A method for detecting a charging fault of a charging base of a scrubber

Abstract

The application relates to the field of floor washing machines and discloses a charging fault detection method for a charging base of a floor washing machine, which comprises the following steps: receiving charging terminal voltage data and floor washing machine working mode data forwarded by a home gateway; intercepting static charging data from the charging terminal voltage data based on a charging period; intercepting voltage data during a self-cleaning operation period from the charging terminal voltage data based on a self-cleaning period; obtaining historical reference charging data corresponding to the floor washing machine, calculating the voltage deviation degree of the static charging data and the historical reference charging data; calculating the voltage fluctuation variance in the voltage data during the self-cleaning operation period; judging whether a contact fault occurs in the charging terminal of the charging base based on the voltage deviation degree and the voltage fluctuation variance and determining the corresponding fault type when it is judged that the contact fault occurs; generating a control instruction corresponding to the fault type; and issuing the control instruction to the home gateway to enable the floor washing machine to perform a corresponding safety response.

Description

Technical Field

[0001] This application relates to the field of floor scrubbers, and more particularly to a method for detecting charging faults in the charging base of a floor scrubber. Background Technology

[0002] A floor scrubber is a household cleaning device that integrates cleaning and water absorption functions, and is widely used in daily cleaning of household floors.

[0003] As household cleaning equipment becomes increasingly intelligent, floor scrubbers are typically equipped with a charging dock for recharging. The charging dock connects to the floor scrubber's battery via charging terminals, replenishing the battery power. Some floor scrubbers also feature a self-cleaning function for the roller brush. After returning to its original position, the machine initiates a self-cleaning program, where a motor drives the roller brush to rotate, and a water pump works in conjunction to remove any remaining dirt.

[0004] As the interface for power transmission between the charging base and the floor scrubber, the contact state of the charging terminal affects the charging efficiency and charging safety, so it is necessary to detect charging faults. Summary of the Invention

[0005] This application provides a method for detecting charging faults in the charging base of a floor scrubber, so as to at least partially solve the above-mentioned technical problems.

[0006] To achieve the above objectives, this application provides a method for detecting charging faults in the charging base of a floor scrubber, comprising:

[0007] The system receives charging terminal voltage data and floor scrubber operating mode data forwarded by the home gateway; the floor scrubber operating mode data is used to determine the charging period and self-cleaning period of the floor scrubber on the charging base.

[0008] Based on the charging period, static charging data is extracted from the charging terminal voltage data; based on the self-cleaning period, voltage data during self-cleaning operation is extracted from the charging terminal voltage data.

[0009] Obtain historical reference charging data corresponding to the floor scrubber, calculate the voltage deviation between the static charging data and the historical reference charging data; calculate the voltage fluctuation variance in the voltage data during the self-cleaning operation; the historical reference charging data includes multiple fault-free charging data samples of the floor scrubber within a preset historical time period;

[0010] Based on the voltage deviation and the voltage fluctuation variance, it is determined whether the charging terminal of the charging base has a contact failure, and when a contact failure is determined, the corresponding fault type is identified.

[0011] Generate a control command corresponding to the fault type; send the control command to the home gateway to cause the floor scrubber to perform the corresponding safety response.

[0012] In this embodiment of the application, based on the charging terminal voltage and floor scrubber working mode data forwarded by the home gateway, static charging data and voltage data during self-cleaning operation are extracted during the charging period and self-cleaning period, respectively. The voltage deviation and voltage fluctuation variance are used to determine whether the abnormal contact is due to oxidation and scaling or vibration and loosening. This improves the accuracy of charging terminal contact fault detection and enables the issuance of corresponding safety response commands based on the fault type.

[0013] Other features and advantages of this application will be described in detail in the following detailed description section. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is an exemplary system framework for a charging fault detection method for the charging base of a floor scrubber provided in an exemplary embodiment of this application.

[0016] Figure 2 This is a flowchart illustrating the steps of a charging fault detection method for the charging base of a floor scrubber provided in an exemplary embodiment of this application. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the protection scope of this application.

[0018] Floor scrubbers, as household cleaning devices that combine cleaning and water absorption functions, are typically equipped with a charging dock for recharging. The charging dock connects to the conductive contacts on the scrubber's body via charging terminals, replenishing the scrubber's battery. During the recharging process, the scrubber also features a self-cleaning function for its roller brush. Under certain conditions, after returning to its charging dock, the motor drives the roller brush to rotate, working in conjunction with a water pump to remove residual dirt. The charging terminals, as the crucial interface for power transmission between the charging dock and the scrubber, directly affect charging efficiency and safety.

[0019] In actual use, the charging terminals are often in humid and dusty home environments, and water stains and cleaning agent residue generated during the self-cleaning process may splash onto the terminal surface; the terminals may experience increased contact resistance or intermittent circuit breakage due to oxidation and scaling or fatigue deformation of the spring contacts caused by frequent insertion and removal, which may affect charging efficiency and thermal safety.

[0020] Reference Figure 1 The charging fault detection method for the charging base of a floor scrubber provided in this application embodiment operates in an environment including a cloud server, a home gateway, a charging base, and a floor scrubber. The charging base is equipped with charging terminals, and the floor scrubber body has conductive contacts that mate with the charging terminals. A voltage acquisition unit is installed inside the charging base to collect voltage signals from the charging terminals in real time. The home gateway is connected to the charging base via a local communication network and communicates with the cloud server via a wide area network. The floor scrubber's operating mode data is reported from the floor scrubber's motherboard to the home gateway through the communication interface of the charging base, and then forwarded by the home gateway to the cloud server.

[0021] Reference Figure 2 A method for detecting charging faults in the charging base of a floor scrubber, applied to a cloud server, includes:

[0022] Step 101: Receive charging terminal voltage data and floor scrubber operating mode data forwarded by the home gateway; the floor scrubber operating mode data is used to determine the charging period and self-cleaning period of the floor scrubber on the charging base.

