Intelligent door lock health monitoring and maintenance method and door lock

Abstract

The embodiments of the present disclosure provide a smart door lock health monitoring and maintenance method and a door lock. The smart door lock health monitoring and maintenance method and the door lock can finely locate the stroke interval where the abnormality occurs by obtaining a motor driving current signal, a position detection signal and a displacement detection signal of the smart door lock during execution of a current lock body action, dividing the current lock body action into multiple action stages according to the motor driving current signal, the position detection signal and the displacement detection signal, extracting lock body health features of each action stage, further obtaining a health baseline of the smart door lock after installation, obtaining a target abnormality score of each action stage according to the deviation between the lock body health features and the health baseline, determining an abnormality type and a maintenance reminder of the smart door lock according to the target abnormality score, accurately identifying early degradation signs of the lock body health state according to the action stage, realizing early warning before the fault occurs, and improving the timeliness of the smart door lock monitoring and maintenance.

Description

Technical Field

[0001] This disclosure relates to the field of smart lock technology, and in particular to a smart door lock health monitoring and maintenance method and a door lock. Background Technology

[0002] With the increasing popularity of smart door locks in homes, apartments, hotels, office buildings, and industrial settings, motor-driven lock bodies, clutch-driven lock bodies, automatic lock bodies, and lock bodies with deadbolts or multiple bolt mechanisms are widely used. Existing smart door locks typically use components such as motors, reduction gear sets, transmission links, bolts, square rods, clutches, Hall switches, micro switches, or photoelectric detectors to achieve unlocking, locking, deadbolting, status detection, and abnormal alarms.

[0003] However, most of the above-mentioned detection methods are real-time status checks or alarms triggered after a fault. For example, a fault message is only output when the motor stalls, the current is too high, the bolt is not in position, the switch is abnormal, or locking fails. Such solutions mainly focus on whether the current action is successful, and pay less attention to the health changes of the lock mechanism during long-term use, resulting in delayed monitoring and maintenance. Summary of the Invention

[0004] The main objective of this disclosure is to provide a smart door lock health monitoring and maintenance method and door lock to solve the above-mentioned problems.

[0005] To achieve the above objectives, a first aspect of this disclosure provides a method for health monitoring and maintenance of a smart door lock, comprising: Acquire the motor drive current signal, position detection signal, and displacement detection signal of the smart door lock during the current lock body action process; The current lock body action is divided into multiple action stages based on the motor drive current signal, the position detection signal, and the displacement detection signal, and the lock body health features of each action stage are extracted respectively. Obtain the health baseline of the smart lock after installation, and obtain the target abnormality score for each action stage based on the deviation between the lock body health characteristics and the health baseline. The anomaly type of the smart lock and maintenance reminders are determined based on the target anomaly score.

[0006] In some embodiments, obtaining the target abnormality score for each of the action stages based on the deviation between the lock body health characteristics and the health baseline includes: For each of the aforementioned action stages, calculate the normalized deviation values ​​between the different types of lock body health characteristics and the health baseline; The multiple normalized deviation values ​​are weighted and summed to obtain the resistance anomaly score corresponding to the action stage. The maximum value among the multiple resistance anomaly scores and the normalized deviation value of the last action stage is taken as the instantaneous anomaly score corresponding to the current lock body action. For each of the aforementioned action stages, the average value of the second historical health features of the most recent Q valid second historical lock body actions is calculated. A window deviation value is obtained based on the deviation between the average value of the second historical health features and the health baseline. The maximum value of the multiple window deviation values ​​is taken as the long-term drift score corresponding to the current lock body action, where Q is a positive integer. The immediate anomaly score and the long-term drift score are determined as the target anomaly score.

[0007] In some embodiments, determining the anomaly type and maintenance reminder of the smart lock based on the target anomaly score includes: The anomaly type of the smart lock is determined based on the instantaneous anomaly score, the long-term drift score, and the abnormal action phase. Maintenance reminders are determined based on the immediate anomaly score, the long-term drift score, and the anomaly type.

