Bucket wheel machine power supply and distribution management method based on multi-motor cooperation and power balance

Abstract

The present application relates to the technical field of power management of bucket wheel machine, and particularly relates to a power supply and distribution management method of bucket wheel machine based on multi-motor cooperation and power balance, comprising the following steps: acquiring real-time load current and current operation efficiency of multiple motors in the bucket wheel machine; calculating a dynamic virtual impedance value for each motor drive circuit through an improved multi-objective optimization function; converting the dynamic virtual impedance value into corresponding cooperative power instructions and issuing the instructions to power drivers of each motor, so that the bus voltage of each motor is redistributed. The present application can significantly reduce power fluctuation in the system, improve motor operation efficiency, adapt to real-time adjustment requirements under different load conditions, improve overall stability and energy efficiency, enhance real-time adjustment capability when the power grid fluctuates violently, and ensure control accuracy and stability.

Description

Technical Field

[0001] This invention relates to the field of bucket wheel excavator power management technology, and in particular to a bucket wheel excavator power supply and distribution management method based on multi-motor coordination and power balance. Background Technology

[0002] As a key piece of equipment in a continuous loading and unloading system, a bucket wheel excavator typically operates with multiple high-power motors, each responsible for different subsystems such as traveling, slewing, pitching, and bucket wheel drive. These motors require a stable and coordinated power supply from the power grid at the excavator's connection point. However, due to fluctuations in the on-site operating environment, the asymmetry of the power grid itself, and the suddenness of load changes, problems often arise such as uneven power distribution among the motors, overload or no-load operation of individual motors, and severe voltage fluctuations, affecting the equipment's operating efficiency and service life.

[0003] In existing technologies, centralized scheduling with fixed parameters or motor power allocation according to preset priorities are often used. However, these methods lack a comprehensive assessment of the real-time status of the power grid and the operating efficiency of each motor, making it difficult to achieve dynamic optimization control of multi-motor systems in complex environments. Some studies have introduced virtual impedance control methods to guide power allocation by simulating series impedance to adjust the bus voltage of each motor. However, most of these methods are based on static models or ideal power grid conditions and do not consider the dynamic characteristics of motor loads, power grid disturbances, or changes in the operating location. Summary of the Invention

[0004] This invention provides a power supply and distribution management method for bucket wheel excavators based on multi-motor collaboration and power balance. This method is designed for multi-motor systems of bucket wheel excavators, can sense the grid status, take into account both power fluctuation suppression and energy efficiency optimization, and integrates historical experience and field data for dynamic feedforward adjustment, so as to achieve power balance and system stability improvement under multi-motor collaborative operation.

[0005] The power supply and distribution management method for bucket wheel excavators based on multi-motor coordination and power balance includes the following steps:

[0006] S1: Real-time monitoring of the actual voltage of the power grid at the connection point of the bucket wheel excavator, and acquisition of the real-time load current and current operating efficiency of multiple motors in the bucket wheel excavator;

[0007] S2: Based on the deviation between the actual voltage and the rated voltage of the power grid, the real-time load current of each motor and its current operating efficiency, a dynamic virtual impedance value is calculated for each motor drive circuit through an improved multi-objective optimization function; the objective of the improved multi-objective optimization function is to simultaneously minimize the total power fluctuation of the system and maximize the overall operating energy efficiency of the system.

[0008] S3: Convert the dynamic virtual impedance value into a corresponding collaborative power command and send it to the power driver of each motor. By adjusting the driver output, it is equivalent to applying virtual impedance, so that the bus voltage of each motor is redistributed, and the power of multiple motors is balanced under non-ideal grid voltage.

[0009] S4: Record and associate the real-time operating position information of the bucket wheel excavator with the sequence of dynamic virtual impedance values ​​calculated in S2, and construct and continuously update an operating trajectory-virtual impedance mapping database; when the bucket wheel excavator runs to the same or similar operating position again, the dynamic virtual impedance value is preloaded based on the operating trajectory-virtual impedance mapping database as the initial value or feedforward instruction calculated in S2.

[0010] Optionally, the actual voltage of the power grid is obtained through a three-phase voltage sensor of the bucket wheel excavator, and the three-phase voltage sensor continuously collects the actual voltage signal of the power grid;

[0011] The real-time load current of the multiple motors is obtained by a Hall effect current sensor installed in each motor. The three-phase current of each motor is detected in real time, and the magnitude of the current vector is calculated as the real-time load current.

[0012] Optionally, S1 further includes obtaining the real-time motor speed through a motor speed encoder, calculating the current load rate of the motor in combination with the real-time load current, querying a pre-stored three-dimensional MAP of efficiency-load rate-speed for the current model of motor, and interpolating to obtain the current operating efficiency.

[0013] Optionally, S2 establishes an improved multi-objective optimization function including a power fluctuation suppression term, an energy efficiency optimization term, and a voltage regulation term; the multi-objective optimization function is solved by a constrained iterative optimization algorithm to obtain a set of optimal dynamic virtual impedance values.

[0014] Optionally, the power fluctuation suppression term is used to minimize the total power fluctuation of the system, which is achieved by penalizing the degree to which the power of each motor deviates from its historical average value;

[0015] The energy efficiency optimization item is used to maximize the overall operating energy efficiency of the system, which is achieved by penalizing the operating state of inefficient motors;

[0016] The voltage regulation item is used to limit the magnitude of virtual impedance regulation, so as to avoid excessive voltage regulation from affecting the stability of motor operation.

[0017] Optionally, the iterative optimization algorithm adopts the particle swarm optimization algorithm, which initializes the particle swarm, iteratively updates the particle velocity and position, introduces a penalty function for voltage regulation deviation to handle the constraints, and dynamically records the individual optimal and global optimal impedance value vectors. Finally, it outputs the optimal dynamic virtual impedance configuration result that satisfies the constraints of power fluctuation suppression, energy efficiency optimization and voltage stability.

