Filter element fault diagnosis classification method, water purifier and filter element fault diagnosis classification device

Abstract

The application discloses a filter core fault diagnosis classification method, a water purifier and a filter core fault diagnosis classification device, relates to the technical field of water purification equipment, and applies the filter core fault diagnosis classification method to a multistage filter core system. The multistage filter core system comprises a multistage filter core and a driving module. The driving module is used for providing power for liquid delivered to the multistage filter core. The driving module comprises a motor. The filter core fault diagnosis classification method comprises the following steps: controlling the driving module to operate in a set operation mode, and acquiring current characteristics of the motor in the set operation mode; and determining a specific filter core position of the multistage filter core in which a fault occurs according to the current characteristics in the operation mode. The technical problem of how to timely and reliably determine the specific filter core position of the multistage filter core in which a fault occurs is solved without additionally setting a detection device.

Description

Technical Field

[0001] This application relates to the field of water purification equipment technology, and in particular to a filter cartridge fault diagnosis and classification method, a water purifier, and a filter cartridge fault diagnosis and classification device. Background Technology

[0002] Water purifiers and other devices are equipped with multi-stage filter cartridges. Different filter cartridges have different filtration functions, and during use, the degree of aging and clogging of different filter cartridges is also different. This makes it impossible to determine which stage of the filter cartridge is faulty when a filter cartridge fails.

[0003] Currently, the relevant technologies for fault detection of multi-stage filter elements generally require setting up sensors for each stage of the filter element to detect its status. This approach not only increases costs and equipment complexity due to the need to install additional sensors and other detection devices, but also makes it impossible to determine the specific location of the faulty filter element in a timely and reliable manner because it requires separate status detection for each stage of the filter element. Summary of the Invention

[0004] The main purpose of this application is to provide a filter cartridge fault diagnosis and classification method, a water purifier, and a filter cartridge fault diagnosis and classification device, aiming to solve the technical problem of how to timely and reliably determine the specific location of a faulty filter cartridge in a multi-stage filter cartridge without additional detection devices.

[0005] On one hand, a filter element fault diagnosis and classification method is provided, applied to a multi-stage filter element system. The multi-stage filter element system includes multiple filter elements and a drive module. The drive module provides power to the liquid delivered to the multi-stage filter elements and includes a motor. The filter element fault diagnosis and classification method includes the following steps: The control drive module operates according to the set operating mode and acquires the current characteristics of the motor in the set operating mode; Based on the current characteristics under the operating mode, the specific location of the faulty filter element in the multi-stage filter element is determined.

[0006] In one embodiment, determining the specific location of the faulty filter element among the multi-stage filter elements based on the current characteristics under the operating mode specifically includes the following steps: Combine the phase current data of the motor under the operating mode into a multi-mode feature vector; The multi-mode feature vectors are classified to determine the specific location of the faulty filter element in the multi-stage filter element based on the classification results.

[0007] In one embodiment, classifying the multi-mode feature vectors to determine the specific location of the faulty filter element in the multi-stage filter element based on the classification results specifically includes the following steps: The first classifier is used to classify the multi-mode feature vectors to analyze and determine the locations of candidate filter elements that may fail in the multi-stage filter elements, and the first classification result is obtained. Calculate and determine the confidence level of the candidate filter locations in the first classification results; When the confidence level is not less than a preset threshold, the candidate filter element location is determined as the specific filter element location where the failure has occurred. When the confidence level is less than a preset threshold, the first classification result is analyzed by the second classifier to obtain the second classification result. The first classification result and the second classification result are combined to calculate the probability of failure at each candidate filter element location. The specific filter element location where the failure occurs is determined from the probability of failure at each candidate filter element location.

[0008] In one embodiment, the first classifier is a random forest classifier, and the second classifier is a neural network classifier.

[0009] In one embodiment, the control drive module operates according to a set operating mode and acquires the current characteristics of the motor in the set operating mode, specifically including the following steps: The system acquires the current phase current data of the motor, controls the drive module to run at the set operating speed according to the current phase current data of the motor, acquires the operating status of the filter element, and acquires the phase current data of the motor in each operating mode.

[0010] In one embodiment, the multi-stage filter element includes at least a first-stage filter element, a second-stage filter element, and a third-stage filter element connected in series; the step of acquiring the operating status of the filter element and acquiring the phase current data of the motor in each operating mode specifically includes the following steps: When the control drive module is running at the first operating speed, it shall at least obtain the operating status of the first-stage filter element and obtain the phase current data of the motor in the first operating mode; When the control drive module is running at the second operating speed, at least the operating status of the second-stage filter element and the phase current data of the motor in the second operating mode are obtained. When the control drive module is running at the third operating speed, at least the operating status of the third-stage filter element and the phase current data of the motor in the third operating mode should be obtained. When the control drive module is running at the fourth operating speed, it acquires the operating status of each filter element and the phase current data of the motor in the fourth operating mode. The first operating speed is less than the second operating speed, the third operating speed is less than the fourth operating speed, and the fourth operating speed is determined based on the current phase current data of the motor.

[0011] In one embodiment, the operating state of the filter element is obtained in the following manner: Obtain the correspondence between motor phase current data and filter element operating status, and obtain the current phase current data of the motor. Based on the current phase current data of the motor and the correspondence between the motor phase current data and the filter element operating status, obtain the operating status of the filter element. And / or, acquire detection data from the sensor module, and determine the operating status of the filter element based on the detection data, wherein the sensor module includes at least one or a combination of multiple of a water quality sensor, a flow sensor, a temperature sensor, and a sound sensor.

[0012] In one embodiment, before executing the step of the control drive module operating according to a set operating mode and obtaining the current characteristics of the motor in the set operating mode, the filter element fault diagnosis and classification method further includes the following steps: In response to the running command, the current phase current data of the motor is acquired and processed. Obtain the mapping relationship between motor phase current data and filter element faults; Based on the processed current phase current data of the motor and the mapping relationship between the motor phase current data and the filter element fault, when a multi-stage filter element fails, a difference test is initiated.

[0013] In one embodiment, processing the current phase current data of the motor specifically includes the following steps: The current phase current data of the motor is preprocessed; Calculate the time-domain characteristics based on the pre-processed current phase current data of the motor; Perform a Fourier transform on the preprocessed current phase current data of the motor to obtain the spectrum, and obtain the frequency domain characteristics calculated based on the spectrum; Wavelet transform is performed on the preprocessed current phase current data of the motor to obtain the time-frequency distribution, and time-frequency domain features calculated based on the time-frequency distribution are obtained. The time-domain features, frequency-domain features, and time-frequency-domain features are combined to form a feature vector to be determined. The feature vector to be determined is then processed to obtain the target feature vector. When a multi-stage filter element malfunction is determined based on the processed current phase current data of the motor and the mapping relationship between the motor phase current data and the filter element malfunction, a differential test is initiated, which specifically includes the following steps: Based on the mapping relationship between the target feature vector, the motor phase current data, and the filter element fault, when a multi-stage filter element fails, a difference test is initiated.

[0014] In one embodiment, when a multi-stage filter element is determined to have a fault based on the processed current phase current data of the motor and the mapping relationship between the motor phase current data and the filter element fault, a differential test is initiated, which specifically includes the following steps: Acquire normal operating condition samples and various filter element failure operating condition samples, label various failure types, determine the mapping relationship between motor phase current data and filter element failure, and construct a third classifier containing a classification model; When the third classifier determines that a multi-stage filter element has failed based on the processed current phase current data of the motor, the mapping relationship between the motor phase current data and the filter element failure, a differential test is initiated. The third classifier is an SVM classifier.

[0015] In one embodiment, the filter element fault diagnosis and classification method further includes the following steps: Establish communication between the multi-stage filter system and the cloud platform; In response to the algorithm optimization command, the mapping relationship between motor phase current data and filter element faults from different user terminals is obtained and aggregated through the cloud platform, and the classification model is optimized and updated based on the obtained mapping relationship between motor phase current data and filter element faults from different user terminals.

[0016] In one embodiment, after performing the step of determining the specific location of the faulty filter element in the multi-stage filter element based on the current characteristics under the operating mode, the method further includes the following steps: Calculate the probability of filter failure at the specific location of the failure and the remaining life of the filter at the specific location of the failure, and generate a diagnostic report based on the calculation results; The control display interface shows the diagnostic report and maintenance information for the faulty filter element.

[0017] In one embodiment, the filter element fault diagnosis and classification method further includes the following steps: In response to system initialization commands, obtain classification models and configuration information; After determining the specific location of the faulty filter element in the multi-stage filter system, the following steps are also included: Outputs a filter replacement reminder for the specific location of the faulty filter element; In response to the user's filter replacement confirmation command, the configuration information is updated.

