is achieved through multi-parameter threshold judgment. In energy-saving mode, the system load is stable, pressure fluctuations are small, and power changes are slow. The criterion is the modal energy change rate R. E Low; Load power change rate R W Low; normal oil temperature with no sudden change trend; reduce motor output power, slow down swashplate adjustment rate, enable low-frequency control parameters to reduce energy consumption and extend system life; high response mode with rapid load changes, frequent starts and stops, and obvious impact fluctuations; the criterion characteristic is: R W R P High, the high-frequency energy of the IMF component increases, the modal energy changes rapidly, and the protection mode exhibits high oil temperature, abnormal pressure fluctuations, and continuous high-frequency energy accumulation. The criterion characteristic is: R T Activities exceeding limits or rapid temperature rise, R E High and concentrated at high frequencies, R P In case of drastic changes or abnormalities, the control strategy is to reduce the maximum power output, activate the temperature protection mechanism, limit the variable response speed, issue an alarm, or enter standby mode.

[0036] The state transition compensation mapping step uses a smoothing algorithm to process the current system state transition. This transition processing is used to handle system oscillations or discontinuities that may occur during the transition.

[0037] Specifically, the state switching compensation mapping step is used to avoid system oscillations, sudden changes or discontinuities during mode switching. When the conditions for triggering mode switching are met, the system will not directly change the control parameters, but will gradually process the current control state through a smooth transition algorithm to ensure the continuity of control commands and the stability of response. Time window filtering or first-order inertial filtering is introduced for key control quantities (such as swashplate commands, motor torque, target displacement, etc.). Interpolation functions (such as cubic splines, sigmoid functions, etc.) are used to gradually map the mode parameters. During the switching process, the controller enters a "buffer" state, delays for a certain period of time (τ), and then fully takes over the control strategy of the new mode.

[0038] A typical form of smoothing process is as follows: u(t)=(1-λ(t))·u old +λ(t)·u new ; in: u(t): Current control command; u old u new : Control quantity for switching between the two modes; λ(t)∈[0,1]: a weight function that changes smoothly over time and satisfies the boundary conditions λ(0)=0,λ(τ)=1; This compensation mapping ensures a smooth transition during the switching process, without causing problems such as pressure shocks, motor jumps, or system instability, and is suitable for various dynamic control environments.

[0039] The smoothing algorithm can use linear interpolation, quadratic interpolation, or higher-order interpolation methods to perform time series interpolation on key control variables, achieving flexible adjustment during the transition period. The smooth expression form of the control command is as follows: u(t)=(1-λ(t))·u old +λ(t)·u new ; The weighting function λ(t) can be set as: Linear interpolation: Quadratic interpolation (ease-in, ease-out): Cubic interpolation or S-curve interpolation: Where τ is the interpolation transition period; The appropriate mode can be flexibly selected based on the different switching scenarios. Linear interpolation is suitable for situations where the load changes little, while quadratic or cubic interpolation is more suitable for high response or protection mode switching, ensuring that variables such as swashplate control, pump displacement, and motor commands are gradually adjusted. Through this smoothing process, the system can smoothly transition between different control modes, avoiding hydraulic shocks or motor overload problems caused by output command jumps.

[0040] Smoothing algorithms include linear interpolation, quadratic interpolation, or higher-order interpolation methods, used to smooth transitions.

[0041] Specifically, to avoid oscillations and discontinuities during control mode switching, this invention employs a smoothing algorithm to process control commands during transition. The smoothing algorithm may include linear interpolation, quadratic interpolation, or higher-order interpolation methods, smoothly adjusting key variables based on time or state weights to achieve a gradual transition of control quantities from the old mode to the new mode, ensuring a stable and smooth switching process and avoiding transient shocks.

[0042] The control strategy employs the optimal control algorithm in each operating mode, minimizing energy consumption and maintaining system stability by adjusting the ratio of servo motor speed to pump displacement in real time.

