A variable displacement and variable frequency speed regulation cooperative control high efficiency energy saving plunger pump system

Abstract

The application relates to the technical field of electromechanical hydraulic control, and discloses a high-efficiency energy-saving plunger pump system with variable displacement and variable frequency speed regulation cooperative control, which comprises a multi-dimensional sensing unit, an electrical driving unit, an electromechanical hydraulic executing unit and an integrated controller for driving and control. A main control logic unit synchronously obtains pressure and oil temperature based on a homologous memory space, calculates a dynamic volume elastic modulus, obtains optimal motor rotating speed and optimal swash plate inclination angle through minimum energy functional solution, superimposes a torque differential signal to the optimal swash plate inclination angle, converts the dynamic volume elastic modulus into an equivalent dynamic stiffness coefficient and injects the equivalent dynamic stiffness coefficient into an excitation current control node, generates an asymmetric constraint boundary according to a power flow direction, adopts a spline function to interpolate a trajectory when the swash plate crosses a dead zone, injects a torque current for keeping the trajectory to generate a comprehensive instruction, sends the comprehensive instruction into the asymmetric constraint boundary to execute trajectory filtering and amplitude limiting clamping, drives a four-quadrant inverter main loop and an electro-hydraulic proportional servo valve, and transient impact of a physical transmission shaft system is inhibited.

Description

Technical Field

[0001] This invention relates to the field of electromechanical-hydraulic control technology, specifically to a high-efficiency and energy-saving plunger pump system with coordinated control of variable displacement and variable frequency speed regulation. Background Technology

[0002] The combined system of variable frequency motor and variable displacement axial piston pump has been applied in industrial transmission. In existing drive and control architectures, electrical drives and hydraulic controllers typically exist as independent hardware nodes. This separate hardware architecture relies on an external communication bus for data interaction, resulting in inherent data transmission delays. When facing transient load conditions, the controller struggles to achieve strict timing alignment between the actual physical quantities in the hydraulic domain and the underlying state variables in the electrical domain, leading to physical lag in system command response and affecting the overall dynamic control accuracy of the machine.

[0003] Meanwhile, existing control strategies often treat the speed regulation of the variable frequency motor and the displacement regulation of the piston pump as two independent closed loops. This approach ignores the nonlinear variation of the dynamic bulk modulus of the fluid under different temperatures and pressures. Because the dynamic stiffness parameters of the hydraulic system are not incorporated into the control loop in the electrical domain, the electromagnetic impedance output by the motor cannot adaptively match the actual mechanical load stiffness. When pipeline pressure or external load changes abruptly, the mechanical transmission shaft system is prone to torsional oscillations and displacement overshoot, increasing not only the system's steady-state energy consumption but also mechanical fatigue losses.

[0004] Furthermore, traditional control systems lack dynamic boundary protection logic for cross-domain power flow when facing extreme impact conditions. When the system enters the generation quadrant under reverse load, unrestricted trajectory jumps can trigger transient energy backlash at the physical actuators. Particularly during the plunger pump swashplate's crossing of the zero-point dead zone, the gaps and friction between the mechanical transmission pairs make it difficult for conventional linear adjustment commands to maintain a tight fit between the transmission components. This can lead to slight vibrations and hardware shocks in the actuators at zero points, reducing the system's operational stability in the nonlinear operating range. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a high-efficiency and energy-saving plunger pump system with coordinated control of variable displacement and variable frequency speed regulation. It aims to solve the problems of timing lag in multi-domain data transmission of electromechanical and hydraulic systems, physical stiffness mismatch and mechanical shaft torsional vibration caused by independent control architecture, and impact and vibration of actuators caused by lack of dynamic boundary protection during extreme impact conditions and zero-position dead zone crossing.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solution: a high-efficiency and energy-saving plunger pump system with variable displacement and variable frequency speed regulation coordinated control, comprising a multi-dimensional sensing unit, an electrical drive unit with a four-quadrant inverter main circuit, an electromechanical-hydraulic actuator unit with a variable frequency motor, an electro-hydraulic proportional servo valve and a swashplate, and an integrated drive and control controller with an integrated main control logic unit.

[0007] The main control logic unit synchronously acquires the actual working pressure and real-time oil temperature based on the same memory space, calculates the dynamic bulk elastic modulus, and then obtains the optimal motor speed and optimal swashplate tilt angle through minimum energy functional solution. The main control logic unit superimposes the extracted torque differential signal onto the optimal swashplate tilt angle and converts the dynamic bulk elastic modulus into an equivalent dynamic stiffness coefficient, which is then injected into the excitation current control node. The main control logic unit dynamically generates asymmetric constraint boundaries based on the torque differential polarity and active power flow direction. When the swashplate crosses the zero tilt dead zone, it uses spline function interpolation to interpolate the trajectory and forcibly injects transient holding torque current to generate comprehensive commands. The main control logic unit sends the integrated instructions to the asymmetric constraint boundary for trajectory filtering and amplitude limiting clamping, and finally converts them into digital pulse sequences and drive current instructions to control the four-quadrant inverter main circuit and the electro-hydraulic proportional servo valve.

[0008] Preferably, the main control logic unit extracts discrete digital quantities and performs smoothing using a digital filtering algorithm; it bypasses the external bus by using a direct memory mapping method and directly reads the underlying state variables synchronously from the shared register address segment; the operation of reading the shared register address segment is triggered by the rising edge of the global hardware clock, so that the electrical domain state variables can maintain physical timing alignment with the hydraulic domain acquired data.

[0009] In one specific embodiment, the main control logic unit uses the calculated dynamic bulk elastic modulus as a dynamic correction factor to update the preset volumetric efficiency matrix in real time; before performing matrix element traversal operations, it checks whether the dynamic bulk elastic modulus is greater than a set minimum stiffness safety threshold. When the dynamic bulk elastic modulus is lower than the minimum stiffness safety threshold, the dynamic bulk elastic modulus is clamped to the minimum stiffness safety threshold; wherein, the minimum stiffness safety threshold is defined as a safety lower limit parameter set to avoid division overflow anomalies under high temperature and low pressure or sensor disconnection fault conditions.

[0010] In a further embodiment, the main control logic unit uses the target pressure and target flow rate as external rigid boundary constraints, and solves for the extreme points of the minimum energy functional through a gradient descent iterative algorithm. In each iteration, it continuously calculates the gradient vector formed by the partial derivatives of the objective function with respect to the optimal motor speed and the optimal swashplate angle. When the L2 norm of the gradient vector is less than the set convergence tolerance, it determines that the functional optimization has converged, and outputs the optimal motor speed and the optimal swashplate angle in steady state.

[0011] Preferably, the main control logic unit inputs the real-time read electromagnetic torque estimate into a preset nonlinear tracking differentiator to extract the torque differential signal; multiplies the torque differential signal by a set hydraulic compensation gain to generate a transient tilt angle compensation amount, and performs direct feedforward superposition processing of the transient tilt angle compensation amount and the optimal swashplate tilt angle in the same memory space to drive the valve core of the electro-hydraulic proportional servo valve to generate a displacement action.

[0012] In one specific embodiment, the main control logic unit uses the mechanical impedance transformation relationship to map the dynamic bulk elastic modulus to the equivalent dynamic stiffness coefficient of the hydraulic side relative to the shaft system; it uses the equivalent dynamic stiffness coefficient to generate the excitation current adjustment amount and calculates the comprehensive direct-axis current command, and writes the comprehensive direct-axis current command into the excitation current control node where the field-oriented control algorithm is located through the internal bus matrix.

[0013] In a further embodiment, the main control logic unit constructs an equal-power physical envelope model in the same memory space, unifies the physical dimensions of the mechanical rotor kinetic energy increment caused by the frequency command change and the fluid hydraulic pressure increment caused by the swashplate tilt angle change, and constrains the system's comprehensive transient power perturbation estimate to be less than or equal to the equal-power dissipation boundary threshold set by the system; wherein, the equal-power dissipation boundary threshold is defined as the maximum instantaneous overload power nominal value allowed by the four-quadrant inverter main circuit hardware when the system is in the electric quadrant.

[0014] Preferably, when the system enters the power generation quadrant, the main control logic unit expands the allowable frequency drop boundary proportionally to the degree of change in negative torque mutation caused by the torque differential signal; simultaneously, it forcibly takes over the response trajectory of the electromechanical-hydraulic actuator in the instruction domain and switches it to a nonlinear cutoff curve to reduce the tilt angle of the swashplate at a speed positively correlated with the torque mutation rate.

[0015] In one specific embodiment, the main control logic unit monitors the integrated swashplate tilt command within the integrated instruction in real time. Only when it is determined that the integrated swashplate tilt command is less than or equal to the mechanical zero-bias dead zone tolerance and the extracted operating frequency is greater than the set minimum frequency threshold for maintaining mechanical engagement, the spline function smooth reconstruction within the instruction domain is activated. The minimum frequency threshold for maintaining mechanical engagement is defined as a constant with a value greater than zero, determined through physical experiments based on the inertial bias requirements of the rotating components.

[0016] In a further embodiment, when the main control logic unit uses space vector pulse width modulation technology to generate physical control signals, it sets hardware-level undervoltage protection for the denominator of the real-time DC bus voltage measurement value. When the real-time DC bus voltage measurement value drops to near the set minimum safe voltage threshold, the calculation denominator is forcibly clamped. A digital current regulator is set in the hydraulic actuation domain and a high-frequency low-amplitude flutter signal is actively superimposed to output the drive current command. The minimum safe voltage threshold is defined as the undervoltage protection action critical point of the inverter bus capacitor.

[0017] This invention provides a high-efficiency and energy-saving plunger pump system with coordinated control of variable displacement and variable frequency speed regulation. It has the following beneficial effects: 1. This invention synchronously acquires the underlying state variables of the hydraulic and electrical domains through the same memory space of the integrated drive and control controller, and triggers direct read and write operations of the memory using the rising edge of the global hardware clock. This hardware architecture and extraction method eliminates the data transmission lag caused by cross-physical device external bus communication, realizes the physical timing alignment of multi-dimensional physical quantities of electromechanical and hydraulic systems, and improves the system's instruction response speed and data synchronization accuracy when facing load transients.

