Sewage suction motor control method, device and equipment of bathing machine

Abstract

This application discloses a method, device, and equipment for controlling the suction motor of a shower machine, relating to the field of smart home appliance technology. The method includes: real-time acquisition of multi-source heterogeneous sensor data of the shower machine's suction motor based on asynchronous sampling frequency; inputting the multi-source heterogeneous sensor data into a multimodal fusion network based on an attention mechanism, and outputting a fused feature vector to identify the current stall fault type of the suction motor; if the stall fault type is the first type, determining the stall fault level based on the current operating current exceeding the standard and / or the current coil temperature rise rate, and matching a corresponding basic low power threshold based on the stall fault level; controlling the suction motor to operate according to the matched basic low power threshold, superimposing a preset breathing micro-desorption waveform, and adjusting the water supply flow rate of the shower machine's water pump; after detecting that the stall fault has been resolved, restoring the operating power of the suction motor to the rated operating power, and restoring the rated water supply flow rate of the water pump.

Description

Technical Field

[0001] This application relates to the technical field of smart home appliances, and in particular to a method, device and equipment for controlling the vacuum motor of a shower machine. Background Technology

[0002] With the increasing popularity of portable bathing aids, these devices use a vacuum motor to generate negative pressure, simultaneously recycling bathing wastewater and solving the wastewater disposal problem in in-home bathing scenarios. However, in actual use, caregivers often place the showerhead too close to the user's skin, causing blockage of the showerhead's return water channel. This, in turn, blocks the vacuum motor's air extraction channel, leading to a stalled motor.

[0003] To address the issue of a sharp increase in operating current caused by a stalled vacuum motor, traditional designs employ a universal stall protection logic. This logic directly stops the vacuum motor upon detecting a stall fault to cut off the current and prevent overcurrent burnout. However, this universal shutdown protection logic has serious flaws in the context of shower machines. Once the vacuum motor stops running, wastewater recycling is immediately interrupted, while the shower head continues to output clean water. The already sprayed wastewater cannot be recycled in time, leading to environmental pollution. Furthermore, due to the lack of effective perception of the multi-source heterogeneous operating status of the vacuum motor and accurate differentiation of stall types, existing technologies can only adopt a black-and-white shutdown strategy. This means that when a stalled vacuum motor occurs in a shower machine, it is completely impossible to simultaneously address both overcurrent protection for the vacuum motor and wastewater overflow prevention.

[0004] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main objective of this application is to provide a method, device, and equipment for controlling the sewage suction motor of a shower machine, which aims to solve the technical problem that existing sewage suction motor stall protection causes sewage overflow, and cannot simultaneously meet the requirements of sewage suction motor overcurrent protection and maintaining the recovery and overflow prevention.

[0006] To achieve the above objectives, this application proposes a method for controlling the vacuum motor of a shower machine, the method comprising: Real-time acquisition of multi-source heterogeneous sensor data from the vacuum motor of the shower machine based on asynchronous sampling frequency; The multi-source heterogeneous sensing data is input into a multimodal fusion network based on an attention mechanism, and the fused feature vector is output to identify the current stall fault type of the vacuum motor. If the stall fault type is the first type, the stall fault level is determined based on the over-limit magnitude of the current operating current and / or the temperature rise rate of the current coil temperature, and the corresponding basic low power threshold is matched based on the stall fault level. The vacuum motor is controlled to operate according to the matched basic low power threshold, and a preset breathing micro-desorption waveform is superimposed and the water supply flow rate of the shower machine water pump is adjusted. After the stall fault is detected and resolved, the working power of the sewage suction motor is restored to the rated working power, and the rated water supply flow of the water supply pump is restored. If the stall fault type is the second type, then the current operating state of the suction motor is maintained.

[0007] In one embodiment, the multi-source heterogeneous sensing data includes at least operating current, coil temperature, housing vibration spectrum, and transient negative pressure waveform at the air inlet; the step of acquiring the multi-source heterogeneous sensing data of the shower machine's vacuum motor in real time based on asynchronous sampling frequency includes: An electrical thermal characteristic sequence is obtained by acquiring the operating current and coil temperature using the first sampling frequency; The flow field characteristic sequence is obtained by acquiring the shell vibration spectrum and the transient negative pressure waveform at the air inlet using the second sampling frequency; The electrical thermal feature sequence and the flow field feature sequence are timestamped and normalized to output a standardized feature matrix.

[0008] In one embodiment, the step of inputting the multi-source heterogeneous sensing data into a multimodal fusion network based on an attention mechanism and outputting a fused feature vector to identify the stall fault type of the current vacuum motor includes: The standardized feature matrix is ​​input into the feature extraction layer of the multimodal fusion network for feature encoding, and the initial feature vectors of each modality are output. The initial feature vector is input into the attention mechanism layer of the multimodal fusion network, where dynamic weights are assigned and weighted fusion is performed to output a fused feature vector. The fused feature vector is input into the classification and recognition layer of the multimodal fusion network for mapping calculation, and the current stall fault type of the sewage suction motor is output.

[0009] In one embodiment, the step of determining the stall fault level based on the extent of the current operating current exceeding the limit and / or the rate of temperature rise of the current coil temperature, and matching a corresponding basic low-power threshold based on the stall fault level, includes: Calculate the extent by which the current operating current exceeds the stall current threshold and the rate of temperature rise of the coil. The above-mentioned exceedance range and temperature rise rate are input into a preset level determination model, and the stall fault level is output. Look up the pre-stored mapping table between stall fault levels and basic low power thresholds, and output the basic low power threshold corresponding to the current stall fault level.

[0010] In one embodiment, the step of controlling the suction motor to operate according to the matched basic low power threshold and superimposing a preset breathing micro-desorption waveform includes: Using the matched baseline low power threshold as the power baseline, a breathing-type micro-desorption waveform with a preset collapse depth and recovery cycle is generated. The power baseline is superimposed with the breathing micro-desorption waveform to output the target drive power waveform, and the operation of the suction motor is controlled according to the target drive power waveform to generate alternating negative pressure in the pipeline of the shower machine to break the vacuum adsorption between the shower machine nozzle and the skin.

[0011] In one embodiment, the step of adjusting the water supply flow rate of the shower machine's water supply pump includes: The equivalent average power of the target driving power waveform is calculated in real time. Calculate the maximum allowable sewage recovery flow rate of the shower machine's water supply pump based on the equivalent average power. The preset ratio of the maximum allowable wastewater recycling flow rate is determined as the target water supply flow rate; The target water supply flow rate is compared with the first rated flow rate corresponding to the first gear and the second rated flow rate corresponding to the second gear of the water supply pump, and the target gear is determined from the first gear and the second gear according to the comparison result; the first rated flow rate is less than the second rated flow rate; Output a gear switching signal to the water supply pump to control the water supply pump to run continuously at the target gear.

[0012] In one embodiment, the steps of restoring the operating power of the vacuum motor to its rated operating power and restoring the rated water supply flow rate of the water supply pump after detecting that the stall fault has been resolved include: During the operation of the suction motor at a basic low power threshold, the operating current and the transient negative pressure waveform at the air inlet are continuously monitored; When the operating current drops below the recovery current threshold and the air inlet negative pressure recovers to the normal operating negative pressure range, a fault clearance signal is output. Based on the fault clearance signal, a command to restore rated operating power is output to the sewage suction motor, and a command to restore rated water supply flow rate is output to the water supply pump.

[0013] In one embodiment, the method further includes a timeout protection step: Timing is started during the process of controlling the vacuum motor to operate according to the basic low power threshold; When the duration of the basic low-power threshold operation reaches the preset timeout threshold and no stall fault clearance signal is received, a forced shutdown command is output to control the sewage suction motor and water supply pump to stop, and an alarm prompt signal is output.

[0014] Furthermore, to achieve the above objectives, this application also proposes a vacuum motor control device for a bath machine, the vacuum motor control device for the bath machine comprising: The data acquisition module is used to acquire multi-source heterogeneous sensor data of the vacuum motor of the shower machine in real time based on asynchronous sampling frequency; The fault identification module is used to input the multi-source heterogeneous sensor data into a multimodal fusion network based on an attention mechanism, and output a fused feature vector to identify the stall fault type of the current sewage suction motor. The fault matching module is used to determine the level of the stall fault based on the over-limit magnitude of the current operating current and / or the temperature rise rate of the current coil if the stall fault type is the first type, and to match the corresponding basic low power threshold based on the stall fault level. The fault handling module is used to control the sewage suction motor to operate according to the matched basic low power threshold, superimpose a preset breathing micro-desorption waveform and adjust the water supply flow of the shower machine water pump. The status recovery module is used to restore the working power of the sewage suction motor to the rated working power and the rated water supply flow of the water supply pump after the stall fault is detected to be resolved. The status maintenance module is used to maintain the current operating status of the sewage suction motor if the stall fault type is the second type.