[0023] Specifically, the home gateway, acting as a local data aggregation node, connects to the charging dock via WiFi. The charging dock's built-in voltage acquisition unit collects the voltage values ​​from the charging terminals, generating charging terminal voltage data. This charging terminal voltage data refers to the voltage sampling values ​​at the metal terminals on the charging dock that make contact with the floor scrubber's conductive contacts. The floor scrubber's operating mode data is generated and reported by the mainboard when its status changes, including identification information for the current mode, such as charging mode and self-cleaning mode identifiers, as well as timestamps for each mode. The cloud server parses the mode identifiers and corresponding timestamps in the floor scrubber's operating mode data to determine the time boundaries for the charging and self-cleaning periods. The charging period refers to the time interval during which the floor scrubber is stationary and charging after returning to its original position, while the self-cleaning period refers to the time interval during which the floor scrubber executes the roller brush self-cleaning program.

[0024] Step 102: Based on the charging period, extract the static charging data from the charging terminal voltage data; based on the self-cleaning period, extract the voltage data during the self-cleaning operation from the charging terminal voltage data.

[0025] Specifically, the cloud server uses the time boundary of the charging period as a clipping window to extract voltage sampling data belonging to the charging period from the charging terminal voltage data, which is used as static charging data. The aforementioned static charging data reflects the voltage change pattern of the floor scrubber in a static charging state, and its fluctuations mainly originate from the slow change in terminal contact resistance. Using the time boundary of the self-cleaning period as a clipping window, voltage sampling data belonging to the self-cleaning period is extracted from the same charging terminal voltage data, which is used as voltage data during self-cleaning operation. The voltage data during self-cleaning operation reflects the voltage fluctuation of the charging terminals under the condition of vibration generated by the motor driving the roller brush rotation, and its fluctuation amplitude is related to the tightness of the terminal contact.

[0026] Step 103: Obtain the historical reference charging data corresponding to the floor scrubber, and calculate the voltage deviation between the static charging data and the historical reference charging data. The historical reference charging data includes multiple fault-free charging data samples of the floor scrubber within a preset historical time period. Calculate the voltage fluctuation variance in the voltage data during self-cleaning operation. Specifically, the cloud server maintains a historical charging log library for each floor scrubber. The historical reference charging data is a set of charging voltage data selected from this log library when the floor scrubber is charging normally within a preset historical time period. The preset historical time period can be set according to actual needs. After receiving the static charging data of the current charging session, the cloud server calls the historical reference charging data corresponding to the floor scrubber, aligns the two in time, and calculates the voltage deviation. The voltage deviation reflects the overall deviation between the current static charging voltage and the normal state voltage. An increase in its value indicates that the terminal contact resistance may increase due to oxidation and scaling. The cloud server calculates the voltage fluctuation variance for the voltage data during self-cleaning operation. The voltage fluctuation variance reflects the stability of the terminal contact under motor vibration. An increase in its value indicates that the terminals may be mechanically loose.

[0027] Step 104: Based on the voltage deviation and the voltage fluctuation variance, determine whether the charging terminal of the charging base has a contact fault, and determine the corresponding fault type when a contact fault is determined.

[0028] Specifically, the cloud server compares the calculated voltage deviation and voltage fluctuation variance with preset deviation and variance judgment thresholds, respectively, to identify two types of contact faults. When the voltage deviation is normal but the voltage fluctuation variance is abnormal, it tends to be judged as a vibration and loosening fault. When the voltage deviation is abnormal but the voltage fluctuation variance is normal, it tends to be judged as an oxidation and scaling fault.

[0029] Step 105: Generate a control command corresponding to the fault type; send the control command to the home gateway to enable the floor scrubber to perform the corresponding safety response.

[0030] Specifically, the cloud server generates corresponding control commands based on the determined fault type. These control commands can be power limiting commands to reduce the charging current, or alarm commands to suspend charging and prompt the user for maintenance. After receiving the control commands, the home gateway forwards them to the charging base or the floor scrubber's mainboard through the local communication link, and the floor scrubber executes the corresponding safety response actions, such as reducing the charging power, illuminating the fault indicator light, or issuing a voice prompt.

[0031] The above technical solution extracts static charging data and self-cleaning operation data based on the charging terminal voltage and floor scrubber operating mode data forwarded by the home gateway. The voltage data during static charging and self-cleaning operation is extracted during the charging and self-cleaning periods, respectively. The voltage deviation and voltage fluctuation variance are used to identify contact abnormalities caused by oxidation and scaling and contact abnormalities caused by vibration and loosening. This improves the accuracy of charging terminal contact fault detection and can issue corresponding safety response commands according to the fault type.

[0032] In one implementation, determining whether a contact fault has occurred at the charging terminal of the charging base based on the voltage deviation and the voltage fluctuation variance, and determining the corresponding fault type when a contact fault is determined, includes:

[0033] When the voltage deviation is less than the deviation judgment threshold and the voltage fluctuation variance is greater than the variance judgment threshold, a contact fault is determined to have occurred at the charging terminal, and the fault type is identified as a first-type contact anomaly. The first-type contact anomaly includes loose terminal contacts of the floor scrubber or loose spring contacts on the charging base. Specifically, the deviation judgment threshold is a boundary value determined based on the historical voltage deviation distribution under normal terminal contact conditions, used to distinguish whether the static voltage deviation exceeds the normal fluctuation range; the variance judgment threshold is a boundary value determined based on the natural fluctuation amplitude of the terminal voltage under vibration-free conditions, used to distinguish whether the voltage fluctuation during self-cleaning vibration is abnormal. When the voltage deviation is small, it indicates that the contact resistance of the terminal under static conditions has not increased significantly and the degree of oxidation and scaling is low. However, a large voltage fluctuation variance indicates that intermittent contact disconnection has occurred between the terminals under motor vibration, consistent with the voltage characteristics of mechanical loosening faults. The first-type contact anomaly can specifically manifest as elastic fatigue deformation of the conductive contacts on the floor scrubber body due to long-term insertion and removal, resulting in insufficient contact pressure between the contacts and the charging terminal, or elastic attenuation of the charging base spring contacts due to repeated bending.