[0018] 2. This invention transforms the dynamic bulk elastic modulus calculated based on the fluid compression derived model into an equivalent dynamic stiffness coefficient, which is then directly injected into the excitation current control node of the frequency converter. At the same time, it combines the minimum energy functional model to calculate the optimal motor speed and optimal swashplate tilt angle under theoretical steady state. While reducing the overall steady-state energy consumption of the system, it forces the physical stiffness change of the hydraulic side to be aligned with the electromagnetic impedance of the electrical domain, thereby suppressing the torsional vibration and displacement overshoot phenomenon that is prone to occur in the mechanical transmission shaft system under sudden load conditions from the physical channel.

[0019] 3. This invention uses dynamically generated asymmetric constraint boundaries to filter the comprehensive command execution trajectory and clamp the highest rate of change. When determining that the swashplate is crossing the zero-tilt dead zone, it employs spline function interpolation of the motion trajectory and forcibly injects transient holding torque current. This control logic limits the transient energy recoil amplitude when the system is forced into the generator quadrant, maintains the physical contact state of the mechanical transmission pair teeth within the zero-position dead zone, and avoids actuator jitter and hardware shock caused by command abrupt changes or dead zone friction. Attached Figure Description

[0020] Figure 1 This is a diagram of the overall system architecture of the present invention; Figure 2 This is a flowchart illustrating the overall control method of the present invention. Figure 3 The following is a comparative line graph of a specific embodiment of the present invention, wherein subgraph (A) is a waveform diagram of the system dynamic response and subgraph (B) is a diagram of the mechanical impact evolution process. Detailed Implementation

[0021] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, 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.

[0022] See attached document Figure 1 This invention provides a high-efficiency and energy-saving plunger pump system with coordinated control of variable displacement and variable frequency speed regulation, including an integrated drive and control controller, an electrical drive unit, an electromechanical-hydraulic actuator unit, and a multi-dimensional sensing unit.

[0023] The integrated drive and control controller is configured with a multi-core system-on-a-chip (SoC) hardware architecture, which integrates a digital signal processor (DSP) core and a field-programmable gate array (FPGA). The DSP core and FPGA share the same memory space at the physical level, forming a homogeneous memory architecture. The system's hydraulic calculation logic module, proportional valve control loop module, and motor field-oriented control module all reside in this homogeneous memory space. These modules directly perform register-level data read / write and interaction through the chip's internal bus matrix.

[0024] The electrical drive unit includes a four-quadrant inverter main circuit. An active front-end module is installed at the input of the four-quadrant inverter main circuit, which is connected to the external AC power grid. The inverter output of the four-quadrant inverter main circuit is electrically connected to the stator winding of the variable frequency motor. The active front-end module supports bidirectional power flow, and when the variable frequency motor is operating in the generating quadrant, it inverts the electrical energy on the DC bus and feeds it back to the external AC power grid.

[0025] The electromechanical-hydraulic actuator includes a variable frequency motor, a bidirectional swashplate axial piston pump, and an electro-hydraulic proportional servo valve. The power output shaft of the variable frequency motor is mechanically connected to the power input shaft of the bidirectional swashplate axial piston pump. The bidirectional swashplate axial piston pump employs a symmetrical distribution plate structure internally. The electro-hydraulic proportional servo valve is installed on the variable displacement control mechanism side of the bidirectional swashplate axial piston pump. The signal receiving end of the electro-hydraulic proportional servo valve is connected to the digital pulse width modulation signal output channel of the integrated drive and control controller.

[0026] The multi-dimensional sensing unit includes a pressure sensor and a temperature sensor. The pressure sensor is located on the main hydraulic output pipeline of the bidirectional swashplate axial piston pump. The temperature sensor is located inside the system's hydraulic oil tank. The signal output pins of the pressure and temperature sensors are directly connected to the high-speed analog-to-digital converter (ADC) interface of the integrated drive and control controller via physical hardwiring. The integrated drive and control controller acquires external physical environment data through this high-speed ADC interface and sends it to its shared memory space.

[0027] See attached document Figure 2 This invention provides a control method for a high-efficiency and energy-saving plunger pump system with coordinated control of variable displacement and variable frequency speed regulation, comprising the following steps: S1, Multi-dimensional physical quantity synchronous acquisition: Based on the same memory space of the integrated drive and control controller, synchronously acquire the actual working pressure and real-time oil temperature of the hydraulic domain, the low-level state variables of the electrical domain, and the target pressure and target flow data of the command domain. S2, Fluid Thermodynamic Mapping and Steady-State Benchmark Optimization: Substitute the actual working pressure and real-time oil temperature into the fluid compression derived model to calculate the dynamic bulk modulus and update the volumetric efficiency matrix. Obtain the theoretical steady-state optimal motor speed and optimal swashplate tilt angle through minimum energy functional solution. S3, Cross-domain bidirectional nonlinear decoupling and noise-resistant feedforward compensation: The extracted smooth electromagnetic torque differential signal is superimposed on the optimal swashplate tilt angle as a feedforward prior signal for advance compensation, and the dynamic bulk elastic modulus is converted into an equivalent dynamic stiffness coefficient and injected into the frequency converter excitation current control node for damping adaptive adjustment. S4, Four-quadrant asymmetric envelope reconstruction and zero-crossing dead zone smoothing: Based on the torque differential polarity and power flow direction, the frequency and swashplate tilt angle are dynamically generated as asymmetric constraint boundaries. When the swashplate crosses the zero-tilt dead zone, spline function interpolation is used to interpolate the trajectory and a holding torque current is forcibly injected. S5, Hardware Topology Driven Execution and Boundary Clamping: The integrated command generated by cross-domain compensation and adaptive adjustment is sent to the asymmetric constraint boundary for trajectory filtering and clamping limitation, and the digital drive pulse is output to the four-quadrant inverter main circuit and electro-hydraulic proportional servo valve.

[0028] The above steps will be described in detail below with reference to specific embodiments and accompanying drawings.

[0029] The multi-dimensional physical quantity synchronous acquisition mechanism relies on the underlying hardware architecture of the aforementioned integrated drive and control controller. It unifies discrete signals from different physical domains into a common memory space, providing data support for subsequent nonlinear decoupling. Step S1 specifically includes the following sub-steps: S101 obtains the actual working pressure and real-time oil temperature of the hydraulic domain through a hardware interface.

[0030] The selection of actual working pressure and real-time oil temperature as input parameters is based on the physical causal relationship in fluid thermodynamics where these two directly determine the dynamic bulk modulus of hydraulic oil. As a preferred approach, the system utilizes a pressure sensor installed on the main pipeline and a temperature sensor installed in the oil tank to sense changes in the physical environment and output analog voltage signals. These analog voltage signals are directly transmitted to the on-chip high-speed analog-to-digital converter (ADC) of the integrated drive and control controller via a hard-wired physical path. The on-chip ADC performs oversampling within a set control cycle, converting the continuous analog voltage signal into a discrete digital quantity. Based on the general signal processing principle that industrial field signals are often accompanied by inverter pulse-width modulation harmonics and high-frequency white noise from mechanical oscillations, to eliminate high-frequency electromagnetic interference in industrial environments, the digital signal processor core inside the integrated drive and control controller extracts the discrete digital quantity from the data buffer register of the high-speed ADC and smooths it using a digital filtering algorithm, outputting the actual working pressure for the current control cycle. With real-time oil temperature Specifically, the digital filtering algorithm uses the following moving average filtering formula: ; In the formula, For the first The filtered output value after each sampling period corresponds to the actual working pressure. Or real-time oil temperature , The length of the sliding window. For the first The original analog-to-digital conversion sampled values ​​for each sampling period. The physical meaning of the above filtering process is to eliminate transient glitches and interference picked up by the sensor circuitry, ensuring the steady-state convergence of subsequent thermodynamic parameter calculations.

[0031] For the wiring topology of the aforementioned analog-to-digital conversion peripheral hardware circuit and the conventional writing of the moving average filter code, those skilled in the art can refer to the basic design specifications of digital signal processing. The specific process is a well-known technology in this field and will not be elaborated here.

[0032] S102 directly extracts the underlying state variables of the electrical domain through a shared register. Conventional distributed control architectures rely on industrial fieldbuses for cross-device data interaction, inevitably incurring protocol packet conversion overhead and transmission phase delay. This step bypasses external bus protocols based on the shared memory architecture of the integrated drive and control controller. The integrated drive and control controller internally runs a field-oriented control algorithm module. When driving the variable frequency motor, this module calculates the internal electromagnetic state of the motor in real time and continuously refreshes and stores the calculation results in a set on-chip shared register address segment. Based on the above memory sharing mechanism, the main control logic unit adopts a direct memory mapping method to bypass the external bus and directly read the estimated electromagnetic torque value of the current cycle synchronously from the shared register address segment. Stator d-axis current components Stator q-axis current component Operating frequency and the direction of active power flow The d-axis is the linear axis, and the q-axis is the horizontal axis.

[0033] The aforementioned electrical parameters were selected for synchronous extraction because they directly reflect the degree of load abrupt changes and dynamic damping characteristics of the electromechanical coupling system. Active power flow is used to characterize the transient power feedback state of the variable frequency motor. To ensure the alignment of operating conditions for multi-source data, the main control logic unit's operation of reading shared registers is strictly triggered by the rising edge of the global hardware clock. Memory read / write operations are directly executed through the chip's internal bus matrix, eliminating data lag caused by cross-physical device communication and ensuring strict physical timing alignment between electrical domain state variables and hydraulic domain acquired data.