[0015] In addition, to achieve the above objectives, this application also proposes a vacuum motor control device for a bath machine, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the vacuum motor control method for the bath machine as described above.

[0016] This application proposes a method, device, and equipment for controlling the vacuum motor of a shower machine. The method includes: real-time acquisition of multi-source heterogeneous sensor data of the vacuum motor of the shower machine based on asynchronous sampling frequency; inputting the multi-source heterogeneous sensor data into a multimodal fusion network based on an attention mechanism, and outputting a fused feature vector to identify the current stall fault type of the vacuum motor; if the stall fault type is a first type, determining the stall fault level based on the current operating current exceeding the standard range and / or the current coil temperature rise rate, and matching a corresponding basic low power threshold based on the stall fault level; controlling the vacuum motor to operate according to the matched basic low power threshold, superimposing a preset breathing micro-desorption waveform, and adjusting the water supply flow rate of the shower machine's water pump; after detecting that the stall fault has been resolved, restoring the operating power of the vacuum motor to the rated operating power, and restoring the rated water supply flow rate of the water pump; if the stall fault type is a second type, maintaining the current operating state of the vacuum motor. This application solves the problems of existing shower machine vacuum motors stopping due to motor blockage and causing sewage overflow, and the inability to simultaneously protect against overcurrent and sewage recovery by using multi-source heterogeneous data fusion to accurately identify the type and level of motor blockage, power reduction superimposed with breathing-type micro-desorption waveform operation, and synchronous adjustment of water supply flow. It achieves accurate differentiation between real blockage and transient fluctuations to avoid misjudgment, power reduction to replace shutdown to prevent burnout and ensure basic recovery, micro-desorption waveform to actively break vacuum dead suction and promote unblocking, and dynamic coupling balance of water and air to prevent water overflow. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart illustrating an embodiment of the sewage suction motor control method for a bath machine according to this application; Figure 2 This is a flowchart illustrating Embodiment 2 of the method for controlling the vacuum motor of a shower machine according to this application. Figure 3 This is a flowchart illustrating Embodiment 3 of the method for controlling the vacuum motor of a shower machine according to this application. Figure 4 This is a schematic diagram of the module structure of the sewage suction motor control device of the shower machine according to an embodiment of this application; Figure 5This is a schematic diagram of the equipment structure of the hardware operating environment involved in the sewage suction motor control method of the bath machine in the embodiments of this application.

[0020] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0021] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0022] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0023] Because the existing sewage suction motor stalls and stops, causing sewage to overflow, it is impossible to simultaneously address the sewage suction motor overcurrent protection and maintain the need for overflow prevention during recycling.

[0024] This application provides a solution that uses multi-source heterogeneous data fusion to accurately identify the type and level of stalling, power reduction superimposed with breathing-type micro-desorption waveform operation and synchronous adjustment of water supply flow. This solves the problems of existing shower machine vacuum motor stalling and shutdown causing sewage overflow, and the inability of simply reducing power to take into account overcurrent protection and sewage recycling. It achieves accurate differentiation between real stalling and transient fluctuations to avoid misjudgment, power reduction to replace shutdown to prevent burnout and ensure basic recycling, micro-desorption waveform to actively break vacuum dead suction and promote unblocking, and dynamic coupling balance of water and air to prevent water overflow.

[0025] Based on this, this application provides a method for controlling the vacuum motor of a shower machine, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the sewage suction motor control method for the bath machine of this application.

[0026] In this embodiment, the method for controlling the vacuum motor of the shower machine includes steps S10 to S60: Step S10: Real-time acquisition of multi-source heterogeneous sensor data from the vacuum motor of the shower machine based on asynchronous sampling frequency.

[0027] It should be noted that, in this embodiment, asynchronous sampling frequency refers to the frequency of data acquisition based on different time intervals used for different types of sensor data according to their physical quantity change characteristics; multi-source heterogeneous sensor data refers to a collection of sensing information from various types of sensors with different data formats and feature dimensions.

[0028] This embodiment comprehensively and accurately grasps the real-time working status of the sewage suction motor through multi-dimensional data monitoring, avoiding misjudgments caused by fluctuations in a single parameter, and realizing stall identification from single electrical monitoring to multi-dimensional physical field monitoring.

[0029] In one possible implementation, by employing asynchronous sampling frequencies, on the one hand, high-frequency sampling can be used for rapidly changing electrical signals (such as operating current) to ensure that transient details are not lost, and on the other hand, lower-frequency sampling can be used for slowly changing thermal signals (such as coil temperature) to save computing resources, thereby achieving a balance between acquisition efficiency and data accuracy.

[0030] In one possible implementation, the multi-source heterogeneous sensing data includes at least the operating current, coil temperature, housing vibration spectrum, and inlet transient negative pressure waveform.

[0031] Step S20: Input the multi-source heterogeneous sensing data into a multimodal fusion network based on an attention mechanism, and output a fused feature vector to identify the current stall fault type of the vacuum motor.

[0032] It should be noted that, in this embodiment, the multimodal fusion network based on the attention mechanism refers to a deep learning model that can automatically learn the importance of different modal data and assign corresponding dynamic weights for feature integration; the fusion feature vector refers to the low-dimensional data representation containing global state information obtained by weighted fusion of different modal features; the stall fault type refers to the abnormal category divided according to the fault characteristics, including at least continuous skin-attacking stall and transient fluctuation type.

[0033] This embodiment extracts deep features of each mode through a deep network and uses an attention mechanism to focus on the key mode information most relevant to the current fault, suppresses noise interference, and outputs a fused feature vector to identify the current stall fault type of the sewage suction motor, thereby greatly improving the robustness and accuracy of fault identification and effectively distinguishing between real stall and transient interference.

[0034] In one possible implementation, the system inputs a standardized feature matrix into the feature extraction layer of the network for feature encoding, outputs initial feature vectors for each modality, and then inputs the initial feature vectors into the attention mechanism layer to assign dynamic weights and perform weighted fusion.

[0035] In one specific implementation, the system inputs an electrical thermal characteristic sequence containing the operating current and coil temperature, and a flow field characteristic sequence containing the housing vibration spectrum and the transient negative pressure waveform of the air inlet into the network. The network assigns a weight of 0.5 to the current anomaly and a weight of 0.3 to the negative pressure anomaly. After fusion, the output identification result is the first type (continuous skin-mounted stall), and the fault confidence is 0.95.

[0036] Step S30: If the stall fault type is the first type, the stall fault level is determined based on the over-limit magnitude of the current operating current and / or the temperature rise rate of the current coil temperature, and the corresponding basic low power threshold is matched based on the stall fault level.

[0037] It should be noted that in this embodiment, the first type refers to the continuous skin-contact stall fault, that is, the substantial blockage of the suction motor's air extraction channel caused by the nozzle being in close contact with the skin; the exceedance range refers to the difference or proportion of the current operating current exceeding the rated stall current threshold; the temperature rise rate refers to the value of the coil temperature rise per unit time; the stall fault level refers to the classification index characterizing the severity of the stall, such as mild, moderate, and severe; the basic low power threshold refers to the minimum safe power value that allows the suction motor to operate and maintain a basic negative pressure without causing overcurrent burnout.

[0038] This embodiment achieves a shift from rudimentary shutdown protection to refined power reduction protection by quantitatively assessing the severity of the fault. This ensures that the vacuum motor coil does not burn out due to overheating, while maintaining basic wastewater recovery capabilities and avoiding wastewater overflow problems caused by direct shutdown.

[0039] In one possible implementation, upon receiving a first-type stall signal, the system immediately acquires the current operating current I and coil temperature T, calculates the over-limit amplitude ΔI = (I - I_th) / I_rated, where I_th is the preset stall current threshold and I_rated is the rated current; simultaneously, it calculates the temperature change rate dT / dt over the past 0.5 seconds. The system inputs ΔI and dT / dt into a preset fuzzy inference engine, which contains three membership functions (low, medium, and high) and three inference rules, outputting a value between 0 and 2, corresponding to mild, moderate, and severe fault levels, respectively. Then, the system looks up a preset mapping table based on the level: mild corresponds to a basic low-power threshold of 60% of the rated power, moderate corresponds to 35%, and severe corresponds to 15%.