[0018] On the other hand, a water purifier is provided, including a multi-stage filter system, the multi-stage filter system comprising: A multi-stage filter element is used for filtering liquids, wherein the multi-stage filter element includes at least a first-stage filter element, a second-stage filter element, and a third-stage filter element connected in series. The drive module includes a variable frequency pump, a motor, and a motor drive circuit. The variable frequency pump is connected to the first-stage filter element and is used to provide power for the liquid delivered to the multi-stage filter element. The motor is driven and connected to the variable frequency pump and is used to provide power for the variable frequency pump. The motor drive circuit is driven and connected to the motor drive circuit and is used to drive the motor to work. A current sampling circuit is used to collect the phase current data of the motor; The controller stores and executes the filter element fault diagnosis and classification program, and implements the steps of the filter element fault diagnosis and classification method as described in the above embodiment when executing the program.

[0019] In one embodiment, the multi-stage filter system further includes a user interface for displaying diagnostic reports and filter maintenance information. The diagnostic report includes at least the probability of filter failure at the specific location of the failure and the remaining lifespan of the filter at the specific location of the failure.

[0020] On the other hand, a filter cartridge fault diagnosis and classification device is provided, which includes: a memory, a processor, and a filter cartridge fault diagnosis and classification program stored in the memory and executable on the processor. The filter cartridge fault diagnosis and classification program is configured to implement the steps of the filter cartridge fault diagnosis and classification method as described in the above embodiments.

[0021] One or more technical solutions proposed in this application have at least the following technical effects: The control drive module operates according to a set operating mode and acquires the current characteristics of the motor in the set operating mode. Based on the current characteristics in the operating mode, the specific location of the faulty filter element in the multi-stage filter element is determined. No additional detection device is installed; instead, the specific location of the faulty filter element in the multi-stage filter element is determined based on the current characteristics of the motor in the set mode. This effectively solves the technical problem of not being able to determine the specific location of the faulty filter element in the multi-stage filter element in a timely and reliable manner. By reducing the use of additional detection devices, it not only reduces costs, equipment complexity, and failure rate, but also improves system reliability. Because current signals can be acquired and processed in real time, the specific location of the faulty filter element in a multi-stage filter system can be directly determined by acquiring the current characteristics of the motor under a set operating mode. This effectively reduces errors, improves the accuracy of fault diagnosis, and enables precise diagnosis of the specific location of the faulty filter element in a multi-stage filter system. Furthermore, it addresses the issues of premature or delayed filter element replacement. Attached Figure Description 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.

[0022] 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.

[0023] Figure 1 A flowchart is provided for an embodiment of the filter element fault diagnosis and classification method of this application; Figure 2 A partial flowchart is provided for an embodiment of the filter element fault diagnosis and classification method of this application; Figure 3 One of the partial flowcharts provided for an embodiment of the filter element fault diagnosis and classification method of this application; Figure 4 A flowchart illustrating an embodiment of step S200 of this application is provided; Figure 5 A flowchart illustrating an embodiment of step S110 of this application is provided; Figure 6 A second partial flowchart is provided for an embodiment of the filter element fault diagnosis and classification method of this application; Figure 7 A partial flowchart is provided for another embodiment of the filter element fault diagnosis and classification method of this application; Figure 8 A flowchart illustrating an embodiment of step S310 of this application is provided; Figure 9 The third schematic diagram of a partial process is provided for an embodiment of the filter element fault diagnosis and classification method of this application; Figure 10 A flowchart illustrating an embodiment of step S330 of this application is provided; Figure 11 The current waveform is shown under normal operating conditions. Figure 12 The current waveform diagram for a PP cotton filter element experiencing a blockage fault; Figure 13 The current waveform diagram shows the current when the activated carbon filter element becomes clogged. Figure 14 The current waveform diagram for a blocked RO reverse osmosis membrane; Figure 15 A flowchart illustrating an embodiment of step S220 of this application is provided; Figure 16 Fourth partial flowchart provided for an embodiment of the filter element fault diagnosis and classification method of this application; Figure 17Fifth schematic diagram of a partial process flow provided for an embodiment of the filter element fault diagnosis and classification method of this application; Figure 18 A partial flowchart is provided for yet another embodiment of the filter element fault diagnosis and classification method of this application; Figure 19 A partial flowchart is provided for another embodiment of the filter element fault diagnosis and classification method of this application; Figure 20 A partial flowchart of an embodiment of the filter element fault diagnosis and classification method of this application is provided (Figure 6). Figure 21 A partial flowchart is provided for another embodiment of the filter element fault diagnosis and classification method of this application; Figure 22 A schematic diagram of a module provided for one embodiment of the water purifier of this application; Figure 23 This is a schematic diagram of one embodiment of the multi-stage filter element of this application; Figure 24 This is a schematic diagram of one embodiment of the current sampling module of this application.

[0024] Explanation of icon numbers: 100. Multi-stage filter cartridge; 110. PP cotton filter cartridge; 120. Activated carbon filter cartridge; 130. RO reverse osmosis membrane; 140. Mineralization filter cartridge; 210. Variable frequency pump; 220. Motor; 230. Motor drive circuit; 240. Analog-to-digital converter module; 300. Current sampling circuit; 410. Signal processing module; 420. Feature extraction module; 430. Fault classification module; 440. Control module; 500. User interface.

[0025] 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

[0026] 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.

[0027] 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.

[0028] Water purifiers and similar devices are equipped with multi-stage filter cartridges. Different cartridges have different filtration functions, and multi-stage filters can achieve progressive purification, offering higher filtration efficiency and a wider range of applications compared to single-stage filters. However, during use, the aging and clogging levels of different cartridges vary, which makes it difficult to pinpoint which stage of the filter has malfunctioned when a filter fails.

[0029] Currently, related technologies for multi-stage filter failure detection employ a uniform replacement reminder strategy, replacing all filters regardless of their condition. Alternatively, separate sensors are needed for each filter stage to monitor its status. Adding extra sensors and other detection devices not only increases product cost and complexity but also increases potential failure points due to improper installation. Furthermore, it fails to reliably pinpoint the location of the faulty filter within the multi-stage system, impacting user experience and maintenance efficiency. When a filter failure occurs, the inability to determine its exact location can lead to premature or delayed replacement of certain filters.

[0030] Some embodiments of this application propose a filter cartridge fault diagnosis and classification method, a water purifier, and a filter cartridge fault diagnosis and classification device, which are used to determine the specific location of a faulty filter cartridge in a multi-stage filter cartridge in a timely and reliable manner without the need for additional sensors or other detection devices.

[0031] In this application, the filter cartridge fault diagnosis and classification method is mainly applied to water purifiers, water purification systems, and other devices and systems with water purification functions, with the controller as the primary execution unit. The controller can be located within the water purifier, a control device connected to the water purifier, or independently of the various components of the water purifier. It can execute various appropriate actions and processes based on programs stored in read-only memory (ROM) or programs loaded from storage devices into random access memory (RAM). The RAM also stores various programs and data required for water purification. The controller and storage modules such as ROM and RAM are interconnected via a bus; the storage modules can also include storage devices such as magnetic tapes or hard disks. Input / output (I / O) interfaces are also connected to the bus. Typically, the following systems can be connected to the I / O interface: input devices such as touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, and gyroscopes as input modules; output devices such as liquid crystal displays (LCDs), speakers, and vibrators as output modules; and communication devices. Communication devices allow water purifiers to communicate wirelessly or wiredly with other devices to exchange data.

[0032] In this application, for ease of description, the main focus is on control devices such as controllers and control devices with control functions as the executing entities. The filter element fault diagnosis and classification method is applied to a multi-stage filter element system, which includes multiple filter elements and a drive module. The multi-stage filter elements adsorb or filter impurities in the liquid. The drive module provides power to the liquid delivered to the multi-stage filter elements and includes a motor.

[0033] like Figure 1 As shown, the filter element fault diagnosis and classification method includes the following steps: Step S100: The control drive module operates according to the set operating mode and acquires the current characteristics of the motor in the set operating mode.

[0034] The current characteristics of a motor, as a key parameter reflecting its operating status, can be used to determine the motor load, speed, torque, etc. The drive module of a multi-stage filter system is used to provide power for the liquid delivered to the multi-stage filter. When the motor is working, the current characteristics of the motor can indirectly reflect the state of the filter, such as blockage or damage, and can then be used to further determine the specific location of the faulty filter in the multi-stage filter.

[0035] For example, current signals can be acquired using current detection components such as Hall effect sensors, current sensors, or shunt resistors. Taking a three-phase motor as an example, the current detection component covers the three-phase circuit of the motor to achieve current acquisition. Based on the acquired motor phase current data and other current signals, time-domain analysis is performed by calculating the mean and peak values ​​of the current. The current signal is then converted into a spectrum using a Fast Fourier Transform (FFT), and harmonic components are analyzed to identify abnormal frequency characteristics, thereby obtaining the current characteristics of the motor under the set operating mode.

[0036] Different filter elements exhibit varying current characteristics when the drive module operates in different modes. By controlling the drive module's motor to run in multiple preset modes, the system's operating conditions can be altered, thereby stimulating the characteristics of different filter elements under varying operating conditions. These different operating modes allow the characteristics of faulty filter elements to be more clearly revealed, facilitating subsequent feature extraction and analysis.

