High-speed dewatering dynamic balance control method for large-capacity washing and dyeing machine

Abstract

This invention relates to the field of laundry and dyeing machine control technology, and discloses a dynamic balance control method for high-speed dehydration in a large-capacity laundry and dyeing machine. The method includes the following steps: acquiring continuous vibration data based on real-time rotational speed data using a segmented sampling strategy, converting it into discrete sampling signals, filtering the discrete sampling signals to locate the eccentric phase angle, extracting the triaxial acceleration signal to calculate the triaxial composite amplitude, extracting subsequent observation data after passing through a fluid stability window to calculate the phase drift rate, constraining the output target water injection quality based on this, constructing a PI control topology with a vibration suppression loop and a capacity constraint loop in parallel, superimposing forward and reverse duty cycles to generate a total control command, calculating the real-time proportional gain, and using a discrete accumulation mechanism to combine the total control command with the target water injection quality to convert it into a pulse water injection command for balanced closed-loop regulation until the laundry and dyeing machine reaches a steady state. This invention filters out broadband noise, eliminates phase observation lag and dead zone of the water replenishment valve, and prevents resonance and overflow.

Description

Technical Field

[0001] This invention relates to the field of washing and dyeing machine control technology, specifically a dynamic balance control method for high-speed dehydration in large-capacity washing and dyeing machines. Background Technology

[0002] High-speed dehydration in high-capacity laundry machines is prone to eccentric vibration due to the random distribution of flexible loads. Traditional passive counterweight solutions increase the overall weight and installation cost, while suspended damping solutions are susceptible to fatigue failure due to long-term exposure to alternating stress. To overcome mechanical limitations, existing technologies have introduced dynamic water injection balancing schemes, but these still have shortcomings.

[0003] Conventional methods often use fixed-frequency sampling and direct reading of acceleration peak values, which makes it difficult to remove broadband noise and ignores the nonlinear growth property of centrifugal force, resulting in distortion of eccentric phase and amplitude determination. Their compensation logic does not consider the fluid redistribution delay and the dead zone of the water supply valve response. Premature observation before reaching steady state can easily cause phase misjudgment, and weak control commands are also easily filtered by the dead zone. In addition, existing algorithms lack quantification of the non-steady-state drift characteristics of the load and cavity volume constraints, which can easily lead to overcompensation resonance and cavity overflow when the load distribution changes drastically. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a dynamic balance control method for high-speed dehydration in large-capacity washing and dyeing machines. This method solves the problem that existing technologies lack a collaborative response mechanism to broadband mechanical noise, fluid distribution delay, and actuator dead zones, making it difficult to achieve accurate and safe dynamic balance control under high-speed, unsteady conditions.

[0005] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a dynamic balance control method for high-speed dehydration in large-capacity washing and dyeing machines, comprising the following steps: The incremental encoder acquires real-time speed data greater than or equal to the initial speed threshold, triggering the triaxial measurement node to acquire continuous vibration data and output analog current signals; Based on real-time speed data, the sampling frequency is increased for different speed ranges, segmented sampling is implemented, and the analog current signal is converted into a discrete sampling signal. The Kalman filter is used to reduce noise in the discrete sampled signal and extract the peak point sequence to locate the off-center phase angle. The basic vibration displacement amplitude is obtained by dividing the extracted axial acceleration by the square of the current mechanical angular velocity, multiplying it by the preset amplitude correction coefficient of the corresponding speed sub-interval, and then converting and synthesizing the triaxial composite amplitude. After the target water supply valve is activated, wait for a preset rotation period as a fluid stability window, extract subsequent observation data, and calculate the phase drift rate by the ratio of the absolute value of the eccentric phase difference to the rotation period time within adjacent effective observation periods. Subtract the product of the fluid penalty coefficient and the phase drift rate from the constant 1, take the larger value of the cutoff zero value, and multiply it by the reference perturbation mass to dynamically constrain the output target water injection mass. A PI control topology consisting of a vibration suppression loop and a capacity constraint loop in parallel is constructed. The positive valve duty cycle, which outputs when the triaxial composite amplitude approaches the target value of zero amplitude, is superimposed with the reverse duty cycle output when the maximum water injection capacity threshold is reached to generate the total control command. The real-time proportional gain is calculated by exponentiating the product of the negative weight sensitivity coefficient and the phase drift rate with the natural constant as the base and multiplying it by the initial calibration gain. The total control command is combined with the target water injection mass to convert it into the theoretical water replenishment mass. The discrete accumulation mechanism accumulates and stores the theoretical water replenishment mass into the flow integral logic when it is less than the minimum opening equivalent. When it is greater than or equal to the minimum opening equivalent, the target water replenishment valve is located by combining the eccentric phase angle to issue a pulse water injection command and deduct the mass equivalent. Continuously perform closed-loop verification and control until a steady state is reached, and persist global operating condition data to the time series database.