[0040] In one specific implementation, the rated operating current of the vacuum motor is 1.2A, and the stall current threshold is set to 2.5A. When the system detects that the current operating current reaches 3.5A, which is significantly higher than the standard, the stall fault level is determined to be severe. The basic low power threshold is matched to 20W. At this time, the operating current of the vacuum motor drops to about 1.4A, which is lower than the overcurrent damage threshold and can output a basic negative pressure of -0.02MPa.

[0041] Step S40: Control the vacuum motor to operate according to the matched basic low power threshold, superimpose the preset breathing micro-desorption waveform and adjust the water supply flow of the shower machine water pump.

[0042] It should be noted that, in this embodiment, the breathing micro-desorption waveform refers to a control signal waveform superimposed on the basic low power, which has the characteristics of periodic power collapse and recovery, and is used to generate alternating negative pressure in the pipeline; the water supply flow rate refers to the volume of clean water output by the water supply pump of the bath machine per unit time.

[0043] In this embodiment, under the premise of ensuring the safe operation of the sewage suction motor, the alternating negative pressure generated by the breathing micro-desorption waveform breaks the vacuum adsorption state between the nozzle and the skin, actively promoting the unblocking of blockages; at the same time, in order to adapt to the reduced sewage recovery capacity, the system correspondingly reduces the water supply flow rate to prevent a small amount of water from overflowing due to the water supply flow rate exceeding the recovery capacity, thereby achieving intelligent coordinated control of air path unblocking and water path throttling.

[0044] In one possible implementation, the system uses a basic low power threshold as the power baseline, superimposes a breathing micro-desorption waveform with a preset collapse depth and recovery period, and proportionally reduces the water supply flow rate according to the real-time equivalent average power of the suction motor. The system can achieve flow rate regulation by adjusting the PWM duty cycle of the water supply pump.

[0045] In another possible implementation, the breathing micro-desorption waveform can be an asymmetric sawtooth wave with a slow rising edge and a steep falling edge to produce a "slow inhalation and rapid exhalation" effect, which is more conducive to breaking vacuum adsorption.

[0046] In one specific implementation, the system controls the suction motor to operate at a low power threshold of 20W, and superimposes a breathing-type micro-desorption waveform with a frequency of 1Hz and an amplitude of ±10%, so that the negative pressure in the air path generates a slight vibration to break the dead suction; at the same time, the system reduces the water supply flow of the shower machine's water pump to 60% of the normal rated water supply flow to adapt to the reduced sewage recovery capacity.

[0047] Step S50: After detecting that the stall fault has been cleared, restore the working power of the sewage suction motor to the rated working power and restore the rated water supply flow of the water supply pump.

[0048] It should be noted that, in this embodiment, "stalling fault relief" means that the physical blockage of the suction motor's air extraction channel is removed, and the operating parameters of the suction motor return to the normal operating range; "rated operating power" refers to the standard power value designed for the suction motor to operate under normal, non-stalled conditions; and "rated water supply flow rate" refers to the standard clean water flow rate designed to be output by the water supply pump under normal bathing conditions.

[0049] This embodiment ensures that the blockage has been cleared and the sewage suction motor can operate safely and stably, thereby quickly restoring the shower machine's efficient sewage recycling capacity and normal water supply. It avoids affecting the shower efficiency and experience due to prolonged low-power operation, achieving a seamless connection between the protection mechanism and normal operation.

[0050] In one possible implementation, the system determines that the stall fault has been cleared based on the operating current dropping below the recovery current threshold.

[0051] In one specific implementation, the system continuously monitors the operating current while the vacuum motor is running at low power. When the current collected for three consecutive times drops below 2A, it is determined that the stall fault has been resolved. The system immediately restores the operating power of the vacuum motor to the rated 75W and simultaneously restores the water supply flow rate of the water pump to 100% of the rated water supply flow rate, returning to normal operation.

[0052] Step S60: If the stall fault type is the second type, then maintain the current operating state of the sewage suction motor.

[0053] It should be noted that in this embodiment, the second type refers to transient fluctuation anomalies, that is, transient changes in current or negative pressure caused by non-substantial blockage such as water splashing or short-term water flow fluctuations; the current operating state refers to the normal or rated operating state of the sewage suction motor before the second type of fault is identified.

[0054] This embodiment aims to avoid system misjudgment and over-response caused by transient interference, prevent the shower machine from frequently reducing power or stopping, thus interrupting the normal showering process, ensuring the continuity of showering and a good user experience, and achieving intelligent fault tolerance.

[0055] In one possible implementation, while maintaining the current operating state of the suction motor, the system can activate an anti-shake timer to continuously observe abnormal characteristics to confirm whether the fault type needs to be reassessed.

[0056] In one specific implementation, when the system detects a brief jump in the operating current but no substantial blockage in the negative voltage waveform, it is determined to be of the second type (transient fluctuation type). At this time, the system does not reduce the power of the suction motor, but maintains the suction motor to continue running at the rated operating power of 75W, and starts a 100-millisecond anti-shake timer. If the jump disappears, it continues to run normally, thereby avoiding the false power reduction caused by transient water splashes.

[0057] This application accurately identifies the type of stall fault in the suction motor through a multimodal fusion network. For continuous stall faults, it employs a strategy of power reduction superimposed with micro-desorption waveforms and coordinated water adjustment. For transient fluctuation faults, it maintains operation and automatically resumes operation after the fault is cleared. This achieves accurate identification and intelligent fault tolerance of stall faults, avoiding malfunctions caused by transient interference. In the event of actual stall, the power reduction operation balances overcurrent protection of the suction motor with basic sewage recovery capacity. At the same time, the micro-desorption waveform actively breaks the vacuum dead suction to promote unblocking, and the water supply flow is adjusted in conjunction to achieve water-air balance, preventing sewage overflow and ensuring the safety, continuity, and good experience of the bathing process.

[0058] Furthermore, the multi-source heterogeneous sensing data includes at least the operating current, coil temperature, housing vibration spectrum, and transient negative pressure waveform at the air inlet; the step of acquiring the multi-source heterogeneous sensing data of the shower machine's vacuum motor in real time based on the asynchronous sampling frequency includes A201~A203: Step A201: Obtain the electrical thermal characteristic sequence by acquiring the operating current and coil temperature through the first sampling frequency.

[0059] It should be noted that in this embodiment, the first sampling frequency refers to the high-frequency data acquisition rate set for physical quantities that change rapidly in electrical and thermal aspects; the operating current refers to the current value consumed in real time by the motor windings during the operation of the vacuum cleaner motor; the coil temperature refers to the real-time temperature value of the vacuum cleaner motor winding coils; and the electrical thermal characteristic sequence refers to the set of sampling data of operating current and coil temperature arranged in chronological order.

[0060] This embodiment captures the sudden increase in electrical load and heat accumulation effect caused by stalling of the vacuum motor, enabling rapid location of the abnormality through core electrical parameters in the early stage of stalling. The idea is to use high-precision current and temperature sensors to monitor the core electrical parameters of the vacuum motor at a high sampling rate, thereby forming continuous time-series data.

[0061] In one possible implementation, using a higher first sampling frequency to sample the operating current can capture the current spike at the moment of stall and the high-frequency components in the current ripple, which are key features for distinguishing different types of stall; while using the same first sampling frequency (or downsampling and processing) for the coil temperature can ensure that temperature changes correspond precisely to current events in time.

[0062] Step A202: Obtain the flow field characteristic sequence by acquiring the shell vibration spectrum and the transient negative pressure waveform of the air inlet through the second sampling frequency.