[0037] Current characteristics can reflect the working state of a filter element under specific operating conditions, such as the magnitude and fluctuation of the current. For example, there can be one or more operating modes. When there are multiple operating modes, it is necessary to acquire the current characteristics of the motor under each of these modes. The current characteristics of the motor differ under different operating modes (e.g., different motor speeds, different water consumption). While filter elements in the same location may have similar current characteristics under the same operating mode, filter elements in different locations will exhibit significant differences under different operating modes. Acquiring the current characteristics of the motor under multiple operating modes can eliminate interference factors from a single operating mode, distinguishing filter element faults in different locations and improving fault location accuracy. Under different operating modes, the current characteristics of a faulty filter element will differ from those of a normal filter element. In each operating mode, the phase current data of the filter element is detected and recorded in real time. By recording and acquiring these characteristics, a basis for subsequent fault diagnosis can be provided.

[0038] Step S200: Based on the current characteristics in the operating mode, determine the specific location of the faulty filter element in the multi-stage filter element.

[0039] Filter cartridge failures include, but are not limited to, clogging, damage (such as aging or damage from external impact), misalignment, and cartridge failure. Taking clogging as an example, when a motor experiences clogging, the resulting change in water flow resistance affects the motor load. The motor's phase current data reflects this load change, and the motor load is closely related to the degree of filter cartridge clogging: when the filter cartridge is not clogged, the water flow resistance is low, requiring less power to drive the liquid through the cartridge, resulting in a lower motor drive current; as the filter cartridge becomes clogged, the water flow resistance increases, requiring the motor to output greater torque to maintain a constant flow rate or pressure, leading to a significant increase in current. The rate of change in current reflects the speed of filter cartridge clogging; for example, a rapid increase in current indicates worsening clogging, while a slow change indicates a slower clogging process. The current rate of change in current can be used to determine the current degree of filter cartridge clogging. Taking sudden blockages and damage as examples, abnormal states that may occur can be identified based on the acquired rate of change of current. For example, when the phase current of the motor rises sharply in a short period of time and exceeds the preset current threshold, it can be determined that the filter element may be suddenly blocked. When the filter element is damaged, the fluid resistance drops sharply, causing the motor load to fluctuate. The motor current may oscillate irregularly, and the energy of the low-frequency component in the frequency domain may rise abnormally. Therefore, when the motor current fluctuates irregularly and the frequency characteristics are abnormal, it can be determined that the filter element may be damaged.

[0040] For multi-stage filter systems, the impact on water flow resistance varies depending on the location of the malfunctioning filter within the system. Multi-stage filters can include pre-filters (such as PP cotton) and post-filters (such as RO membranes). Taking a pre-filter as an example, because it is located at the front and has a greater impact on the overall flow rate, a malfunction in the pre-filter will result in a significant increase in motor current data, such as phase current. Similarly, with post-filters, which have smaller pores and higher resistance, a malfunction in the post-filter will cause a rapid increase in motor current within a short period.

[0041] As an example, the current characteristics (such as time-domain and frequency-domain parameters) of each filter element under normal operating conditions can be acquired under different operating modes (such as different motor operating speeds, different water consumption, etc.) to determine the current range of each filter element under normal conditions; and the current characteristics of each filter element under different operating modes when a fault occurs can also be acquired. Current characteristics can be further obtained by acquiring current signals such as the phase current of the motor in real time. Based on the acquired current signals, when an abnormal current situation is determined, such as the current exceeding the normal current range or the current characteristics differing from those under normal conditions, a fault is identified. The acquired current characteristics can be compared with the current characteristics of each filter element when a fault occurs to determine the specific location of the faulty filter element; alternatively, a correspondence between filter element faults and current characteristics can be established, and based on this correspondence (such as the mapping relationship between motor phase current data and filter element faults), the specific location of the faulty filter element can be determined from the acquired current characteristics.

[0042] As another example, the current characteristics of each stage of the filter element when it fails under different operating modes can be obtained to determine the correspondence between the fault and the current characteristics (such as the mapping relationship between motor phase current data and filter element fault). Based on the correspondence between the fault and the current characteristics, the specific location of the filter element that has failed can be determined from the obtained current characteristics.

[0043] As another example, the impact of filter failure at different locations in a multi-stage filter system on current characteristics can be established. After obtaining the current characteristics of each filter failure under different operating modes, the specific location of the faulty filter can be determined based on the impact of filter failure at different locations in the multi-stage filter system on current characteristics.

[0044] Compared to solutions that use multiple pressure sensors and flow meters for each stage of the filter cartridge, which are costly and complex, prone to creating additional fault points due to improper installation, and unable to accurately pinpoint the location of the faulty filter cartridge, this application's embodiment eliminates the need for additional pressure sensors or flow meters. Instead, it utilizes the water purifier's built-in drive module and current sampling circuit. By real-time acquisition of the motor current, the current characteristics of the motor in the set mode can be determined, and further, the location of the faulty filter cartridge in the multi-stage filter cartridge can be determined based on the current characteristics during operation. This effectively solves the technical problem of not being able to reliably and timely determine the location of the faulty filter cartridge in the multi-stage filter cartridge. By reducing the use of additional detection devices, it not only lowers costs, reduces equipment complexity and failure rate, but also improves system reliability. Because current signals can be acquired and processed in real time, the location of the faulty filter cartridge in the multi-stage filter cartridge can be directly determined by acquiring the current characteristics of the motor in the set operating mode, effectively reducing errors and improving the accuracy of fault diagnosis. In some scenarios, the fault diagnosis accuracy can even reach over 90%. This allows for precise diagnosis of the specific location of the faulty filter element, and further addresses the issues of replacing the filter element too early or too late.

[0045] like Figure 2 , Figure 3 As shown, in one embodiment, step S100, controlling the drive module to operate according to a set operating mode and acquiring the current characteristics of the motor in the set operating mode, specifically includes the following steps: Step S110: Obtain the current phase current data of the motor, control the drive module to run at the running speed of the set operating mode according to the current phase current data of the motor, and obtain the operating status of the filter element, as well as the phase current data of the motor in each operating mode.

[0046] The current characteristics of a motor are directly related to its operating speed. For example, when a filter cartridge becomes clogged, increasing water flow resistance, the motor speed may decrease due to increased load. If the speed is not controlled, the current change will simultaneously reflect the effects of speed fluctuations and load changes, making it difficult to accurately attribute the problem to a filter cartridge malfunction. By real-time detection and acquisition of the motor's phase current data and controlling the drive module to operate according to a set operating mode, deviations in current characteristics caused by speed fluctuations can be avoided. Furthermore, the motor's current characteristics are related to its operating speed. By controlling the drive module to operate according to a set operating mode based on the current phase current data, deviations in current characteristics caused by speed fluctuations can be reduced to some extent, and errors in determining the specific location of the faulty filter cartridge can be minimized. For instance, when water flow resistance increases due to filter cartridge clogging, the motor speed may decrease due to increased load. In this case, the current change will simultaneously reflect the effects of speed fluctuations and load changes. If the speed is not controlled, the current change will simultaneously include the effects of speed fluctuations and load (such as filter cartridge clogging, water quality, etc.), making it impossible to distinguish between gradual filter cartridge clogging caused by a filter cartridge malfunction and temporary resistance fluctuations caused by water quality issues.

[0047] The operating status of a filter element includes current flow rate, pressure, whether it is clogged, usage time, and cumulative water consumption. Taking clogging caused by the filter element itself as an example, when this type of clogging occurs, there is a temporal correlation between changes in the filter element's resistance and changes in the motor's current characteristics. By obtaining the filter element's operating status to determine whether a malfunction has occurred, the correlation between the malfunction and current characteristics can be established (such as the mapping relationship between motor phase current data and filter element malfunction). This can be used to eliminate non-structural malfunction factors such as increased resistance due to normal aging of the filter element and instantaneous clogging caused by temporary large particulate impurities in the water.

[0048] For example, the motor is a three-phase motor with controllable operating speed, and the motor's operating speed varies in different operating modes. Based on the current phase current data of the motor, the drive module is controlled to operate according to the set operating mode, and the operating status of the filter element and the phase current data of the motor in each operating mode are obtained. This can further determine the specific location of the faulty filter element in the multi-stage filter system, thereby reducing errors, improving the accuracy of fault diagnosis, and achieving precise diagnosis of the specific location of the faulty filter element.

[0049] In other implementations, the specific location of the faulty filter element can be determined by combining the operating status of the filter element, and further, it can be determined whether to output fault reminders, filter element replacement alarms, and other prompts; the specific settings can be configured according to actual conditions without limitation.

[0050] like Figure 4As shown, in one embodiment, step S200, determining the specific location of the faulty filter element in the multi-stage filter element based on the current characteristics under the operating mode, specifically includes the following steps: Step S210: Combine the phase current data of the motor in the operating mode into a multi-mode feature vector.

[0051] Multi-mode feature vectors are used to form high-dimensional vectors by concatenating the current features (such as time-domain parameters, frequency-domain parameters, time-frequency domain features, etc.) of a motor under different operating modes.

[0052] Step S220: Classify the multi-mode feature vectors to determine the specific location of the faulty filter element in the multi-stage filter element based on the classification results.