[0008] In some embodiments, the current lock body action includes locking or unlocking, and the plurality of action stages are divided into at least a first stage, a second stage, a third stage, a fourth stage and a fifth stage. When the current lock body action is the locking action, the first stage, the second stage, the third stage, the fourth stage and the fifth stage are, in sequence, the motor start stage, the idle travel stage, the transmission load stage, the lock tongue engagement stage and the position confirmation stage. When the current lock body action is the unlocking action, the first stage, the second stage, the third stage, the fourth stage and the fifth stage are, in sequence, the motor start stage, the transmission engagement stage, the lock tongue retraction stage, the clutch reset stage and the position confirmation stage. The process of determining the anomaly type of the smart lock based on the immediate anomaly score, the long-term drift score, and the abnormal action phase includes: When the long-term drift scores corresponding to the second stage, the third stage, and the fourth stage are all greater than the first score threshold, and the average current integral and stage consumption time in X consecutive effective lock body actions show an upward trend, and there is no sudden increase in the average current integral and stage consumption time of a single action stage, the abnormal type of the smart door lock is determined to be lubrication degradation trend, where X is a positive integer; When the instantaneous abnormal score corresponding to the fourth stage is greater than the second score threshold, and the stage consumption time of the fourth stage rises more than the first preset ratio relative to the baseline, and the stage average current rises more than the second preset ratio relative to the baseline, while the second stage and the third stage do not reach the same abnormal level, the abnormal type of the smart door lock is determined to be strike box offset, door frame deformation, or increased lock tongue engagement resistance. When the instantaneous anomaly score corresponding to the second stage or the third stage is greater than the third score threshold, and the grating displacement shows zero speed or short-term retreat in the second stage or the third stage, and the current curve shows a peak surge, it is determined that the anomaly type of the smart door lock is an anomaly in the gear, linkage or transmission mechanism. When the displacement of the latch grating has reached more than the third preset proportion of the total stroke pulses of the grating, and the Hall position signal has not stabilized within the preset confirmation time, or the deviation between the Hall position and the grating end point exceeds the preset multiple of the baseline deviation for Y consecutive times, the abnormal type of the smart door lock is determined to be that there is a deviation in the Hall position, the magnet position, or the latch end point position, where Y is a positive integer. If the current in the first stage, second stage, third stage, fourth stage, or fifth stage is not abnormal, but the stage duration increases by more than a fourth preset ratio, and the battery voltage drops by more than a voltage threshold during the operation, the abnormality type of the smart door lock is determined to be an increase in battery internal resistance or a decrease in power supply capacity.

[0009] In some embodiments, determining the maintenance reminder based on the immediate anomaly score, the long-term drift score, and the anomaly type includes: When the long-term drift score is greater than the first score threshold and there are X consecutive effective lock body actions, or when there are Y consecutive slight deviations in the same period, the maintenance reminder is determined to prompt the user to check whether the door is closed properly, whether there are foreign objects on the lock tongue, and whether the strike box is rubbing. When the long-term drift score is greater than the first score threshold and the abnormality type is lubrication degradation trend or increased lock tongue engagement resistance, the maintenance reminder is determined to prompt the user or maintenance personnel to perform cleaning, lubrication, lock box calibration or door frame inspection. When the instantaneous anomaly score is greater than the third score threshold, or when the anomaly type is transmission anomaly, Hall position deviation, frequent retries, or near stall, the maintenance reminder is determined to prompt for maintenance as soon as possible, and the number of automatic retries is limited or the system enters protection mode.