[0034] When the voltage deviation is greater than or equal to the deviation judgment threshold and the voltage fluctuation variance is less than or equal to the variance judgment threshold, a contact fault is determined to have occurred at the charging terminal, and the fault type is identified as a second type of contact anomaly, which includes terminal oxidation and scaling. Specifically, when the voltage deviation exceeds the deviation judgment threshold, it indicates that even under static conditions without vibration, the terminal contact resistance has increased significantly, and the static charging voltage continues to deviate from the normal level; while the voltage fluctuation variance is small, it indicates that the terminal did not experience intermittent contact disconnection during self-cleaning vibration, and the terminal connection remains tight. This combination of increased voltage deviation and normal fluctuation variance is consistent with the voltage characteristics of a stable increase in contact resistance caused by the thickening of the oxide film or dirt layer on the terminal surface; when the charging terminal is exposed to humid air and self-cleaning moisture for a long time, oxides or scale may gradually form on the surface, forming a high-resistivity film, causing a stable positive deviation in the charging voltage.

[0035] By using the above technical solution and the dual-indicator judgment logic of voltage deviation and voltage fluctuation variance, terminal loosening faults and oxidation scaling faults can be effectively distinguished in terms of voltage signal characteristics. This enables the cloud server to take different processing strategies according to different fault causes, avoiding unnecessary downtime or maintenance omissions caused by using a uniform response method for different types of faults.

[0036] In one implementation, obtaining the historical reference charging data corresponding to the floor scrubber includes:

[0037] The historical charging logs of the floor scrubber within a preset sliding time window prior to the current time point are extracted from the cloud server. Specifically, the historical charging log library on the cloud server stores the charging session records of each floor scrubber using the device identifier as an index. Each record includes voltage data for the charging period, voltage data for the self-cleaning period, working mode flags, and abnormal alarm flags. The preset sliding time window is, for example, 90 days; when the interval between the last charging of the floor scrubber is long, historical records exceeding this time window are no longer included in the reference range to maintain the timeliness of historical reference charging data.

[0038] Traverse the historical charging logs, remove invalid logs that contain abnormal alarm indicators or charging interruption indicators, and obtain a candidate log set.

[0039] Specifically, the cloud server iterates through the historical charging logs one by one, checking each log for abnormal alarm markers, such as over-temperature alarms, over-current alarms, and abnormal charging termination markers. Logs with these markers are considered to contain voltage data under fault conditions and cannot be used as a normal benchmark, so they are discarded. A charging interruption marker indicates that the charging was not completed normally, such as when the user removed the floor scrubber midway or the power was cut off; the voltage data in this log is incomplete. After removing invalid logs containing the above two types of markers, the remaining logs constitute a candidate log set.

[0040] The current firmware version number of the floor scrubber is obtained, and log records that do not match the firmware version number are removed from the candidate log set. Specifically, firmware upgrades may cause changes in charging management strategies, voltage sampling accuracy, or communication protocols, and the statistical characteristics of charging voltage data collected under different firmware versions may differ. The cloud server extracts the current firmware version number from the device information reported by the floor scrubber, compares it with the firmware version number of each log record in the candidate log set, and removes log records with inconsistent version numbers.

[0041] The remaining log records after filtering are used as initial samples to obtain the multiple fault-free charging data samples; the multiple fault-free charging data samples together constitute the historical reference charging data.

[0042] Specifically, the remaining log records after abnormal log removal and firmware version matching are the initial samples. Each initial sample contains complete static charging voltage data for a normal charging session. These initial samples together constitute the historical reference charging data. The quantity and quality of the historical reference charging data directly affect the accuracy of voltage deviation calculation. When the number of available samples is less than the preset minimum number of samples, the cloud server can extend the span of the sliding time window to supplement the samples.

[0043] Through the above technical solution, the cloud server extracts fault-free charging data samples from historical charging logs by combining abnormal alarm indicators, charging interruption indicators, and firmware version number matching, which serves as historical reference charging data. This ensures the accuracy of voltage deviation calculation and avoids interference from fault logs and cross-version data on the accuracy of contact fault judgment.

[0044] In one implementation, obtaining the plurality of fault-free charging data samples by using the remaining log records after filtering as initial samples includes:

[0045] The initial terminal voltage characteristic in the charging terminal voltage data is analyzed. Specifically, the initial terminal voltage characteristic refers to the terminal voltage value at the beginning of the charging circuit establishment after the floor scrubber is first connected to the charging base. This voltage value is mainly affected by the current remaining power of the floor scrubber's battery; the lower the remaining power, the lower the initial charging voltage. The cloud server selects the average or median voltage value within a preset sampling window after the start of charging from the charging terminal voltage data of the current charging session as the initial terminal voltage characteristic.

[0046] The corresponding historical starting terminal voltage is extracted from each of the initial samples. Specifically, for each fault-free charging data sample in the historical reference charging data, the cloud server extracts the starting terminal voltage value of that historical charging session in the same way and records it as the historical starting terminal voltage.

[0047] The absolute difference between the historical starting terminal voltage and the starting terminal voltage characteristic is calculated. Specifically, for each historical starting terminal voltage, the cloud server calculates the absolute difference between it and the current starting terminal voltage characteristic. This difference reflects the degree of difference between the historical sample and the current charging session in terms of the initial battery capacity.

[0048] When the absolute difference is less than the tolerance threshold, the corresponding initial sample is retained; when the absolute difference is greater than or equal to the tolerance threshold, the corresponding initial sample is discarded to eliminate the influence of initial charge differences on the idle charging data. Specifically, the tolerance threshold is a preset allowable deviation value used to determine whether the initial charge of a historical sample is sufficiently close to the current charging session. When the absolute difference is less than the threshold, it indicates that the initial battery charge state of the historical sample is close to the current state, and its idle charging voltage data is comparable to the current data in terms of charge, so it is retained. When the absolute difference reaches or exceeds the threshold, it indicates that the initial charge of the historical sample differs significantly from the current state. If it is included in the historical reference charging data, the calculated voltage deviation may be too large due to charge differences rather than contact faults, so it is discarded.

[0049] The set of retained initial samples is used as the plurality of fault-free charging data samples. Specifically, initial samples with absolute differences within the allowable range are retained to form the final plurality of fault-free charging data samples; this sample set eliminates voltage offset interference caused by differences in the initial battery charge, making the voltage deviation more accurately reflect changes in the terminal contact state.