[0034] S103 synchronously reads and latches target instruction data through the instruction domain interface. The integrated drive and control controller receives work instructions from the host computer using the set industrial communication interface. The system is equipped with a global hardware timer. At the moment the global hardware timer interrupt is triggered, the main control logic unit latches the work instructions issued by the host computer into the instruction control block area in the same memory space and extracts them as the target pressure. With target traffic The technical purpose of this latching mechanism is to prevent target data tearing or overwriting anomalies caused by asynchronous issuance of new instructions from the host computer within a single control cycle. Thus, the system will filter the actual workload... With real-time oil temperature Directly read electromagnetic torque estimation value and various underlying electrical parameters, as well as target pressure. With target traffic By relying on the timestamp of the global hardware timer to be uniformly fixed on the same time section of the current control cycle, the same source synchronous acquisition of the three-dimensional physical variables is completed, providing a time-difference-free input data source for the next stage of thermodynamic mapping.

[0035] Based on the input data source determined in step S1, the system performs fluid thermodynamic mapping and steady-state benchmark optimization. The essence of this process is to eliminate the physical property shifts in the fluid caused by temperature fluctuations and high-pressure compression, thereby constructing a realistic dynamic load model at the system control level and providing benchmark parameters for cross-domain decoupling. Step S2 specifically includes the following sub-steps: 201. Calculation of dynamic bulk modulus based on a fluid compression derived model. The selection of actual working pressure and real-time oil temperature as input parameters for the derived model is based on the physical causal relationship in fluid thermodynamics where ambient temperature and working pressure jointly determine the spacing between micro-molecules within the fluid, thus directly altering the fluid's macroscopic compressibility. The integrated drive and control controller has a pre-built fluid compression derived model, and the main control logic unit uses the aforementioned filtered actual working pressure... With real-time oil temperature The fluid compression derived model is substituted into the equation for nonlinear solution. Based on the general physical principle of fluid high-pressure stiffness enhancement and high-temperature stiffness weakening, the specific calculation uses the following dynamic stiffness calculation formula: ; In the formula, For dynamic bulk modulus, To reference the initial fluid bulk modulus under ambient temperature and standard atmospheric pressure, This is the positive pressure influence coefficient. It is a negative temperature influence coefficient. The set reference ambient temperature. The values ​​of and are both real numbers greater than 0, and their specific values ​​are determined through curve fitting from multidimensional isothermal and isobaric compression physical calibration experiments.

[0036] The physical meaning of this dynamic stiffness calculation formula lies in quantifying the true state of the fluid: an increase in pressure leads to an exponential increase in fluid stiffness, while an increase in temperature causes thermal expansion, resulting in a decrease in stiffness. This formula extracts characteristic parameters reflecting the fluid's true compressive strength, establishing a physical basis for accurate prediction of the overall mechanical efficiency of the system.

[0037] S202, updating the volumetric efficiency matrix based on the dynamic bulk modulus. Conventional plunger pump control largely assumes that volumetric efficiency is a two-dimensional static lookup table that varies with speed and pressure, neglecting the significant flow loss caused by fluid compression under high-pressure conditions. In this embodiment, the system uses the calculated dynamic bulk modulus as a dynamic correction factor to update the preset volumetric efficiency matrix in real time. As a preferred method, the volumetric efficiency matrix is ​​reconstructed using the following matrix element correction formula: ; In the formula, For the updated volumetric efficiency matrix, the first... Line 1 The elements of the column correspond to the current overall volumetric efficiency, where Index for working pressure nodes, For the motor speed node index; These are the initial elements at the corresponding positions in the offline calibrated volumetric efficiency matrix. The structural constant related to the dead zone volume of the plunger pump cavity is set to a range of (0, 1), with the specific value depending on the overlap angle of the pump body's distribution plate and the physical dimensions of the plunger cavity. The technical purpose of this matrix element correction formula is to make explicit mathematical compensation for the implicit flow loss that cannot be effectively output due to the compression of the fluid volume in the high-pressure closed cavity.

[0038] To ensure the completeness of the matrix update algorithm and the stability of the control system, the underlying software of the main control logic unit includes division-by-zero checks and singularity protection logic before performing the matrix element traversal operation. Specifically, this involves a mandatory check of the denominator terms. Is it greater than the set minimum stiffness safety threshold? When the stiffness falls below the minimum stiffness safety threshold, it is forcibly clamped to that threshold to prevent division overflow anomalies under extreme high temperature and low pressure conditions or sensor disconnection failures. The minimum stiffness safety threshold is defined as the lower limit parameter set to prevent division overflow anomalies under extreme high temperature and low pressure conditions or sensor disconnection failures. Furthermore, to prevent out-of-bounds lookup matrix errors that could cause system-level memory address overflows, the main control logic unit indexes the working pressure nodes. Motor speed node index All of them have maximum addressing space constraint logic.

[0039] S203, based on the minimum energy functional theory, calculates the steady-state optimal motor speed and optimal swashplate angle. After obtaining the accurate real-time volumetric efficiency matrix, the system enters the steady-state optimization phase. The optimization goal is to find the combined operating point of the variable frequency motor and piston pump while satisfying the target request in the instruction domain, thereby minimizing the total global energy consumption of the drive system. Based on the principles of energy conservation and electromechanical-hydraulic coupling dissipation, the main control logic unit constructs a minimum energy functional model in the same memory. The control program then uses the aforementioned target pressure... With target traffic As external rigid boundary constraints, the extreme points of this minimum energy functional model are solved using a gradient descent iterative algorithm. Specifically, the objective function formula of the minimum energy functional model is set as follows: ; In the formula, To characterize the functional evaluation value of the total power loss of the system, To find the optimal speed for the independent variable motor, Find the optimal tilt angle for the swashplate as the independent variable. Based on the aforementioned dynamic bulk modulus, the dynamic comprehensive volumetric efficiency... This is a function of mechanical loss power that includes the copper and iron losses of the variable frequency motor and the mechanical friction of the plunger pump shaft system. and The weighting coefficient is a dimensionless coefficient, and the sum of the values ​​of the weighting coefficients is equal to 1. The allocation ratio is set according to the rated power level of the system.

[0040] To ensure the completeness of the algorithmic logic and prevent arithmetic divergence caused by extremely low volumetric efficiency during functional solution, the system optimizes the division denominator terms in the objective function formula. A mandatory lower limit constraint has been added. If the obtained efficiency value is lower than the set limit shutdown efficiency threshold, it will be directly clamped to the limit threshold and the warning status will be suspended simultaneously.

[0041] The reason for choosing the objective function of the minimum energy functional is that it comprehensively encompasses the output power loss on the hydraulic side affected by compressibility and the physical friction loss on the electromechanical side. After multiple rounds of iterative optimization until the gradient of the objective function changes close to zero, the system output makes... Minimize the corresponding combination of independent variables, i.e., the optimal motor speed under the theoretical steady state in the current physical condition. With optimal swashplate angle .

[0042] To avoid getting stuck in an infinite loop or failing to converge, the main control logic unit incorporates a dynamic optimization step size parameter and a gradient convergence tolerance for the aforementioned gradient descent iterative algorithm. Specifically, the control program continuously calculates the objective function relative to the given gradient in each iteration. and The gradient vector is formed by the partial derivatives of the function. When the L2 norm of the gradient vector is less than the set convergence tolerance, the functional optimization is determined to be converged and the iteration is stopped. If the iterative operation reaches the set maximum number of iterations boundary and still has not converged, the solution is forcibly interrupted and the optimization result of the previous control cycle is output.

[0043] Obtain the optimal motor speed under the aforementioned theoretical steady state With optimal swashplate angle Subsequently, the system enters the cross-domain bidirectional nonlinear decoupling and noise-resistant feedforward compensation stage based on this steady-state reference parameter. Conventional electromechanical-hydraulic coupling systems exhibit a significant time-scale difference due to the fast response of electrical components and the slow response of hydraulic components. Directly executing commands according to the steady-state reference can easily lead to dynamic instability during sudden load changes. Based on the objective physical law of the aforementioned response time difference, this embodiment breaks down the physical domain barrier through the interaction of the same-source memory of the integrated drive and control controller. Step S3 specifically includes the following sub-steps: S301 extracts the torque differential signal based on a nonlinear tracking differentiator. During system operation, the high-frequency switching action of the insulated-gate bipolar transistor in the inverter of the variable frequency motor inevitably generates pulse-width modulation harmonics in the electrical domain. Based on the objective physical conditions of this industrial site, the estimated electromagnetic torque value is directly and synchronously extracted from the shared register. It inevitably contains a large amount of high-frequency measurement noise, which is detrimental to... When performing direct differential calculations, the amplitude of the high-frequency measurement noise contained therein is drastically amplified, generating huge numerical spike pulses. Once these numerical spike pulses are introduced into the control loop as feedforward compensation, they will cause the valve core of the electro-hydraulic proportional servo valve to produce severe high-frequency chatter, making it impossible for the hydraulic system to stably build up pressure.

[0044] To isolate the aforementioned pulse width modulation high-frequency harmonic noise and obtain an accurate dynamic rate of torque change, the main control logic unit reads the estimated electromagnetic torque value from the shared register in real time. The input is fed into a preset nonlinear tracking differentiator. Based on the general principle in classical filtering theory that phase and amplitude attenuation cannot be simultaneously achieved, the nonlinear tracking differentiator is chosen instead of a traditional linear low-pass filter because a traditional low-pass filter inevitably introduces significant system phase lag while smoothing noise, and this phase lag completely cancels out the advance suppression physical effect required for feedforward compensation.

[0045] As a preferred approach, the specific difference equation structure of this nonlinear tracking differentiator in the discrete domain adopts the following discrete tracking differential formula: ; In the formula, and Representing the first The and the first Smooth torque tracking signal for each control cycle, and Representing the first The and the first The torque differential signal for each control cycle. The discrete integration step size is strictly aligned with the interrupt cycle of the global hardware clock to ensure strict synchronous computation of multi-source data in terms of timing. For the first The estimated value of the electromagnetic torque input in each control cycle. For acceleration factor, The filter factor that determines the noise suppression strength This is the fastest control synthesis function.

[0046] For the fastest control synthesis function The specific internal mathematical expressions and computer discrete code writing can be implemented by referring to the basic literature on nonlinear active disturbance rejection control theory. The process of denoising calculation is a well-known technology in this field and will not be described in detail here.