[0018] Optionally, S3 specifically includes:

[0019] S31, Conversion Calculation Process: For each motor, its dynamic virtual impedance value is converted into the corresponding voltage correction amount;

[0020] S32, Instruction generation process: Based on the voltage correction amount, generate the current motor's cooperative power instruction, which is the corrected voltage setting value;

[0021] S33, Command issuance process: The coordinated power command is issued to the power driver of the corresponding motor in real time;

[0022] S34, Equivalent impedance realization process: After receiving the coordinated power command, the power driver adjusts the output voltage through its internal regulating driver so that the actual voltage of the motor input terminal is equal to the coordinated power command, which is equivalent to the actual impedance with the series impedance value of the dynamic virtual impedance value in the motor drive circuit, thereby realizing the redistribution of the voltage of each motor bus.

[0023] Optionally, S3 further includes voltage differential allocation, which, through S31-S34, enables:

[0024] The smaller the dynamic virtual impedance value of a motor that is in an overload trend, the smaller its voltage correction amount, and the higher the bus voltage and power supply it can obtain.

[0025] Motors under light load tend to have a larger dynamic virtual impedance value, which in turn results in a larger voltage correction, allowing them to obtain a relatively lower bus voltage and power supply, thus achieving power balance among multiple motors under non-ideal grid voltage conditions.

[0026] Optionally, S4 specifically includes:

[0027] S41, through the timing data synchronization module in the bucket wheel excavator control system, the bucket wheel excavator's traveling position, rotation angle and pitch angle are fused to generate a unique three-dimensional working position coordinate, and the three-dimensional working position coordinate is associated with and stored with the dynamic virtual impedance value sequence of each motor under the same timestamp;

[0028] S42, adopts a data compression and storage strategy based on key location points, discretizes the continuous operation trajectory into multiple key location points, and stores a set of representative dynamic virtual impedance values ​​for each key location point. The representative set of dynamic virtual impedance values ​​includes the cluster center values ​​obtained after performing density clustering analysis on multiple sets of impedance values ​​recorded at the same location point in multiple operation cycles.

[0029] S43, when the bucket wheel excavator passes through the recorded key location point again, the newly calculated dynamic virtual impedance value is compared and fused with the cluster center value stored in the database, and the impedance value set of the key location point is updated using a weighted average algorithm. The weight of the newly calculated dynamic virtual impedance value depends on the timeliness of the new data and the access frequency of the operation location point.

[0030] S44, preload execution: When the bucket wheel excavator reaches a certain working position, the k-nearest neighbor algorithm is used to find several historical position points in the database that are closest to the current position. The dynamic virtual impedance values ​​corresponding to the several historical position points are weighted and averaged, and the calculation result is transmitted to S2 as the preload value.

[0031] Optionally, S4 further includes a feedforward-feedback fusion mechanism, specifically including in S2, using the pre-loaded dynamic virtual impedance value in the database as the basis for the feedforward instruction, and fusing it with the correction value calculated based on real-time data. During fusion, a feedforward weighting coefficient is set, which is adaptively adjusted according to the stability of the current grid voltage.

[0032] When the grid voltage fluctuation is less than the threshold, increase the weight of the preload value;

[0033] When the grid voltage fluctuates drastically, increase the weight of the real-time calculated value.

[0034] The beneficial effects of this invention are:

[0035] 1. This invention constructs an improved multi-objective optimization function that integrates power fluctuation suppression, energy efficiency optimization, and voltage stability constraints. It dynamically calculates the virtual impedance value of each motor and performs voltage regulation by simulating series impedance through a power driver. This effectively achieves balanced power distribution among multiple motors under unstable grid voltage conditions. Compared with traditional fixed power settings or proportional control methods, this solution can significantly reduce power fluctuations within the system, improve motor operating efficiency, adapt to real-time adjustment requirements under different load conditions, and enhance overall stability and energy efficiency.

[0036] 2. This invention introduces a time-series correlation mechanism between three-dimensional working coordinates (travel position, rotation angle, pitch angle) and the dynamic virtual impedance of the motor, constructing an impedance mapping database with representative compressed characteristics. Through key location point discretization and density clustering analysis, typical impedance patterns are extracted and compressed for storage, reducing data redundancy. At the same time, a weighted update strategy based on access frequency and data timeliness is proposed, enabling the database to have progressive learning capabilities and evolve over a long period of time with equipment aging, material changes, and other operating conditions, thereby enhancing the adaptability of the control system.

[0037] 3. During operation, this invention uses the K-nearest neighbor algorithm to achieve similarity matching of work locations, calls historical impedance data to generate preloaded virtual impedance initial values, and introduces a feedforward-feedback fusion strategy in the optimization calculation. The fusion ratio of the preloaded value and the real-time calculated value is adaptively adjusted according to the grid voltage fluctuation. When the grid is stable, historical experience is used first to improve calculation efficiency; when the grid fluctuates drastically, the real-time adjustment capability is enhanced to ensure control accuracy and stability. The fusion mechanism effectively improves the system's response speed and generalization ability, and has good application in complex industrial environments. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a schematic diagram of the management method flow according to an embodiment of the present invention;

[0040] Figure 2 This is a schematic diagram of the voltage redistribution process according to an embodiment of the present invention. Detailed Implementation

[0041] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. For some well-known technologies, those skilled in the art may also use other alternative methods to implement the invention. Moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0042] like Figures 1-2 As shown, the power supply and distribution management method for bucket wheel excavators based on multi-motor cooperation and power balance includes the following steps:

[0043] S1: Real-time monitoring of the actual voltage of the power grid at the connection point of the bucket wheel excavator, and acquisition of the real-time load current and current operating efficiency of multiple motors in the bucket wheel excavator.