[0063] It should be noted that, in this embodiment, the second sampling frequency refers to the data acquisition rate set for the fluid dynamics and mechanical vibration characteristics, which needs to cover the frequency band of the vibration signal and the transient characteristics of the negative pressure fluctuation; the shell vibration spectrum refers to the frequency component distribution of the vibration on the shell surface when the suction motor is running, which is usually obtained by Fourier transform after collecting the time-domain vibration signal through a MEMS accelerometer or piezoelectric vibration sensor. The dominant frequency and amplitude distribution of the vibration under different stall conditions are significantly different; the transient negative pressure waveform at the air inlet refers to the real-time curve of the negative pressure at the air inlet of the suction motor changing with time, which is collected by a thin-film or piezoresistive pressure sensor, reflecting the dynamic change of the suction negative pressure actually generated by the suction motor; the flow field characteristic sequence refers to the set of sampled data of the shell vibration spectrum and the transient negative pressure waveform at the air inlet arranged in chronological order, which characterizes the mechanical vibration and air pressure state in the fluid channel of the suction motor.

[0064] This embodiment indirectly reflects the physical unobstructedness of the suction motor's air extraction channel from two dimensions: mechanical vibration and gas flow field. This can effectively distinguish between substantial physical blockage and electrical transient interference. By using acceleration sensors and negative pressure sensors to monitor the flow field and mechanical state of the suction motor during operation, since air blockage will directly cause negative pressure changes and abnormal pump vibration, the shell vibration spectrum can reveal whether the suction motor has experienced abnormal mechanical collisions or resonance. Meanwhile, the negative pressure waveform at the air inlet directly indicates whether the sewage recovery channel is unobstructed and the ability to establish negative pressure. These characteristics can serve as auxiliary verification for determining the stall, thereby avoiding false alarms caused by relying solely on current judgment.

[0065] In one possible implementation, the second sampling frequency can be lower than the first sampling frequency because the response speed of the flow field and mechanical vibration has a certain lag relative to electrical changes.

[0066] In one specific implementation, the system collects the vibration spectrum of the vacuum motor housing and the transient negative pressure waveform of the air inlet at a second sampling frequency of 50Hz. When the nozzle is in close contact with the skin, causing the air extraction channel to be blocked, the transient negative pressure waveform of the air inlet will fluctuate violently, and an abnormal low-frequency resonance peak will appear in the vibration spectrum of the housing. Based on this, the system collects a flow field feature sequence containing information on abrupt changes in the flow field.

[0067] Step A203: Perform timestamp alignment and normalization on the electrical thermal feature sequence and the flow field feature sequence to output a standardized feature matrix.

[0068] It should be noted that in this embodiment, timestamp alignment refers to the operation of synchronizing and matching the timelines of multi-source data collected at different sampling frequencies according to their absolute occurrence times; normalization refers to the data processing method of mapping data with different dimensions and numerical ranges to a unified numerical range (such as between 0 and 1) through linear transformation or statistical standardization; and the standardized feature matrix refers to a two-dimensional data array that can be used as input to a neural network, formed by splicing multiple features according to dimensions after time alignment and numerical normalization. Only through precise timestamp alignment can the network learn the causal relationship between current changes and vibration and negative pressure changes, avoiding misjudgments caused by time offset; while normalization ensures that the features of each modality are numerically comparable, so that the weights allocated by the attention mechanism can truly reflect the importance of each modality rather than being dominated by dimensions.

[0069] This embodiment eliminates the differences in sampling frequency and numerical magnitude between heterogeneous sensors, enabling the multimodal fusion network to perform feature extraction and fusion calculations efficiently and stably. This prevents large numerical features from overshadowing small numerical features, accelerates model convergence, and improves recognition accuracy.

[0070] In one possible implementation, the system first receives an electrical thermal feature sequence (containing timestamp t_e and value v_e) and a flow field feature sequence (containing timestamp t_f and value v_f). Using nearest neighbor matching or linear interpolation, all data points are unified onto a common time grid. For example, using the time step of the highest sampling frequency (e.g., 10kHz) as a reference, low-frequency data (e.g., temperature, negative pressure) are interpolated and filled. Then, the system calculates the mean and standard deviation for each feature channel and performs Z-score standardization, i.e., (x - mean) / std, so that the mean of each channel is 0 and the standard deviation is 1. Finally, the system arranges the standardized data into a matrix according to the feature channels, with a matrix size of (time window length) × (number of feature channels).

[0071] In one possible implementation, the normalization process can employ the maximum-minimum normalization method to linearly transform all feature data to the interval [0,1], specifically using the formula (x - min) / (max - min).

[0072] This application acquires the electrical and thermal characteristic sequences and flow field characteristic sequences of a sewage suction motor through asynchronous sampling frequencies, and performs timestamp alignment and normalization on both sequences to output a standardized feature matrix. The technical advantages are: it achieves a unified representation of multi-dimensional heterogeneous sensor data, eliminates the differences in sampling frequencies and numerical dimensions between different sensors, provides high-quality, standardized input for multi-modal fusion networks, and thus effectively improves the efficiency of data fusion and the accuracy and robustness of sewage suction motor stall fault identification.

[0073] Furthermore, referring to Figure 2 The second embodiment of the sewage suction motor control method for the bath machine in this application provides a flowchart, based on the above. Figure 2 The embodiment shown further refines the step S20, "inputting the multi-source heterogeneous sensing data into a multimodal fusion network based on an attention mechanism and outputting a fused feature vector to identify the current stall fault type of the vacuum motor," including steps A301 to A303: Step A301: Input the standardized feature matrix into the feature extraction layer of the multimodal fusion network for feature encoding, and output the initial feature vector of each modality.

[0074] It should be noted that, in this embodiment, the standardized feature matrix refers to a two-dimensional data array that can be used as input to a neural network, formed by concatenating multiple features according to dimensions after timestamp alignment and normalization; the multimodal fusion network refers to a deep learning model architecture that can simultaneously process and integrate multiple different modal data features; the feature extraction layer refers to the network layer in the multimodal fusion network used to reduce the dimensionality and abstract the representation of the original input data, and is used to map the modal data in the original standardized feature matrix to a high-dimensional latent space to extract more discriminative deep features; feature encoding refers to the nonlinear transformation process of converting the original input data into a high-dimensional feature vector with semantic information; the initial feature vector of each modality refers to the feature representation output for the four modes of operating current, coil temperature, shell vibration spectrum, and air inlet transient negative pressure waveform, respectively. Each feature vector is a fixed-length numerical sequence (e.g., a 128-dimensional floating-point number) representing the essential attribute of the modality within a given time window.

[0075] This embodiment automatically extracts deep abstract features that are highly correlated with the operating status of the sewage suction motor from complex multi-source data, which can achieve the effects of removing data redundancy, highlighting key status information, and reducing the complexity of subsequent calculations. Specifically, the standardized feature matrix is ​​convolutionally calculated or sequence modeled using the preset network parameters in the feature extraction layer, and the shallow sensing signals are converted into deep feature expressions, thereby generating the initial feature vectors corresponding to the electrical thermal mode and the flow field mode, respectively.

[0076] In one possible implementation, the feature extraction layer can employ a one-dimensional convolutional neural network to capture local dependency features of the time-series data.

[0077] In one possible implementation, the feature extraction layer may also employ a long short-term memory network to extract long-term dependency features from time-series data.

[0078] In one specific implementation, the system inputs a standardized feature matrix containing operating current, coil temperature, housing vibration spectrum, and transient negative pressure waveform at the air inlet into the feature extraction layer of the multimodal fusion network. The feature extraction layer uses a one-dimensional convolution kernel to slide calculation in the time dimension to extract the time-dependent features of sudden rise in operating current and sudden change in negative pressure, and outputs the initial feature vector representing the electrical-thermal mode and the initial feature vector representing the flow field mode, respectively.

[0079] Step A302: Input the initial feature vector into the attention mechanism layer of the multimodal fusion network, assign dynamic weights, perform weighted fusion, and output the fused feature vector.

[0080] It should be noted that, in this embodiment, the attention mechanism layer refers to the network structure in the multimodal fusion network used to evaluate the importance of different modal features and assign corresponding weights; assigning dynamic weights refers to adaptively calculating the attention score of each modal feature based on the context information of the current input data; weighted fusion refers to the feature integration process of multiplying the initial feature vector of each modality with its corresponding dynamic weight and then concatenating or summing the results; and the fused feature vector refers to the comprehensive feature representation that integrates key information from multiple modalities and is used for the final decision.

[0081] This embodiment maps the initial feature vectors of each modality to query and key-value pairs through an attention mechanism layer, calculates the correlation score between each modality as dynamic weights, and then multiplies each initial feature vector with the corresponding dynamic weights and fuses them to generate a fused feature vector containing global dominant fault information. This enables the adaptive highlighting of the most critical feature modalities for current fault judgment based on the actual operating status of the current sewage suction motor, suppressing noise from interfering or irrelevant modalities, improving the accuracy of fault feature representation, and enhancing the model's anti-interference ability and robustness.