[0047] To ensure the integrity of the algorithm logic and the security against out-of-bounds errors, an acceleration factor is used. The value range is set to real numbers greater than 0. Its specific value is physically calibrated based on the rated maximum electromagnetic torque ramp-up rate of the variable frequency motor, and is used to determine the highest output amplitude of the rigid constraint torque differential signal. Filtering factor. The value is set to be greater than or equal to the discrete integral step size. The value of the error must be an integer multiple of the error value. A larger value indicates a stronger ability to smooth high-frequency harmonics, but the tracking speed will suffer a corresponding nonlinear degradation. Furthermore, to prevent the internal steepest descent function from generating singularities such as division by zero or high-frequency chattering when handling extremely small errors, the algorithm has a static dead-zone tolerance parameter. When the absolute value of the tracking error... When the tolerance is less than the dead zone, force to The output item is clamped to zero.

[0048] The technical purpose and physical meaning of the aforementioned discrete tracking differential formula is that the output state approximates the estimated input electromagnetic torque in a mathematical manner with a nonlinear, steepest descent. While ensuring that the smooth torque tracking signal can reproduce the true low-frequency torque profile without static error and significant phase lag, the precise time derivative of this torque profile is synchronously output from the torque differential signal. Through the aforementioned underlying data processing, a smooth torque differential signal with a high signal-to-noise ratio is successfully extracted without sacrificing the real-time dynamic response, thus providing a basis for subsequent analysis of the optimal swashplate angle. The generation of advance compensation control variables establishes a highly reliable feedforward data source.

[0049] S302 generates transient tilt angle compensation based on the torque differential signal and performs feedforward advance compensation. After successfully extracting the torque differential signal, it proceeds to the calculation stage of the forward electro-hydraulic decoupling channel. Conventional closed-loop control architectures mostly rely on the error feedback from pipeline pressure sensors for adjustment. Due to the objective physical inertia of fluid volume compression and mechanical swashplate drive, the response speed of the hydraulic actuator lags significantly behind the load change, leading to transient pressure loss and slow response when the system faces sudden load increases. Based on the objective law of response time difference in different physical domains, this embodiment breaks the limitations of traditional intra-domain closed loops and constructs a cross-domain feedforward intervention mechanism.

[0050] The reason for choosing the torque differential signal as the feedforward input is that the transient change of the electromagnetic torque of the variable frequency motor can reflect the instantaneous load fluctuation at the mechanical shaft end without any physical delay, and its sensitivity and timeliness are far superior to those of the pressure sensor at the end of the hydraulic line. The system extracts the first differential signal from step S301 above. Torque differential signal for each control cycle As a feedforward prior signal, multiplied by a set hydraulic compensation gain, the transient tilt angle compensation amount for feedforward intervention is generated. The specific calculation uses the following transient compensation formula: ; In the formula, For the first Transient tilt angle compensation amount per control cycle For the first The torque differential signal for each control cycle This is the cross-domain hydraulic compensation gain. The value of this cross-domain hydraulic compensation gain is a real number greater than 0.

[0051] To avoid overcompensation or undercompensation traps under extreme operating conditions caused by relying on a single fixed gain, the main control logic unit constructs a two-dimensional weight mapping table based on the current actual working pressure and the operating frequency of the variable frequency motor, and adjusts the weight within a set range according to the real-time operating conditions. The gain of the cross-domain hydraulic compensation is determined by internal dynamic interpolation optimization.

[0052] After obtaining the transient tilt compensation, the main control logic unit compares it with the optimal swashplate tilt angle in the same memory. Direct feedforward superposition is performed. The system generates the composite swashplate tilt command using the following feedforward superposition formula: ; In the formula, For the first The final output of the combined swashplate tilt command in each control cycle This is the optimal swashplate tilt angle based on the steady-state optimization output of the minimum energy functional theory. The technical purpose of the above-mentioned positive electro-hydraulic logic is to drive the valve core of the electro-hydraulic proportional servo valve to produce a displacement action in advance before the actual working pressure of the pipeline drops significantly or overshoots. It utilizes the high-frequency rapid sensing capability of the electrical domain to compensate for the low-frequency execution lag of the hydraulic components, thereby establishing dynamic alignment on the cross-domain physical time axis and achieving proactive suppression of system load pulsation and pressure fluctuations.

[0053] Based on the completeness of the algorithm logic and the safety protection requirements of the mechanical mechanism, in order to prevent excessive transient tilt angle compensation caused by drastic load changes, which could lead to excessive mechanical impact on the swashplate, the main control logic unit outputs a comprehensive swashplate tilt angle command downstream. Previously, a hard physical clamping logic was added. Specifically, the system enforces verification. Is it located in Within the absolutely closed interval. and These represent the minimum and maximum allowable reverse tilt angle extremes of the bidirectional swashplate axial piston pump on the mechanical structure limit stop, respectively. If the feedforward superposition calculation result exceeds either of these boundary limits, the system directly saturates and clamps the comprehensive swashplate tilt angle command to the corresponding boundary extreme. This limiting logic ensures from the algorithm's underlying layer that the cross-domain feedforward intervention mechanism is always strongly constrained within a safe and controllable mechanical physical tolerance range, avoiding the introduction of control noise that could trigger secondary mechanical shocks.

[0054] S303 maps the dynamic bulk modulus of elasticity to an equivalent dynamic stiffness coefficient and injects it into the excitation current control node. After completing the feedforward superposition of the aforementioned forward electro-hydraulic channel, the system synchronously executes the decoupling operation of the reverse hydraulic-electric channel. In conventional electromechanical-hydraulic coupling systems, under high-pressure and heavy-load conditions, changes in the physical state of the hydraulic oil can cause a nonlinear shift in the overall mechanical resonant frequency of the system. If the electrical damping on the variable frequency motor side remains constant, the stiffness mismatch between the machine side and the hydraulic side can easily trigger severe torsional vibration of the power transmission shaft system. Based on the general dynamic principle of impedance matching, maintaining dynamic stiffness matching across the entire electromechanical-hydraulic system is a physical prerequisite for eliminating shaft torsional vibration. This embodiment constructs a reverse hydraulic-electric adaptive adjustment mechanism through a homogeneous memory architecture.

[0055] Select the dynamic bulk elastic modulus of the current control cycle obtained from the aforementioned step S201. The input parameter for reverse intervention is because the change in the fluid's own compressibility directly constitutes the source of the alternating load stiffness of the piston pump's mechanical main bearing. As a preferred approach, the main control logic unit extracts this dynamic bulk elastic modulus from the same memory source. The mechanical impedance transformation relationship is used to map it into the equivalent dynamic stiffness coefficient of the hydraulic side relative to the shaft system. The specific calculation uses the following equivalent stiffness mapping formula: ; In the formula, The equivalent dynamic stiffness coefficient obtained by mapping, For dynamic bulk modulus, The mechanical coupling conversion coefficient is determined based on the effective displacement and mechanical transmission ratio of the bidirectional swashplate axial piston pump. Its value is greater than 0 and depends on the pump body's mechanical geometric parameters. This is the reference mechanical damping constant formed by the bearings and seals when the system is in a zero-pressure, no-load state.

[0056] The technical purpose of the aforementioned stiffness equivalent mapping formula is to transform simple fluid thermodynamic variables into mechanical dynamic parameters that can be directly correlated with the electromagnetic model. After obtaining the equivalent dynamic stiffness coefficient, the system initiates adaptive adjustment of electrical damping. To avoid the system relying solely on a single stiffness extreme value for adjustment, which could lead to erroneous damping injection during unsteady cold starts or sensor malfunctions, the main control logic unit is equipped with multi-dimensional enable logic. The system will comprehensively evaluate whether the real-time oil temperature has reached the steady-state threshold and whether the active power flow is stable. Only when all multi-dimensional boundary conditions are met will the following damping injection operation be permitted.

[0057] The dynamic electromagnetic damping characteristics of a variable frequency motor are essentially controlled by the magnetic flux linkage on the stator d-axis, i.e., by the magnitude of the excitation current. To enable the motor's electrical damping to adaptively track the high-frequency fluctuations of fluid rigidity, the main control logic unit uses the calculated equivalent dynamic stiffness coefficient to generate the excitation current adjustment and calculates the comprehensive direct-axis current command. Specifically, the following adaptive damping injection formula is used: ; In the formula, For the first The combined direct-axis current command for each control cycle. This is the reference excitation current request value calculated by the underlying layer of the field-oriented control algorithm to maintain the current rotational speed. For damping current conversion gain, This is the equivalent dynamic stiffness coefficient. This is the reference nominal stiffness calibrated offline under the system's rated operating conditions. The value range is a real number greater than 0, and its specific value is determined by physical calibration based on the magnetization curve and torque constant of the motor.

[0058] The physical meaning of the above adaptive damping injection formula is that when the fluid stiffness increases sharply due to a sudden high pressure in the system (i.e., When the fluid stiffness drops, the system synchronously increases the excitation current component of the d-axis to strengthen the electromagnetic coupling stiffness between the motor stator and rotor; conversely, when the fluid stiffness drops, the excitation current is correspondingly weakened. Based on the homogeneous memory architecture of the integrated drive and control controller, the main control logic unit, through an internal bus matrix, within one instruction control cycle, processes the aforementioned comprehensive direct-axis current command. It is directly written into the excitation current control node register where the field-oriented control algorithm is located. This mechanism forces the electromagnetic impedance to be aligned with the physical stiffness of the hydraulic system, absorbing and suppressing torsional overshoot of the mechanical shaft system at its source.

[0059] Based on the safety protection requirements of the motor's physical boundaries, to prevent excessive injection of excitation current from causing severe stator overheating or the core falling into a deep magnetic saturation region, the main control logic unit sets strict anti-saturation limiting logic before writing the comprehensive direct-axis current command into the register. Specifically, the system reads the rated maximum excitation current threshold and the minimum excitation current threshold for maintaining field weakening operation of the variable frequency motor in real time, and forcibly determines... Whether it is within the safe operating area defined by the two thresholds. If the calculated current command exceeds the above upper limit due to extreme pressure spikes in the pipeline, the system will directly cut off the increment and clamp it to the maximum excitation current threshold. The above complete cross-domain bidirectional nonlinear decoupling logic enables the system to eliminate pressure surges in the forward direction through feedforward swashplate compensation and suppress mechanical resonance in the reverse direction through damping injection, thus completely closing the dynamic coordinated control of the electromechanical and hydraulic domains.