[0053] For example, the operating mode can be set to one or more. When there are multiple operating modes, it is necessary to obtain the current characteristics of the motor under multiple set operating modes.

[0054] For example, if the pore size of the post-filter is smaller than that of the pre-filter, in low-speed mode, the motor drives the water flow at a lower speed. In the initial stage of clogging in the pre-filter, the water flow resistance increases slowly. To maintain a constant flow rate, the motor torque increases slightly, and the current shows a gradual, small increase. The time-domain characteristic shows a steady increase, and the frequency domain harmonic components show no significant abnormalities. In high-speed mode, the motor needs to drive a large flow of water. Clogging of the pre-filter causes a sudden increase in resistance. At this time, the motor will frequently adjust its output torque to maintain the speed, resulting in sharp fluctuations in the current. The time-domain characteristic shows irregular pulses, and the frequency domain may show a surge in high-order harmonic components. The resistance change of the post-filter is more pronounced in high-speed mode. Therefore, a slight increase in current in low-speed mode could be due to clogging of the pre-filter, the post-filter, or water quality issues. Further analysis using motor phase current data in high-speed operation mode is necessary. If sharp current fluctuations occur, water quality can be ruled out, and the problem is identified as a clogged pre-filter. If the current accumulates and rises with abnormal high-frequency waveforms, the problem is identified as a faulty post-filter.

[0055] By acquiring the current characteristics of the motor under various set operating modes and combining them into a multi-mode feature vector, and by classifying the multi-mode feature vector, interference factors under a single operating mode can be eliminated, filter element faults at different filter element locations can be distinguished, and fault location accuracy can be improved.

[0056] In one embodiment, the multi-stage filter element includes at least a first-stage filter element, a second-stage filter element, and a third-stage filter element connected in series.

[0057] A multi-stage filter cartridge may include a pre-filter for primary filtration, a mid-filter for deep water purification, and a post-filter for adsorbing residual odors. The number of pre-filters, mid-filters, and post-filters can be one or more. When the multi-stage filter cartridge includes a first-stage filter, a second-stage filter, and a third-stage filter, they are connected in series, with the first-stage filter being a pre-filter, the second-stage filter being a mid-filter, and the third-stage filter being a post-filter. When the multi-stage filter cartridge includes a fourth-stage filter or at least another first-stage filter in addition to the first-stage, second-stage, and third-stage filters, they are connected in series, with the first-stage filter being a pre-filter, the second-stage filter, the third-stage filter, etc., being mid-filters, and the last-stage filter serving as a post-filter. Specific configurations can be determined based on actual needs and are not limited here.

[0058] like Figure 5 , Figure 6 As shown, step S110, which involves obtaining the operating status of the filter element and the phase current data of the motor in each operating mode, specifically includes the following steps: Step S111: When the control drive module is running at the first operating speed, at least the operating status of the first-stage filter element and the phase current data of the motor in the first operating mode are obtained. Step S112: When the control drive module is running at the second operating speed, at least the operating status of the second-stage filter element and the phase current data of the motor in the second operating mode are obtained. Step S113: When the control drive module is running at the third operating speed, at least the operating status of the third-stage filter element and the phase current data of the motor in the third operating mode are obtained. Step S114: When the control drive module is running at the fourth operating speed, obtain the operating status of each filter element and the phase current data of the motor in the fourth operating mode. The first operating speed is less than the second operating speed, the third operating speed is less than the fourth operating speed, and the fourth operating speed is determined based on the current phase current data of the motor.

[0059] For example, the motor can be operated according to different running speeds by controlling the drive module, thus operating in corresponding running modes. When the motor operates in different running modes, the current characteristics of different levels of filter elements, such as those experiencing blockages, will differ. The order is: First running speed < Second running speed < Third running speed < Fourth running speed. When the drive module operates at the first running speed, it can acquire the operating status of the first-stage filter element, as well as the operating status of at least one other level of filter elements. When the drive module operates at the second running speed, it can acquire the operating status of the second-stage filter element, as well as the operating status of at least one other level of filter elements. When the drive module operates at the third running speed, it can acquire the operating status of the third-stage filter element, as well as the operating status of at least one other level of filter elements. Because the fourth running speed is determined based on the current phase current data of the motor and can be dynamically adjusted, when the drive module operates at the fourth running speed, it can acquire the operating status of the first-stage filter element, as well as the operating status of all levels of filter elements. By controlling the drive module to operate at different speeds, the motor can work in the corresponding operating mode. By collecting phase current data of the motor operating in different operating modes, differentiated testing can be achieved. This allows for the accurate location of the faulty filter element based on the obtained filter element operating status and phase current data.

[0060] The aforementioned steps S111 to S113 can be performed sequentially in order of increasing running speed, or sequentially in order of decreasing running speed, or one or more of them can be selected to be executed according to the actual situation.

[0061] Taking the control drive module operating at the first operating speed as an example, which acquires at least the operating status of the first-stage filter element and the phase current data of the motor in the first operating mode, the pre-filter element is responsible for intercepting large particles of impurities and is prone to clogging due to impurity accumulation. Therefore, when the control drive module operates at a lower speed, such as the first operating speed, the water flow is slow, and clogging of the pre-filter element will directly lead to a decrease in the inlet water flow. Thus, when the control drive module operates at the first operating speed, it can acquire the operating status of the first-stage filter element, as well as the operating status of at least one other filter element. When the motor operates at low speed, if the pre-filter element malfunctions, the motor current will change significantly. The clogging status of the pre-filter element can be identified through current fluctuations, preventing excessive load on the middle and post-filter elements due to pre-filter clogging.

[0062] Taking the control drive module operating at the third operating speed as an example, which involves acquiring the operating status of at least the third-stage filter and the phase current data of the motor in the third operating mode, since the later-stage filter elements are mainly used for deep filtration, they can become clogged due to the accumulation of fine impurities. Therefore, when the control drive module operates at a higher speed, such as the third operating speed, the water flow rate is faster. If the later-stage filter elements become clogged, it will lead to unstable water flow. Therefore, when the control drive module operates at the third operating speed, the operating status of the third-stage filter can be acquired, as well as the operating status of at least one other filter element. When the motor operates at a higher speed, if a later-stage filter element malfunctions, it will cause current fluctuations or abnormal increases. Therefore, identifying current characteristics such as high-frequency fluctuations can reflect the clogging status of the later-stage filter elements.

[0063] Taking the operation of the control drive module at the fourth operating speed as an example, the operating status of each filter element and the phase current data of the motor in the fourth operating mode are obtained. Since the fourth operating speed is determined based on the current phase current data of the motor and is mainly used to change the water flow speed, it is necessary to detect the status of each filter element in order to further achieve precise location of the specific filter element where the fault occurs.

[0064] In addition, the aforementioned fourth operating speed is determined based on the current phase current data of the motor. In addition to being set to be greater than the third operating speed, it can also be controlled by switching the operating speed from low speed to high speed, and flexibly adjusting the operating state of the filter element at different filter element positions when switching the operating speed of the driving module; or, the driving module can be controlled by switching the operating set time according to different operating speeds, and flexibly adjusting the operating state of the filter element at different filter element positions when switching the operating speed of the driving module.

[0065] The operating status of the filter element is obtained in the following ways: In one embodiment, the correspondence between motor phase current data and filter element operating status is obtained, and the current phase current data of the motor is obtained. The operating status of the filter element is obtained based on the current phase current data of the motor and the correspondence between the motor phase current data and the filter element operating status.

[0066] The operating status of a filter element includes normal operation, clogging, and damage. When the motor experiences clogging, the resulting change in water flow resistance affects the motor load, which is reflected in the motor's phase current data. Based on the correlation between motor phase current data and filter element operating status, the filter element's operating status can be directly determined by acquiring the current phase current data. Because current signals can be acquired and processed in real time, acquiring the motor's current characteristics under a set operating mode allows for timely and reliable determination of the filter element's current operating status. Furthermore, by reducing the need for additional detection devices, this not only lowers costs, reduces equipment complexity and failure rates, but also improves system reliability.

[0067] In another embodiment, detection data from a sensor module is acquired, and the operating status of the filter element is determined based on the detection data. The sensor module includes at least one or a combination of multiple of a water quality sensor, a flow sensor, a temperature sensor, and a sound sensor.

[0068] Water quality sensors can combine indicators such as changes in influent and effluent TDS values ​​to help determine the operating status of filter elements such as RO membranes; flow sensors, combined with flow rate changes, can help determine the clogging status of filter elements; water temperature affects motor performance, and temperature sensors can improve the accuracy of determining the correlation between motor phase current data and filter element operating status; sound sensors can analyze the operating sound characteristics of pumps such as variable frequency pumps in the drive module to help identify fault types.

[0069] In other embodiments, the operating status of the filter element can be obtained based on the current phase current data of the motor, the correspondence between the motor phase current data and the operating status of the filter element, and the operating status of the filter element can be further determined by combining the detection data of the sensor module, so as to further improve the accuracy of the determination of the filter element status.