[0050] By using the above technical solution, samples with significant differences in initial charge levels compared to the current charging session are further removed from historical reference charging data, thus eliminating the interference of initial charge level differences on the comparability of static charging data.

[0051] In one implementation, calculating the voltage deviation between the static charging data and the historical reference charging data includes:

[0052] Using the starting sampling point of the idle charging data as a time reference, the historical reference charging data is shifted and aligned along the time axis. Specifically, both the idle charging data and the historical reference charging data contain voltage sampling data arranged in chronological order. However, due to the different charging start times in different charging sessions, the timestamps of the two sets of data are offset. The cloud server uses the first sampling point of the current idle charging data as the alignment reference and shifts the voltage data of each fault-free charging data sample of the historical reference charging data along the time axis to align its first sampling point with the first sampling point of the idle charging data, thus establishing a unified time coordinate system for point-by-point comparison.

[0053] In the aligned time coordinate system, the first voltage time series data of the static charging data is obtained, and the corresponding second voltage time series data is extracted from the historical reference charging data. Specifically, after time alignment, the cloud server extracts the voltage values ​​arranged in chronological order from the static charging data to form the first voltage time series data. For the historical reference charging data, if it contains multiple fault-free charging data samples, the voltage values ​​of multiple samples at the same time point can be averaged or the median value to synthesize a second voltage time series data representing the normal charging voltage trend.

[0054] The system iterates through each time node in the first voltage time series data, calculating the absolute difference between the first voltage value and the corresponding second voltage value at each time node, generating a voltage difference set. Specifically, starting from the aligned time axis, the cloud server calculates the absolute difference between the current voltage value in the first voltage time series data and the corresponding voltage value in the second voltage time series data for each time node, and aggregates the absolute differences of all time nodes into a voltage difference set. The magnitude of each element in the voltage difference set reflects the deviation between the current charging voltage and the historical normal charging voltage at that time node.

[0055] Each absolute difference in the voltage difference set is assigned a time-increasing weight; the magnitude of the time-increasing weight is positively correlated with the chronological order of the corresponding time node. Specifically, during charging, the change in terminal contact resistance has a cumulative effect as charging time progresses, and voltage deviations appearing later reflect the true trend of terminal contact state changes better than deviations appearing earlier. The cloud server assigns a time-increasing weight to each time node in the voltage difference set, with later time nodes corresponding to larger weights. The time-increasing weight can use a linear increasing function, i.e., the weight is proportional to the time node index, or a non-linear increasing function, such as an exponential function, can be used to make the weight of later nodes grow faster.

[0056] Based on the aforementioned time-series increasing weights, all the absolute differences are weighted and summed, and a weighted average is calculated. This weighted average is used as the voltage deviation. Specifically, the cloud server multiplies each absolute difference in the voltage difference set by its corresponding time-series increasing weight, sums the results, and then divides the weighted sum by the sum of all weights to obtain the weighted average, which is the voltage deviation. The voltage deviation is a numerical indicator that comprehensively reflects the overall deviation between the current stationary charging voltage and the historical normal charging voltage. The design of tilting the weights towards later nodes makes this indicator more sensitive to the terminal contact degradation that gradually appears in the later stages of charging.

[0057] The above technical solution uses a time-series increasing weight to calculate the voltage deviation by weighted averaging. This makes the voltage deviation index more sensitive to the terminal contact degradation that gradually appears in the later stages of charging. Compared with the equal weighting method, it can detect the increasing contact resistance trend caused by terminal oxidation and scaling earlier.

[0058] In one implementation, calculating the voltage fluctuation variance in the voltage data during the self-cleaning operation includes:

[0059] Step 601: A preset high-pass filter operator is used to filter the voltage data during self-cleaning operation. This filters out the first reference voltage trend generated by the continuous discharge of the floor scrubber battery, resulting in contact noise data caused by motor vibration. Specifically, during self-cleaning operation, even with normal terminal contact, the continuous discharge of the floor scrubber battery causes the charging terminal voltage to slowly decrease, forming a slowly decreasing voltage trend curve, i.e., the first reference voltage trend. This first reference voltage trend is a low-frequency component, while the terminal contact fluctuations caused by motor vibration are a high-frequency component superimposed on this trend. The cloud server uses a preset high-pass filter operator, such as a first-order high-pass digital filter or a moving window de-trending method, to filter the voltage data during self-cleaning operation, removing the low-frequency first reference voltage trend and retaining the high-frequency voltage fluctuation components to obtain contact noise data. The contact noise data mainly originates from the voltage jitter introduced by the periodic vibration generated when the motor rotates, transmitted through the terminal contact surface to the voltage acquisition channel. Its amplitude directly reflects the tightness of the terminal contact.

[0060] The variance of each voltage value in the contact noise data is calculated, and this variance is used as the voltage fluctuation variance to obtain the mechanical loosening characteristics. Specifically, the cloud server calculates the variance of all voltage values ​​in the contact noise data, which is the average of the squares of the deviations of each voltage value from its mean. This variance is the voltage fluctuation variance, reflecting the severity of the fluctuations in the contact noise data. The tighter the terminal contact, the smaller the amplitude of vibration transmitted to the voltage signal, and the smaller the voltage fluctuation variance. The looser the terminal contact, the greater the amplitude of voltage fluctuations caused by vibration, and the larger the voltage fluctuation variance. Therefore, the voltage fluctuation variance can be used as a mechanical loosening characteristic to characterize the degree of degradation in the reliability of the terminal connection.

[0061] In one implementation, the high-pass filter operator can be a second-order Butterworth high-pass filter with a cutoff frequency set to a value lower than the motor's fundamental frequency, so as to effectively filter out the battery discharge trend while retaining the motor vibration frequency components.

[0062] In another implementation, the extraction of the first reference voltage trend can be achieved by using a polynomial fitting method instead of high-pass filtering. First, a low-order polynomial trend line is fitted to the voltage data during self-cleaning operation, and then the contact noise data is obtained by subtracting the trend line from the original voltage data.