[0006] This invention provides a dynamic balance control method for high-speed dehydration in large-capacity washing and dyeing machines. It has the following beneficial effects: 1. This invention employs segmented frequency conversion sampling and Kalman filtering to convert the acceleration signal into a composite amplitude at the displacement level, thereby filtering out mechanical noise and eliminating centrifugal force nonlinearity judgment errors.

[0007] 2. This invention sets a fluid stability window to eliminate phase observation lag, and uses flow integral logic to overcome the valve action dead zone, thereby achieving precise pulse water injection and reducing transmission mechanism losses.

[0008] 3. This invention constructs a competitive dual-loop control strategy, dynamically adjusts the compensation step size and proportional gain based on the phase drift rate to prevent resonance, and uses a capacity constraint loop to truncate commands to avoid overflow. Attached Figure Description

[0009] Figure 1 This is a flowchart of the dynamic balance control method for high-speed dehydration of a large-capacity washing and dyeing machine according to the present invention. Figure 2 This is a flowchart of the vector perturbation feature extraction and variable step size calculation logic of the present invention; Figure 3 This is a flowchart of the observation gradient-driven competitive dual-loop balance control logic of the present invention. Detailed Implementation

[0010] 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.

[0011] Please see the appendix Figure 1 This invention provides a dynamic balance control method for high-speed dehydration in a large-capacity washing and dyeing machine, comprising the following steps: Step S1: Multi-dimensional signal frequency conversion sampling. During the high-speed dehydration stage, based on the input real-time rotational speed data, a segmented sampling strategy is used to acquire the triaxial vibration signal. This step specifically includes the following sub-steps: Step S101: During the high-speed dehydration stage, the incremental encoder performs real-time speed measurement on the washing and dyeing machine to obtain real-time rotational speed data. In the specific calculation process, the incremental encoder captures the pulse signals generated by the spindle rotation, and the control device calculates the ratio of the pulse increment to the corresponding time scale to determine the rotational speed. To ensure the robustness of the algorithm logic, the control program pre-determines the state of the time scale before calculation. If the time scale is detected to be close to zero (e.g., the spindle is stopped or the sensor feedback is abnormal), the real-time speed data is reset to avoid the denominator being zero, which could cause the algorithm to crash.

[0012] When the real-time rotational speed data is greater than or equal to the set initial rotational speed threshold (e.g., 100 r / min), the multi-dimensional signal sampling program is triggered. This initial rotational speed threshold is set based on the critical rotational speed at which the flexible clothing load begins to adhere to the wall and generate effective centrifugal force. In actual judgment, a double-limit hysteresis logic can be used to prevent the sampling program from frequently starting and stopping due to small fluctuations in rotational speed at the critical point.

[0013] Step S102: After triggering the multi-dimensional signal sampling program, continuous vibration data is acquired by a triaxial measurement node located at the transmission bearing. In this embodiment, the triaxial measurement node is rigidly fixed along the three orthogonal directions of the transmission bearing: the X-axis, Y-axis, and Z-axis. To ensure the physical accuracy of subsequent spatial vector synthesis, the signals acquired by the triaxial measurement node are synchronously driven by the same reference clock to achieve time alignment. Under the constraint of hardware response parameters (e.g., 10μs), the high-speed sampling module reads the continuous vibration data according to the hardware sampling rate (e.g., 50kHz).

[0014] During the reading process, the high-speed sampling module uses a built-in anti-aliasing filter to preprocess the original sequence and finally outputs an analog current signal that meets the current signal range (e.g., 4-20mA), thereby effectively suppressing electromagnetic interference in the industrial field.

[0015] Step S103: The control program determines and implements a segmented sampling strategy based on real-time rotational speed data. The core of this strategy is to ensure that the useful vibration signal is always within the Nyquist bandwidth, thereby preserving the complete frequency gradient characteristics at different rotational speeds.