[0044] S11, Data source for actual grid voltage: Actual grid voltage at the bucket wheel excavator connection point. Real-time monitoring is performed by three-phase voltage sensors installed in the high-voltage incoming cabinet of the bucket wheel excavator or on the low-voltage side of the transformer. The voltage sensors collect voltage signals and transmit them to the central controller via fieldbus. This is represented as:

[0045] ;in, This represents the real-time three-phase voltage vector of the power grid. This represents the instantaneous voltage values ​​of phases A, B, and C. Indicates the sampling time.

[0046] S12, Real-time Load Current Acquisition of Multiple Motors: The three-phase load current of each motor is acquired by a Hall effect current sensor installed on its power cable. The acquired three-phase current signal is transmitted to the central controller, which calculates the load current amplitude of each motor. This is used for subsequent power scheduling. It is represented as:

[0047] ;in, Indicates the first The load current vector amplitude of the trolley motor. Indicates the first The instantaneous values ​​of the three-phase currents A / B / C of the trolley motor.

[0048] S13, Current motor operating efficiency The following steps were used to calculate the result:

[0049] S131, the current motor speed is acquired by a speed encoder installed on each motor shaft. ;

[0050] S132, based on the current load current Calculate the current load rate based on the motor's rated parameters. : ;in, Indicates the first Rated current of the motor;

[0051] S133, in the efficiency-load rate-speed three-dimensional MAP mapping table for motor model matching, using the current... and Interpolation improves runtime efficiency: ;in, Indicates the first Taiwan motor's current operating efficiency, This represents a three-dimensional interpolation function. To obtain the operating efficiency of a motor based on its current load rate and speed, a three-dimensional efficiency mapping table (MAP) needs to be constructed. This table, based on experimental or manufacturer-provided data for the motor model, records the correspondence between different load rates, speeds, and corresponding efficiency values. The goal of three-dimensional interpolation is to find the efficiency value in this discrete data table based on the actually measured load rate. and rotational speed Estimate the efficiency value corresponding to its location. .

[0052] The entire interpolation process includes the following key steps:

[0053] 1. Organize the motor efficiency curves into a three-dimensional data table:

[0054] Horizontal axis: Load factor;

[0055] Vertical axis: Rotational speed;

[0056] Depth axis: corresponding operating efficiency;

[0057] The data can be viewed as a table of function values ​​on a two-dimensional grid.

[0058] Bilinear interpolation is used, which is suitable for situations where both load rate and speed are between two adjacent points.

[0059] 2. Let the current collected value be... The goal is to solve The bilinear interpolation process is as follows:

[0060] 2.1: Determine The load rate range [0.6, 0.8];

[0061] 2.2: Determine The speed range ,[900,1000];

[0062] 2.3: Obtain the efficiency values ​​of the four corresponding corner points:

[0063] ;

[0064] 2.4: First, interpolate along the rotational speed direction to obtain two intermediate values:

[0065] ;

[0066] ;

[0067] 2.5: Interpolate along the load rate direction to obtain the final efficiency: ;

[0068] The three-dimensional efficiency MAP can be generated through experimental calibration, simulation modeling, or efficiency curves provided by the motor manufacturer; the interpolation program is deployed in the central controller as part of the real-time scheduling system.

[0069] Table 1 Examples of 3D MAP Mapping

[0070]

[0071] Table Explanation:

[0072] The rows represent different load rates (0.2, 0.4, 0.6, 0.8, 1.0).

[0073] The columns represent different motor speeds (500rpm to 1250rpm).

[0074] The cell value represents the motor operating efficiency at the corresponding load rate and speed.

[0075] If collected:

[0076] Current load rate is 0.63, current speed is 940 rpm.

[0077] Then it can be found in this table:

[0078] Between the rows [0.6, 0.8],

[0079] Between columns [750, 1000],

[0080] Bilinear interpolation was performed on the four corresponding corner points (0.85, 0.89, 0.88, 0.92) to obtain the final interpolation efficiency value.

[0081] S2: Based on the deviation between the actual voltage and the rated voltage of the power grid, the real-time load current of each motor and its current operating efficiency, a dynamic virtual impedance value is calculated for each motor drive circuit through an improved multi-objective optimization function; the objective of the improved multi-objective optimization function is to simultaneously minimize the total power fluctuation of the system and maximize the overall operating energy efficiency of the system.

[0082] The dynamic virtual impedance value is calculated based on an improved multi-objective optimization function, while obtaining the actual voltage of the power grid. Real-time load current of each motor and current operating efficiency Subsequently, the central controller calculates the dynamic virtual impedance value corresponding to each motor drive circuit based on the following improved multi-objective optimization function. .

[0083] S21, the improved multi-objective optimization function is expressed as:

[0084] ;

[0085] in, To comprehensively optimize the objective function, This represents the total number of motors. For power fluctuation suppression weighting coefficients, To optimize the weighting coefficients for energy efficiency, This is the voltage regulation weighting coefficient. For the first Power fluctuation of each motor For the first Real-time operating efficiency of each motor. For the first The expected regulation of the bus voltage of each motor. For the first The optimal dynamic virtual impedance of each motor.

[0086] In multi-objective optimization, normalization or dimensional balancing of weights makes the objectives comparable. Weight coefficients. Dimensional compensation has been performed; power fluctuation amount The unit is W, which squares to W², therefore the main dimension of this term is power squared. have The unit is used to normalize the power fluctuation term. Dimensionless or related Value range matching, have The units are adjusted to balance the dimensions of the voltage term. Through weight dimension compensation, each term participates in optimization at the same numerical level.

[0087] Each optimization sub-item:

[0088] (1) Power fluctuation suppression term: This item is used to minimize the total power fluctuation of the system.