[0010] In some embodiments, the lock body health characteristics include at least one of current-related characteristics, time-related characteristics, position displacement characteristics, consistency characteristics, and environmental compensation characteristics; The current-related characteristics include at least one of the following: starting peak current, steady-state average current, maximum current, root mean square current value, current integral value, location or time of current peak occurrence, segmented current average value, current fluctuation amplitude, and current rise slope before stall protection. The time-related features include at least one of the following: total action duration, time consumed in each action stage, arrival confirmation delay, number of retries, and number of action timeouts. The position displacement features include at least one of the following: starting position, ending position, actual travel length, displacement velocity, displacement acceleration, whether the displacement curve has pauses, backtracking or jitter, and the deviation between the Hall position signal and the grating displacement endpoint. The consistency features include at least one of the following: curve similarity when the same action is performed multiple times, resistance difference during unlocking and locking processes, characteristic deviation under different door opening directions, installation directions, or different lock tongue combinations, and phase relationship between current curve and displacement curve. The environmental compensation characteristics include at least one of the following: current characteristics after temperature compensation, operating time after battery voltage compensation, and correction coefficients under low temperature, high humidity, or low battery conditions.

[0011] In some embodiments, prior to obtaining the health baseline of the smart lock after installation, the method further includes: After the smart lock is installed, the first historical health features corresponding to each action stage of the first historical lock body action of the smart lock for N consecutive times are obtained, where N is a positive integer; If the first historical lock body action has at least one of the following conditions: low battery, door not closed tightly, user interference, motor retry, action timeout, Hall effect not in position, or stall protection, the corresponding first historical health feature will be removed. When the number of remaining first historical health features reaches a preset number, the average value and standard deviation of each first historical health feature in each action stage are calculated respectively, and the health baseline is obtained based on the average value and standard deviation, and the health baseline is saved.

[0012] In some embodiments, the method further includes: In the lock body health baseline, if the standard deviation corresponding to the first historical health feature of the same type is lower than a preset proportion of the corresponding average value, then the lower limit of the standard deviation of the first historical health feature is set to a preset proportion of the corresponding average value. After determining the health baseline, the first historical health features corresponding to each action stage of the M first historical lock body actions are saved. The M first historical lock body actions are used for subsequent trend analysis, where M is a positive integer. To achieve the above objectives, a second aspect of this disclosure provides a door lock, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the smart door lock health monitoring and maintenance method described in the first aspect embodiment.

[0013] To achieve the above objectives, a third aspect of this disclosure provides a storage medium, which is a computer-readable storage medium storing a computer program that, when executed by a processor, implements the smart door lock health monitoring and maintenance method described in the first aspect embodiment.

[0014] The beneficial effects of the embodiments disclosed herein include: By acquiring the motor drive current signal, position detection signal, and displacement detection signal of the smart lock during the current lock body operation, the current lock body operation is divided into multiple operation stages based on these signals. The lock body health characteristics of each operation stage are extracted, enabling precise location of the travel range where abnormalities occur. Furthermore, the health baseline of the smart lock after installation is obtained. The target abnormality score for each operation stage is obtained based on the deviation between the lock body health characteristics and the health baseline. The abnormality type of the smart lock and maintenance reminders are determined based on the target abnormality score. This allows for accurate identification of early signs of lock body health degradation at each operation stage, enabling early warning before malfunctions occur and improving the timeliness of smart lock monitoring and maintenance. Attached Figure Description

[0015] Figure 1 This is an architecture diagram of a monitoring and maintenance system in a smart door lock provided in an embodiment of this disclosure; Figure 2 A flowchart of a smart door lock health monitoring and maintenance method provided in this disclosure embodiment; Figure 3 A flowchart of step S203 provided in an embodiment of this disclosure; Figure 4 A flowchart for establishing a health baseline provided in this disclosure embodiment; Figure 5 This is an overall flowchart provided for an embodiment of the present disclosure. Detailed Implementation

[0016] The accompanying drawings in the embodiments clearly and completely describe the technical solutions in the embodiments of this disclosure. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0017] It is understood that in the specific embodiments of this disclosure, which involve the retrieval of relevant data, when the above embodiments of this disclosure are applied to specific products or technologies, permission or consent from the subject is required, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards.

[0018] In this disclosure, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.