[0063] The above technical solution separates the battery discharge trend from the voltage fluctuation caused by motor vibration through high-pass filtering, so that the voltage fluctuation variance reflects the vibration transmission characteristics caused by terminal mechanical loosening, thus avoiding interference from the battery discharge trend in the judgment of terminal loosening.

[0064] In one implementation, calculating the variance of each voltage value in the contact noise data and using the variance as the voltage fluctuation variance includes:

[0065] The floor scrubber's operating mode data is analyzed to obtain the real-time motor speed data during the self-cleaning operation. Specifically, the floor scrubber's operating mode data includes not only a mode identifier and timestamp, but also motor operating parameters, among which the real-time speed data records the actual motor speed value at each moment during the self-cleaning period; the cloud server analyzes this data to obtain the correspondence between time and speed during the self-cleaning operation.

[0066] Based on the real-time rotational speed data, the fundamental frequency range of the motor during operation is mapped.

[0067] Specifically, there is a conversion relationship between the vibration frequency generated by the motor rotation and the motor speed; the cloud server calculates the corresponding vibration fundamental frequency based on real-time speed data moment by moment and combines it with the fluctuation range of the motor speed to determine the fundamental frequency vibration range of the motor vibration during self-cleaning operation; this frequency range defines the distribution range of normal vibration in the frequency domain, and the voltage fluctuation energy falling within this range can be attributed to normal motor vibration, while the voltage fluctuation energy falling outside this range may come from other interference sources, such as water splashes.

[0068] The contact noise data undergoes time-frequency transformation to generate a corresponding frequency domain energy distribution map. Specifically, the cloud server performs time-frequency transformation on the contact noise data, such as short-time Fourier transform or wavelet transform, converting the voltage fluctuation signal in the time domain into an energy distribution in the frequency domain, generating a frequency domain energy distribution map. This map, with time as the horizontal axis, frequency as the vertical axis, and energy intensity as color depth or brightness, displays the changes in voltage fluctuation energy of different frequency components over time during self-cleaning operation.

[0069] The resonant energy percentage of the frequency bands falling within the fundamental frequency range in the frequency domain energy distribution map is calculated. Specifically, the cloud server calculates the sum of the frequency band energies falling within the fundamental frequency range in each time window, and the sum of the total energy of all frequency bands in that time window, on the frequency domain energy distribution map. The ratio of the former to the latter is calculated to obtain the resonant energy percentage for each time window. Finally, the average of the resonant energy percentages of each time window during the entire self-cleaning operation is taken as the overall resonant energy percentage. A higher resonant energy percentage indicates that the voltage fluctuations in the contact noise data are mainly caused by normal motor vibration; a lower percentage indicates that there is a large amount of voltage fluctuation interference caused by non-mechanical resonance factors, such as water splashes or detergent residue.

[0070] The initial variance of each voltage value in the contact noise data is calculated, and the initial variance is weighted and corrected using the resonant energy ratio. The corrected result is used as the voltage fluctuation variance. Specifically, when the resonant energy ratio is less than a preset purity threshold, the contact noise data is determined to be affected by non-mechanical resonance factors such as water splashes, and the value of the voltage fluctuation variance is reduced to decrease the possibility of misjudging water splash interference as a type I contact anomaly. Specifically, the cloud server first calculates the initial variance of the contact noise data, i.e., the uncorrected standard variance. If the initial variance is directly used as the voltage fluctuation variance, when self-cleaning water stains or residual liquid splash onto the charging terminal, liquid conduction or electrolytic effects may generate additional voltage fluctuations, leading to an abnormally large initial variance, which may be misjudged as terminal mechanical loosening. The cloud server uses the proportion of resonant energy to correct the initial variance. When the proportion of resonant energy is greater than or equal to the purity threshold, it indicates that the voltage fluctuation is mainly caused by the normal vibration of the motor. The correction coefficient is close to 1, and the initial variance remains basically unchanged. When the proportion of resonant energy is less than the purity threshold, it indicates that there is a large amount of non-resonant interference. The correction coefficient is less than 1, and the initial variance is corrected downward to reduce the value of voltage fluctuation variance and reduce misjudgments caused by water stain interference.

[0071] The above technical solution utilizes the real-time speed mapping of the motor to obtain the fundamental frequency range of vibration. By using time-frequency transformation and resonance energy ratio calculation, the influence of normal vibration and water stain interference on voltage fluctuations is distinguished. The initial variance is weighted and corrected, which effectively reduces the misjudgment of the first type of contact anomaly caused by non-mechanical factors such as water stain splashing and cleaning fluid residue.

[0072] In one implementation, before step S102 extracts the static charging data from the charging terminal voltage data, the method further includes:

[0073] Based on the floor scrubber's operating mode data, the end time of the self-cleaning period and the start time of the charging period are obtained, and the interval between the end of self-cleaning and the start of static charging is calculated. Specifically, after the floor scrubber executes the self-cleaning program, water stains may remain on the surface of the charging base and the charging terminal area. If the static charging voltage is measured immediately during the charging period, the conductive path or electrolytic effect formed by the water stains may introduce additional voltage fluctuations into the static charging data. The cloud server extracts the end time of the self-cleaning period and the start time of the charging period from the floor scrubber's operating mode data, calculates the time difference between the two, and obtains the interval. The interval reflects the static drying time from the end of the floor scrubber's self-cleaning process to the start of charging.

[0074] If the interval is less than a preset water stain drying threshold, the current stationary charging data is marked as candidate data for water stain interference, and the voltage deviation calculation is paused. Specifically, the water stain drying threshold is a time threshold determined based on the time required for the terminal area to air dry naturally under typical ambient temperature and humidity, such as 5 minutes or 10 minutes. When the interval is less than the water stain drying threshold, it indicates that the self-cleaning residual water stains have not been fully dried, and the collected stationary charging voltage data at this time may contain water stain interference components. The cloud server marks the current stationary charging data as candidate data for water stain interference and pauses the voltage deviation calculation process, without performing fault detection analysis on this batch of data.