[0089] No. Power fluctuation of individual motors Defined as:

[0090] ;

[0091] ;

[0092] in: For the first The instantaneous active power of each motor is calculated by the central controller based on the real-time collected motor bus voltage and load current. The voltage signal comes from the three-phase voltage sensor installed in the high-voltage incoming cabinet of the bucket wheel excavator or the low-voltage side of the transformer, and the current signal comes from the Hall effect current sensor installed on the motor power cable. After the controller synchronously samples the voltage and current signals, it calculates the active power at that moment by combining the power factor. The historical average power of the motor is calculated using a sliding time window. The central controller performs a weighted average of the instantaneous power of the motor within a certain time sliding window. The length of this window can be set according to the working cycle of the bucket wheel excavator to reflect the recent load level and smooth out random fluctuations. The real-time effective value of the motor bus voltage is given by... The grid voltage signal measured by the three-phase voltage sensor is projected onto the bus side of each motor after phasor transformation and filtering by the central controller. If each motor is powered by a frequency converter, the value can be directly obtained by the bus voltage detection module fed back by the driver. The phase difference between current and voltage is obtained by the central controller through phase comparison of the voltage and current sampling waveforms of the motor. The phase angle is calculated by the time lag of the current waveform relative to the voltage waveform and used for real-time calculation of power factor and active power.

[0093] The above approach, by penalizing the square of the power deviation, aims to bring the system toward power balance and reduce load fluctuations among multiple motors.

[0094] (2) Energy efficiency optimization items: This item is used to maximize the overall energy efficiency of the system by penalizing inefficient motors, thus prioritizing power allocation to high-efficiency motors during the optimization process. This achieves optimal energy utilization across the entire system and improves the overall efficiency of the bucket wheel excavator under complex power grid conditions.

[0095] (3) Voltage regulation item: ;in: , Indicates the first Reference value for the bus voltage of each motor. Indicates the first The current bus voltage of each motor is formed by the difference between the two. This reflects the impact of the deviation between the actual grid voltage and the rated voltage on the adjustment of virtual impedance. By suppressing the voltage deviation caused by virtual impedance adjustment, excessive voltage changes can be avoided, leading to unstable motor operation.

[0096] The core improvements to the multi-objective optimization function are reflected in three aspects:

[0097] 1. Voltage regulation and suppression term: Achieve a balance between power optimization and operational stability. Traditional multi-objective optimization functions usually only consider minimizing system power fluctuations and maximizing energy efficiency. In the process of pursuing power balance or energy efficiency improvement, they tend to neglect the dynamic stability control of motor bus voltage, resulting in excessive voltage fluctuations in some motors, which affects operational safety and response performance.

[0098] Based on this, the present invention adds a voltage regulation suppression term. By constraining the voltage change amplitude caused by virtual impedance adjustment, the motor bus voltage is kept within a reasonable range, enhancing the safety constraints and steady-state controllability of the optimization process and avoiding over-adjustment.

[0099] 2. The definition of power fluctuation is improved, more accurately reflecting the dynamic imbalance of the system. In the power fluctuation item, this invention does not directly use the absolute value or standard deviation of the current motor power as a measure, but uses the difference between the real-time power and the historical average power to define the power fluctuation. It can dynamically reflect the trend of motor load changes rather than a single instantaneous value, and is more sensitive to sudden load disturbances and non-steady-state operation. It helps to identify potential risks in the system and avoid misjudging stable low power as an anomaly.

[0100] In summary, while introducing voltage regulation constraints to improve system stability, a more dynamically sensitive power fluctuation definition and an adaptive weighting strategy are used to achieve triple optimization of safety, energy efficiency and flexible control in the multi-motor power collaborative scheduling process.

[0101] S22, the constrained optimization problem is solved using an iterative algorithm. The main constraints include:

[0102] ;

[0103] ;

[0104] in, Indicates the permissible voltage regulation tolerance. Indicates the upper and lower limits of virtual impedance. The main determination is based on the motor model parameters and the output adjustment capability of its matching driver. Specifically, the following aspects can be evaluated and set:

[0105] First, the virtual impedance should not cause the bus voltage drop to exceed the allowable range of the motor. It is necessary to ensure that the output voltage remains within ±10% of the rated voltage under the action of impedance. Excessive impedance will limit the rate of current change, resulting in a slower motor response. Therefore, the virtual impedance should be less than the upper limit of the allowable dynamic change of current due to its electromagnetic inertia.

[0106] Second, the accuracy of the driver's PWM control determines the minimum impedance value that can be finely adjusted, i.e., the lower limit of the virtual impedance. The voltage regulation range supported by the driver determines the maximum magnitude that can be adjusted through virtual impedance, thus limiting the upper limit of virtual impedance.

[0107] Third, the lower impedance limit is set as the equivalent impedance value corresponding to the minimum effective output change that the driver control accuracy can achieve; the upper impedance limit is calculated based on the maximum allowable voltage drop ratio (not exceeding 10%) combined with the rated current.

[0108] In this invention, the Particle Swarm Optimization (IPSO) algorithm is used to solve constrained multi-objective optimization problems containing multiple physical objectives and constraints.

[0109] The objective optimization function is: ;

[0110] The termination condition is: ;

[0111] Particle swarm optimization basic structure: Each particle represents a vector of virtual impedance values ​​to be optimized.

[0112] Each particle has:

[0113] Location : Represents the current virtual impedance value;

[0114] speed : Indicates the search direction;

[0115] Individual optimal position ;

[0116] Global optimal position ;

[0117] The iterative solution process is as follows:

[0118] Step 1: Initialize the virtual impedance of each particle in the particle swarm. Initialize the velocity vector by randomly selecting values ​​within the allowed range. Set inertia weights Learning factors For each particle, calculate the objective function. And determine whether the constraints are satisfied; initialize the individual optimal pbest and the global optimal gbest.