[0082] In one possible implementation, the attention mechanism layer can employ a self-attention mechanism to compute dynamic weights at different time steps within the same modality.

[0083] In one possible implementation, the attention mechanism layer may also employ a cross-modal attention mechanism to compute dynamic weights between different modalities.

[0084] In one possible implementation, the attention mechanism layer computes an unnormalized score for each modality by passing the four initial feature vectors through a shared fully connected layer (also known as an attention scoring network). Then, a Softmax function is used to convert the scores into dynamic weights in the form of a probability distribution. Specifically, for the initial feature vector h_i of the i-th modality, the score s_i = W_a*tanh(U_a*h_i) is computed, where W_a and U_a are learnable parameter matrices; then the weights α_i = exp(s_i) / Σexp(s_i). Finally, the feature vectors F = Σα_i*h_i (weighted summation) are fused.

[0085] Step A303: Input the fused feature vector into the classification and recognition layer of the multimodal fusion network for mapping calculation, and output the current stall fault type of the sewage suction motor.

[0086] It should be noted that, in this embodiment, the classification and identification layer refers to the network layer in the multimodal fusion network used to map high-dimensional features to specific fault categories; the mapping calculation refers to the operation process of transforming the fused feature vector into a probability distribution through linear transformation and nonlinear activation; the stall fault type refers to the abnormal category label divided according to the fault characteristics, including at least the first type (nozzle skin stall) and the second type (instantaneous load fluctuation); in addition, the classification and identification layer can also output the fault confidence of the stall fault type, which refers to the quantitative evaluation value of the system's credibility of the identified stall fault type.

[0087] In this embodiment, the classification and recognition layer maps the fused feature vector to the same dimension as the preset number of fault categories through a fully connected layer, and then converts the output into a probability distribution through a softmax activation function. The category with the highest probability that exceeds the preset threshold is selected as the stall fault type, and its corresponding probability value is the fault confidence level. This transforms the complex fused features into an intuitive fault judgment result, accurately distinguishing between real stall and transient fluctuations.

[0088] In one possible implementation, the classification layer uses a support vector machine (SVM) or linear discriminant analysis (LDA) instead of a fully connected layer of a neural network. In this case, the fused feature vector is first input into the SVM classifier, which outputs the class label and the distance to the decision boundary. Then, a sigmoid function is used to convert the distance into confidence.

[0089] Additionally, it should be noted that in one possible implementation, the classification and recognition layer can also use a multilayer perceptron combined with a softmax function for mapping calculation.

[0090] This application encodes the standardized feature matrix into initial feature vectors for each modality through a feature extraction layer, then assigns dynamic weights through an attention mechanism layer and performs weighted fusion to obtain a fused feature vector. Finally, a classification and recognition layer maps and calculates the stall fault type and fault confidence of the sewage suction motor. This achieves in-depth mining and adaptive integration of multimodal data, effectively highlighting the most critical feature modes for the current fault and suppressing irrelevant noise interference. As a result, it accurately distinguishes between real stall and transient fluctuations and outputs reliable confidence, providing an accurate and robust decision-making basis for differentiated control strategies.

[0091] Furthermore, the step of determining the stall fault level based on the extent of the current operating current exceeding the limit and / or the rate of temperature rise of the current coil temperature, and matching the corresponding basic low-power threshold based on the stall fault level, includes A401~A403: Step A401: Calculate the extent by which the current operating current exceeds the stall current threshold and the rate of temperature rise of the coil.

[0092] It should be noted that, in this embodiment, the current operating current refers to the current value of the vacuum motor drive current collected at the current moment (or within the most recent time window) after the system determines that the stall fault type is the first type; the stall current threshold refers to the preset critical current value for determining that the vacuum motor is in a stall state; the exceedance range refers to the difference or ratio between the current operating current and the stall current threshold; the coil temperature refers to the real-time temperature value of the vacuum motor winding coil; and the temperature rise rate refers to the rise in coil temperature per unit time, reflecting the speed of heat accumulation.

[0093] This embodiment calculates the over-limit range by comparing the difference between the current operating current and the preset stall current threshold, and calculates the temperature rise rate by using the ratio of the continuously sampled coil temperature difference to the time difference, thereby intuitively reflecting the overload degree and heating trend of the sewage suction motor; the purpose is to quantify the current electrical overload and heat accumulation urgency of the sewage suction motor, so as to accurately classify the stall fault level.

[0094] In one possible implementation, the system first acquires the current operating current I, with a preset stall current threshold of I_th and a rated current of I_rated. The excess range is then defined as the relative excess range ΔI = (I - I_th) / I_rated, a dimensionless value that eliminates dimensional differences between suction motors of different power. Simultaneously, the system acquires the coil temperature sequence over a past time interval Δt (e.g., 0.5 seconds or 1 second) and calculates the temperature rise rate dT / dt = (T_end - T_start) / Δt using linear regression or first-order difference.

[0095] Step A402: Input the exceedance range and temperature rise rate into the preset level determination model, and output the stall fault level.

[0096] It should be noted that, in this embodiment, the preset level determination model refers to a predefined function or rule system that can map the input exceedance magnitude and temperature rise rate into discrete stall fault levels. The stall fault level refers to a quantitative classification of the severity of stalling, such as dividing it into three levels: mild, moderate, and severe, or using numerical scores (e.g., 1-5 points). The level determination model can be based on expert rule lookup tables, fuzzy inference systems, or classifiers trained using machine learning (e.g., decision trees, logistic regression, etc.).

[0097] This embodiment comprehensively considers both electrical overload and thermal accumulation as dangerous conditions to accurately classify the severity of stall faults, avoiding bias caused by single-parameter evaluation and achieving refined fault assessment.

[0098] In one possible implementation, the severity rating model is a fuzzy inference system containing two input variables (excess magnitude and temperature rise rate) and one output variable (fault severity). The membership functions of the input variables are set to three fuzzy sets: low, medium, and high, respectively, and the output variable also corresponds to three fuzzy sets: mild, moderate, and severe.

[0099] In another possible implementation, the severity rating model is a decision tree that performs a binary judgment based on whether the exceedance magnitude and temperature rise rate exceed their respective thresholds. For example, if the exceedance magnitude is >0.5 or the temperature rise rate is >10℃ / s, it is rated as severe; if the exceedance magnitude is >0.2 and ≤0.5, or the temperature rise rate is >5℃ / s and ≤10℃ / s, it is rated as moderate; otherwise, it is rated as mild.

[0100] Step A403: Locate the pre-stored mapping table between stall fault levels and basic low power thresholds, and output the basic low power threshold corresponding to the current stall fault level.

[0101] It should be noted that, in this embodiment, the pre-stored mapping table between stall fault levels and basic low-power thresholds refers to a lookup table pre-stored in the system memory, which corresponds each discrete stall fault level to a specific basic low-power threshold value. The basic low-power threshold refers to the minimum safe power value that allows the vacuum motor to operate and maintain a basic negative pressure while ensuring that the vacuum motor does not burn out due to overcurrent.

[0102] In this embodiment, the current stall fault level is used as the index key. The system traverses and queries the pre-stored mapping table to locate the power parameter corresponding to the level and outputs it as the basic low power threshold. That is, the most suitable safe operating power is dynamically matched according to the severity of the fault, so as to ensure that the sewage suction motor is not burned out while maximizing the sewage recycling capacity and achieving the best balance between protection and recycling needs.

[0103] In one possible implementation, the pre-stored mapping table can be stored in the controller's memory in the form of an array or key-value pairs; the pre-stored mapping table can also be replaced by a function model that calculates the basic low-power threshold based on the fault level.

[0104] This application calculates the extent of excessive operating current and the rate of coil temperature rise, determines the level of stall fault through a level judgment model, and then matches the corresponding basic low power threshold by looking up the mapping relationship table. This achieves multi-dimensional comprehensive quantification and refined classification of the severity of stall fault, avoids the bias of single parameter evaluation, and dynamically matches the most suitable safe operating power according to the severity of the fault. Thus, while ensuring that the sewage suction motor is not burned out, the basic sewage recovery capacity is preserved to the maximum extent, achieving the best balance between equipment safety protection and sewage recovery needs.