[0060] After the cross-domain bidirectional nonlinear decoupling and feedforward compensation in step S3, the system has achieved stable dynamic disturbance rejection under normal load fluctuation conditions. However, in more extreme forced impact conditions (such as rapid load release), single damping adaptive adjustment cannot completely absorb transient recoil energy. This section represents the second substantial technical breakthrough of this invention, focusing on the mechanism by which the system achieves energy recovery and prevents pressure loss and buckling under forced impact conditions by dynamically modifying the motion boundary. Step S4 specifically includes the following sub-steps: S401 determines the operating quadrant based on multi-dimensional state characteristics and generates a conventional equal-power energy consumption boundary. Based on the general theories of energy conservation and power flow reversal in electrical machinery, the variable frequency motor is in the motoring quadrant when actively driving the load, continuously consuming active power from the grid. However, when the hydraulic system faces forced operating conditions such as rapid depressurization, the fluid kinetic energy is converted into mechanical recoil torque, dragging the main shaft and forcing the motor to cross the zero point and enter the generating quadrant, feeding back active power to the active front-end module. To accurately define this physical inflection point, the main control logic unit abandons static judgments relying on a single speed or pressure extreme value, and adopts a dynamic multi-dimensional fusion logic based on torque differential polarity and active power flow direction.

[0061] To ensure strict consistency of multi-source data determination across the time dimension, the torque differential signal output from step S301 and the active power flow direction directly read from the shared register in step S102 are extracted. Both are latched by the same global hardware clock interrupt. The main control logic unit constructs the following quadrant determination logic function in the shared memory: ; In the formula, This is the quadrant status indicator variable for the current control cycle. Its output value of 1 corresponds to the motor quadrant, and its output value of -1 corresponds to the generator quadrant. This represents the direction of active power flow. A positive value indicates that active power flows from the external AC power grid into the motor, while a negative value indicates that active power is inverted and fed back to the power grid. The first one, smoothed by a nonlinear tracking differentiator The torque differential signal for each control cycle. and These are the set power zero-drift tolerance band threshold and torque differential anti-jitter dead zone threshold. Both thresholds are small positive real numbers slightly greater than zero, and their specific values ​​are determined offline based on the noise floor of the underlying current sensor and the inherent standby losses of the inverter. The technical purpose of setting these tolerance bands is to prevent frequent quadrant flag jumps and logic dead loops caused by high-frequency white noise near the zero-power crossover point. The physical meaning of the above judgment logic is that only when the system exhibits a clear power reverse flow trend and has a definite negative torque abrupt acceleration is it rigidly confirmed that the system has been forced into the generating quadrant, thereby ensuring high robustness of the operating condition boundary determination.

[0062] As a preferred approach, after determining the current operating quadrant, the system generates a conventional equal-power energy consumption boundary based on this, serving as a baseline for subsequent controller output actions. Based on the principle of maximum transient capacity limitation in electromechanical energy conversion, when the actuator rapidly responds to command changes, the rate of change of the driving variable cannot increase indefinitely; otherwise, it will cause overcurrent interruption in the inverter main circuit or damage to the pump's mechanical shaft system. The main control logic unit constructs an equal-power physical envelope model in the same memory space based on the current transient physical field. Specifically, the following equal-power boundary constraint formula is used: ; In the formula, For the first Estimated value of the system's overall transient power perturbation generated by the instructions to be implemented within each control cycle; The equivalent mechanical moment of inertia constant, referred to the motor shaft end, is always greater than zero and depends on the determination of the physical parameters of the unit's mechanical rotor assembly. For the first The operating frequency of each control cycle; For the first The frequency change command that the system intends to output within each control cycle; This is the maximum theoretical displacement constant of the bidirectional swashplate axial piston pump. For the first The actual working pressure of each control cycle. For the first The command to be output within each control cycle, specifying the change in swashplate angle.

[0063] To limit the aforementioned energy perturbations within the system hardware safety margin, the main control logic unit is equipped with judgment condition logic: mandatory constraints. ,in, The equal power dissipation boundary threshold set for the system has a value in the electric quadrant (i.e., When the load is 100 ohms, take the nominal value of the maximum instantaneous overload power allowed by the hardware of the four-quadrant inverter main circuit.

[0064] If the forward-looking evaluation by the main control logic unit reveals... Instead of executing a rigid one-sided truncation, the system initiates a proportional shrinkage logic: forcibly shrinking... and Synchronization multiplied by attenuation coefficient This ensures that the dynamic responses on both the electromechanical and hydraulic sides maintain a physical coupling ratio, avoiding cross-domain misalignment caused by unilateral limiting.

[0065] The technical purpose and physical meaning of this boundary constraint formula is to unify the physical dimensions of the increase in mechanical rotor kinetic energy caused by frequency command changes and the increase in fluid hydraulic work quantity caused by swashplate tilt angle changes, and reconstruct them into a jointly constrained energy surface in mathematical space. Through this conventional equal-power energy consumption boundary, the system can pre-determine whether the subsequently generated integrated frequency and tilt angle change commands exceed the electromagnetic and mechanical work limits of the physical hardware, thus providing a solid benchmark constraint for the safe execution of cross-domain collaborative control actions.

[0066] S402 relaxes the frequency descent constraint based on multidimensional state characteristics and switches the swashplate response to a nonlinear cutoff curve. Based on the general physical principles of fluid dynamics and motor regenerative power generation, when the hydraulic pipeline experiences a rapid load release (e.g., a high-flow directional valve opening instantaneously), the high-pressure compressed fluid accumulated inside the pipeline will undergo a violent volume expansion. This fluid expansion energy drives the axial piston pump in reverse, transforming it into a hydraulic motor operating condition, which in turn drives the variable frequency motor spindle to rotate at high speed. During this physical abrupt change, the electromagnetic torque will inevitably undergo a rapid negative abrupt change, forcing the system to cross the zero energy point and enter the power generation quadrant.

[0067] The aforementioned quadrant state indicator variables and torque differential signals are chosen as input parameters for dynamic reconstruction because they can most quickly and intuitively characterize the burst intensity of recoil energy and the inversion trend of active power in the physical domain. Based on the determination results of the aforementioned multi-dimensional state characteristics in the same memory, the main control logic unit dynamically modifies the maximum allowable frequency descent rate of the system. As a preferred approach, an asymmetric frequency constraint relaxation formula is specifically adopted: ; In the formula, For the first The maximum allowable frequency drop rate limit for the reconfigured system within a control cycle. The baseline maximum frequency decay rate is set for offline operation. This baseline value corresponds to the safe physical boundary in the electric quadrant. The quadrant state indicator variable for the current control cycle generated in step S401 above. The torque differential signal is extracted by a nonlinear tracking differentiator. To relax the sensitivity coefficient for frequency, The value range is a real number greater than 0, and its specific value is determined by physical matching between the dynamic resistance energy dissipation power of the main circuit of the frequency converter and the absorption margin of the bus capacitor.

[0068] To prevent the calculation under extreme step recoil... The divergence caused the bus voltage to exceed the limit, so the system added an absolute frequency change rate clamping logic constraint to the output stage of this formula. ,in This is the critical value for hardware-level bus overvoltage protection. The physical meaning of this asymmetric frequency constraint relaxation formula is that when the system is in the electric quadrant (…),… When the system enters the generator quadrant, the algebraic terms in the latter half are zeroed, and the frequency descent rate is limited by the reference boundary to prevent motor stall; however, once the system determines that it has entered the generator quadrant... The permissible frequency drop boundary is expanded proportionally to the severity of the negative torque abrupt change. This mechanism utilizes the relaxed kinematic boundary to effectively guide the hydraulic mechanical energy generated by the recoil back to the AC power grid via inverter inversion, fundamentally resolving the problem of transient energy accumulation.

[0069] Simultaneously, to cut off the continuous injection of recoil energy from the hydraulic physical source and prevent mechanical pressure loss and jerking, the system abandons the linear feedforward superposition strategy in the aforementioned positive electro-hydraulic channel. The main control logic unit forcibly takes over the response trajectory of the piston pump actuator in the command domain, switching it to a nonlinear throttling curve. Based on the mathematical characteristics of the exponential decay function—having a large rate of change in the initial stage and a smooth convergence at the end—a new swashplate command is specifically generated using a nonlinear throttling reconstruction formula: ; In the formula, For the first The final output of the combined swashplate tilt command in each control cycle To determine the optimal swashplate tilt angle based on the steady-state optimization output of the minimum energy functional, For the first Transient tilt angle compensation amount per control cycle This is the combined swashplate tilt command from the previous control cycle. The nonlinear cutoff attenuation coefficient is... This is the discrete integral step size of the control system. The value of is greater than 0, and it is physically calibrated based on the mechanical throttling characteristics of the plunger pump's distribution plate and the bandwidth of the servo valve.

[0070] To ensure the integrity of the discrete algorithm during long-term execution decay, the main control logic unit has a lower limit dead zone to prevent overflow. When the tolerance decays to near zero, the exponential calculation is forcibly interrupted and the process is smoothly transitioned to the subsequent zero-crossing dead zone handling logic.

[0071] The technical purpose and physical meaning of the aforementioned nonlinear cutoff and reconstruction formula is that when subjected to an impact and entering the power generation quadrant, the swashplate angle is immediately and rapidly reduced at an exponential rate positively correlated with the torque mutation rate, quickly suppressing the instantaneous oil intake of the pump body's internal cavity; while in the terminal region where the angle is close to zero, the decay rate automatically slows down. While ensuring the extremely high-speed limiting feedback energy, it avoids the secondary mechanical cavitation and hydraulic water hammer effects caused by the swashplate slamming into zero with excessive acceleration, ensuring a smooth transition under extreme forced operating conditions.