[0119] Step 2, Particle Iterative Update: For each iteration, perform the following updates on all particles in sequence:

[0120] 2.1 Update the velocity vector:

[0121] This formula is used in the particle swarm optimization algorithm to update the search direction of each particle, indicating how the current particle will move.

[0122] The first option is to retain the speed (inertia) of the previous generation;

[0123] The second factor pushes the particle closer to its historical best position (individual experience).

[0124] The third factor is to push the particle closer to the optimal position of the entire swarm (swarm cooperation).

[0125] in, Represents a random number in the interval [0,1]. The inertia factor (using a linearly decreasing method to improve convergence speed) Indicates the first The particle in the first The speed of generation Indicates the first The particle in the first The position of the generation (corresponding to the first) (virtual impedance value of each motor) Indicates the first The impedance value updated for each particle Indicates the first The historical best position of each particle (individual optimal solution), Represents the position of the current global optimal position. The virtual impedance value corresponding to each motor. This is an inertia weighting factor used to balance global and local search capabilities. This serves as an individual learning factor, adjusting the speed at which particles approach their historical best. This is a group learning factor that adjusts the speed at which particles approach the global optimum. Index of the current iteration number This refers to the total number of particles or the total number of motors.

[0126] 2.2 Update location (impedance value): This formula represents the update method for the current particle (i.e., the virtual impedance value of the motor): based on the original position, the newly updated velocity is added to obtain a new candidate solution;

[0127] 2.3 Constraint Handling (Penalty Function Method): If a particle violates the constraint ( ), in its fitness function Add a penalty item above: ;in A large penalty coefficient is used to ensure that particles that violate the constraints do not become the optimal solution. Indicates the first The adjustment amount of the bus voltage of each motor (the difference between the expected value and the actual value). The maximum allowable deviation threshold for bus voltage; This is a penalty coefficient used to increase the fitness penalty for particles that violate voltage constraints. The original objective function value (without penalty) This is the fitness function value after adding the constraint penalty term, used for actual iterative selection;

[0128] Step 3, update individual best and global best: If the current particle's fitness is better than its historical best, then update pbest; if the current particle's fitness is better than the global best, then update gbest.

[0129] Step 4, Termination Judgment: The maximum number of iterations is met; or the global optimal objective function value converges to the set threshold; or there is no significant improvement after several iterations.

[0130] Step 5, Output the optimal solution: Output the virtual impedance vector corresponding to gbest. This is the optimal dynamic virtual impedance that should be applied to each motor in the current control cycle.

[0131] S3: Convert the dynamic virtual impedance value into a corresponding collaborative power command and send it to the power driver of each motor. By adjusting the driver output, it is equivalent to applying virtual impedance, so that the bus voltage of each motor is redistributed, and the power of multiple motors is balanced under non-ideal grid voltage.

[0132] After achieving the optimal dynamic virtual impedance values ​​for each motor After the solution is obtained, the following four processes are executed in sequence to achieve voltage correction, power command generation, driver adjustment and power distribution control, thereby realizing power balance among multiple motors.

[0133] S31, Conversion Calculation Process, Virtual Impedance to Voltage Correction: For the first... Each motor, and its corresponding dynamic virtual impedance value Converted to voltage correction amount The calculation is as follows: ;in, For the first Voltage correction amount for each motor. For the first Real-time load current vector amplitude of each motor For the first The dynamic virtual impedance of each motor is obtained by S2 optimization. This formula is based on Ohm's law and converts impedance changes into a quantity that can be directly used for voltage regulation.

[0134] The purpose of S31 is to convert the optimized dynamic virtual impedance value into an executable quantity in actual control, namely, a voltage correction value, so that the subsequent driver can perform control. Based on Ohm's law, namely: Voltage = Current × Impedance; here, voltage refers to the desired adjustment, that is, a voltage correction value artificially introduced into the motor bus voltage to simulate the voltage drop effect caused by a virtual resistor in series in the circuit. The essence of this conversion is to transform the control concept of impedance regulation into an actual control command for voltage adjustment. Because in actual engineering, we cannot frequently insert resistors into the drive circuit, but we can adjust the voltage output to make the motor feel an impedance at its front end. In other words, by issuing a set value slightly lower than the normal voltage to the driver, it's as if a portion of the voltage has dropped across the series impedance, achieving the purpose of power redistribution.

[0135] S32, Command Generation Process: Generating Coordinated Power Commands. Since a variable virtual impedance element cannot be physically inserted into the actual circuit, its effect is simulated indirectly by adjusting the driver's output voltage to make the motor perceive impedance insertion. Based on the aforementioned voltage correction... , generate the first Corrected voltage setting value for each motor This is used to indicate the direction of driver output adjustment. The calculation is as follows: ;in, For the first The reference voltage setting value for each motor under ideal power grid conditions serves as the default output reference value for the driver. This represents the voltage command value after virtual impedance correction. This formula reflects the voltage drop behavior caused by simulated series impedance in actual control and is the core embodiment of the virtual impedance control concept. The subtraction operation in the above formula essentially replaces series impedance adjustment with driver voltage regulation, forming the core of the virtual impedance concept; real-time current is used in the calculation. It reflects the current load status; the voltage drop is achieved through subtraction, which is equivalent to inserting an impedance element in the motor circuit; by setting the voltage correction value, the driver is guided to output as needed, ultimately affecting the actual voltage and power obtained by the motor.

[0136] S33, Command issuance process, real-time transmission to the power driver: the corrected voltage setting value The data is sent in real time to the power driver of the corresponding motor via the fieldbus interface, and the driver performs the corresponding voltage output adjustment operation.