[0105] Furthermore, the step of controlling the suction motor to operate according to the matched basic low power threshold and superimposing a preset breathing micro-desorption waveform includes A501~A502: Step A501: Using the matched basic low power threshold as the power baseline, a breathing-type micro-desorption waveform with a preset collapse depth and recovery cycle is generated. Step A502: Superimpose the power baseline with the breathing micro-desorption waveform to output the target drive power waveform, and control the operation of the suction motor according to the target drive power waveform to generate alternating negative pressure in the pipeline of the shower machine to break the vacuum adsorption between the shower machine nozzle and the skin.

[0106] It should be noted that in this embodiment, the power baseline refers to the basic low power threshold as the DC component or average value, which is the reference value of the breathing micro-desorption waveform; the breathing micro-desorption waveform refers to the control signal waveform superimposed on the basic low power threshold, which has periodic power collapse and recovery characteristics, and is used to generate alternating negative pressure in the pipeline. The preset collapse depth refers to the amplitude of the downward fluctuation of power in the breathing micro-desorption waveform; the recovery period refers to the length of time in the breathing micro-desorption waveform after power collapse to recover to the baseline and complete one fluctuation cycle; superposition refers to the algebraic addition of the breathing micro-desorption waveform as the AC component with the DC component of the power baseline to obtain the dynamic drive power command that changes with time. The target drive power waveform refers to the power-time function curve generated after superposition and used for real-time control of the sewage suction motor drive circuit. Alternating negative pressure refers to the periodically changing negative pressure generated by the sewage suction motor driven by the target drive power waveform, whose amplitude and frequency are consistent with the power waveform. Vacuum adsorption refers to the tight negative pressure adsorption state formed between the nozzle and the skin when the nozzle is in close contact with the skin, causing substantial blockage of the air extraction channel.

[0107] In this embodiment, while the suction motor operates in a reduced-power mode to avoid overcurrent burnout, the actively applied power fluctuations cause the suction motor to output alternating negative pressure, thereby generating tiny pressure pulsations between the nozzle and the skin. This repeatedly pushes the nozzle away from the skin and then draws it back in, gradually breaking the vacuum adsorption state between the nozzle and the skin. This achieves a self-recovery effect of "protecting while clearing blockages," avoiding the interruption of bathing caused by passively waiting for the blockage to be cleared.

[0108] In one possible implementation, the breathing micro-desorption waveform can be a triangular wave, a sawtooth wave, or a trapezoidal wave. The asymmetric characteristics of the sawtooth wave (slow rise and steep fall) can produce a "slow release and rapid absorption" effect, which is more conducive to breaking vacuum adsorption.

[0109] In one possible implementation, the preset collapse depth can be dynamically adjusted according to the current stall fault level. The higher the stall fault level, the greater the preset collapse depth is set, so as to generate a stronger alternating negative pressure pulling force.

[0110] In one possible implementation, the recovery cycle can be adaptively adjusted based on the feedback of the transient negative pressure waveform at the air inlet, so that the frequency of the breathing micro-desorption waveform approaches the physical resonant frequency inside the pipeline, thereby improving the efficiency of micro-desorption.

[0111] Furthermore, referring to Figure 3 The third embodiment of the sewage suction motor control method for the bath machine in this application provides a flowchart, based on the above. Figure 3 The embodiment shown further refines the step of "adjusting the water supply flow rate of the shower machine water pump" in step S40, including steps A601 to A605: Step A601: Calculate the equivalent average power of the target drive power waveform in real time.

[0112] It should be noted that in this embodiment, the target drive power waveform refers to the dynamic power curve used to actually drive the sewage suction motor, which is formed by superimposing the power baseline and the breathing micro-desorption waveform; the equivalent average power refers to the equivalent constant power value obtained by integrating and averaging the target drive power waveform that fluctuates over time over one or more periods.

[0113] This embodiment aims to transform the dynamically fluctuating power curve into a steady-state index that can intuitively reflect the actual energy consumption level of the sewage suction motor, so as to accurately calculate the maximum allowable load of the water supply pump.

[0114] In one possible implementation, the system directly uses the baseline low power threshold as the equivalent average power because the breathing micro-desorption waveform is typically designed as a symmetrical waveform (such as a sine wave or a triangular wave), with a time average of zero. Therefore, the superimposed average power equals the power baseline. In another possible implementation, when the breathing micro-desorption waveform is an asymmetrical waveform (e.g., a sawtooth wave with a slow rise and steep fall), the system numerically integrates the target drive power waveform over a complete waveform cycle and then divides by the cycle duration to obtain the equivalent average power.

[0115] Step A602: Calculate the maximum allowable sewage recovery flow rate of the shower machine water supply pump based on the equivalent average power.

[0116] It should be noted that, in this embodiment, the shower machine water supply pump refers to the power device in the shower machine system responsible for drawing cleaning water and supplying water to the shower head; the maximum allowable sewage recovery flow rate refers to the maximum water flow threshold that ensures sewage can be completely recovered without overflow under the suction capacity provided by the current equivalent average power of the sewage suction motor.

[0117] This embodiment establishes a dynamic balance constraint between the suction capacity of the sewage suction motor and the water supply capacity of the water supply pump to prevent sewage overflow caused by the water supply flow exceeding the suction capacity of the sewage suction motor and to ensure that the bathing machine can still operate safely in a low-power stall state.

[0118] In one possible implementation, the mapping relationship between equivalent average power and maximum permissible wastewater recovery flow rate can be calculated using a pre-fitted quadratic function curve.

[0119] In another possible implementation, the system uses an air inlet negative pressure sensor to provide real-time feedback and dynamically adjusts the flow rate using an adaptive algorithm. For example, when the suction motor is running at a certain power, the water supply flow rate is gradually increased until the sewage tank level begins to rise abnormally or a small amount of water accumulates at the nozzle. This critical flow rate is recorded as the maximum allowable recovery flow rate under the current power, and the mapping table is updated.

[0120] Step A603: Determine the preset ratio of the maximum allowable wastewater recycling flow rate as the target water supply flow rate.

[0121] It should be noted that in this embodiment, the preset ratio value refers to a safety factor less than 1, typically between 0.7 and 0.9, used to provide a certain safety margin based on the maximum recovery capacity to cope with instantaneous fluctuations or performance drift of the suction motor. The target water supply flow rate refers to the actual clean water flow rate output by the commanded water supply pump. This value should be less than or equal to the maximum allowable sewage recovery flow rate multiplied by this ratio to ensure that the sewage does not exceed the recovery capacity of the suction motor at any time.

[0122] This embodiment effectively addresses the risk of sewage overflow by reserving sufficient suction safety margin during low-power operation of the sewage suction motor, thus mitigating fluctuations in pipeline resistance or instantaneous water flow impacts.

[0123] In one possible implementation, the preset ratio value can be dynamically adjusted according to the current stall fault level, with the preset ratio value being smaller as the stall fault level increases.

[0124] Step A604: Compare the target water supply flow rate with the first rated flow rate corresponding to the first gear and the second rated flow rate corresponding to the second gear of the water supply pump, and determine the target gear from the first gear and the second gear according to the comparison result; the first rated flow rate is less than the second rated flow rate.

[0125] It should be noted that, in this embodiment, the first gear of the water supply pump refers to the low flow rate operation gear of the water supply pump; the first rated flow rate refers to the fixed water flow rate output by the water supply pump when it is continuously operating in the first gear; the second gear of the water supply pump refers to the high flow rate operation gear of the water supply pump; the second rated flow rate refers to the fixed water flow rate output by the water supply pump when it is continuously operating in the second gear; and the target gear refers to the safe operating gear that the water supply pump is about to switch to, selected based on the comparison results.

[0126] In this embodiment, given the hardware condition that the water supply pump only has discrete fixed speed settings, a safe speed setting is selected that meets bathing needs without causing sewage overflow due to excessive water supply. This setting operates at a speed closest to and not exceeding the pump's suction capacity, balancing bathing experience with overflow prevention. The system acquires a first rated flow rate and a second rated flow rate, compares the continuous target water supply flow rate value with these two discrete fixed rated flow rates, and selects the highest speed setting whose rated flow rate is not greater than the target water supply flow rate as the target speed setting. If the rated flow rates of both speed settings are greater than the target water supply flow rate, the lower speed setting is forcibly selected as the target speed setting, ensuring that the water supply flow rate is limited within a safe range.