[0016] Specifically, when the real-time speed data is in the first speed range (e.g., [100, 200) r / min), the control sampling frequency is set to the first sampling frequency (e.g., 200Hz); when it is in the second speed range (e.g., [200, 300) r / min), it is set to the second sampling frequency (e.g., 400Hz); when it is in the third speed range (e.g., [300, 400) r / min), it is set to the third sampling frequency (e.g., 600Hz); and when it is in the fourth speed range (e.g., [400, 500) r / min), it is set to the fourth sampling frequency (e.g., 800Hz). During range switching, the boundaries of each speed range adopt a left-closed, right-open inclusion logic to ensure the continuity of the sampling frequency switching.

[0017] Based on the segmented sampling frequency determined above, the high-speed sampling module converts the 50kHz analog current signal generated in step S102 into a discrete sampling signal and outputs it by performing equal-interval decimation or moving average downsampling.

[0018] Step S2: Data Filtering and Eccentricity Parameter Reference Conversion. The discrete sampled signal is filtered to locate the eccentricity phase angle, and then converted into a three-axis composite amplitude based on the three-axis acceleration signal components for balance control judgment. This step specifically includes the following implementation path: Step S201: Within the algorithm response constraint interval (e.g., 1ms-3ms), the control program receives the discrete sampling signal generated in step S103. Based on the state-space estimation principle, a Kalman filter is invoked to filter and reduce noise on the discrete sampling signal of the triaxial measurement node within one complete mechanical rotation cycle.

[0019] The discrete time step of the Kalman filter is dynamically resampled and aligned according to changes in real-time rotational speed data to ensure that the filter gain remains locked to the unbalanced vibration fundamental frequency signal at different sampling frequencies. In the specific processing, the Kalman filter effectively eliminates high-frequency disturbances caused by motor electromagnetic pulses and bearing friction by recursively estimating the vibration state of the washing and dyeing machine, and outputs a filtered discrete signal.

[0020] Step S202: Call the phase extraction algorithm to extract the peak point sequence from the filtered discrete signal generated in step S201. To avoid false triggering caused by a single noise spike, the peak point sequence extraction adopts multi-point cross-correlation verification logic. That is, after identifying the local extrema of the time-domain waveform, it is necessary to combine the same phase points of adjacent periods for weighted judgment.

[0021] The eccentric phase angle of the washing and dyeing machine is located and recorded based on the peak point sequence. During this process, the control program calculates the time offset of the peak point relative to the zero-position pulse provided by the incremental encoder on the time-domain coordinate axis and converts it into a spatial angle. This eccentric phase angle establishes a geometric mapping relationship between the spindle rotation coordinate system and the physical load eccentricity point.

[0022] Step S203: For the eccentricity parameters required for balance control, a space vector algorithm is used for benchmark conversion. This algorithm is based on the principle of particle dynamics, mapping the dynamic acceleration signal to a quasi-static displacement offset, thereby eliminating the amplitude nonlinearity caused by the increase of centrifugal force with the square of the rotational speed, ensuring that the control gain at different rotational speeds is physically comparable. The calculation logic is to extract the triaxial acceleration signal from the filtered discrete signal generated in step S201.

[0023] After obtaining the components of acceleration in each axis, the control program divides each axial acceleration in the triaxial acceleration signal by the square of the current mechanical angular velocity calculated based on real-time rotational speed data. Before performing the division operation, the control program monitors the real-time rotational speed data. If the rotational speed is determined to be below the preset safety detection limit or the spindle is detected to have stopped rotating, the division calculation logic is disabled to prevent the denominator from approaching zero and causing overflow. Through this normalization process, the basic vibration displacement amplitudes of the X, Y, and Z axes are obtained.

[0024] Step S204: Further combining the structural resonance characteristics of the washing and dyeing machine in different speed ranges, a preset amplitude correction coefficient is applied to the basic vibration displacement amplitude generated in step S203. This embodiment takes into account the difference in dynamic stiffness of the washing and dyeing machine's suspension structure at specific frequencies. Within the four speed sub-ranges corresponding to the global operating range (e.g., [100, 500] r / min), namely the first sub-range (e.g., [100, 200] r / min), the second sub-range (e.g., [200, 300] r / min), the third sub-range (e.g., [300, 400] r / min), and the fourth sub-range (e.g., [400, 500] r / min), progressively increasing correction coefficients are preset.