[0137] S34, Equivalent impedance realization process, PWM output realizes virtual impedance effect: Power driver receives voltage setting command Then, the output voltage is adjusted by internal PWM modulation (regulation driver) so that the actual voltage at the motor input terminal is equal to the set value: From an electrical equivalence perspective, this process is equivalent to connecting an impedance value in series in the motor drive circuit. The virtual component is a structure that does not require hardware connection of actual impedance components, but is implemented through software instructions and internal driver control.

[0138] S35, through the above mechanism, different motors obtain differentiated supply voltages based on their current load status and optimization calculation results:

[0139] For motors with heavy loads, due to their virtual impedance The smaller value results in a smaller voltage correction amount. The smaller the reference voltage, the better. The voltage is higher, resulting in a higher actual supply voltage and thus more power supply.

[0140] For motors with lighter loads, The larger the voltage, the greater the voltage correction, resulting in a corresponding decrease in the supply voltage and thus a reduction in power supply.

[0141] Ultimately, a load-voltage coupling mechanism based on virtual impedance regulation is formed, enabling dynamic power balance distribution among multiple motors under grid voltage fluctuations or asymmetry conditions.

[0142] S4: Record and associate the real-time operating position information of the bucket wheel excavator with the sequence of dynamic virtual impedance values ​​calculated in S2, and construct and continuously update an operating trajectory-virtual impedance mapping database; when the bucket wheel excavator runs to the same or similar operating position again, the dynamic virtual impedance value is preloaded based on the operating trajectory-virtual impedance mapping database as the initial value or feedforward instruction calculated in S2.

[0143] S41, Position-Impedance Correlation Mechanism: During the operation of the bucket wheel excavator, the system uses the timing data synchronization module built into the controller to fuse data from three types of position sensors to generate unique operating coordinates.

[0144] ;in, The travel position of the bucket wheel excavator along the track is obtained by the encoder of the traveling mechanism. The slewing angle of the bucket wheel excavator is obtained by a slewing encoder. The pitch angle of the bucket wheel excavator is indicated by a pitch encoder. Represents the three-dimensional position coordinates of the operation (uniquely identifying an operation posture);

[0145] This coordinate is compared with the corresponding sequence of dynamic virtual impedance values ​​of each motor at the same timestamp. Perform corresponding storage to realize the timing correlation between location and impedance.

[0146] S42, Mapping Database Construction Method: To avoid redundant data accumulation, a compression strategy based on key location points is adopted to construct the trajectory database. Specifically:

[0147] The continuous trajectory is discretized into a finite number of critical operation locations. ;

[0148] For each location point, record multiple sets of virtual impedance value sequences over multiple work cycles. ;

[0149] Density clustering algorithm is used to extract cluster centers to form representative impedance patterns:

[0150] ;in, Represents the cluster center. Key location points A representative set of virtual impedance values ​​is used for subsequent calls. At a critical operating point of the bucket wheel excavator... The system recorded multiple sets of dynamic virtual impedance values ​​obtained during various runs. To simplify storage and extract typical impedance behaviors, density clustering was used to classify these data, and then the most representative center point in each class was taken as the representative impedance value at that location. .

[0151] Assuming at location point At this location, the following three sets of impedance values ​​were recorded:

[0152] First time: [0.9, 1.2, 1.0];

[0153] Second time: [0.8, 1.3, 1.1];

[0154] Third time: [0.85, 1.25, 1.05];

[0155] If the clustering results indicate that they belong to the same cluster, then:

[0156] The average value of this set represents the cluster center that indicates the impedance behavior at that location.

[0157] The core of S42 is to establish an efficient and reliable mapping database between operating positions and virtual impedances, facilitating subsequent experience preloading and feedforward control. During continuous operation, bucket wheel excavators generate a large amount of continuous operating trajectory and impedance data. If position-impedance combinations are recorded every few seconds, tens of thousands of records could be generated daily, consuming significant storage resources and reducing system query efficiency over long-term operation. Therefore, it is necessary to compress this data into key points, retaining only representative positions and corresponding impedances to reduce data volume while preserving representative characteristics of the control effect. Within the bucket wheel excavator trajectory, some positions are frequently traversed, of high importance, or undergo drastic state changes. These points are defined as critical operating positions, such as operating points with fixed stacking angles or fixed bucket wheel lifting heights; or points near the critical points of slewing and pitching movements. The entire trajectory is discretized into these key points, and impedance values ​​are recorded and analyzed only at these points.

[0158] Even at the same location, the virtual impedance values ​​will vary slightly across multiple work cycles due to differences in the power grid, motor load, and material conditions. If all these historical values ​​were directly saved, the data volume would still be enormous. However, if the core features of these values ​​could be extracted, then only a representative impedance value could be retained instead of all the values. This is the role of density clustering algorithms: analyzing the impedance values ​​recorded across multiple work cycles at the same location, grouping similar values ​​into one category, and finding the center of these values. This center point is considered to be the typical impedance pattern commonly used at this location.

[0159] In summary:

[0160] The original work trajectory is very long, so key location points are extracted through discretization;

[0161] Each location point has impedance records from multiple operating cycles, which are then clustered and compressed into representative impedance values.

[0162] The resulting cluster centers are stored in a database as a data source for subsequent "preloading" or "learning updates".

[0163] S43, Mapping Database Update Strategy: When the bucket wheel excavator runs to a recorded key location again... At that time, the currently calculated virtual impedance value will be... Compared with historical representative values The data is then merged and updated using a weighted average method: ;in, , where is the weighting coefficient. The frequency of access to the timestamps and locations of new data is positively correlated, reflecting the adaptability of the update; this update strategy enables the impedance database to achieve progressive self-learning capability.

[0164] S43 updates the representative virtual impedance value of each key location point through a progressive learning strategy. When the bucket wheel excavator passes through a recorded key location point again, it not only recalculates the new virtual impedance value, but also merges the latest calculation result into the original database, so that the database content is continuously fine-tuned and optimized according to the actual operation.