[0072] S403 triggers B-order spline functions based on multidimensional boundary conditions for spatial interpolation and forced injection of transient holding torque current. The distribution plate crossing the zero-tilt dead zone is chosen as the specific operating range for triggering spatial interpolation and current injection because within this small mechanical stroke, the effective displacement of the hydraulic pump drops sharply to zero and the load stiffness exhibits a step-like collapse. Without active intervention, the gear meshing clearance of the electromechanical transmission shaft will experience severe reverse knocking due to the loss of hydraulic load torque support. The main control logic unit monitors the current control cycle (the [number]th cycle) in real time. The final output of the integrated swashplate tilt command (from each control cycle) .

[0073] To avoid false triggering of the algorithm due to high-frequency white noise jitter in a single position sensor signal, the main control logic unit incorporates a multi-dimensional weighted triggering logic based on actual working pressure and motor operating frequency. Specifically, triggering only occurs when the system determines... And the extracted first Operating frequency per control cycle Greater than the set minimum frequency threshold for maintaining mechanical engagement Only then are the following spatial interpolation and current compensation operations activated. Among them, The mechanical zero-bias dead zone tolerance is calibrated offline. Its range is slightly larger than the angular displacement of the dead zone physically blocked by the distribution plate. It is determined based on the actual mechanical machining clearance of the pump body. The value of is greater than 0, determined through physical experiments based on the inertial bias requirements of the rotating component. Once the trigger condition is met, the system immediately stops the aforementioned nonlinear interception trajectory and uses a B-order spline function to perform a high-order smooth reconstruction of the motion trajectory within the dead zone within the command domain. Based on the fluid continuity theorem and the requirement of second-order differentiability of the kinematics of mechanical components, the use of a B-order spline function can ensure the absolute continuity of the angular velocity and angular acceleration of the swashplate when crossing the zero point, thereby completely eliminating the water hammer impact caused by sudden changes in oil pressure. As a preferred method, the following cubic B-spline spatial interpolation formula is used to generate the transition tilt angle command within the dead zone: ; In the formula, The tilt angle instruction for smooth transition of the reconstructed output within the dead zone; To construct the B-spline control polygon of the first The coordinates of each feature control point are obtained by solving the matrix equations jointly with the initial tilt angle and initial angular velocity of the cut-in dead zone point and the opposite boundary displacement and target velocity boundary conditions of the target crossing point. The basis functions are cubic B-spline functions. The normalized internal interpolation time variable parameter is internally calculated as follows: ,in Based on the current system timestamp extracted from the global hardware timer, This is the initial timestamp when the swashplate trajectory enters the dead zone. The expected total duration for the algorithm.

[0074] Due to the mandatory consideration of the completeness of the underlying algorithm logic, in the division operation of the above normalized parameters, the main control logic unit is equipped with singularity protection logic to enforce the constraint that the denominator term meets the requirement. , Step size for discrete integration The overflow threshold is set to be a positive integer multiple of the microprocessor's underlying division by zero, which not only avoids arithmetic crashes but also rigidly limits the swashplate's maximum speed across the dead zone from a kinematic perspective, thus physically isolating the mechanical distributor plate from irreversible wear caused by high-speed friction.

[0075] Meanwhile, to prevent time difference out-of-bounds errors caused by thread preemption or timing jitter at the operating system level, the system strictly controls the normalized internal interpolation time variable parameters. Hardware-level clamping in closed intervals completely eliminates the risk of divergence and extrapolation of spline space interpolation formulas outside the domain. Utilizing the smooth splicing mathematical properties of high-order polynomials, a kinematic damping buffer zone is artificially created within an extremely narrow mechanical dead zone to prevent excessive tearing of the lubricating oil film by transient fluid shear forces.

[0076] While the aforementioned spatial interpolation is performed on the hydraulic side, the variable frequency motor side faces a rapid and instantaneous load removal due to the hydraulic displacement returning to zero, which can easily lead to rotor runaway or backlash. To maintain a constant tension in the transmission chain within the zero-load range, the system synchronously injects transient holding torque current at the quadrature axis (q-axis, i.e., the component axis that determines the electromagnetic torque intensity) control node of the field-oriented control algorithm. Specifically, the following quadrature axis collaborative compensation formula is adopted: ; In the formula, The final comprehensive quadrature-axis current command generated and issued to the underlying current closed loop for the current control cycle; , is the original quadrature shaft current request value independently calculated and output by the speed loop regulator in the field-oriented control algorithm to maintain the target speed; , is the transient holding current reference amplitude, which is a real number greater than 0, and the specific value is calibrated offline based on the minimum torsional offset preload torque of the gear reducer and coupling in the drive system; For combined swashplate tilt commands; This refers to the aforementioned mechanical zero-bias dead zone tolerance. The sign function is extracted from the optimal motor speed direction output in the aforementioned steady-state benchmark optimization stage, and the output value is 1 or -1.

[0077] Before performing the parabolic compensation calculation, the underlying logic performs a mandatory check of the mechanical zero-bias dead zone tolerance. It is required that it must be greater than the set minimum positive real number mechanical precision tolerance. This eliminates the potential for arithmetic crashes caused by division by zero at the root. Furthermore, to ensure the completeness of global computation of symbolic functions, the underlying algorithm stipulates that when... When strictly zero, The output is forced to 1 to maintain the basic disturbance rejection bias when the system is statically shut down.

[0078] The physical meaning of this cross-axis collaborative compensation formula is that, based on the current spatial position and depth of the swashplate, the system dynamically constructs a parabolic active current injection envelope that is completely symmetrical about the absolute physical zero tilt angle. As the swashplate gets closer to the physical zero position, the current component injected for compensation approaches its extreme value, thereby generating an artificial constant electromagnetic drag counter-torque that forcibly maintains the tight fit of the physical tooth surfaces between the mechanical transmission pairs. As the swashplate gradually moves away from the dead zone center and approaches the two side boundaries, this compensation amount decays smoothly to zero in a quadratic manner, and the system re-establishes the conventional electromechanical-hydraulic load coupling feedback.

[0079] To prevent logical conflicts between the forced current injection and the overcurrent hardware protection action triggered by sudden load changes in the inverter main circuit, the underlying software incorporates multi-source timing authentication logic. The system forcibly compares the timestamp of the sudden quadrature-axis current surge with the timestamp of the swashplate entering the dead zone position flag. Only when both are determined to be perfectly aligned at the global clock level is the injection of the transient holding torque current allowed to be legalized and effective. Furthermore, to ensure the thermal safety of the inverter's insulated-gate bipolar transistors (IGBTs) under extremely high-frequency switching conditions, the main control logic unit outputs a comprehensive quadrature-axis current command to the downstream bus. The system is constantly constrained by the hardware threshold saturation clamping constraint of the absolute upper limit of the inverter's nominal maximum transient peak overload current. The above-mentioned joint intervention mechanism ensures that when the system faces extreme operating conditions such as high-energy bidirectional interference, it not only achieves a smooth pressure transition at the hydraulic circuit level, but also completely eliminates the source of mechanical jerking and noise in the electromagnetic domain.

[0080] After zero-dead-zone smoothing and asymmetric boundary reconstruction, the system has generated comprehensive kinematic commands adapted to extreme impact conditions. To ensure that the aforementioned cross-domain discrete digital commands can be seamlessly and safely converted into physical actions by the underlying electrical and hydraulic actuators, and to avoid secondary oscillations caused by overshoot at the actuator end, the system enters the hardware topology-driven execution and boundary clamping stage. Step S5 specifically includes the following sub-steps: The S501 module performs trajectory filtering and maximum rate-of-change clamping based on a real-time asymmetric constraint boundary module. Based on the general physical principles of mechanical fatigue limits of actuators and electrical overload boundaries of drives in industrial transmission systems, any discrete step command without smoothing and limiting may induce transient destructive stress or current avalanche at the physical execution end. The current control cycle frequency change command and the reconstructed integrated swashplate tilt angle command are chosen as the direct inputs to this module because they represent the final spatiotemporal kinematic requirements of the system after cross-domain decoupling, and their derivative characteristics directly map the maximum servo response capability of the physical hardware.

[0081] The main control logic unit synchronously inputs the calculated current control cycle frequency change command and the reconstructed integrated swashplate tilt angle command into the real-time asymmetric constraint boundary module set at the bottom layer.

[0082] As a preferred approach, a discrete trajectory amplitude limiting filter formula is specifically used to truncate and smoothly reconstruct the original kinematic commands: ; ; In the formula, and These represent the system timestamps of the current and previous discrete control cycles, respectively. This represents the actual permissible frequency change after asymmetric clamping, and its sign determines whether the rotational speed increases or decreases. This is the maximum allowable rate of frequency decay limit. For the first The frequency change command that the system intends to output within each control cycle; The maximum frequency rise acceleration allowed by the drive system is always greater than 0. The specific value is determined by physical calibration based on the combined moment of inertia of the variable frequency motor rotor and the pump body main shaft and the transient peak output torque of the inverter. and These represent the combined frequency commands actually output to the underlying hardware after trajectory filtering in the current control cycle and the previous control cycle, respectively.

[0083] Simultaneously, for the swashplate command in the hydraulic domain, the module employs similar boundary protection logic to calculate the final output swashplate tilt angle command: ; In the formula, and These represent the final output swashplate tilt commands for the current and previous control cycles, respectively. This is the combined swashplate tilt command after the preceding compensation is completed. This is the maximum single-cycle angular displacement deflection extreme value constant allowed by the physical structure of the electro-hydraulic proportional servo valve. Its value is greater than 0 and is calculated from the servo valve's factory-set mechanical full-scale stroke and maximum control bandwidth. This filtering architecture transforms nonlinear boundary constraints into pure algebraic limit operations, completely avoiding micro-jitter in the instruction execution cycle caused by complex conditional branching logic.

[0084] The technical purpose and physical meaning of the aforementioned trajectory filtering and amplitude limiting clamping formulas are to rigidly constrain the ideal control quantity in the digital control domain within the kinematic and thermodynamic limit boundaries of the physical actuator. Through asymmetric AND function combinations, the system not only ensures the rapid release of recoil energy in the power generation quadrant, but also prevents hardware overspeed runaway and fluid cavitation induced by drastic jumps in high-frequency commands.