[0043] Specifically, an optimal control algorithm is introduced in each operating mode. This algorithm takes the current system state as input and calculates the optimal ratio between the servo motor speed and the pump displacement in real time. This ratio is then used as a control command to be issued to the actuator, achieving dynamic optimization of system operation. The optimal control strategy is based on a preset objective function, which typically includes the following two types of optimization objectives: System stability constraints: Keep state variables such as pressure, flow rate, and torque within acceptable ranges, suppress oscillations, and improve the smoothness of control response; Energy minimization objective: Where P(t) represents the instantaneous power of the system, and the ineffective energy consumption is reduced by reducing the redundancy matching between the servo motor output and the pump displacement under non-load conditions.

[0044] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A multi-mode switching control method for a servo motor-driven variable displacement piston pump, characterized in that, Includes the following steps: Collect multidimensional operating status parameters and load power of the variable displacement piston pump system; Dynamic mode decomposition is performed on the load power signal to extract multi-scale modal features representing the system's operating state; Based on the extracted modal features and current state parameters, the current operating mode of the system is determined. If the switching conditions are met, the preset state switching compensation mapping is executed to complete the mode switching; In the new mode, the corresponding control strategy is executed, and the pump displacement and servo motor speed are dynamically adjusted through feedback control; Based on the performance indicators during system operation, the controller parameters are adjusted in real time to monitor and regulate the system's response and energy efficiency.

2. The multi-mode switching control method for a servo motor-driven variable displacement piston pump according to claim 1, characterized in that: The status parameters include pressure, speed, oil temperature, and pump displacement. The load power signal is acquired and calculated through the current sensor and speed sensor of the servo motor. The load power signal represents the actual power output of the servo motor load.

3. The multi-mode switching control method for a servo motor-driven variable displacement piston pump according to claim 1, characterized in that: The load power signal is calculated using the following formula: W = k × I × N; Where: k is the power constant; I is the current signal; N is the rotational speed signal.

4. The multi-mode switching control method for a servo motor-driven variable displacement piston pump according to claim 1, characterized in that: The modal feature extraction step includes performing empirical mode decomposition on the load power signal, decomposing the original power signal into multiple eigenmode functions with local time-varying characteristics and a trend term.

5. The multi-mode switching control method for a servo motor-driven variable displacement piston pump according to claim 4, characterized in that: The intrinsic mode functions are subjected to wavelet packet transform to obtain the energy distribution of each mode component, and the operating mode of the current system is determined based on the changing trend of the energy distribution.

6. The multi-mode switching control method for a servo motor-driven variable displacement piston pump according to claim 1, characterized in that: The criteria for determining mode switching include: the rate of change of modal energy, the rate of change of load power, the rate of change of pump outlet pressure, and the change of oil temperature. A threshold is set to determine whether the system needs to switch modes.

7. The multi-mode switching control method for a servo motor-driven variable displacement piston pump according to claim 1, characterized in that: The switching modes include energy-saving mode, high-response mode, and protection mode. Energy-saving mode corresponds to stable load conditions, high-response mode corresponds to drastic load changes, and protection mode corresponds to abnormal conditions.

8. The multi-mode switching control method for a servo motor-driven variable displacement piston pump according to claim 1, characterized in that: The state switching compensation mapping step uses a smoothing algorithm to process the current system state transition. This transition processing is used to generate system oscillations or discontinuities during the switching process.

9. A multi-mode switching control method for a servo motor-driven variable displacement piston pump according to claim 8, characterized in that: Smoothing algorithms include linear interpolation, quadratic interpolation, or higher-order interpolation methods, used to smooth transitions.

10. A multi-mode switching control method for a servo motor-driven variable displacement piston pump according to claim 1, characterized in that: The control strategy employs the optimal control algorithm in each operating mode, minimizing energy consumption and maintaining system stability by adjusting the ratio of servo motor speed to pump displacement in real time.

Images