[0070] In other embodiments, in addition to identifying faulty filter cartridges, the fault type and severity can be further subdivided based on the detection results. For example, fault types can be classified into different fault types such as physical blockage of the filter cartridge, microbial contamination, and material aging; the filter cartridge status can be divided into four levels—normal, slightly abnormal, moderately faulty, and severely faulty—based on the severity of the blockage fault; combined fault diagnosis of filter cartridges can be performed to identify complex fault situations where multiple filter cartridges have problems simultaneously; non-filter cartridge fault identification can also be performed to distinguish filter cartridge faults from other system problems such as pump faults and pipeline blockages; the specific settings can be configured according to actual conditions and are not limited here.

[0071] like Figure 7As shown, in one embodiment, before performing step S100, controlling the drive module to operate according to the set operating mode, and obtaining the current characteristics of the motor in the set operating mode, the filter element fault diagnosis and classification method further includes the following steps: Step S310: In response to the running command, obtain the current phase current data of the motor and process the current phase current data of the motor.

[0072] An operation command is an instruction that triggers the start of operation of drive modules, etc., and is used to automatically trigger the operation of devices with multi-stage filter systems, such as water purifiers, thereby driving water and other liquids through the filter elements. Operation commands can be hardware signals automatically triggered by the circuit after detecting the start of multi-stage filter operation; they can also be software commands triggered by a timed software mechanism; or they can be interactive signals triggered by the user through local devices such as touchscreens and buttons during human-machine interaction, or manually selected and triggered by remote devices such as user interfaces or terminal devices (APPs). Real-time detection and acquisition of the motor's current phase current data are performed, and this data is then processed.

[0073] Step S320: Obtain the mapping relationship between motor phase current data and filter element faults.

[0074] For example, the mapping relationship between motor phase current data and filter failure can be determined based on motor phase current data at different levels of filter failure determined by experiments or from historical databases.

[0075] Step S330: Based on the processed current phase current data of the motor and the mapping relationship between the motor phase current data and the filter element fault, when a multi-stage filter element fails, start the difference test.

[0076] Differential testing refers to the process of controlling the driving module's operating speed after identifying a fault in a multi-stage filter element. This allows the motor to operate according to a set mode, enabling targeted testing of different filter elements to accurately locate the faulty element. When a preliminary judgment is made based on real-time acquired motor current data indicating a fault in a multi-stage filter element (such as abnormal current waveform), the differential testing is automatically initiated.

[0077] By dynamically detecting and acquiring motor phase current data, the system determines whether a multi-stage filter element is currently faulty. This faulty filter element is then used to initiate differentiated testing. The differentiated testing further pinpoints the location of the faulty filter element within the multi-stage filter system based on the current characteristics under the operating mode. This effectively eliminates non-filter element faults, improves the accuracy of fault diagnosis, and significantly enhances the reliability of the diagnostic results regarding the specific location of the faulty filter element within the multi-stage filter system.

[0078] like Figure 8As shown, in one embodiment, the processing of the current phase current data of the motor in step S310 specifically includes the following steps: Step S311: Preprocess the current phase current data of the motor.

[0079] Preprocessing is used to preprocess the current phase current data of the motor, and noise is removed by filtering algorithm to obtain the preprocessed current phase current data of the motor.

[0080] Step S312: Calculate the time-domain characteristics based on the preprocessed current phase current data of the motor.

[0081] Time-domain features can be extracted from the time series of the current phase current data of the motor to reflect the changing characteristics of the current phase current data over time. Time-domain features include current peak (lpeak), current mean (lmean), current mean (Irms), current waveform factor (CF), and current rise rate (Irise). Specifically, current peak represents the maximum value of the current waveform; current mean represents the average value of the current waveform; current mean represents the root mean square value of the current waveform; current waveform factor represents the ratio of the peak value to the effective value; and current rise rate represents the rate at which the current increases from zero to the peak value.

[0082] Step S313: Perform Fourier transform on the preprocessed current phase current data of the motor to obtain the spectrum and obtain the frequency domain characteristics based on the spectrum calculation.

[0083] Frequency domain characteristics can be derived from the time domain to the frequency domain using Fourier transform to analyze the energy distribution of different frequency components. Frequency domain characteristics include the fundamental amplitude (A1), harmonic component amplitudes (A2, ..., An), total harmonic distortion (THD), harmonic characteristic ratio (HCR), and spectral centroid (FC). Specifically, the fundamental amplitude represents the amplitude of components that are not fundamental to the current; the harmonic component amplitudes represent the amplitudes of each harmonic component; the total harmonic distortion represents the ratio of harmonic components to the fundamental component; the harmonic characteristic ratio represents the ratio of a specific harmonic to the fundamental frequency; and the spectral centroid represents the weighted center frequency of the spectral energy distribution.

[0084] Steps S312 and S313 can be executed sequentially or simultaneously, and there is no limitation on this.

[0085] Step S314: Perform wavelet transform on the preprocessed current phase current data of the motor to obtain the time-frequency distribution, and obtain the time-frequency domain features calculated based on the time-frequency distribution.

[0086] Time-frequency domain features can be used to analyze the distribution characteristics of signals in the time-frequency two-dimensional plane. Time-frequency domain features include wavelet energy distribution (WE), wavelet entropy (WEnt), and empirical mode decomposition (EMD) features. Among them, wavelet energy distribution is used to represent the energy distribution of current waveforms in different frequency bands; wavelet entropy is used to represent an index describing the complexity of current waveforms; and empirical mode decomposition features are used to represent the intrinsic mode function characteristics of current signals.

[0087] Step S315: Combine time-domain features, frequency-domain features, and time-frequency-domain features to form a feature vector to be determined. Perform data processing on the feature vector to be determined to obtain the target feature vector.

[0088] For example, the eigenvector to be determined can be normalized. This can be done by converting the eigenvalues ​​to a standard normal distribution with a mean of 0 and a standard deviation of 1, or by using min-max normalization to scale the eigenvector to a uniform interval. Normalizing the eigenvector scales different features (such as mean current, harmonic amplitude, time-frequency entropy, etc.) to a uniform interval (e.g., [0, 1] or [-1, 1]), giving different types of features (such as time-domain, frequency-domain, and time-frequency-domain features) the same weight, and preventing features with excessively large values ​​from directly affecting the analysis results.

[0089] After performing data processing such as normalization on the given eigenvectors, Principal Component Analysis (PCA) can be used for dimensionality reduction to obtain the target eigenvectors. This reduces data storage space and computational complexity, improving the efficiency of subsequent analysis and processing.

[0090] The aforementioned step S330, based on the processed current phase current data of the motor and the mapping relationship between the motor phase current data and filter element faults, determines when a multi-stage filter element fault occurs, and initiates a difference test, specifically including the following steps: Based on the mapping relationship between the target feature vector, motor phase current data and filter element failure, when a multi-stage filter element fails, a differential test is initiated.

[0091] This significantly improves the accuracy of filter element fault detection and enhances the precision of detecting whether multi-stage filter elements are faulty. Differential testing is only initiated when multi-stage filter elements are faulty, reducing the consumption of meaningless testing resources and improving the accuracy of fault diagnosis results.

[0092] like Figure 9 , Figure 10 As shown, in one embodiment, step S330, determining when a multi-stage filter element fails based on the processed current phase current data of the motor and the mapping relationship between the motor phase current data and the filter element fault, initiates a difference test, specifically including the following steps: Step S331: Obtain normal operating condition samples and various filter element failure operating condition samples, label various failure types, determine the mapping relationship between motor phase current data and filter element failures, and construct a third classifier containing a classification model.

[0093] To obtain samples of normal operating conditions, one can control the multi-stage filter system to operate under different operating modes, or control the multi-stage filter system to operate under normal conditions such as rated voltage, flow rate, and water temperature, and obtain the phase current data of the motor during stable operation. To obtain samples of various filter failure conditions, one can obtain them through experiments simulating filter clogging and filter damage; alternatively, one can obtain them from phase current data recorded during long-term operation of water purifiers and other equipment when failures occur, or from historical databases containing historical usage records.

[0094] For example, such as Figure 11 As shown, under normal operating conditions, the fundamental current is distinct, harmonic components are few, and the waveform is stable. The pre-filter includes a PP cotton filter, a pre-activated carbon filter, a metal mesh, etc. Figure 12 As shown, taking a PP cotton filter as an example, the dashed line represents the current waveform under normal operating conditions; the solid line represents the current waveform when a blockage occurs. When a blockage occurs, the current characteristics are as follows: the current waveform shows increased high-frequency fluctuations, and the fundamental amplitude slightly increases. Intermediate filter elements include activated carbon filters, RO reverse osmosis membranes, nanofiltration membranes, etc. Figure 13 As shown, taking an activated carbon filter as an example, the dashed line represents the current waveform under normal operating conditions; the solid line represents the current waveform when a blockage occurs. When blockage occurs, the current characteristics are as follows: the third harmonic in the current harmonic components increases significantly, and waveform distortion increases; for example... Figure 14 As shown, taking the RO reverse osmosis membrane as an example, the dashed line represents the current waveform under normal operating conditions; the solid line represents the current waveform when a blockage occurs. When a blockage occurs, the current characteristics are: a significant increase in the average current, a decrease in the waveform factor, and an enhancement of low-frequency components. Post-filters include post-activated carbon filters, mineralization filters, and weakly alkaline filters. Taking the mineralization filter as an example, when a blockage occurs, the current characteristics are: a significant change in the phase of the current waveform, and a change in the proportion of harmonic characteristics.