[0165] Existing representative value: The database originally stored a representative virtual impedance value for this location;

[0166] Newly acquired value: The new impedance value obtained during the current operation based on parameters such as real-time voltage, current, and efficiency.

[0167] By using a weighted average method, existing experience and the current new value are combined to update a new representative value.

[0168] S44, Preloading Execution Process: When the bucket wheel excavator reaches any position Invoke the preloading mechanism:

[0169] The k-nearest neighbor algorithm is used in the database to find the element closest to the current coordinates. Key location points:

[0170] ; This represents the k-nearest neighbor algorithm. This indicates the current coordinates of the bucket wheel excavator's operating position;

[0171] Extract the corresponding representative impedance set Perform a weighted average to calculate the preloaded impedance value:

[0172] ;in, For the first The weighting coefficients for the nearest neighboring locations are set using inverse distance normalization. This is the set of preloaded virtual impedance values ​​for the current location. This indicates the number of nearest neighbor locations that participate in the weighted average. Indicates the first A set of representative virtual impedance values ​​for key historical locations.

[0173] The purpose of preloading is to provide an estimated value (preloading value) in advance when the bucket wheel excavator moves to a new working position, instead of calculating the virtual impedance from scratch.

[0174] The specific steps are as follows:

[0175] Obtain the current three-dimensional operating position of the bucket wheel excavator;

[0176] Find the few recorded locations that are closest to this location in the mapping database using the k-nearest neighbor algorithm;

[0177] Extract representative virtual impedance values ​​corresponding to these similar locations;

[0178] These impedance values ​​are weighted according to their similarity (the closer the values, the higher the weight).

[0179] These values ​​are weighted and averaged to obtain a pre-loaded set of virtual impedance values, which can be used to guide the next step of control in advance.

[0180] This is equivalent to allowing the system to make a judgment in advance, providing a historical experience value as a reference before fully understanding the current operating conditions, thereby improving response speed and reducing the real-time calculation burden.

[0181] S45, Feedforward-Feedback Fusion: While the preloading in S44 can improve speed, it also carries the risk that historical experience may not always be applicable to the current grid or equipment state. Therefore, preloaded values ​​are not used blindly; instead, they are fused with real-time optimization calculation results to achieve a balanced control between feedforward and feedback. Specifically, in the virtual impedance optimization calculation of S2, a feedforward-feedback fusion strategy is introduced: ;in, The feedforward weighting coefficient is adaptively set based on the current grid voltage fluctuation; if the grid voltage fluctuation is less than the set fluctuation threshold... Then increase If the grid voltage fluctuates drastically, then reduce... Increase the proportion of real-time calculated values. This represents the set of impedance values ​​recalculated in S2 based on the current state.

[0182] Grid voltage fluctuation threshold The value is taken as 2% to 5% of the rated voltage. When the grid voltage fluctuates within this range, it is considered that the voltage is relatively stable, indicating that the load disturbance and power supply fluctuation are small, and the preloaded impedance value of the historical data has high reliability. When the fluctuation exceeds this range, it indicates that there is a significant voltage imbalance or instantaneous disturbance, the historical model may be invalid, and more reliance on real-time calculation results is required.

[0183] Feedforward weights The value range is 0.3 to 0.8. When the grid voltage fluctuation is < At that time, the system environment is stable, and the preloaded impedance has high reliability. Take the larger value to emphasize the feedforward role of historical experience;

[0184] When the grid voltage fluctuation is ≥ At that time, the system fluctuated significantly. Take a smaller value to enhance the sensitivity of real-time feedback.

[0185] The specific mechanism of the S45 is:

[0186] In S2, a new impedance value will still be calculated based on the current real-time voltage, current, efficiency, etc.

[0187] The system determines the fusion method based on the current grid voltage fluctuations:

[0188] If the voltage is stable, it means the environment is similar to historical conditions, so the preloaded value should be trusted more.

[0189] If the voltage fluctuates drastically, it indicates that the current situation differs greatly from historical data, and more reliance should be placed on real-time calculated values.

[0190] The final virtual impedance is generated by proportionally weighting and fusing the preloaded value and the real-time value, with the fusing weights being... (i.e., the feedforward weighting coefficient) will be adjusted automatically.

[0191] This mechanism enables the control system to be both experience-driven and reality-corrected, relying on historical data to improve efficiency under stable operating conditions, and enhancing real-time perception to ensure safety under fluctuating operating conditions.

[0192] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0193] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A power supply and distribution management method for bucket wheel excavators based on multi-motor coordination and power balance, characterized in that, Includes the following steps: S1: Real-time monitoring of the actual voltage of the power grid at the connection point of the bucket wheel excavator, and acquisition of the real-time load current and current operating efficiency of multiple motors in the bucket wheel excavator; S2: Based on the deviation between the actual voltage and the rated voltage of the power grid, the real-time load current of each motor and its current operating efficiency, a multi-objective optimization function is established to calculate a dynamic virtual impedance value for each motor drive circuit; the objective of the multi-objective optimization function is to simultaneously minimize the total power fluctuation of the system and maximize the overall operating energy efficiency of the system. S3: Convert the dynamic virtual impedance value into a corresponding collaborative power command and send it to the power driver of each motor. By adjusting the driver output, it is equivalent to applying virtual impedance, so that the bus voltage of each motor is redistributed, and the power of multiple motors is balanced under non-ideal grid voltage. S4: Record and associate the real-time operating position information of the bucket wheel excavator with the sequence of dynamic virtual impedance values ​​calculated in S2, and construct and continuously update an operating trajectory-virtual impedance mapping database; when the bucket wheel excavator runs to the same or similar operating position again, the dynamic virtual impedance value is preloaded based on the operating trajectory-virtual impedance mapping database as the initial value or feedforward instruction calculated in S2.