[0127] In one possible implementation, when the target water supply flow rate is exactly between the first rated flow rate and the second rated flow rate, the system preferentially selects the first level as the target level to ensure that overflow prevention safety is prioritized.

[0128] Step A605: Output a gear switching signal to the water supply pump to control the water supply pump to run continuously at the target gear.

[0129] It should be noted that, in this embodiment, the gear switching signal refers to the level command signal used to control the water supply pump to switch windings or change tap connections to change the operating gear; continuous operation refers to the state in which the water supply pump maintains constant power supply and operates continuously at the selected gear.

[0130] This embodiment translates logical decisions into actual hardware actions, enabling the water pump to output water stably at the selected safety level. This avoids the water hammer effect and mechanical wear caused by frequent switching of fixed-frequency water pumps, achieves stable and reliable water supply, and effectively prevents sewage overflow.

[0131] In one possible implementation, the gear switching signal can be used to electrically switch between different tap windings of the water pump by controlling the coil activation of a double-pole double-throw relay.

[0132] This application calculates the target water supply flow rate by measuring the equivalent average power of the target drive power waveform. This flow rate is then compared with the rated flow rate of two fixed speed settings of the water supply pump to select a safe target speed setting. A speed setting switching signal is then output to control the water supply pump to operate continuously at the target speed setting. Even with the hardware limitation of the water supply pump having only discrete fixed speed settings and no linear speed adjustment, the flow comparison logic accurately maps continuous safe water supply demand to the closest and within-limit physical speed setting. This ensures that the water supply flow rate does not exceed the low-power suction capacity of the air pump, thus preventing sewage overflow, and avoids the water hammer effect and mechanical wear caused by frequent pump on / off cycles, guaranteeing the stable and safe operation of the shower machine in a stalled state.

[0133] Furthermore, the steps of restoring the operating power of the sewage suction motor to its rated operating power and restoring the rated water supply flow rate of the water supply pump after detecting that the stall fault has been cleared include A701~A703: Step A701: During the operation of the suction motor at a basic low power threshold, continuously monitor the operating current and the transient negative pressure waveform at the air inlet.

[0134] It should be noted that in this embodiment, the alternating negative pressure generated by the breathing micro-desorption waveform in real time is used to detect whether the vacuum adsorption between the nozzle and the skin is successfully broken, thereby timely capturing the changes in electrical and flow field parameters at the moment the stall condition is relieved. Specifically, the system's monitoring of the transient negative pressure waveform at the air inlet can include extracting the amplitude of its peaks and troughs to determine whether the alternating negative pressure tends to stabilize due to unobstructed pipeline.

[0135] Step A702: When the operating current drops below the recovery current threshold and the air inlet negative pressure recovers to the normal operating negative pressure range, a fault clearance signal is output.

[0136] It should be noted that, in this embodiment, the recovery current threshold refers to the preset upper limit of the current for determining that the stall state of the vacuum motor has been eliminated and normal operation has resumed; the air inlet negative pressure refers to the absolute pressure value of the air inlet of the vacuum motor's suction channel; the normal operating negative pressure range refers to the range of air inlet negative pressure of the vacuum motor under normal operating conditions without stall; and the fault clearance signal refers to the control flag used to indicate that the stall fault of the vacuum motor has been eliminated and to trigger the system to resume normal operation.

[0137] This embodiment uses cross-verification based on both electrical and flow field conditions to accurately determine whether the micro-desorption operation has successfully removed the nozzle from the vacuum adsorption, avoiding misjudgments caused by fluctuations in a single parameter and ensuring that the equipment performs a recovery operation after the blockage is truly cleared.

[0138] Step A703: Based on the fault clearance signal, output a command to restore rated operating power to the sewage suction motor and output a command to restore rated water supply flow to the water supply pump.

[0139] It should be noted that in this embodiment, the rated operating power refers to the standard full-load operating power of the sewage suction motor under normal, unblocked conditions; the rated water supply flow rate refers to the standard water supply value of the water supply pump under normal bathing operation mode.

[0140] In one possible implementation, to avoid instantaneous full-load current surges, when the system outputs commands to restore rated operating power and rated water supply flow, it can use a ramp-up method to control the power and flow to gradually and smoothly reach the rated values.

[0141] This application continuously monitors the operating current and inlet negative pressure of the suction motor during low-power operation. When both return to normal ranges, it determines that the stall has been cleared and outputs a signal, thereby controlling the suction motor and water pump to return to their rated operating state. It achieves accurate identification of the true stall clearance state through cross-verification of electrical and flow field parameters, avoiding misjudgments and premature recovery caused by fluctuations in a single parameter. Thus, after ensuring that the equipment is safely free from the risk of stalling, it quickly restores the full-power shower function, ensuring the continuity of showering and the safety of equipment operation.

[0142] Furthermore, the method also includes a timeout protection step: Timing is started during the process of controlling the vacuum motor to operate according to the basic low power threshold; When the duration of the basic low-power threshold operation reaches the preset timeout threshold and no stall fault clearance signal is received, a forced shutdown command is output to control the sewage suction motor and water supply pump to stop, and an alarm prompt signal is output.

[0143] It should be noted that, in this embodiment, the timeout protection step refers to a safety fallback mechanism set up to prevent the equipment from being in an abnormal stalled and derated operating state for an extended period of time; the preset timeout threshold refers to the maximum time limit that the suction motor is allowed to continuously attempt desorption in a low-power stalled state as preset by the system; the stall fault clearance signal refers to a control indicator used to indicate that the stall fault of the suction motor has been eliminated and to trigger the system to resume normal operation; the forced shutdown command refers to a hard control signal issued by the system requiring the equipment to immediately stop operating; and the alarm prompt signal refers to the warning information output by the system to indicate to the user that the equipment has an abnormality and requires manual intervention.

[0144] In this embodiment, an internal timer is simultaneously activated when entering the reduced-power breathing operation phase. The accumulated running time is compared with a preset timeout threshold in real time. If no feedback signal indicating that the stall has been cleared is received after the timeout, the automatic desorption is deemed to have failed. The system directly cuts off the drive output of the suction motor and the water supply pump, and triggers an audible and visual or interface alarm to prompt the user to manually remove the nozzle. This aims to limit the maximum ineffective running time of the suction motor at low power, terminate operation in a timely manner when the breathing micro-desorption strategy fails to break the vacuum adsorption, avoid the equipment being in abnormal operating conditions for a long time or even experiencing extreme thermal runaway, and prompt the user to intervene actively.

[0145] In one possible implementation, the preset timeout threshold can be set differently based on the current stall fault level; the higher the stall fault level, the shorter the preset timeout threshold.

[0146] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the sewage suction motor control method of the bath machine of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0147] This application also provides a vacuum motor control device for a bath machine; please refer to [reference needed]. Figure 4 The vacuum motor control device of the bath machine includes: Data acquisition module 10 is used to acquire multi-source heterogeneous sensor data of the vacuum motor of the shower machine in real time based on asynchronous sampling frequency; The fault identification module 20 is used to input the multi-source heterogeneous sensing data into a multimodal fusion network based on an attention mechanism, and output a fused feature vector to identify the stall fault type of the current sewage suction motor. The fault matching module 30 is used to determine the level of the stall fault based on the over-limit magnitude of the current operating current and / or the temperature rise rate of the current coil if the stall fault type is the first type, and to match the corresponding basic low power threshold based on the stall fault level. The fault handling module 40 is used to control the sewage suction motor to operate according to the matched basic low power threshold, superimpose the preset breathing micro-desorption waveform and adjust the water supply flow of the shower machine water supply pump. The status recovery module 50 is used to restore the working power of the sewage suction motor to the rated working power and restore the rated water supply flow of the water supply pump after the stall fault is detected to be resolved. The status holding module 60 is used to maintain the current operating status of the sewage suction motor if the stall fault type is the second type.

[0148] The wastewater suction motor control device for a shower machine provided in this application adopts the wastewater suction motor control method for a shower machine in the above embodiments. This solves the technical problem that existing wastewater suction motor stall protection causes sewage overflow, failing to simultaneously address overcurrent protection and maintain the need for overflow prevention during recycling. Compared with the prior art, the beneficial effects of the wastewater suction motor control device for a shower machine provided in this application are the same as those of the wastewater suction motor control method for a shower machine provided in the above embodiments. Furthermore, other technical features of the wastewater suction motor control device for a shower machine are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0149] This application provides a vacuum motor control device for a bathing machine. The vacuum motor control device for the bathing machine includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the vacuum motor control method for the bathing machine in the above embodiment 1.