[0095] Step S332: When the third classifier determines that a multi-stage filter element has failed based on the processed current phase current data of the motor, the mapping relationship between the motor phase current data and the filter element failure, a differential test is initiated; wherein, the third classifier adopts an SVM classifier.

[0096] A training set is constructed by acquiring samples of normal operating conditions and various filter element failure conditions. An SVM classifier M1 is then trained. The SVM classifier M1 learns the differences between the feature vectors of normal and faulty filters, enabling it to classify new feature vectors obtained from the current phase current data of the motor. This classification is used to determine the presence of filter element failures and quickly filter out potentially faulty situations. When the SVM classifier determines that the filter element at the specific location corresponding to the feature vector is faulty, it indicates that the features of that filter element are significantly different from those of normal filters. However, at this point, it is not possible to determine which specific filter element has failed. A differential testing procedure is then implemented to further obtain more information about the fault, allowing for a more accurate location of the faulty filter element.

[0097] Because the mapping relationship between motor phase current data and filter element faults can be non-linear and complex, a third classifier can be trained to learn complex mapping rules for real-time analysis of current characteristics and rapid output of the fault type corresponding to those current characteristics, improving detection efficiency and accuracy. Current anomalies may be caused by non-filter element factors such as motor faults, voltage fluctuations, and water quality issues. First, the mapping relationship is used to initially screen for multi-stage filter element faults, then differential testing is initiated. This ensures that differential testing is only activated when multi-stage filter element faults occur, reducing unnecessary consumption of testing resources and improving the accuracy of fault diagnosis. Differentiated testing can also distinguish between single-stage and multi-stage faults, avoiding the need for complete filter element replacement and reducing maintenance costs.

[0098] like Figure 15 As shown, in one embodiment, step S220, classifying the multi-mode feature vectors to determine the specific location of the faulty filter element in the multi-stage filter element based on the classification results, specifically includes the following steps: Step S221: Use the first classifier to classify the multi-mode feature vectors to analyze and determine the locations of candidate filter elements that may fail in the multi-stage filter elements, and obtain the first classification result; Step S222: Calculate and determine the confidence level of the candidate filter location in the first classification result.

[0099] The confidence level corresponds to the set of filter element locations where failures may occur. By determining the confidence level of candidate filter element locations, their corresponding failure probabilities can be calculated. Confidence level screening can narrow down the fault location range, reduce misjudgments, and further improve the accuracy of locating specific filter element locations where failures occur in multi-stage filter elements.

[0100] Step S223: When the confidence level is not less than the preset threshold, the candidate filter element location is determined as the specific filter element location where the fault occurred; Step S224: When the confidence level is less than a preset threshold, the first classification result is analyzed by the second classifier and the second classification result is obtained. The first classification result and the second classification result are combined to calculate the probability of failure at each candidate filter element location. The specific filter element location where the failure occurs is determined from the probability of failure at each candidate filter element location.

[0101] Understandably, the current characteristics recorded and acquired under various operating modes are combined into a multi-mode feature vector. This feature vector contains comprehensive information about the filter element under different operating conditions, providing a more complete reflection of the filter element's working status. Compared to features under a single operating mode, the multi-mode feature vector contains richer information, which helps improve the accuracy of fault diagnosis.

[0102] like Figure 16 , Figure 17 As shown, in one embodiment, the first classifier is a random forest classifier, and the second classifier is a neural network classifier.

[0103] Random forest is an ensemble learning algorithm consisting of multiple decision trees. Each decision tree classifies the input feature vector, and the results are used to determine the final first classification. The first classifier, a random forest classifier, is used to exclude correctly functioning filters from the multi-level filter pool, narrowing the selection to the locations of potentially faulty filters, thus improving the accuracy of locating faulty filter positions. A training set is constructed using known faulty filter samples and their corresponding multi-modal feature vectors to train the random forest classifier M2. The random forest classifier M2 learns the mapping relationship between different faulty filter types and multi-modal feature vectors, enabling it to classify new feature vectors and determine the type of faulty filter. Based on the multi-modal feature vectors, a preliminary type determination of the faulty filter is made, providing a foundation for further analysis.

[0104] The confidence score P reflects the reliability of the classification result. In the random forest classifier, the confidence score of the candidate filter positions in the first classification result is calculated and determined. When the confidence score is lower than a preset threshold, it indicates that the candidate filter positions in the first classification result determined by the random forest classifier M2 are not reliable enough and there may be misclassification. At this time, the neural network classifier M3 is started for further analysis. Neural networks have powerful nonlinear fitting and adaptive learning capabilities, and can handle complex nonlinear relationships. By learning from a large amount of sample data, they can discover potential patterns and features in the data, thereby improving the accuracy of fault diagnosis.

[0105] By comprehensively considering the classification results of the random forest classifier M2 and the neural network classifier M3, a weighted average method can be used to determine the specific location of the faulty filter among the various candidate filter locations. This approach integrates the advantages of multiple classifiers, outputting a more accurate and reliable faulty filter type and failure probability, providing a basis for subsequent repair and maintenance decisions.

[0106] like Figure 18 As shown, in one embodiment, the filter element fault diagnosis and classification method further includes the following steps: Step S410: Establish communication between the multi-stage filter system and the cloud platform; Step S420: In response to the algorithm optimization instruction, the mapping relationship between motor phase current data and filter element faults from different user terminals is obtained and summarized through the cloud platform, and the classification model is optimized and updated based on the obtained mapping relationship between motor phase current data and filter element faults from different user terminals.

[0107] The cloud platform is a service platform based on cloud computing technology, providing interfaces for data storage, computation, analysis, and application services. It enables remote management and interaction of device data, stores the mapping relationship between motor phase current data from various user terminals and filter element faults, and provides computational resources for algorithm optimization and model updates. By combining the mapping relationship between motor phase current data from different user terminals obtained through communication with the cloud platform and filter element faults, the classification model can be optimized. This continuously improves the accuracy of filter element status judgment, accurately determining whether multi-stage filter elements have failed and pinpointing the specific location of the faulty filter element within the multi-stage filter system.

[0108] like Figure 19 , Figure 20 As shown, in one embodiment, after performing step S200, which determines the specific location of the faulty filter element in the multi-stage filter element based on the current characteristics in the operating mode, the following steps are also included: Step S510: Calculate the failure probability of the filter element at the specific location of the failure and the remaining life of the filter element at the specific location of the failure, and generate a diagnostic report based on the calculation results. Step S520: The control display interface displays the diagnostic report and maintenance information for the faulty filter element.

[0109] A diagnostic report is generated based on the probability of filter failure at the specific location of the malfunction and the remaining lifespan of the filter at that location. This report, along with maintenance information for the malfunctioning filter, is then displayed on the user interface. This allows users to quickly and intuitively understand the filter's status, whether it has malfunctioned, the remaining lifespan of the malfunctioning filter, and filter maintenance information.

[0110] In addition to being displayed through the user interface, users can also view diagnostic reports and maintenance information through mobile apps, laptops, tablets, or other mobile devices, or fixed devices such as desktop computers.

[0111] In one embodiment, the filter element fault diagnosis and classification method further includes the following steps: In response to system initialization commands, obtain classification models and configuration information; Initialization commands are signals that trigger the reset or clearing of system parameters. These can be hardware signals automatically triggered by the circuit after a filter replacement is detected; software commands triggered at regular intervals; or interactive signals triggered by the user via local devices such as touchscreens or buttons during human-machine interaction, or manually selected and triggered by remote devices such as user interfaces or terminal devices (APPs). The classification model can be built and updated based on the mapping relationship between motor phase current data and filter failures; configuration information includes configuration parameters, operating status, and replacement time for multiple filter stages.

[0112] like Figure 21 As shown, after determining the specific location of the faulty filter element in the multi-stage filter element in step S200, the following steps are also included: Step S610: Output a replacement reminder for the filter element at the specific location of the malfunction; Step S620: In response to the user's filter replacement confirmation command, update the configuration information.

[0113] This can remind users to replace faulty filters in a timely manner and update configuration information promptly after filter replacement, reducing misdiagnosis of faults due to inadequate configuration information updates.

[0114] like Figure 22 As shown, some embodiments of this application also provide a water purifier, which includes a multi-stage filter system 100, the multi-stage filter system 100 including a multi-stage filter 100, a drive module, a current sampling circuit 300 and a controller.

[0115] The multi-stage filter element 100 is used for filtering liquids, and the multi-stage filter element 100 includes at least a first-stage filter element, a second-stage filter element, and a third-stage filter element connected in series.