[0025] The amplitude correction factor includes a first correction factor, a second correction factor, a third correction factor, and a fourth correction factor. The amplitude correction factor is determined based on the structural amplification factor obtained from the centrifugal force simulation test. For example, the value range is between [1.0-1.5].

[0026] The corrected triaxial components are obtained by multiplying the basic vibration displacement amplitude by the amplitude correction coefficient of the corresponding sub-interval. Finally, based on the Euclidean norm principle, the corrected triaxial components are synthesized using a space vector algorithm, that is, the arithmetic square root of the sum of squares of the triaxial components is calculated, and the final triaxial composite amplitude is output.

[0027] See appendix Figure 2 Step S3: Vector perturbation feature extraction and variable step size calculation. After the fluid stability window, the vector observation gradient parameters are extracted, and the target injection quality for balance control is dynamically calculated and output by measuring the phase drift rate. The specific implementation path of this step is as follows: Step S301: The control program acquires the triaxial composite amplitude generated in step S204 and the eccentric phase angle generated in step S202 for multiple consecutive rotation cycles. Based on fluid dynamics transport theory, the spreading speed of fluid in a high-speed rotating cavity is affected by both liquid viscosity and centrifugal acceleration. To avoid misjudgment by the control logic during fluid dynamic steady-state switching, the control program forcibly inserts a fluid stability window between two valid gradient observation actions.

[0028] Specifically, after the target water supply valve performs its action, the control program suspends the current observation logic and forces the user to wait for a number of preset rotation cycles (e.g., set to [5-10] cycles).

[0029] The set rotation period is determined based on the physical inner diameter of the balance ring and the momentum diffusion time of the fluid under centrifugal field calibration, aiming to ensure that the extracted vibration characteristics truly reflect the mass distribution state after water replenishment. After the injected water flow completes redistribution under centrifugal force, the control program extracts the subsequent observation data corresponding to the next cycle.

[0030] Step S302: The control program calculates the phase drift rate of the washing and dyeing machine based on the phase components in the subsequent observation data generated in step S301, using the phase drift formula. The phase drift rate, calculated by measuring the rate of change of the angle between the eccentric vectors within adjacent effective observation periods, reflects the dynamic activity of load redistribution inside the drum and serves as a stability criterion for determining whether further water replenishment is feasible. The phase drift formula is: ; In the formula: For the first Phase drift rate for each effective observation period; The first position is located based on the eccentric phase angle generated in step S202. The eccentric phase of one effective observation period, i.e., the current eccentric phase; The first position is located based on the eccentric phase angle generated in step S202. The eccentric phase of one effective observation period, i.e., the previous eccentric phase; The rotation cycle time of the washing and dyeing machine at the current speed is calculated based on the real-time rotation speed data generated in step S101. It is the absolute value symbol.

[0031] In the specific computational robustness handling, considering the transient fluctuations that may occur in the spindle speed, the control program will monitor in real time. The state is such that if the calculated rotation period is detected to be close to zero (e.g., sensor malfunction causing step loss), the phase drift rate of the previous valid moment is automatically locked to prevent logic crashes caused by a zero denominator. Simultaneously, to suppress phase glitches in discrete sampling, A sliding weighted average can be used instead of a single-cycle instantaneous value to improve the smoothness of the drift rate calculation.

[0032] Step S303: Based on the phase drift rate generated in step S302, the control program calls the perturbation step size formula to dynamically constrain and output the target water injection quality for a single compensation. This embodiment adopts a variable step size control strategy, introducing a feedback suppression operator through the phase drift rate, thereby maintaining high water injection efficiency when the load tends to be stable and reducing the water injection increment when the load is in an unsteady state. The perturbation step size formula is: ; In the formula: For the first The target water injection quality for each effective observation period; The pre-calibrated baseline perturbation quality; This is the preset fluid penalty coefficient; The first generated by step S302 Phase drift rate for each effective observation period; 0 is the function for maximizing values; 0 is used to cut off zero values ​​to prevent the target water injection quality from becoming negative; 1 is the reference constant for calculating the step size attenuation ratio.

[0033] As a preferred approach, the reference perturbation mass (e.g., set to [20-50] g) is set based on the percentage of the single-cavity volume of the balance ring and the minimum balance accuracy allowed by the controller. The fluid penalty coefficient (e.g., set to [0.5-0.8]) is determined based on the mapping relationship between the structural damping ratio of the washing and dyeing machine and the load phase sensitivity.