2. The bucket wheel excavator power supply and distribution management method based on multi-motor coordination and power balance according to claim 1, characterized in that, The actual voltage of the power grid is obtained through the three-phase voltage sensor of the bucket wheel excavator, and the three-phase voltage sensor continuously collects the actual voltage signal of the power grid. The real-time load current of the multiple motors is obtained by a Hall effect current sensor installed in each motor. The three-phase current of each motor is detected in real time, and the magnitude of the current vector is calculated as the real-time load current.

3. The bucket wheel excavator power supply and distribution management method based on multi-motor cooperation and power balance according to claim 1, characterized in that, S1 further includes obtaining the real-time motor speed through a motor speed encoder, calculating the current load rate of the motor in combination with the real-time load current, querying a pre-stored three-dimensional MAP of efficiency-load rate-speed for the current model of motor, and interpolating to obtain the current operating efficiency.

4. The bucket wheel excavator power supply and distribution management method based on multi-motor coordination and power balance according to claim 1, characterized in that, S2 establishes a multi-objective optimization function including power fluctuation suppression, energy efficiency optimization, and voltage regulation terms; the multi-objective optimization function is solved by a constrained iterative optimization algorithm to obtain a set of optimal dynamic virtual impedance values.

5. The bucket wheel excavator power supply and distribution management method based on multi-motor coordination and power balance according to claim 4, characterized in that, The power fluctuation suppression term is used to minimize the total power fluctuation of the system, which is achieved by penalizing the degree to which the power of each motor deviates from its historical average value; The energy efficiency optimization item is used to maximize the overall operating energy efficiency of the system, which is achieved by penalizing the operating state of inefficient motors; The voltage regulation item is used to limit the magnitude of virtual impedance regulation, so as to avoid excessive voltage regulation from affecting the stability of motor operation.

6. The bucket wheel excavator power supply and distribution management method based on multi-motor coordination and power balance according to claim 5, characterized in that, The iterative optimization algorithm adopts the particle swarm optimization algorithm. It initializes the particle swarm, iteratively updates the particle velocity and position, introduces a penalty function for voltage regulation deviation to handle the constraints, and dynamically records the individual optimal and global optimal impedance value vectors. Finally, it outputs the optimal dynamic virtual impedance configuration result that meets the constraints of power fluctuation suppression, energy efficiency optimization and voltage stability.

7. The bucket wheel excavator power supply and distribution management method based on multi-motor cooperation and power balance according to claim 1, characterized in that, S3 specifically includes: S31, Conversion Calculation Process: For each motor, its dynamic virtual impedance value is converted into the corresponding voltage correction amount; S32, Instruction generation process: Based on the voltage correction amount, generate the current motor's cooperative power instruction, which is the corrected voltage setting value; S33, Command issuance process: The coordinated power command is issued to the power driver of the corresponding motor in real time; S34, Equivalent impedance realization process: After receiving the coordinated power command, the power driver adjusts the output voltage through its internal regulating driver so that the actual voltage of the motor input terminal is equal to the coordinated power command, which is equivalent to the actual impedance with the series impedance value of the dynamic virtual impedance value in the motor drive circuit, thereby realizing the redistribution of the voltage of each motor bus.

8. The bucket wheel excavator power supply and distribution management method based on multi-motor coordination and power balance according to claim 7, characterized in that, S3 further includes voltage differential allocation, which, through S31-S34, enables: The smaller the dynamic virtual impedance value of a motor that is in an overload trend, the smaller its voltage correction amount, and the higher the bus voltage and power supply it can obtain. Motors under light load tend to have a larger dynamic virtual impedance value, which in turn results in a larger voltage correction, allowing them to obtain a relatively lower bus voltage and power supply, thus achieving power balance among multiple motors under non-ideal grid voltage conditions.

9. The bucket wheel excavator power supply and distribution management method based on multi-motor cooperation and power balance according to claim 1, characterized in that, S4 specifically includes: S41, through the timing data synchronization module in the bucket wheel excavator control system, the bucket wheel excavator's traveling position, rotation angle and pitch angle are fused to generate a unique three-dimensional working position coordinate, and the three-dimensional working position coordinate is associated with and stored with the dynamic virtual impedance value sequence of each motor under the same timestamp; S42, adopts a data compression and storage strategy based on key location points, discretizes the continuous operation trajectory into multiple key location points, and stores a set of representative dynamic virtual impedance values ​​for each key location point. The representative set of dynamic virtual impedance values ​​includes the cluster center values ​​obtained after performing density clustering analysis on multiple sets of impedance values ​​recorded at the same location point in multiple operation cycles. S43, when the bucket wheel excavator passes through the recorded key location point again, the newly calculated dynamic virtual impedance value is compared and fused with the cluster center value stored in the database, and the impedance value set of the key location point is updated using a weighted average algorithm. The weight of the newly calculated dynamic virtual impedance value depends on the timeliness of the new data and the access frequency of the operation location point. S44, preload execution: When the bucket wheel excavator reaches a certain working position, the k-nearest neighbor algorithm is used to find several historical position points in the database that are closest to the current position. The dynamic virtual impedance values ​​corresponding to the several historical position points are weighted and averaged, and the calculation result is transmitted to S2 as the preload value.

10. The bucket wheel excavator power supply and distribution management method based on multi-motor cooperation and power balance according to claim 9, characterized in that, S4 also includes a feedforward-feedback fusion mechanism, specifically including in S2, using the pre-loaded dynamic virtual impedance value in the database as the basis for the feedforward command, and fusing it with the correction value calculated based on real-time data. During fusion, a feedforward weighting coefficient is set, which is adaptively adjusted according to the stability of the current grid voltage. When the grid voltage fluctuation is less than the threshold, increase the weight of the preload value; When the grid voltage fluctuates drastically, increase the weight of the real-time calculated value.

Images