[0116] The multi-stage filter element 100 may include a pre-filter for primary filtration, a mid-filter for deep water purification, and a post-filter for adsorbing residual odors. The number of pre-filters, mid-filters, and post-filters may be one or more. When the multi-stage filter element 100 includes a first-stage filter element, a second-stage filter element, and a third-stage filter element, the first-stage filter element may be a pre-filter element, the second-stage filter element may be a mid-filter element, and the third-stage filter element may be a post-filter element, following a series connection. When the multi-stage filter element 100 includes a fourth-stage filter element or at least another first-stage filter element in addition to the first-stage, second-stage, and third-stage filter elements, the first-stage filter element may be a pre-filter element, the second-stage filter element, the third-stage filter element, etc., may be mid-filters, with the last-stage filter element serving as the post-filter element. Specific configurations can be determined based on actual conditions and are not limited here.

[0117] like Figure 23 As shown, the pre-filters include a PP cotton filter 110, a pre-activated carbon filter 120, and a metal mesh. Taking the PP cotton filter 110 as an example, when clogging occurs, its current characteristics are as follows: the current waveform shows increased high-frequency fluctuations, and the fundamental amplitude slightly increases. The mid-filters include an activated carbon filter 120, an RO reverse osmosis membrane 130, and a nanofiltration membrane. Taking the activated carbon filter 120 as an example, when clogging occurs, its current characteristics are as follows: the third harmonic in the current harmonic components increases significantly, and waveform distortion increases. Taking the RO reverse osmosis membrane 130 as an example, when clogging occurs, its current characteristics are as follows: the average current increases significantly, the waveform factor decreases, and the low-frequency components are enhanced. The post-filters include a post-activated carbon filter 120, a mineralization filter 140, and a weakly alkaline filter. Taking the mineralization filter 140 as an example, when clogging occurs, its current characteristics are as follows: the current waveform phase changes significantly, and the harmonic characteristic ratio changes.

[0118] The drive module provides power to the liquid delivered to the filter element, enabling the filter element to adsorb or filter impurities from the liquid. The drive module includes a variable frequency pump 210, a motor 220, and a motor drive circuit 230. The variable frequency pump 210 is connected to the first-stage filter element and provides power to the liquid delivered to the multi-stage filter element 100. The motor 220 is driven and connected to the variable frequency pump 210, providing power to the variable frequency pump 210. The motor drive circuit 230 is driven and connected to the motor 220, driving the motor 220 to operate. Besides the variable frequency pump 210, any pump body, such as a centrifugal pump or gear pump, can be used; no limitation is made here. The motor 220 is driven and connected to the variable frequency diaphragm pump, providing power to the variable frequency diaphragm pump. The motor drive circuit 230 is driven and connected to the motor 220, driving the motor 220 to operate. The motor drive circuit 230 provides drive current to the variable frequency motor 220, controls the speed and torque of the motor 220, and can also be used to detect the status of the motor 220. The controller drives the motor 220 through the motor drive circuit 230 and controls the running speed of the motor 220 to make the drive module operate according to the set operating mode.

[0119] The current sampling circuit 300 is used to collect the phase current data of the motor 220. The current sampling circuit 300 uses a dual-resistor sampling method to detect the three-phase current of the motor 220 in real time, obtaining current signals such as the phase current data of the motor 220. The current sampling circuit 300 is connected to the controller and is used to feed back the real-time detected phase current data and other current signals to the controller. The controller includes a signal processing module 410, a feature extraction module 420, a fault classification module 430, and a control module 440. The signal processing module 410 processes the received phase current data of the motor 220 and classifies it; the feature extraction module 420 extracts current features; and the fault classification module 430 determines whether a fault has occurred and the specific location of the faulty filter element in the multi-stage filter element 100. The control module 440 controls the drive module to automatically provide power to the liquid being delivered to the filter element.

[0120] like Figure 24As shown, for example, the current sampling module includes a motor phase line, a sampling resistor, a differential amplifier, and an analog-to-digital converter module 240. The motor phase line is used to transmit the electrical energy required for driving. The sampling resistor is connected to the motor phase line and can use the FOC (Field-Oriented Control) dual-resistor sampling algorithm to detect the phase current and other current data of the motor 220 in real time. The differential amplifier is connected to the sampling resistor and is used to amplify and process the current data. The analog-to-digital converter module 240 is connected to the differential amplifier and the signal processing unit, respectively, and is used to convert the analog voltage signal output by the differential amplifier into a digital signal. The signal processing unit performs digital filtering (such as sliding window filtering) and fast Fourier transform (FFT) on the received digital signal to extract the effective value of the phase current, the peak current, the rate of change (ΔI / Δt), and abnormal fluctuations of the motor 220, providing data support for the motor 220 control algorithm and filter fault diagnosis. The current sampling module is used to detect the phase current data and other operating parameters of the motor 220 in real time. The controller controls the drive module to work, so that the drive module can automatically provide power for the liquid delivered to the filter element.

[0121] The controller stores and executes the filter element fault diagnosis and classification program. When executing the program, it implements the steps of the filter element fault diagnosis and classification method as described in the above embodiment.

[0122] Compared to solutions that use multiple pressure sensors, flow meters, and other detection devices for each multi-stage filter element 100, which suffer from high cost and complexity, increased potential for failure due to improper installation, and inability to accurately pinpoint the location of the faulty filter element, the embodiments of this application do not require additional pressure sensors or flow meters. Instead, they utilize the water purifier's built-in drive module and current sampling circuit 300 to determine the current characteristics of the motor 220 in the set mode by real-time acquisition of the motor 220 current. Furthermore, based on the current characteristics in the operating mode, the specific location of the faulty filter element in the multi-stage filter element 100 is determined. This effectively solves the technical problem of not being able to determine the location of the faulty filter element in the multi-stage filter element 100 in a timely and reliable manner. By reducing the use of additional detection devices, not only are costs reduced, equipment complexity and failure rate decreased, but system reliability is also improved. Because current signals can be acquired and processed in real time, the specific location of the faulty filter element in the multi-stage filter element 100 can be directly determined by acquiring the current characteristics of the motor 220 in a set operating mode. This effectively reduces errors and improves the accuracy of fault diagnosis. In some scenarios, the fault diagnosis accuracy can even reach over 90%. This allows for precise diagnosis of the specific location of the faulty filter element and further addresses the issues of premature or delayed filter element replacement.

[0123] In one embodiment, the multi-stage filter 100 system further includes a user interface 500 for displaying diagnostic reports and filter maintenance information. The diagnostic report includes at least the probability of filter failure at the specific location of the failure and the remaining lifespan of the filter at the specific location of the failure.

[0124] This allows users to easily and intuitively understand the status of the filter element, whether it has malfunctioned, the remaining lifespan of a malfunctioning filter element, and filter element maintenance information.

[0125] The water purifier provided in this application employs the filter cartridge fault diagnosis and classification method described in the above embodiments, which solves the technical problem of how to timely and reliably determine the specific location of a faulty filter cartridge in a multi-stage filter system without additional detection devices. Compared with the prior art, the beneficial effects of the water purifier provided in this application are the same as those of the filter cartridge fault diagnosis and classification method provided in the above embodiments, and other technical features of the water purifier are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0126] On the other hand, this application provides a filter cartridge fault diagnosis and classification device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a filter cartridge fault diagnosis and classification program executable by the at least one processor, and the filter cartridge fault diagnosis and classification program is configured to implement the steps of the filter cartridge fault diagnosis and classification method as described in the above embodiment.

[0127] The filter cartridge fault diagnosis and classification device in this application embodiment 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), and in-vehicle terminals (such as in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Specific configurations can be made according to actual conditions and should not impose any limitations on the functionality and scope of use of this application embodiment.

[0128] A filter cartridge fault diagnosis and classification device may include a processing unit that can perform various appropriate actions and processes based on a program stored in a read-only memory or a program loaded from a storage device into a random access memory. The random access memory also stores various programs and data required for the operation of the filter cartridge fault diagnosis and classification device. The processing unit, read-only memory, and random access memory are interconnected via a bus. Input / output interfaces are also connected to the bus. Typically, the following systems can be connected to the input / output interface: input devices including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices including, for example, magnetic tapes, hard disks, etc.; and communication devices. The communication device allows the filter cartridge fault diagnosis and classification device to communicate wirelessly or wiredly with other devices to exchange data. Although the figure shows a filter cartridge fault diagnosis and classification device 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.

[0129] 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, the embodiments disclosed in this application include a filter cartridge fault diagnosis classification program product, which includes a filter cartridge fault diagnosis classification program carried on a computer-readable medium, the filter cartridge fault diagnosis classification program containing program code for performing the methods shown in the flowcharts. In such embodiments, the filter cartridge fault diagnosis classification program can be downloaded and installed from a network via a communication device, or installed from a storage device, or installed from a read-only memory. When the filter cartridge fault diagnosis classification program is executed by a processing device, it performs the functions defined above in the methods of the embodiments disclosed in this application.

[0130] The filter cartridge fault diagnosis and classification device provided in this application, employing the filter cartridge fault diagnosis and classification method described in the above embodiments, can solve the technical problem of how to timely and reliably determine the specific location of a faulty filter cartridge in a multi-stage filter system without additional detection devices. Compared with the prior art, the beneficial effects of the filter cartridge fault diagnosis and classification device provided in this application are the same as those of the filter cartridge fault diagnosis and classification method provided in the above embodiments, and other technical features of this filter cartridge fault diagnosis and classification device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0131] 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.