[0034] See appendix Figure 3 Step S4: Observe the gradient-driven competitive dual-loop balance control. Construct a PI control topology consisting of a vibration suppression loop and a capacity constraint loop, and generate the overall control command by combining the phase drift rate. The specific implementation path of this step is as follows: Step S401: The control architecture constructs a PI control topology consisting of a main control loop (vibration suppression loop) and a competing loop (balance loop) with a capacity constraint loop connected in parallel. Based on the feedback control principle, the vibration suppression loop serves as the main feedback loop for adjusting the physical balance of the washing and dyeing machine, with the control target being that the triaxial composite amplitude generated in step S204 tends towards the zero amplitude target value (e.g., set to 0 mm). The zero amplitude target value is preset based on the theoretical displacement state when the washing and dyeing machine reaches absolute dynamic balance.

[0035] Within each control cycle, the control program calculates the real-time deviation between the triaxial composite amplitude and the zero target amplitude value. A proportional (P) element provides a rapid response to the current deviation, while an integral (I) element accumulates and compensates for historical deviations to eliminate steady-state error. When calculating the integral term, the control program presets integral limiting logic to prevent integral saturation due to persistent deviation. Finally, it calculates and outputs the positive valve duty cycle.

[0036] Step S402: The capacity constraint loop uses the cumulative water injection volume corresponding to the current compensation phase as its state input. The physical significance of this loop is to establish a safety boundary parallel to the vibration suppression loop, preventing the fluid from exceeding the physical bearing limit of the balance loop. When the cumulative water injection volume is greater than or equal to the preset maximum water injection capacity threshold, the capacity constraint loop activates the competition mechanism.

[0037] The maximum water injection capacity threshold is set based on the physical volume limit of a single chamber of the balance ring and the liquid level rise height of the fluid under centrifugal force, for example, it is set to [90%-95% of the rated volume of a single chamber].

[0038] Once the threshold is triggered, the capacity constraint loop calculates and outputs the reverse duty cycle based on the exceeded capacity gradient. Subsequently, the control architecture algebraically sums this reverse duty cycle with the forward valve duty cycle generated in step S401 via a control adder. In this synthetic logic, the reverse duty cycle, as a negative feedback disturbance, can actively cancel the output component of the forward valve duty cycle, thereby completing the forced truncation of the water injection duration before the total control command is generated.

[0039] Step S403: To address the physical time delay issue in dynamic balance control and enable the algorithm to adapt to changing operating conditions, the control program obtains the phase drift rate generated in step S302 and uses the gain adjustment formula to calculate the real-time proportional gain of the vibration suppression loop in the PI control topology. This calculation step is based on the principle of exponential decay and aims to actively reduce control sensitivity when the phase drift rate increases (i.e., the load is in a non-steady state). The gain adjustment formula is: ; In the formula: For the vibration suppression ring in the first Real-time proportional gain for each effective observation period; The initial calibration gain is the initial calibration value. It is a natural constant; The preset weight sensitivity coefficient; The first generated by step S302 Phase drift rate for each effective observation period.

[0040] As a preferred approach, the initial calibration gain (e.g., within the range of [0.1-1.0]) is determined based on the loop gain margin of the washing and dyeing machine under stable dehydration conditions. The weight sensitivity coefficient (e.g., within the range of [0.5-2.0]) is preset based on the characteristic time constant of load redistribution and is used to control the rate of gain decay.

[0041] As the phase drift rate increases, this formula drives the underlying algorithm to exponentially compress the positive feedback strength. Based on this, the overall control command will synchronously increase the anti-overflow truncation weight of the capacity constraint loop in step S402 during the synthesis process.

[0042] Step S5: Discrete Accumulation Execution and Balanced Closed-Loop Verification. The total control command is converted into pulsed water injection commands through a discrete accumulation mechanism, and a closed-loop traversal is continuously performed until a steady-state equilibrium is reached. Finally, the data is serialized and stored. The specific implementation path of this step is as follows: Step S501: The control program acquires the overall control command generated in step S402 and the target injection water mass generated in step S303. Based on the principles of mass conservation and flow integral, the control program maps and correlates the two to calculate the theoretical makeup water mass that should theoretically be injected into the balance loop within the current control cycle.