[0019] With the increasing popularity of smart door locks in homes, apartments, hotels, office buildings, and industrial settings, motor-driven lock bodies, clutch-driven lock bodies, automatic lock bodies, and lock bodies with deadbolts or multiple bolt mechanisms are widely used. Existing smart door locks typically use components such as motors, reduction gear sets, transmission links, bolts, square rods, clutches, Hall switches, micro switches, or photoelectric detectors to achieve unlocking, locking, deadbolting, status detection, and abnormal alarms.

[0020] In existing products, smart door locks can generally detect whether the door is closed, whether the bolt is extended or retracted, whether the motor is stalled, whether the battery power is too low, whether the unlocking or locking timeout has occurred, whether there is any tampering, trial unlocking, or abnormal events such as the door not being closed properly.

[0021] However, most of the above-mentioned detection methods are real-time status checks or alarms triggered after a fault. For example, a fault message is only output when the motor stalls, the current is too high, the bolt is not in position, the switch is abnormal, or locking fails. Such solutions mainly focus on whether the current action is successful, and pay less attention to changes in the health status of the lock mechanism during long-term use.

[0022] Over long-term use, smart door locks are susceptible to damage from factors such as dust, moisture, temperature changes, aging grease, mechanical wear, installation misalignment, door frame deformation, changes in the resistance of the deadbolt and strike plate, misalignment of the bolt and strike box, motor aging, and gear wear. These factors typically do not immediately cause locking or unlocking failures, but rather gradually manifest as early signs such as increased motor drive current, longer action time, uneven bolt movement, distorted displacement curve, jittering of the engagement signal, and abnormal resistance in a certain travel range.

[0023] While existing smart locks may have features such as current protection, Hall effect position detection, photoelectric code detection, or timeout protection, this information is typically only used for closed-loop control or fault protection for a single operation. It doesn't further develop a health characteristic model of the lock body, nor does it perform long-term trend analysis on data from multiple operations. Therefore, users or maintenance personnel often only repair or replace the lock after it has become noticeably stuck, frequently fails, or completely malfunctions.

[0024] Reference Figure 1 , Figure 1 This is an architecture diagram of the monitoring and maintenance system in the smart door lock provided in this embodiment of the disclosure, specifically: The motor and drive module includes a motor, drive chip, H-bridge or other power drive circuit, used to drive the latch, clutch, transmission rod or automatic lock body mechanism.

[0025] The current acquisition module is installed in the motor drive circuit to collect the drive current during the lock's operation. Current acquisition can be achieved through a sampling resistor, a Hall current sensor, the current detection port built into the drive chip, or an ADC sampling circuit.

[0026] Hall effect / switch position detection modules include Hall effect sensors, magnets, microswitches or other position detection elements, used to detect whether a latch, transmission component, motor rotor, clutch or critical mechanism has reached a predetermined position.

[0027] The grating / encoded displacement detection module includes a grating ruler, photoelectric encoder, reflective photoelectric sensor, magnetic encoder or motor encoder, used to acquire continuous or segmented displacement information of the latch or transmission mechanism during operation.

[0028] The temperature / battery voltage acquisition module is used to collect ambient temperature or the internal temperature of the door lock, as well as battery voltage signals, to perform environmental compensation correction on characteristics such as current and time, and to eliminate the influence of low temperature and low battery on the detection results.

[0029] The data processing module receives data from the current acquisition module, Hall / switch position detection module, grating / encoded displacement detection module, and temperature / battery voltage acquisition module, processes the data, and outputs it to the health feature model.

[0030] The local / cloud storage module is used to save health baselines, historical action feature data, anomaly records, and model parameters, supporting both local storage and cloud synchronization modes for the door lock.

[0031] The health feature model filters, segments, and extracts features from the collected multi-source signals to generate a health feature vector for each lock action.

[0032] The drift trend analysis module compares the current characteristics with the health baseline, calculates the immediate abnormality score and the long-term drift score, and identifies the slow degradation trend of the lock body's health status.

[0033] The anomaly type identification module is used to classify anomalies into types such as lubrication degradation, increased locking tongue engagement resistance, transmission anomalies, positioning deviation, or power supply reduction based on the action stage and anomaly score.