[0132] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0133] 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 classifying and diagnosing filter element faults, characterized in that, The method for classifying filter cartridge faults includes the following steps: A multi-stage filter cartridge system is applied, comprising multiple filter cartridges and a drive module. The drive module provides power to the liquid being delivered to the multi-stage filter cartridges and includes a motor. The control drive module operates according to the set operating mode and acquires the current characteristics of the motor in the set operating mode; Based on the current characteristics under the operating mode, determine the specific location of the faulty filter element in the multi-stage filter element; The step of determining the specific location of the faulty filter element in the multi-stage filter element based on the current characteristics under the operating mode specifically includes the following steps: Combine the phase current data of the motor under the operating mode into a multi-mode feature vector; A classifier is used to classify the multi-mode feature vectors in order to determine the specific location of the faulty filter element in the multi-stage filter element based on the classification results. The classifier includes a first classifier and a second classifier. The first classifier is used to exclude filter cartridges that are in good working order from the multi-stage filter cartridges and narrow down the screening range to the locations of filter cartridges that may have failed.

2. The filter element fault diagnosis and classification method as described in claim 1, characterized in that, The classification of multi-mode feature vectors to determine the specific location of the faulty filter element in the multi-stage filter element system based on the classification results includes the following steps: The first classifier is used to classify the multi-mode feature vectors to analyze and determine the locations of candidate filter elements that may fail in the multi-stage filter elements, and the first classification result is obtained. Calculate and determine the confidence level of the candidate filter locations in the first classification results; When the confidence level is not less than a preset threshold, the candidate filter element location is determined as the specific filter element location where the failure has occurred. When the confidence level is less than a preset threshold, the first classification result is analyzed by the second classifier to obtain the second classification result. The first classification result and the second classification result are combined to calculate the probability of failure at each candidate filter element location. The specific filter element location where the failure occurs is determined from the probability of failure at each candidate filter element location.

3. The filter element fault diagnosis and classification method as described in claim 2, characterized in that, The first classifier uses a random forest classifier, and the second classifier uses a neural network classifier.

4. The filter element fault diagnosis and classification method as described in claim 1, characterized in that, The control drive module operates according to a set operating mode and acquires the current characteristics of the motor in the set operating mode, specifically including the following steps: The system acquires the current phase current data of the motor, controls the drive module to run at the set operating speed according to the current phase current data of the motor, acquires the operating status of the filter element, and acquires the phase current data of the motor in each operating mode.

5. The filter element fault diagnosis and classification method as described in claim 4, characterized in that, The multi-stage filter element includes at least a first-stage filter element, a second-stage filter element, and a third-stage filter element connected in series; the acquisition of the operating status of the filter element and the acquisition of the phase current data of the motor in each operating mode specifically includes the following steps: When the control drive module is running at the first operating speed, it shall at least acquire the operating status of the first-stage filter element and the phase current data of the motor in the first operating mode. When the control drive module is running at the second operating speed, at least the operating status of the second-stage filter element and the phase current data of the motor in the second operating mode are obtained. When the control drive module is running at the third operating speed, at least the operating status of the third-stage filter element and the phase current data of the motor in the third operating mode should be obtained. When the control drive module is running at the fourth operating speed, it acquires the operating status of each filter element and the phase current data of the motor in the fourth operating mode. The first operating speed is less than the second operating speed, the third operating speed is less than the fourth operating speed, and the fourth operating speed is determined based on the current phase current data of the motor.

6. The filter element fault diagnosis and classification method as described in claim 4, characterized in that, The operating status of the filter element is obtained in the following ways: Obtain the correspondence between motor phase current data and filter element operating status, and obtain the current phase current data of the motor. Based on the current phase current data of the motor and the correspondence between the motor phase current data and the filter element operating status, obtain the operating status of the filter element. And / or, acquire detection data from the sensor module, and determine the operating status of the filter element based on the detection data, wherein the sensor module includes at least one or a combination of multiple of a water quality sensor, a flow sensor, a temperature sensor, and a sound sensor.

7. The filter element fault diagnosis and classification method as described in claim 1, characterized in that, Before executing the steps of the control drive module operating according to the set operating mode and obtaining the current characteristics of the motor in the set operating mode, the filter element fault diagnosis and classification method further includes the following steps: In response to the running command, the current phase current data of the motor is acquired and processed. Obtain the mapping relationship between motor phase current data and filter element faults; Based on the processed current phase current data of the motor and the mapping relationship between the motor phase current data and the filter element fault, when a multi-stage filter element fails, a difference test is initiated.

8. The filter element fault diagnosis and classification method as described in claim 7, characterized in that, The processing of the current phase current data of the motor specifically includes the following steps: The current phase current data of the motor is preprocessed; Calculate the time-domain characteristics based on the pre-processed current phase current data of the motor; Perform a Fourier transform on the preprocessed current phase current data of the motor to obtain the spectrum, and obtain the frequency domain characteristics calculated based on the spectrum; Wavelet transform is performed on the preprocessed current phase current data of the motor to obtain the time-frequency distribution, and time-frequency domain features calculated based on the time-frequency distribution are obtained. The time-domain features, frequency-domain features, and time-frequency-domain features are combined to form a feature vector to be determined. The feature vector to be determined is then processed to obtain the target feature vector. When a multi-stage filter element malfunction is determined based on the processed current phase current data of the motor and the mapping relationship between the motor phase current data and the filter element malfunction, a differential test is initiated, which specifically includes the following steps: Based on the mapping relationship between the target feature vector, the motor phase current data, and the filter element fault, when a multi-stage filter element fails, a difference test is initiated.

9. The filter element fault diagnosis and classification method as described in claim 7, characterized in that, When a multi-stage filter element malfunction is determined based on the processed current phase current data of the motor and the mapping relationship between the motor phase current data and the filter element malfunction, a differential test is initiated, which specifically includes the following steps: Acquire normal operating condition samples and various filter element failure operating condition samples, label various failure types, determine the mapping relationship between motor phase current data and filter element failure, and construct a third classifier containing a classification model; When the third classifier determines that a multi-stage filter element has failed based on the processed current phase current data of the motor, the mapping relationship between the motor phase current data and the filter element failure, a differential test is initiated. The third classifier is an SVM classifier.

10. The filter element fault diagnosis and classification method as described in claim 9, characterized in that, The filter element fault diagnosis and classification method also includes the following steps: Establish communication between the multi-stage filter system and the cloud platform; In response to the algorithm optimization command, the mapping relationship between motor phase current data and filter element faults from different user terminals is obtained and aggregated through the cloud platform, and the classification model is optimized and updated based on the obtained mapping relationship between motor phase current data and filter element faults from different user terminals.

11. The filter element fault diagnosis and classification method according to any one of claims 1 to 10, characterized in that, After performing the step of determining the specific location of the faulty filter element in the multi-stage filter element based on the current characteristics under the operating mode, the following steps are also included: Calculate the probability of filter failure at the specific location of the failure and the remaining life of the filter at the specific location of the failure, and generate a diagnostic report based on the calculation results; The control display interface shows the diagnostic report and maintenance information for the faulty filter element.

12. The filter element fault diagnosis and classification method according to any one of claims 1 to 10, characterized in that, The filter element fault diagnosis and classification method also includes the following steps: In response to system initialization commands, obtain classification models and configuration information; After determining the specific location of the faulty filter element in the multi-stage filter system, the following steps are also included: Outputs a filter replacement reminder for the specific location of the faulty filter element; In response to the user's filter replacement confirmation command, the configuration information is updated.

13. A water purifier, characterized in that, Includes a multi-stage filter system, the multi-stage filter system comprising: A multi-stage filter element is used for filtering liquids, wherein the multi-stage filter element includes at least a first-stage filter element, a second-stage filter element, and a third-stage filter element connected in series. The drive module includes a variable frequency pump, a motor, and a motor drive circuit. The variable frequency pump is connected to the first-stage filter element and is used to provide power for the liquid delivered to the multi-stage filter element. The motor is driven and connected to the variable frequency pump and is used to provide power for the variable frequency pump. The motor drive circuit is driven and connected to the motor drive circuit and is used to drive the motor to work. A current sampling circuit is used to collect the phase current data of the motor; The controller stores and executes a filter element fault diagnosis and classification program, and when executing the program, implements the steps of the filter element fault diagnosis and classification method as described in any one of claims 1 to 12.

14. The water purifier as described in claim 13, characterized in that, The multi-stage filter system also includes a user interface for displaying diagnostic reports and filter maintenance information. The diagnostic report includes at least the probability of filter failure at the specific location of the failure and the remaining lifespan of the filter at the specific location of the failure.

15. A filter cartridge fault diagnosis and classification device, characterized in that, include: The filter cartridge fault diagnosis and classification program is stored in the memory and can run on the processor, the filter cartridge fault diagnosis and classification program being configured to implement the steps of the filter cartridge fault diagnosis and classification method as described in any one of claims 1 to 12.

Images