[0043] Specifically, the theoretical water replenishment quality is calculated by combining the target water injection quality with the duty cycle weight in the overall control command and referring to the calibrated flow coefficient of the water replenishment valve. Given that the water replenishment valve hardware has a valve action dead zone (e.g., set to 20ms) caused by the electromagnetic coil excitation response and valve core mechanical displacement, this results in a minimum physical opening equivalent limit. If the command opening time is shorter than this dead zone time, the valve will be unable to form a stable jet, leading to a loss of control.

[0044] Based on this, the control program deploys a discrete accumulation mechanism on the underlying drive side to establish flow integral logic. The flow integral logic, acting as a virtual buffer unit at the software level, determines the theoretical water replenishment quality within each control cycle. When the theoretical water replenishment quality is less than the preset minimum opening equivalent (e.g., set to [5-10]g), the control program blocks the physical valve's action signal and accumulates the theoretical water replenishment quality into the flow integral logic to form an accumulated value.

[0045] The minimum opening equivalent is calibrated based on the minimum flow rate required for the water supply valve to maintain a stable open state under rated pressure. To prevent integral saturation and data overflow, the flow integral logic has a preset maximum accumulation limit. When the accumulated value fails to trigger execution for an extended period and exceeds the limit, an automatic reset logic will be executed.

[0046] Step S502: When the accumulated value in the flow integral logic is greater than or equal to the minimum opening equivalent, the control program makes a decision based on the eccentric phase angle generated in step S202. As a preferred method, the control program maps the eccentric phase angle to the polar coordinate system of the water supply valve array, selects the target water supply valve with the eccentric pointing to the back side (i.e., within a phase difference of 180°±15°), and issues a pulse water injection command.

[0047] The pulsed water injection command drives the target water supply valve to perform a physical action, and its opening duration is obtained by dividing the current accumulated value by the valve's rated flow rate. Simultaneously with the command issuance, the control program subtracts the executed mass equivalent from the flow integral logic. In this way, the weak control quantity, which would otherwise be filtered by the hardware dead zone, is accumulated in the time domain, ultimately achieving precise balance compensation in the form of discrete pulses.

[0048] In step S503, the overall control architecture continuously traverses the execution paths from steps S1 to S5, performing high-frequency balanced closed-loop verification and control. In this closed-loop logic, after each round of water replenishment is completed, it must pass through the fluid stability window of step S301. Subsequently, the control program monitors the triaxial composite amplitude generated in step S204 in real time.

[0049] As the water replenishment quality counteracts the eccentric torque, when the triaxial composite amplitude shows a continuous convergence trend, and the eccentricity of the washing and dyeing machine calculated based on the rotational speed and dynamic stiffness model stabilizes within the preset displacement bearing threshold (e.g., set to less than or equal to 10μm), the washing and dyeing machine is determined to have reached a steady equilibrium state.

[0050] Among them, the triaxial composite amplitude and the eccentricity of the washing and dyeing machine are both characterized by displacement magnitude. The displacement bearing threshold is determined based on the geometric tolerance and oil film stability requirements of the main bearing of the washing and dyeing machine under high-speed rotation, and is the criterion for ensuring long-term damage-free operation of the equipment.

[0051] Step S504: After reaching the equilibrium steady state generated in step S503, the control architecture actively uploads the extracted global operating condition data to the PC server and persistently writes it to the time series database. During this serialization and storage process, the control program aligns and encapsulates the multi-source data using a unified timestamp to ensure that each record can completely restore the equilibrium state at a specific moment. The global operating condition data includes real-time rotational speed data, triaxial composite amplitude, phase drift rate, target water injection mass, and positive valve duty cycle.