[0034] The driver protection strategy module is used to implement protective measures such as reducing the number of retries, segmented driving, and pausing forced locking while outputting maintenance reminders.

[0035] The maintenance reminder output module is used to output tiered maintenance reminders to users, property management, or maintenance platforms. The reminder methods include local notification sounds, App push notifications, cloud platform work orders, or background alarms.

[0036] based on Figure 1 The architecture shown is referenced. Figure 2 , Figure 2 This is a flowchart of a smart lock health monitoring and maintenance method provided in an embodiment of the present disclosure. The smart lock health monitoring and maintenance method can be applied to smart locks and includes, but is not limited to, steps S201 to S204.

[0037] Step S201: Obtain the motor drive current signal, position detection signal, and displacement detection signal of the smart door lock during the current lock body operation.

[0038] In this step, when the smart door lock performs unlocking, locking, deadbolting, unlocking, self-testing, or maintenance detection actions, an operation signal acquisition is initiated. The acquired content includes, but is not limited to, motor drive current signals, position detection signals, and displacement detection signals.

[0039] It should be noted that the embodiments of this disclosure, by acquiring multi-source signals during the lock body's operation, provide a comprehensive data foundation for subsequent health feature extraction and anomaly analysis, avoiding the uncertainty caused by relying on only a single signal for judgment in traditional solutions.

[0040] In one embodiment, the smart lock can be pre-configured with data acquisition parameters before leaving the factory or after installation. These parameters may include sampling period, acquisition duration, and trigger conditions. For example, the sampling period can be configured to any value between 1ms and 50ms, preferably 10ms. The acquisition duration can be set according to the typical duration of the lock's movement; for example, the locking action typically lasts 0.5 to 3 seconds, and the acquisition duration can be set to 5 seconds to ensure complete capture of the entire movement. The trigger condition can be set to the moment the motor starts; when the MCU detects the start of the motor drive signal, it automatically triggers data acquisition.

[0041] In one embodiment, the acquisition of the motor drive current signal can be achieved in several ways. For example, a sampling resistor can be set in the motor drive circuit, the voltage drop across the sampling resistor can be amplified by a differential amplifier, and then the analog voltage can be converted into a digital current value by an ADC. Alternatively, a Hall current sensor can be used to achieve non-contact current measurement. Alternatively, the current value can be directly read using the current detection port built into the driver chip. For different types of lock bodies, the amplitude range of the current signal may differ; for example, the current range of a lock body driven by a small motor may be between 0.1A and 2A, while the current range of a lock body driven by a large motor may be between 0.5A and 5A. The current sampling range and amplification factor can be automatically configured according to the lock body type.

[0042] In one embodiment, the position detection signal can be acquired using a Hall sensor. Specifically, a magnet can be installed on the latch, transmission component, or key mechanism, and a Hall sensor can be installed at the corresponding position. When the magnet moves with the latch to the vicinity of the Hall sensor, the output level of the Hall sensor changes. The MCU reads the state of the Hall sensor through GPIO to determine whether the latch has reached the predetermined position. Alternatively, position detection can also be achieved using a microswitch, reed switch, magnetic encoder, photoelectric switch, or capacitive position detection element; this disclosure does not impose specific limitations on these methods.

[0043] In one embodiment, the displacement detection signal can be acquired using a grating ruler or a photoelectric encoder. Specifically, a grating ruler can be installed on the latch or transmission mechanism, with photoelectric transmitters and receivers positioned at corresponding locations. When the grating ruler moves with the latch, the photoelectric receiver outputs a pulse signal, and the MCU counts the pulses through a counting interface to obtain the continuous displacement information of the latch. Alternatively, displacement detection can also be achieved using a magnetic encoder, rotary encoder, motor encoder, or displacement estimation based on the back electromotive force of the motor; this disclosure does not impose specific limitations in this regard.