[0075] After the floor scrubber completes the preset drying process, the charging terminal voltage is re-collected and the static charging data is updated to eliminate false voltage anomalies caused by residual water stains from self-cleaning. Specifically, the cloud server sends a waiting command to the home gateway, or the home gateway times the process locally. After the water stain drying threshold time is reached, the charging terminal voltage data is re-collected and new static charging data is captured. The new static charging data replaces the marked water stain interference candidate data, and the voltage deviation calculation continues. The preset drying process can be set with a waiting time according to actual environmental conditions, or the floor scrubber can activate the base heating drying function to accelerate the drying process.

[0076] Through the above technical solution, the cloud server checks the interval between the end of self-cleaning and the start of charging before calculating the voltage deviation. For charging sessions with insufficient intervals, the analysis is paused and the data is collected again after drying. This reduces the interference of residual water stains from self-cleaning on the static charging data and reduces false judgments of voltage anomalies caused by water stains.

[0077] In one implementation, before determining whether a contact failure has occurred at the charging terminal of the charging dock, the method further includes:

[0078] The cumulative charge-discharge cycle count of the floor scrubber is obtained from the cloud server. Specifically, the cloud server maintains a device file for each floor scrubber, where the cumulative charge-discharge cycle count records the total number of complete charge-discharge cycles since the floor scrubber was first activated. Each time a complete charge and subsequent discharge process is detected, the cycle count is incremented by one; the cumulative charge-discharge cycle count is an effective indicator for measuring the service life of the floor scrubber's battery and charging terminals.

[0079] When the cumulative charge-discharge cycle count is greater than or equal to a cycle count threshold, the corresponding aging correction coefficient is obtained. Specifically, the cycle count threshold is a preset cycle count limit. When the cumulative charge-discharge cycle count is lower than this threshold, the floor scrubber is considered to be in a relatively new state, and the terminal contact characteristics have not yet shown obvious aging; the deviation judgment threshold remains unchanged. When the cumulative charge-discharge cycle count reaches or exceeds the cycle count threshold, it indicates that the floor scrubber has entered the aging stage. Due to long-term insertion and removal wear, the normal contact resistance of the terminal contact surface may naturally increase. If the factory-set deviation judgment threshold is still used, normal aging may be misjudged as a fault. The cloud server obtains the aging correction coefficient corresponding to the current cycle count. The aging correction coefficient is a value greater than or equal to 1, used to amplify the deviation judgment threshold to adapt to the terminal characteristics after aging.

[0080] The deviation judgment threshold is generated based on the aging correction coefficient and the initial deviation judgment threshold; the larger the cumulative charge-discharge cycle count, the larger the deviation judgment threshold. Specifically, the cloud server multiplies the initial deviation judgment threshold by the aging correction coefficient to obtain a deviation judgment threshold adapted to the current aging level. The aging correction coefficient is positively correlated with the cumulative charge-discharge cycle count; the larger the cycle count, the larger the aging correction coefficient, and the larger the deviation judgment threshold, meaning a higher tolerance for aging. This threshold adjustment method, which increases with the years of use, avoids false alarms caused by normal aging while retaining the alarm capability when abnormal accelerated degradation occurs at the terminals.

[0081] In one implementation, the relationship between the aging correction factor and the cumulative number of charge-discharge cycles can be mapped using a piecewise linear function. For example, the correction factor increases by 0.05 for every 100 additional cycles, with an upper limit of 1.5, in order to achieve a balance between the correction magnitude and the alarm sensitivity.

[0082] By introducing an aging correction coefficient associated with the cumulative number of charge-discharge cycles, the deviation judgment threshold is adjusted, so that the fault judgment standard gradually relaxes as the terminal ages naturally. This avoids frequent false alarms caused by normal wear and tear and can maintain effective fault detection capability when the terminal undergoes abnormal accelerated degradation.

[0083] In one implementation, after determining that the fault type is a first-type contact anomaly, the method further includes:

[0084] When the charging period occurs before the self-cleaning period, based on the floor scrubber's operating mode data, a recovery charging period is determined whereby the floor scrubber enters normal charging or full-charge float charging mode after the self-cleaning operation ends. Specifically, in some usage scenarios, the floor scrubber enters the charging period after returning to its position, and then executes the self-cleaning program after charging is completed or during charging; in this sequence, the charging period occurs before the self-cleaning period; after the self-cleaning period ends, the cloud server identifies whether the floor scrubber has re-entered normal charging or full-charge float charging mode based on the floor scrubber's operating mode data, and determines the period during which it re-enters the charging state as the recovery charging period; the recovery charging period and the charging period previously determined to be the first type of contact abnormality belong to two different stages of the same charging base returning event.

[0085] The recovery charging voltage data corresponding to the recovery charging period is extracted from the charging terminal voltage data. Specifically, the cloud server uses the time boundary of the recovery charging period as a clipping window to extract the voltage sampling data within that period from the charging terminal voltage data of the same charging session, and uses it as the recovery charging voltage data.

[0086] The verification deviation between the restored charging voltage data and the historical reference charging data is calculated. Specifically, the cloud server performs time-series alignment between the restored charging voltage data and the historical reference charging data and then calculates a weighted average voltage deviation as the verification deviation. If the terminal contact returns to normal during the restoration charging process, the verification deviation should fall back to the normal range. If the terminals are indeed irreversibly loose, the verification deviation will remain at a high abnormal value.

[0087] When the deviation of the verification is less than the deviation judgment threshold, the first type of contact anomaly is confirmed as a momentary fluctuation-type fault induced by vibration, thus eliminating the risk of misjudging battery damage. Specifically, if the verification deviation is less than the deviation judgment threshold, it indicates that the terminal contact status has returned to normal after the self-cleaning vibration ends, and the first type of contact anomaly detected during the self-cleaning period was indeed a momentary contact fluctuation induced by the vibration of the self-cleaning motor, rather than permanent loosening of the terminals or battery damage. The verification process distinguishes momentary vibration fluctuations from mechanical damage, eliminating the risk of misjudging non-contact faults such as changes in battery internal resistance or battery aging as the first type of contact anomaly.

[0088] By using the above technical solution, after determining the first type of contact abnormality, a verification of the deviation during the recovery charging period is introduced to distinguish the instantaneous fluctuation type fault induced by self-cleaning vibration from terminal mechanical damage, thereby eliminating the risk of misjudging battery damage and improving the accuracy of fault characterization.