Claims

1. A dynamic balance control method for high-speed dehydration in a large-capacity washing and dyeing machine, characterized in that: Includes the following steps: Based on the acquired real-time rotational speed data, a segmented sampling strategy is used to obtain continuous vibration data, which is then converted into discrete sampled signals. The discrete sampled signal is filtered to locate the eccentric phase angle, and the triaxial acceleration signal is extracted and converted into the triaxial composite amplitude. After passing through the fluid stability window, the triaxial composite amplitude and the eccentric phase angle of the next cycle are extracted as subsequent observation data. The phase drift rate is calculated based on the subsequent observation data using the phase drift formula. The target water injection quality is output based on the phase drift rate using the perturbation step size formula. A PI control topology consisting of a vibration suppression loop and a capacity constraint loop in parallel is constructed. The duty cycle of the positive valve, which outputs with the triaxial synthesized amplitude approaching the target value of zero amplitude, is superimposed with the reverse duty cycle output based on the acquired cumulative water injection volume to generate a total control command. The real-time proportional gain of the vibration suppression loop is calculated based on the phase drift rate using the gain adjustment formula. The total control command is combined with the target water injection mass to convert it into the theoretical water replenishment mass through a discrete accumulation mechanism. The target water replenishment valve is located and the theoretical water replenishment mass is converted into a pulse water injection command. Balanced closed-loop verification and control are performed until the washing and dyeing machine reaches a balanced steady state.

2. The dynamic balance control method for high-speed dehydration of a large-capacity washing and dyeing machine according to claim 1, characterized in that, The steps involved in acquiring continuous vibration data using a segmented sampling strategy based on the obtained real-time rotational speed data, and converting it into discrete sampled signals, specifically include: The real-time rotation speed data is obtained by the incremental encoder measuring the washing and dyeing machine in real time. When the real-time rotation speed data is greater than or equal to the set initial rotation speed threshold, the multi-dimensional signal sampling program is triggered. After the multi-dimensional signal sampling program is triggered, the continuous vibration data is acquired by the triaxial measurement node arranged at the transmission bearing of the washing and dyeing machine, and the analog current signal is output by reading the continuous vibration data. The segmented sampling strategy is implemented based on the real-time speed data. The control sampling frequency in the first speed range is set as the first sampling frequency, the second sampling frequency in the second speed range, the third sampling frequency in the third speed range, and the fourth sampling frequency in the fourth speed range. The analog current signal is converted into the discrete sampling signal according to the determined sampling frequency. The initial rotational speed threshold is preset based on the critical rotational speed at which the flexible clothing load begins to adhere to the wall and generate effective centrifugal force.

3. The dynamic balance control method for high-speed dehydration of a large-capacity washing and dyeing machine according to claim 2, characterized in that, The specific steps of filtering the discrete sampled signal to locate the off-center phase angle include: The Kalman filter is used to extract the discrete sampled signals of the triaxial measurement nodes within one complete mechanical rotation cycle, and the signals are filtered and denoised to output the filtered discrete signals. The phase extraction algorithm is invoked to extract the peak point sequence from the filtered discrete signal; The eccentric phase angle is located and recorded based on the peak point sequence.

4. The dynamic balance control method for high-speed dehydration of a large-capacity washing and dyeing machine according to claim 3, characterized in that, The specific steps for extracting the triaxial acceleration signal and calculating the triaxial composite amplitude include: The three-axis acceleration signal is extracted from the filtered discrete signal using a space vector algorithm. The acceleration of each axis in the three-axis acceleration signal is divided by the square of the current mechanical angular velocity calculated based on the real-time rotational speed data to obtain the basic vibration displacement amplitude of the X, Y, and Z axes. Within the four speed sub-intervals corresponding to the global operating range, the basic vibration displacement amplitude is multiplied by the preset amplitude correction coefficient corresponding to the speed sub-interval to obtain the corrected triaxial component. The modified triaxial components are synthesized using a space vector algorithm to output the triaxial composite amplitude. The amplitude correction coefficient is a pre-set structural amplification factor obtained from centrifugal force simulation experiments.

5. The dynamic balance control method for high-speed dehydration of a large-capacity washing and dyeing machine according to claim 1, characterized in that, The steps for calculating the phase drift rate based on the subsequent observation data using the phase drift formula specifically include: After the target water supply valve performs its action, the observation logic is suspended and several set rotation cycles are waited for as the fluid stability window. The triaxial composite amplitude and the eccentric phase angle corresponding to the next cycle after the end of the fluid stability window are extracted as the subsequent observation data. The phase drift rate is calculated using the phase drift formula based on the phase components in the subsequent observation data. The calculation logic of the phase drift formula is as follows: Calculate the absolute value of the difference between the current eccentric phase and the previous eccentric phase within two adjacent effective observation periods, and divide the absolute value of the difference by the rotation period time calculated based on the real-time rotation speed data to obtain the phase drift rate; The aforementioned set rotation periods are predetermined based on the physical inner diameter of the balance ring and the momentum diffusion time of the fluid under a centrifugal field.