[0044] In one embodiment, auxiliary information can be collected simultaneously with the aforementioned signals. This auxiliary information may include motor drive voltage or battery voltage V(t), motor PWM duty cycle, drive direction and control phase, start and end times of the action and the duration of each phase, ambient temperature or internal door lock temperature, door status, latch status, deadbolt status, user operation type, whether the action was successful, whether a retry was performed, and whether a timeout or stall occurred. This auxiliary information can be used for subsequent feature compensation and anomaly detection, improving the accuracy of health monitoring.

[0045] For example, in an automatic smart door lock driven by a DC motor, when the user locks the door, the MCU detects the motor start signal via an interrupt and immediately begins data acquisition. The MCU acquires the motor circuit current and battery voltage via an ADC at 10ms intervals, reads the Hall sensor status via GPIO, and reads the grating pulse count via a counting interface. The acquisition process continues until the motor stops rotating and the Hall signal stabilizes for a preset time (e.g., 500ms), after which acquisition ends. The acquired data can be temporarily stored in the MCU's RAM and saved or discarded as needed after processing.

[0046] Step S202: Divide the current lock body action into multiple action stages based on the motor drive current signal, position detection signal and displacement detection signal, and extract the lock body health features of each action stage.

[0047] In this step, based on the collected multi-source signals, a complete lock body action is divided into multiple action stages according to the mechanical motion law, and health features are extracted for each stage.

[0048] It should be noted that, by dividing the lock body action into multiple stages and extracting features in each stage, the travel position where the abnormality occurs can be located in the present embodiment.

[0049] In one embodiment, the division of action stages can be based on the grating displacement ratio. During the self-learning installation stage, the total number of stroke pulses P_total for the lock body movement is recorded. In subsequent actions, the stages are divided according to the ratio of the real-time grating pulse count to P_total. For example, the locking action can be divided into the following five stages: The first stage, S1, is the motor starting stage, corresponding to the latch displacement reaching 0% to 10% of the total stroke. In this stage, the motor starts to rotate from rest, overcoming static friction and the inertia of the transmission system, and the current usually shows a starting peak.

[0050] The second stage, S2, is the idle travel stage, corresponding to the bolt displacement reaching 10% to 40% of the total travel. In this stage, the bolt has begun to move but has not yet made contact with the latch box. The load on the transmission system mainly comes from the frictional resistance of the transmission gears and connecting rods, and the current is relatively stable.

[0051] The third stage, S3, is the transmission load stage, corresponding to the latch displacement reaching 40% to 80% of the total stroke. In this stage, the latch gradually approaches the latch box and may begin to contact the edge of the latch box, the load gradually increases, and the current shows an upward trend.

[0052] The fourth stage, S4, is the latch engagement stage, corresponding to the latch displacement reaching 80% to 100% of the total stroke. In this stage, the latch enters the latch box, overcomes the resistance of the latch box to reach the end point, the load is at its maximum, and the current usually reaches its peak.

[0053] The fifth stage, S5, is the positioning confirmation stage, corresponding to the confirmation time from when the bolt reaches its end point until the Hall signal stabilizes. During this stage, the motor may have stopped or be in a holding state, waiting for the Hall signal to stabilize to confirm the bolt's positioning.

[0054] If the lock body uses different mechanical structures, the actual total stroke pulse count P_total can be recorded during production testing or installation self-learning. The above stage boundaries are automatically calculated based on P_total. For example, if the total locking stroke is 1000 grating pulses, then S1 is 0 to 100 pulses, S2 is 100 to 400 pulses, S3 is 400 to 800 pulses, S4 is 800 to 1000 pulses, and S5 is the arrival confirmation time after the endpoint.

[0055] It should be noted that the above stage division ratios are merely examples. The ratios of the stage boundaries can be adjusted adaptively for different types of lock body mechanical structures. For example, for a top and bottom hook lock body, an additional top and bottom hook linkage stage can be added between S2 and S3; for a multi-latch lock body, the main latch action stage and the oblique latch action stage can be further divided. The stage boundaries can be automatically determined based on the actual mechanical characteristics during production testing or installation self-learning.