[0089] In one implementation, the method further includes:

[0090] When the fault type is determined to be the second type of contact anomaly, a first push instruction containing terminal wiping prompts is generated. Specifically, when the cloud server classifies the current contact fault as a second type of contact anomaly, i.e., terminal oxidation and scaling, based on a joint determination of voltage deviation and voltage fluctuation variance, the first push instruction is generated. The first push instruction includes text information prompting the user that the terminal needs to be wiped and cleaned, and may also include graphic instructions for wiping, such as suggesting the use of a dry soft cloth or lint-free cotton swabs to wipe the terminal surface. The first push instruction is sent to the user management terminal, such as a mobile APP or smart speaker, via a cloud push service.

[0091] Upon confirming that the first type of contact anomaly is a transient fluctuation-type fault, a second push instruction containing a replacement indicator for the base spring is generated. Specifically, when the cloud server confirms that the first type of contact anomaly is a vibration-induced transient fluctuation-type fault rather than mechanical damage, a second push instruction is generated; the second push instruction contains a replacement indicator for the base spring, prompting the user that the elasticity of the charging base spring may have deteriorated and needs to be replaced, and may include spring model and replacement method instructions.

[0092] The first push instruction or the second push instruction is sent to a preset user management terminal to execute the corresponding maintenance procedures for different fault causes. Specifically, depending on the fault type, the cloud server sends the first push instruction to the user management terminal to prompt for wiping maintenance, or sends the second push instruction to the user management terminal to prompt for replacement of the spring contact. After receiving the push instruction, the user management terminal displays the corresponding maintenance reminder and execution instructions on the application interface, guiding the user to perform the maintenance procedure corresponding to the fault cause.

[0093] Through the above technical solution, the cloud server generates corresponding push instructions based on the differentiated judgment results of the fault type, so that oxidation and scaling faults trigger wiping prompts and vibration instantaneous fluctuation faults trigger spring replacement prompts.

[0094] In one implementation, after sending the first push instruction, the method further includes:

[0095] The system receives wiping information from the user management terminal and acquires verification status data reported when the floor scrubber reconnects to the charging dock. Specifically, after receiving the first push instruction and completing the terminal wiping as instructed, the user management terminal sends wiping information to the cloud server, including wiping completion confirmation and a timestamp. When the user reconnects the floor scrubber to the charging dock, and the floor scrubber re-establishes contact with the charging terminals, the charging dock re-collects voltage data and reports it to the cloud server via the home gateway. This batch of data is marked as verification status data. The verification status data includes charging terminal voltage data and operating mode data after reconnection, used to verify the effectiveness of the wiping maintenance.

[0096] Based on the duration corresponding to the charging period, verification standby data is extracted from the verification status data. Specifically, the cloud server uses the duration of the charging period previously used to determine the second type of contact anomaly as a clipping window, and extracts standby charging data of the same duration from the verification status data as verification standby data. Using the same time window length ensures that the conditions for the two voltage deviation calculations are consistent, eliminating deviations caused by differences in time windows.

[0097] Specifically, the cloud server recalculates the verification deviation between the verified static data and the historical reference charging data after aligning the verified static data with the historical reference charging data according to their timelines. If the wiping action effectively removes the oxide layer and dirt from the terminal surface, the verification deviation should fall back to the normal range below the deviation judgment threshold.

[0098] If the verification deviation still exceeds the deviation judgment threshold, the charging terminal is determined to have irreversible terminal corrosion, and a review work order is generated and sent to the user management terminal. Specifically, if the verification deviation still exceeds the deviation judgment threshold after wiping, it indicates that the oxide scale on the terminal surface has developed into deep corrosion, and simple wiping cannot restore the normal contact resistance of the terminal. The cloud server determines this situation as irreversible terminal corrosion, generates a review work order containing the corrosion judgment result and recommended professional repair suggestions, and sends it to the user management terminal, prompting the user to contact after-sales service for terminal replacement or deep cleaning.

[0099] When the verification deviation is less than or equal to the deviation judgment threshold, it is determined that the second type of contact anomaly has been eliminated by the user's wiping action. The verification status data is then added to the historical reference charging data as a fault-free charging data sample. Simultaneously, the oldest historical sample in the historical reference charging data is removed to achieve rolling updates of the baseline status. Specifically, if the verification deviation falls back to within the deviation judgment threshold after wiping, it indicates that the second type of contact anomaly has been successfully eliminated by the user's wiping action, and the terminal contact status has returned to normal. The cloud server marks the verification status data that has passed this verification as a fault-free charging data sample and adds it to the historical reference charging data set of the floor scrubber. At the same time, the oldest historical sample is removed from the historical reference charging data set to maintain a constant sample size. Through this rolling update method of adding and subtracting, the historical reference charging data continuously incorporates the latest normal charging voltage data, allowing the voltage reference to gradually adapt to the normal characteristic drift of the equipment over time, without increasing computational overhead due to the increase in the number of samples.

[0100] With the above technical solution, after the user performs terminal wiping maintenance, the cloud server confirms the maintenance effect by verifying the deviation: when the maintenance is effective, the verification data is included in the historical reference charging data for benchmark rolling updates.

[0101] The above are merely preferred embodiments of this application and are not intended to limit this application in any way. Any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of this application without departing from the scope of the technical solution of this application shall still fall within the scope of the technical solution of this application.

Claims

1. A method for detecting charging faults in the charging base of a floor scrubber, characterized in that, include: Receives charging terminal voltage data and floor scrubber operating mode data forwarded by the home gateway; The floor scrubber's operating mode data is used to determine the charging period and self-cleaning period of the floor scrubber on the charging base; Based on the charging period, static charging data is extracted from the charging terminal voltage data; Based on the self-cleaning period, voltage data during the self-cleaning operation is extracted from the charging terminal voltage data; Obtain historical reference charging data corresponding to the floor scrubber, calculate the voltage deviation between the static charging data and the historical reference charging data; calculate the voltage fluctuation variance in the voltage data during the self-cleaning operation; the historical reference charging data includes multiple fault-free charging data samples of the floor scrubber within a preset historical time period; Based on the voltage deviation and the voltage fluctuation variance, it is determined whether the charging terminal of the charging base has a contact failure, and when a contact failure is determined, the corresponding fault type is identified. Generate a control command corresponding to the fault type; send the control command to the home gateway to cause the floor scrubber to perform the corresponding safety response.