[0085] The S502, based on space vector pulse width modulation, converts the comprehensive frequency command into a digital pulse sequence and drives the inverter main circuit. After confirming that the transient rate of change meets the above boundary conditions, the system compares the current control cycle with the comprehensive frequency command actually output to the underlying hardware after trajectory filtering. Together with the integrated direct-axis current command and integrated quadrature axis current command The core modulation module of the field-oriented control algorithm is seamlessly integrated. Based on the general technical principle of power electronic inverters reconstructing AC sinusoidal waveforms through high-frequency switching sequences, the main control logic unit uses space vector pulse width modulation technology to generate physical control signals.

[0086] Specifically, the amplitude of the synthesized voltage vector calculated by the modulation module is divided by the current DC bus voltage, and the on-time of the inverter's insulated-gate bipolar transistor is calculated using the following duty cycle time allocation formula: ; In the formula, For the first The effective switching time of a certain bridge arm within a control cycle; The fixed period for pulse width modulation is rigidly determined by the underlying overload value of the microcontroller's timer. The magnitude of the synthesized reference voltage vector generated based on the Clark and Park inverse transform in the field-oriented control algorithm; This is a real-time DC bus voltage measurement that is synchronously sampled by the underlying analog-to-digital converter.

[0087] To ensure the alignment of vector modulation operating conditions, The generation timestamp and The sampling trigger time is achieved by cross-linking hardware timers to realize microsecond-level physical synchronization.

[0088] To ensure the integrity of the above division operation algorithm under extreme grid voltage dips or abnormal operating conditions of the inverter's front-end rectifier bridge, the underlying logic uses real-time DC bus voltage measurements. The denominator term is equipped with hardware-level undervoltage singularity protection. When The voltage drops to near the set minimum safe voltage threshold. At the critical point (i.e., the undervoltage protection threshold of the inverter bus capacitor), the system forcibly clamps the calculation denominator to this minimum safe voltage threshold, completely avoiding the arithmetic logic unit crash caused by division by zero in the microprocessor. The physical meaning of this duty cycle time allocation formula is that it precisely maps the voltage vector amplitude in the digital space to the high-level pulse width of the physical switching devices at the microsecond level, approximating a perfect circular rotating magnetic field within the motor stator. To prevent a short circuit between the upper and lower switches on the same bridge arm of the inverter main circuit, which could lead to an explosion, the controller forcibly inserts a nanosecond-level hardware dead time through the underlying timer peripheral before sending the generated 6-channel digital pulse sequence to the physical pins.

[0089] The S503, based on a closed-loop current regulator with chatter compensation, converts the comprehensive swashplate tilt command into a physical drive signal. In the hydraulic actuation domain, the aforementioned smoothed output swashplate tilt command... The force must be converted into an electromagnetic control force capable of overcoming the mechanical spring resistance and the steady-state hydrodynamic force of the fluid, thereby directly driving the pilot coil of the electro-hydraulic proportional servo valve. Based on the physical law that the electromagnet's attraction force is strictly proportional to the amplitude of the current flowing through it, the system maps the physical position command to a given reference value for the closed-loop current. The main control logic unit is equipped with a digital current regulator and actively superimposes a high-frequency, low-amplitude chatter signal. Specifically, the servo drive synthesis formula is used: ; In the formula, This is the drive current command that is ultimately output to the electro-hydraulic proportional servo valve coil during the current control cycle. The tilt tracking position error at the current moment is internally defined as follows: ,in The discrete sampled values ​​of the actual swashplate position fed back by the hydraulic side linear displacement sensor; To accumulate from the initial time 0 at system startup to the current control cycle Discrete time series index; and These are the proportional gain and integral gain coefficients, respectively, and their values ​​are greater than 0. They are obtained by dynamically adaptively setting poles based on the equivalent inductance and resistance parameters of the servo valve solenoid coil. For the amplitude constant of the high-frequency flutter signal, The flutter angular frequency is constant. This refers to the aforementioned discrete integration step size; The value is determined based on the mass of the servo valve core and the stiffness of the return spring. Its upper limit is constrained to prevent macroscopic mechanical vibration that could cause the effective displacement of the pump body.

[0090] To verify the completeness of the closed-loop algorithm logic and avoid control quantity divergence and overshoot caused by the accumulation of high-frequency measurement white noise in the integrator, the above formula is equipped with anti-integral saturation clamping logic. Abandoning a single saturation threshold cutoff, the system adopts a multi-dimensional joint judgment of current limiting state and tracking error polarity: the system forcibly judges the drive current command. Does the rated maximum operating current of the proportional servo valve reach the limit? If it touches this hardware limit and the position error at that time... If the symbol attempts to further exacerbate the saturation trend, the integral term is immediately suspended and frozen. The subsequent accumulation calculation; conversely, if the error polarity reversal guides the system out of the saturation region, the dynamic accumulation calculation of the integrator is immediately restored. The physical meaning and technical purpose of this servo drive synthesis formula are as follows: the first half, based on the classical proportional-integral control law, ensures steady-state tracking of the target tilt angle without static error; the sign trigonometric function in the second half artificially injects a continuously existing high-frequency micro-excitation. This excitation forces the micro-slide valve inside the servo valve to always maintain a dynamic reciprocating high-frequency micro-vibration lubrication state, thereby physically reducing the huge static dry friction force that originally hindered the valve core from starting to an extremely small dynamic friction force, exponentially improving the sensitivity and bandwidth of the hydraulic actuator when crossing zero.

[0091] Finally, to ensure the high-frequency synchronization of cross-physical domain data control instructions at the final execution port, all the aforementioned pulse modulation cycle trigger edges and servo current refresh write cycles are strictly and forcibly aligned by the clock interrupt of the global hardware bus matrix on the chip of the integrated drive and control controller. This completely eliminates the random phase jitter in the micro-instruction issuance and ensures that the actions from the computing core to the peripheral power devices are completely physically anchored on the time axis.

[0092] I. Specific Application Example: Hydraulic Drive System for a 3000-ton Large Injection Molding Machine The aforementioned control method of the present invention can be widely applied to various heavy-duty electro-hydraulic equipment with high-frequency and high-speed load changes and energy feedback. To aid understanding, this embodiment will be described in detail using the main drive hydraulic system of a 3000-ton large servo injection molding machine as an example.

[0093] System basic physical parameter settings: In this application scenario, the hardware settings of the electromechanical hydraulic actuator are as follows: the rated power of the variable frequency motor is 55kW and the rated speed is 1500rpm; the maximum theoretical displacement of the bidirectional swashplate axial piston pump is 250cc / rev; the rated high pressure working boundary of the system is 31.5MPa; the hydraulic oil type is No. 46 anti-wear hydraulic oil.

[0094] Cross-domain collaborative control process in a typical process cycle: The standard working cycle of an injection molding machine includes stages such as rapid mold closing, high-pressure injection, pressure holding, rapid pressure release, and mold opening. The specific adaptive performance of the integrated drive and control controller in each stage is as follows: High-pressure injection and holding stage: At this stage, the system requires extremely high pressure and extremely low flow rate. The actual working pressure surges to approximately 30 MPa, and the fluid is severely compressed. The system calculates the dynamic bulk modulus in step S2, detects the surge in fluid stiffness, updates the volumetric efficiency matrix accordingly, and re-optimizes to obtain a steady-state benchmark with low speed and large tilt angle. Simultaneously, to prevent pressure overshoot during the holding phase, the nonlinear tracking differentiator in step S3 detects the sharp increase in electromagnetic torque and generates a comprehensive swashplate tilt angle command in advance. The feedforward attenuation is increased, and transient damping current is injected into the direct axis (d-axis) of the variable frequency motor, which perfectly suppresses the mechanical scream and pipe vibration of traditional injection molding machines at the moment of pressure holding and switching.

[0095] Rapid pressure relief and mold opening stage: When the pressure holding ends and the reversing valve is fully open, the 30MPa high-pressure oil is released instantaneously, and the backflow energy drags the plunger pump. The system immediately identifies the reverse of active power through multi-dimensional state characteristics and determines that it has entered the generation quadrant. At this time, step S4 triggers asymmetric boundary reconstruction, and the system dynamically relaxes the maximum allowable frequency descent rate limit. This allows the motor speed to be increased by reverse dragging, inverting the fluid potential energy back to the power grid. When the swashplate angle decreases sharply and crosses the mechanical dead zone, the system automatically switches to a cubic B-spline interpolation trajectory and simultaneously injects transient holding torque current into the quadrature axis (q-axis) of the variable frequency motor. This action ensures that the injection molding machine's massive mechanical transmission gears remain tightly meshed even when the hydraulic load is lost, completely eliminating metallic knocking noises.

[0096] II. Experimental Verification and Effect Comparison To verify the actual technical effect of the present invention, a semi-physical hardware-in-the-loop test was conducted based on the integrated drive and control hardware platform, and combined with the attached... Figure 3 The simulated waveforms shown are used for dynamic data comparison. The comparison object is a basic variable frequency hydraulic control system that uses a traditional fixed proportional integral closed loop and a two-dimensional static efficiency lookup table strategy.