[0056] In one embodiment, the unlocking action can be divided into the following five stages according to a similar principle: The first stage, S1, is the motor start-up stage, corresponding to the bolt displacement reaching 0% to 10% of the total stroke.

[0057] The second stage, S2, is the transmission engagement stage, corresponding to the locking tongue displacement reaching 10% to 40% of the total stroke. In this stage, the transmission system engages from rest, overcoming static friction.

[0058] The third stage, S3, is the bolt retraction stage, corresponding to a bolt displacement of 40% to 80% of the total stroke. In this stage, the bolt retracts from the latch box, and the load is relatively large.

[0059] The fourth stage, S4, is the clutch reset stage, corresponding to the locking tongue displacement reaching 80% to 100% of the total stroke. In this stage, the locking tongue fully retracts, and the clutch resets.

[0060] The fifth stage, S5, is the confirmation stage, which corresponds to the confirmation time from when the locking tongue reaches the end point until the Hall signal stabilizes.

[0061] In one embodiment, feature extraction can be performed separately within each stage. The current signal, time signal, and displacement signal within each stage are calculated to generate a health feature vector for that stage.

[0062] For example, in stage S3 of a locking action, the average value of all current sampling points within this stage is extracted as the stage average current; the maximum current value within this stage is extracted as the stage peak current; the cumulative current over time within this stage is extracted as the stage current integral; the duration of this stage from start to finish is extracted as the stage time; and the ratio of displacement change to time within this stage is extracted as the stage average displacement velocity. These features together constitute the lock body health characteristics of stage S3.

[0063] In one embodiment, various data storage strategies can be employed to reduce storage and computational burden. For example, complete current and displacement curves can be saved for detailed analysis. Alternatively, downsampled curves can be saved to reduce data volume. Alternatively, only statistical characteristic values ​​for each stage can be saved without saving the original curves. Alternatively, only abnormal segments and trend statistics can be saved, while normal data only has characteristic values ​​saved. Alternatively, recent data can be saved locally on the door lock, and long-term statistical data can be synchronized to a mobile app or the cloud. A suitable storage strategy can be dynamically selected based on the door lock's storage capacity and computing power.

[0064] Step S203: Obtain the health baseline of the smart lock after installation, and obtain the target abnormality score for each action stage based on the deviation between the lock body health characteristics and the health baseline.

[0065] In this step, a pre-established and stored health baseline is obtained, and the health features extracted from the current action are compared with the health baseline to calculate the target abnormality score reflecting the health status of the lock body.

[0066] It should be noted that this embodiment establishes a health baseline as a reference standard for the normal state, and performs deviation analysis between the current characteristics and the baseline. This effectively identifies subtle changes in the lock body's state and enables early warning. Simultaneously, by calculating anomaly scores, multi-dimensional features are integrated into quantitative indicators, facilitating subsequent anomaly type determination and maintenance decisions.

[0067] In one embodiment, the health baseline is established after the smart lock is installed, using valid action samples from the learning phase. The health baseline includes the normal average value and normal fluctuation range for each action phase and each lock body health characteristic.

[0068] Step S204: Determine the abnormality type of the smart lock and the maintenance reminder based on the target abnormality score.

[0069] In this step, the anomaly type is determined by comprehensively considering the immediate anomaly score, the long-term drift score, and the location of the anomaly during the action phase, and corresponding maintenance reminders are output based on the anomaly type.

[0070] By acquiring the motor drive current signal, position detection signal, and displacement detection signal of the smart lock during the current lock body operation, the current lock body operation is divided into multiple operation stages based on these signals. The lock body health characteristics of each operation stage are extracted, enabling precise location of the travel range where abnormalities occur. Furthermore, the health baseline of the smart lock after installation is obtained. The target abnormality score for each operation stage is obtained based on the deviation between the lock body health characteristics and the health baseline. The abnormality type of the smart lock and maintenance reminders are determined based on the target abnormality score. This allows for accurate identification of early signs of lock body health degradation at each operation stage, enabling early warning before malfunctions occur and improving the timeliness of smart lock monitoring and maintenance.