6. The dynamic balance control method for high-speed dehydration of a large-capacity washing and dyeing machine according to claim 1, characterized in that, The specific steps for outputting the target water injection quality include: The target water injection quality is output using the perturbation step size formula based on the phase drift rate constraint; The calculation logic of the perturbation step size formula is as follows: The attenuation result is obtained by subtracting the product of the fluid penalty coefficient and the phase drift rate from the constant 1. The attenuation result is then subjected to a maximum value function operation with the cutoff zero value to obtain the constraint coefficient. The target water injection mass is obtained by multiplying the reference perturbation mass with the constraint coefficient. The reference perturbation mass is preset based on the percentage of the single-cavity volume of the balance ring and the minimum allowable balance accuracy, and the fluid penalty coefficient is preset based on the mapping relationship between the structural damping ratio of the washing and dyeing machine and the load phase sensitivity.

7. The dynamic balance control method for high-speed dehydration of a large-capacity washing and dyeing machine according to claim 1, characterized in that, The specific steps for generating the master control command include: The vibration suppression ring controls the output of the positive valve duty cycle with the triaxial composite amplitude approaching the target value of zero amplitude; The capacity constraint loop takes the cumulative water injection volume as the state input, and outputs the reverse duty cycle when the cumulative water injection volume is greater than or equal to the maximum water injection capacity threshold. The total control command is generated by directly superimposing the reverse duty cycle and the forward valve duty cycle using a control adder. The zero target value of amplitude is preset based on the theoretical displacement state when the washing and dyeing machine reaches absolute dynamic equilibrium, and the maximum water injection capacity threshold is preset based on the physical volume limit of the single cavity of the balance ring and the liquid surface rise height of the fluid under centrifugal force.

8. The dynamic balance control method for high-speed dehydration of a large-capacity washing and dyeing machine according to claim 1, characterized in that, The specific steps for calculating the real-time proportional gain of the vibration suppression loop include: The real-time proportional gain is dynamically calculated based on the phase drift rate using the gain adjustment formula. The calculation logic of the gain adjustment formula is as follows: Using the natural constant as the base, the product of the negative weight sensitivity coefficient and the phase drift rate is raised to the power of the power, and the result is multiplied by the initial calibration gain to obtain the real-time proportional gain. The initial calibration gain is preset based on the loop gain margin of the washing and dyeing machine under stable dehydration conditions, and the weight sensitivity coefficient is preset based on the characteristic time constant of load redistribution.

9. The dynamic balance control method for high-speed dehydration of a large-capacity washing and dyeing machine according to claim 1, characterized in that, The specific steps for converting into a pulse water injection command include: The overall control command is mapped and correlated with the target water injection quality to calculate the theoretical water replenishment quality; A flow integral logic is established as the discrete accumulation mechanism. When the theoretical water replenishment mass is less than the minimum opening equivalent, the physical valve action is disabled, and the theoretical water replenishment mass is accumulated and stored in the flow integral logic to form an accumulated value. If and only if the accumulated value is greater than or equal to the minimum opening equivalent, the target water supply valve is located in combination with the eccentric phase angle, the pulse water injection command is issued to perform physical water supply action, and the corresponding mass equivalent is deducted from the flow integral logic; The minimum opening equivalent is preset based on the minimum flow rate required for the target water supply valve to maintain a stable open state under rated pressure.

10. The dynamic balance control method for high-speed dehydration of a large-capacity washing and dyeing machine according to claim 1, characterized in that, The specific steps for performing closed-loop verification and control until the washing and dyeing machine reaches a steady equilibrium state include: The steps of the dynamic balance control method are iterated repeatedly to perform the balance closed-loop verification and control until the triaxial composite amplitude converges and the corresponding calculated eccentricity of the washing and dyeing machine is stabilized within the displacement bearing threshold, and the washing and dyeing machine reaches the balance steady state. After reaching the equilibrium steady state, the extracted global operating condition data is actively uploaded to the PC server and persistently written into the time series database. The global operating condition data includes the real-time rotational speed data, the triaxial composite amplitude, the phase drift rate, the target water injection mass, and the positive valve duty cycle; The displacement bearing threshold is preset based on the geometric tolerance and oil film stability requirements of the main bearing of the washing and dyeing machine under high-speed rotation.

Images