[0097] Energy consumption comparison: The system runs continuously for 100 cycles within a standard injection molding cycle (45s per cycle), and the total active power consumption on the grid side is recorded using a high-precision power analyzer. Evaluation indicators Traditional PI control system This invention is a collaborative control system. Optimization and improvement Average comprehensive power consumption per cycle 0.58kWh 0.49kWh Energy saving of 15.5% High-voltage pressure holding section volumetric power loss 4.2kW 2.8kW Losses reduced by 33.3% Rapid decompression stage single-cycle energy recovery 0kJ (heat dissipated through braking resistor) 12.5kJ (inverter feedback to the grid) Achieving 100% green energy return

[0098] Dynamic response waveform comparison: Key dynamic data were extracted using a high-frequency data acquisition card, and referenced in the appendix. Figure 3 The comparison curves in the figure provide detailed evidence: Transient pressure overshoot suppression test: Refer to the appendix Figure 3 Sub-figure A in the figure visually illustrates the dynamic response waveform of the system when faced with a step pressure command. The horizontal axis represents time (s), and the vertical axis represents the actual working pressure (MPa). The extremely thin dashed line represents the target pressure command. At 0.1s, the system triggers an action to step from the target pressure of 5MPa to 28MPa. Observing the response trajectory of the traditional control architecture represented by the dark gray dashed line in the figure, due to the hydraulic side execution delay and physical inertia, the actual working pressure of the traditional system reaches a maximum of 32.5MPa (overshoot as high as 16%), and only barely converges after three obvious decaying oscillation cycles. In contrast, the thick solid black line in the figure represents the response trajectory of the system of this invention. The system of this invention relies on the phase-shiftless torque differential signal for cross-domain feedforward intervention, and the pressure smoothly climbs to 28MPa, with a peak value of only 28.5MPa (overshoot is extremely compressed to 1.7%), and achieves zero second-order oscillations. The waveform exhibits highly ideal critical damping tracking characteristics.

[0099] Zero-crossing dead zone mechanical vibration acceleration test: Data is collected using a triaxial accelerometer mounted on the pump housing under the condition of zero displacement reversal. (See attached document.) Figure 3 Subgraph B in the figure records the mechanical impact evolution process at the instant of recoil crossing zero. The horizontal axis in the figure is time (s), and the vertical axis is mechanical vibration acceleration (g). Observing the evolution trend of the traditional system represented by the dark gray dashed line in the figure, when the displacement reverses at 0.1s, the maximum transient impact acceleration of the traditional system suddenly soars to 4.5g (the waveform shows high-frequency dense oscillation, directly representing the severe water hammer excitation and tooth backlash in the physical domain); the evolution trend of the system of the present invention represented by the pure black thick solid line in the figure shows that after introducing B-order spline trajectory interpolation and cross-axis transient holding torque current, the peak value of the zero-crossing transient impact acceleration of the present invention is sharply reduced to 0.8g, and the oscillation waveform decreases to a stable state in a very short time. The system operates smoothly, fundamentally solving the problem of mechanical fatigue stress concentration under alternating working conditions.

[0100] Experimental Conclusions: The experimental data extracted from the multidimensional attached curves fully demonstrate that the variable displacement and variable frequency speed control system proposed in this invention eliminates steady-state errors through fluid thermodynamic mapping, overcomes the physical domain response time difference barrier through cross-domain bidirectional nonlinear decoupling, and perfectly resolves mechanical stuttering and energy accumulation under extreme forced operating conditions through asymmetric boundary reconstruction. Significant technological advancements have been achieved in energy efficiency, response agility, and mechanical safety, surpassing those of traditional architectures.

Claims

1. A variable displacement and variable frequency speed regulation collaborative control high efficiency energy saving plunger pump system, characterized in that, It includes a multi-dimensional sensing unit, an electrical drive unit with a four-quadrant inverter main circuit, an electromechanical-hydraulic actuator with an electro-hydraulic proportional servo valve and a swashplate, and an integrated drive and control controller with an integrated main control logic unit. The main control logic unit synchronously acquires the actual working pressure and real-time oil temperature, as well as the underlying state variables including the electromagnetic torque estimate, based on the same memory space, calculates the dynamic bulk elastic modulus, and obtains the optimal motor speed and optimal swashplate tilt angle through the minimum energy functional model. The main control logic unit superimposes the extracted torque differential signal onto the optimal swashplate angle and generates a comprehensive swashplate angle command, and converts the dynamic bulk elastic modulus into an equivalent dynamic stiffness coefficient, injects it into the excitation current control node, and generates a comprehensive direct-axis current command. The main control logic unit dynamically generates asymmetric constraint boundaries based on the torque differential polarity and active power flow direction to adjust the frequency change command. When the swashplate crosses the zero tilt dead zone, it uses spline function interpolation to interpolate the trajectory and forcibly injects transient holding torque current to generate a comprehensive quadrature axis current command. The main control logic unit sends the frequency change command and the comprehensive swashplate tilt command to the asymmetric constraint boundary to perform trajectory filtering and amplitude limiting clamping. Finally, it combines the comprehensive direct-axis current command and the comprehensive quadrature-axis current command to convert them into digital pulse sequences and drive current commands to control the four-quadrant inverter main circuit and the electro-hydraulic proportional servo valve.

2. The variable displacement and variable frequency speed governing high efficiency energy saving plunger pump system of claim 1, wherein, The main control logic unit extracts discrete digital quantities and performs smoothing processing using a digital filtering algorithm; By using direct memory mapping, the external bus is bypassed, and the underlying state variables are synchronously read directly from the shared register address segment; The operation of reading the shared register address segment is strictly triggered by the rising edge of the global hardware clock, so that the electrical domain state variables can maintain strict physical timing alignment with the hydraulic domain acquired data.

3. The variable displacement and variable frequency speed governing high efficiency and energy saving plunger pump system, as claimed in claim 1, wherein The main control logic unit uses the calculated dynamic bulk elastic modulus as a dynamic correction factor to refresh the preset volumetric efficiency matrix in real time. Before performing matrix element traversal operations, it is forcibly checked whether the dynamic volumetric elastic modulus is greater than the set minimum stiffness safety threshold. When the dynamic volumetric elastic modulus is lower than the minimum stiffness safety threshold, the dynamic volumetric elastic modulus is forcibly clamped to the minimum stiffness safety threshold. The minimum stiffness safety threshold is defined as a safety lower limit parameter set to avoid division overflow anomalies under extreme high temperature and low pressure conditions or sensor disconnection faults.

4. The variable displacement and variable frequency speed governing high efficiency and energy saving plunger pump system, as recited in claim 1, characterized in that, The main control logic unit solves for the extreme points of the minimum energy functional model using a gradient descent iterative algorithm; In each iteration, the gradient vector formed by the partial derivatives of the objective function of the minimum energy functional model with respect to the optimal motor speed and the optimal swashplate angle is continuously calculated. When the L2 norm of the gradient vector is less than the set convergence tolerance, the functional optimization is determined to be converged, and the optimal motor speed and the optimal swashplate angle under theoretical steady state are output.

5. The high-efficiency energy-saving plunger pump system with variable displacement and variable frequency speed regulation coordinated control according to claim 1, characterized in that, The process by which the main control logic unit superimposes the extracted torque differential signal onto the optimal swashplate angle to generate the comprehensive swashplate angle command includes: The main control logic unit inputs the obtained electromagnetic torque estimation value into a preset nonlinear tracking differentiator to extract the torque differential signal with a high signal-to-noise ratio; The torque differential signal is multiplied by a set hydraulic compensation gain to generate a transient tilt angle compensation amount; In the same memory space, the transient tilt angle compensation amount and the optimal swashplate tilt angle are directly fed forward and superimposed to generate the comprehensive swashplate tilt angle command, which drives the valve core of the electro-hydraulic proportional servo valve to produce a displacement action in advance.

6. The variable displacement and variable frequency speed governing high efficiency and energy saving plunger pump system, as recited in claim 1, characterized in that, The main control logic unit uses the mechanical impedance transformation relationship to map the dynamic bulk elastic modulus to the equivalent dynamic stiffness coefficient of the hydraulic side relative to the shaft system. The equivalent dynamic stiffness coefficient is used to generate the excitation current adjustment amount and calculate the comprehensive direct-axis current command. The comprehensive direct-axis current command is directly written into the excitation current control node where the field-oriented control algorithm is located through the internal bus matrix.

7. The high-efficiency energy-saving plunger pump system with variable displacement and variable frequency speed regulation coordinated control according to claim 1, characterized in that, The main control logic unit constructs an equal power physical envelope model in the same memory space, unifies the physical dimensions of the mechanical rotor kinetic energy increment caused by the frequency command change and the fluid hydraulic pressure caused by the swashplate tilt angle change as the functional quantity increment, and forces the comprehensive transient power perturbation estimate to be less than or equal to the equal power dissipation boundary threshold set by the system. The equal power dissipation boundary threshold is defined as the nominal value of the maximum instantaneous overload power allowed by the hardware of the four-quadrant inverter main circuit when it is in the electric quadrant.

8. The variable displacement and variable frequency speed governing high efficiency and energy saving plunger pump system, as recited in claim 1, characterized in that, When forced into the generating quadrant, the main control logic unit expands the allowable frequency drop boundary proportionally to the severity of the negative torque mutation caused by the torque differential signal. The response trajectory of the electromechanical-hydraulic actuator is forcibly taken over in the command domain, and the response trajectory is switched to a nonlinear cutoff curve to rapidly reduce the swashplate tilt angle at an exponential rate positively correlated with the torque mutation rate.

9. The variable displacement and variable frequency speed governing high efficiency and energy saving plunger pump system, as recited in claim 1, characterized in that, The main control logic unit monitors the integrated swashplate tilt command within the integrated command in real time. Only when the integrated swashplate tilt command is determined to be less than or equal to the mechanical zero-bias dead zone tolerance and the extracted operating frequency is greater than the set minimum frequency threshold for maintaining mechanical engagement, the spline function in the command domain is activated for smooth reconstruction. The minimum frequency threshold for maintaining mechanical engagement is defined as a constant that is determined through physical experiments based on the inertial bias requirements of the rotating components and has a value greater than zero.

10. The high-efficiency energy-saving plunger pump system with variable displacement and variable frequency speed regulation coordinated control according to claim 1, characterized in that, When the main control logic unit uses space vector pulse width modulation technology to generate physical control signals, it sets hardware-level undervoltage singularity protection for the denominator of the real-time DC bus voltage measurement value. When the real-time DC bus voltage measurement value drops to near the set minimum safe voltage threshold, the calculation denominator is forcibly clamped. A digital current regulator is set in the hydraulic actuation domain, and a high-frequency low-amplitude chatter signal is actively superimposed to output the drive current command; The minimum safe voltage threshold is defined as the critical point for the undervoltage protection of the inverter bus capacitor.

Images