[0150] The following is for reference. Figure 5 The diagram illustrates a structural schematic of a vacuum motor control device suitable for implementing embodiments of the present application for a shower machine. The vacuum motor control device for the shower machine in the embodiments of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 5 The illustrated vacuum motor control device for a shower machine is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0151] like Figure 5 As shown, the vacuum motor control device of a shower machine may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory 1002 or a program loaded from a storage device 1003 into a random access memory 1004. The random access memory 1004 also stores various programs and data required for the operation of the vacuum motor control device of the shower machine. The processing unit 1001, the read-only memory 1002, and the random access memory 1004 are interconnected via a bus 1005. An input / output interface 1006 is also connected to the bus. Typically, the following systems can be connected to the input / output interface 1006: input devices 1007 including, for example, a touch screen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. The communication device 1009 allows the vacuum motor control equipment of the shower machine to communicate wirelessly or wiredly with other devices to exchange data. Although the figure shows a vacuum motor control equipment for a shower machine with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented alternatively.

[0152] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from read-only memory 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0153] The wastewater suction motor control device for a shower machine provided in this application adopts the wastewater suction motor control method for a shower machine in the above embodiments. This solves the technical problem that existing wastewater suction motor stall protection causes sewage overflow, failing to simultaneously address overcurrent protection and maintain the need for overflow prevention during recycling. Compared with the prior art, the beneficial effects of the wastewater suction motor control device for a shower machine provided in this application are the same as those of the wastewater suction motor control method for a shower machine provided in the above embodiments. Furthermore, other technical features of this wastewater suction motor control device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0154] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0155] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for controlling the suction motor of a bath machine, characterized in that, The method for controlling the vacuum motor of the bath machine includes: Real-time acquisition of multi-source heterogeneous sensor data from the vacuum motor of the shower machine based on asynchronous sampling frequency; The multi-source heterogeneous sensing data is input into a multimodal fusion network based on an attention mechanism, and the fused feature vector is output to identify the current stall fault type of the vacuum motor. If the stall fault type is the first type, the stall fault level is determined based on the over-limit magnitude of the current operating current and / or the temperature rise rate of the current coil temperature, and the corresponding basic low power threshold is matched based on the stall fault level. The vacuum motor is controlled to operate according to the matched basic low power threshold, and a preset breathing micro-desorption waveform is superimposed and the water supply flow rate of the shower machine water supply pump is adjusted. After the stall fault is detected and resolved, the working power of the sewage suction motor is restored to the rated working power, and the rated water supply flow of the water supply pump is restored. If the stall fault type is the second type, then the current operating state of the suction motor is maintained.

2. The method for controlling the vacuum motor of a bathing machine as described in claim 1, characterized in that, The multi-source heterogeneous sensing data includes at least the operating current, coil temperature, housing vibration spectrum, and transient negative pressure waveform at the air inlet. The steps for real-time acquisition of multi-source heterogeneous sensor data from the vacuum motor of the shower machine based on asynchronous sampling frequency include: An electrical thermal characteristic sequence is obtained by acquiring the operating current and coil temperature using the first sampling frequency; The flow field characteristic sequence is obtained by acquiring the shell vibration spectrum and the transient negative pressure waveform at the air inlet using the second sampling frequency; The electrical thermal feature sequence and the flow field feature sequence are timestamped and normalized to output a standardized feature matrix.

3. The method for controlling the vacuum motor of a bathing machine as described in claim 2, characterized in that, The step of inputting the multi-source heterogeneous sensing data into a multimodal fusion network based on an attention mechanism and outputting a fused feature vector to identify the stall fault type of the current suction motor includes: The standardized feature matrix is ​​input into the feature extraction layer of the multimodal fusion network for feature encoding, and the initial feature vectors of each modality are output. The initial feature vector is input into the attention mechanism layer of the multimodal fusion network, where dynamic weights are assigned and weighted fusion is performed to output a fused feature vector. The fused feature vector is input into the classification and recognition layer of the multimodal fusion network for mapping calculation, and the current stall fault type of the sewage suction motor is output.

4. The method for controlling the vacuum motor of a bathing machine as described in claim 3, characterized in that, The step of determining the stall fault level based on the extent of the current operating current exceeding the limit and / or the rate of temperature rise of the current coil, and matching the corresponding basic low power threshold based on the stall fault level, includes: Calculate the extent by which the current operating current exceeds the stall current threshold and the rate of temperature rise of the coil. The above-mentioned exceedance range and temperature rise rate are input into a preset level determination model, and the stall fault level is output. Look up the pre-stored mapping table between stall fault levels and basic low power thresholds, and output the basic low power threshold corresponding to the current stall fault level.

5. The method for controlling the suction motor of a bathing machine according to claim 1, characterized in that, The step of controlling the suction motor to operate according to the matched basic low power threshold and superimposing a preset breathing micro-desorption waveform includes: Using the matched baseline low power threshold as the power baseline, a breathing-type micro-desorption waveform with a preset collapse depth and recovery cycle is generated. The power baseline is superimposed with the breathing micro-desorption waveform to output the target drive power waveform, and the operation of the suction motor is controlled according to the target drive power waveform to generate alternating negative pressure in the pipeline of the shower machine to break the vacuum adsorption between the shower machine nozzle and the skin.

6. The method for controlling the suction motor of a bath machine according to claim 5, characterized in that, The steps for adjusting the water supply flow rate of the shower machine's water supply pump include: The equivalent average power of the target driving power waveform is calculated in real time. Calculate the maximum allowable sewage recovery flow rate of the shower machine's water supply pump based on the equivalent average power. The preset ratio of the maximum allowable wastewater recycling flow rate is determined as the target water supply flow rate; The target water supply flow rate is compared with the first rated flow rate corresponding to the first gear and the second rated flow rate corresponding to the second gear of the water supply pump, and the target gear is determined from the first gear and the second gear according to the comparison result; the first rated flow rate is less than the second rated flow rate; Output a gear switching signal to the water supply pump to control the water supply pump to run continuously at the target gear.

7. The method for controlling the vacuum motor of a bathing machine according to claim 1, characterized in that, The steps of restoring the operating power of the vacuum motor to its rated operating power and the rated water supply flow rate of the water supply pump after detecting that the stall fault has been cleared include: During the operation of the suction motor at a basic low power threshold, the operating current and the transient negative pressure waveform at the air inlet are continuously monitored; When the operating current drops below the recovery current threshold and the air inlet negative pressure recovers to the normal operating negative pressure range, a fault clearance signal is output. Based on the fault clearance signal, a command to restore rated operating power is output to the sewage suction motor, and a command to restore rated water supply flow rate is output to the water supply pump.

8. The method for controlling the vacuum motor of a bathing machine according to claim 1, characterized in that, The method also includes a timeout protection step: Timing is started during the process of controlling the vacuum motor to operate according to the basic low power threshold; When the duration of the basic low-power threshold operation reaches the preset timeout threshold and no stall fault clearance signal is received, a forced shutdown command is output to control the sewage suction motor and water supply pump to stop, and an alarm prompt signal is output.

9. A vacuum motor control device for a bath machine, characterized in that, The vacuum motor control device of the bath machine includes: The data acquisition module is used to acquire multi-source heterogeneous sensor data of the vacuum motor of the shower machine in real time based on asynchronous sampling frequency; The fault identification module is used to input the multi-source heterogeneous sensor data into a multimodal fusion network based on an attention mechanism, and output a fused feature vector to identify the stall fault type of the current sewage suction motor. The fault matching module is used to determine the level of the stall fault based on the over-limit magnitude of the current operating current and / or the temperature rise rate of the current coil if the stall fault type is the first type, and to match the corresponding basic low power threshold based on the stall fault level. The fault handling module is used to control the sewage suction motor to operate according to the matched basic low power threshold, superimpose a preset breathing micro-desorption waveform and adjust the water supply flow of the shower machine water pump. The status recovery module is used to restore the working power of the sewage suction motor to the rated working power and the rated water supply flow of the water supply pump after the stall fault is detected to be resolved. The status maintenance module is used to maintain the current operating status of the sewage suction motor if the stall fault type is the second type.

10. A vacuum motor control device for a bath machine, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the method for controlling the vacuum motor of a bath machine as described in any one of claims 1 to 8.

Images