2. The method according to claim 1, characterized in that, Based on the voltage deviation and the voltage fluctuation variance, it is determined whether a contact fault has occurred at the charging terminal of the charging base, and when a contact fault is determined, the corresponding fault type is identified, including: When the voltage deviation is less than the deviation judgment threshold and the voltage fluctuation variance is greater than the variance judgment threshold, it is determined that the charging terminal has a contact fault and the fault type is determined to be the first type of contact abnormality; the first type of contact abnormality includes the terminal contact of the floor scrubber being loose or the spring contact of the charging base being loose; When the voltage deviation is greater than or equal to the deviation determination threshold and the voltage fluctuation variance is less than or equal to the variance determination threshold, the charging terminal is determined to have a contact fault and the fault type is determined to be a second type of contact abnormality; the second type of contact abnormality includes terminal oxidation and scaling.

3. The method according to claim 2, characterized in that, Obtain the historical reference charging data corresponding to the floor scrubber, including: Extract the historical charging logs of the floor scrubber within a preset sliding time window prior to the current time point from the cloud server; Traverse the historical charging logs, remove invalid logs that contain abnormal alarm indicators or charging interruption indicators, and obtain a candidate log set; Obtain the current firmware version number of the floor scrubber, and remove log records that do not match the firmware version number from the candidate log set; The remaining log records after filtering are used as initial samples to obtain the multiple fault-free charging data samples; the multiple fault-free charging data samples together constitute the historical reference charging data.

4. The method according to claim 3, characterized in that, The remaining log records after filtering are used as initial samples to obtain the multiple fault-free charging data samples, including: Analyze the starting terminal voltage characteristics in the charging terminal voltage data; Extract the corresponding historical starting terminal voltage from each initial sample; calculate the absolute difference between the historical starting terminal voltage and the starting terminal voltage characteristic; When the absolute difference is less than the tolerance threshold, the corresponding initial sample is retained; when the absolute difference is greater than or equal to the tolerance threshold, the corresponding initial sample is discarded. The set of the retained initial samples is used as the plurality of fault-free charging data samples.

5. The method according to claim 4, characterized in that, Calculating the voltage deviation between the static charging data and the historical reference charging data includes: Using the starting sampling point of the static charging data as a time reference, the historical reference charging data is shifted and aligned along the time axis. In the aligned time coordinate system, the first voltage time series data of the static charging data is obtained and the corresponding second voltage time series data in the historical reference charging data is extracted. Traverse each time node in the first voltage time series data, calculate the absolute difference between the first voltage value and the corresponding second voltage value at each time node, and generate a voltage difference set. Each absolute difference in the voltage difference set is assigned a time-increasing weight; the magnitude of the time-increasing weight is positively correlated with the temporal order of the corresponding time node. Based on the time-series increasing weights, all the absolute differences are weighted and summed, and a weighted average is calculated. The weighted average is then used as the voltage deviation.

6. The method according to claim 5, characterized in that, Calculating the voltage fluctuation variance in the voltage data during the self-cleaning operation includes: The voltage data during the self-cleaning operation is filtered to obtain the contact noise data caused by the motor vibration; Calculate the variance of each voltage value in the contact noise data and use the variance as the voltage fluctuation variance.

7. The method according to claim 6, characterized in that, Calculating the variance of each voltage value in the contact noise data and using the variance as the voltage fluctuation variance includes: The floor scrubber's operating mode data is analyzed to obtain the real-time speed data of the motor during the self-cleaning operation. Based on the real-time rotational speed data, the fundamental frequency range of the motor during operation is obtained; The contact noise data is subjected to time-frequency transformation to generate the corresponding frequency domain energy distribution map; Calculate the proportion of resonant energy in the frequency bands that fall within the fundamental frequency range of the frequency domain energy distribution spectrum, relative to the total energy of all frequency bands. The initial variance of each voltage value in the contact noise data is calculated, and the initial variance is weighted and corrected using the resonant energy ratio. The corrected result is used as the voltage fluctuation variance. When the resonant energy ratio is less than a preset purity threshold, it is determined that the contact noise data is affected by water splashing. The value of the voltage fluctuation variance is reduced to reduce the possibility of misjudging water splashing interference as a type I contact anomaly.

8. The method according to claim 7, characterized in that, Before extracting the static charging data from the charging terminal voltage data, the method further includes: Based on the floor scrubber's working mode data, the end time of the self-cleaning period and the start time of the charging period are obtained, and the interval between the end of self-cleaning and the start of static charging is calculated. If the interval is less than the preset water stain drying threshold, mark the current static charging data as water stain interference candidate data and pause the voltage deviation calculation. After the floor scrubber completes the preset drying process, the charging terminal voltage is re-collected and the static charging data is updated.

9. The method according to claim 8, characterized in that, Before determining whether a contact failure has occurred at the charging terminals of the charging dock, the method further includes: Obtain the cumulative charge-discharge cycle count of the floor scrubber from the cloud server; When the cumulative number of charge-discharge cycles is greater than or equal to the cycle number threshold, the corresponding aging correction coefficient is obtained; The deviation determination threshold is generated based on the aging correction coefficient and the initial deviation determination threshold; the larger the cumulative charge-discharge cycle count, the larger the deviation determination threshold.

10. The method according to claim 9, characterized in that, After determining that the fault type is a first-class contact anomaly, the method further includes: When the charging period occurs before the self-cleaning period, based on the floor scrubber's operating mode data, determine the recovery charging period when the floor scrubber enters normal charging or full-charge float charging state after the self-cleaning operation ends; Extract the recovery charging voltage data corresponding to the recovery charging period from the charging terminal voltage data; Calculate the verification deviation between the recovered charging voltage data and the historical reference charging data; When the deviation of the verification is less than the deviation judgment threshold, the first type of contact abnormality is excluded as battery damage.

Images