[0071] In one embodiment, the lock body health characteristics may include, but are not limited to, at least one of current-related characteristics, time-related characteristics, position displacement-related characteristics, consistency characteristics, and environmental compensation characteristics.

[0072] It should be noted that the embodiments of this disclosure can comprehensively characterize the mechanical state of the lock body during the operation process by extracting multiple types and dimensions of health features, providing rich quantitative indicators for subsequent health baseline establishment and anomaly detection, and avoiding the defect that a single feature is difficult to reflect a complex degradation process.

[0073] In one embodiment, current-related features may include, but are not limited to, at least one of the following: Starting peak current: The maximum current that occurs at the instant the motor starts, reflecting the static friction and inertial load during startup. As lubrication deteriorates or mechanical wear occurs, the starting peak current typically shows an upward trend.

[0074] Steady-state average current: The average current after the motor enters the stable operation stage, reflecting the resistance level of the transmission system during normal operation.

[0075] Maximum current: The maximum current that occurs during the entire operation, usually during the stage of the largest load.

[0076] Root mean square current (RMS): The effective value of the current during the entire operation, which comprehensively reflects the average load level during the operation.

[0077] Integral current value: The cumulative current over time during the entire action process, which comprehensively reflects the electrical energy consumed by the action and is directly related to the work done by the machine.

[0078] Location or time of current peak occurrence: The stage or time point in the operation process at which the maximum current occurs reflects the stroke position corresponding to the maximum load.

[0079] Segmented current average: The average current within each operating stage, reflecting the resistance distribution at different stages.

[0080] Current fluctuation amplitude: The range of current fluctuation during operation reflects the smoothness of the transmission system. Increased fluctuation amplitude may indicate gear wear or increased transmission clearance.

[0081] Current rise slope before stall protection: The rate at which the current rises before the motor stall protection is triggered reflects the severity of the sudden resistance.

[0082] In one embodiment, the time-related features may include, but are not limited to, at least one of the following: Total motion duration: The total time from motor start-up to completion of the motion, reflecting the smoothness of the overall movement. As resistance increases, the total motion duration usually tends to increase.

[0083] Time taken for each action phase: The length of time from the start to the end of each phase, reflecting the resistance characteristics of each phase.

[0084] Position confirmation delay: The waiting time for the Hall position signal to stabilize after the locking tongue reaches the mechanical end point.

[0085] Number of retries: The number of times the motor retryes during a single operation, reflecting the degree of resistance encountered during the action.

[0086] Action timeout count: The number of times the action time exceeds the preset timeout threshold.

[0087] In one embodiment, the position displacement feature may include, but is not limited to, at least one of the following: Starting position: The initial position of the locking tongue at the moment the action begins.

[0088] End position: The final position of the locking tongue at the end of the action.

[0089] Actual travel length: The actual displacement from the starting point to the ending point.

[0090] Displacement velocity: The amount of displacement of the locking tongue per unit time.

[0091] Displacement acceleration: the rate of change of displacement velocity.

[0092] The presence of pauses, regressions, or jitters in the displacement curve reflects whether there is jamming, slippage, or backlash in the transmission system.

[0093] Deviation between Hall position signal and grating displacement endpoint: The difference between the position detected by the Hall sensor and the endpoint position measured by the grating encoder reflects the consistency of the position detection system.

[0094] In one embodiment, the consistency feature may include, but is not limited to, at least one of the following: Curve similarity when the same action is performed multiple times: The degree of similarity between the current curve or displacement curves of multiple identical operations reflects the consistency of the lock's actions. Decreased similarity may indicate that the lock's state is unstable.

[0095] Resistance difference during unlocking and locking: Comparing the resistance in the two directions of unlocking and locking for the same lock body, an increased difference may indicate installation deviation or unidirectional wear of the mechanism.

[0096] Characteristic deviations under different door opening directions, installation directions, or different lock tongue combinations: characteristic changes of the lock body under different usage conditions.

[0097] The phase relationship between the current curve and the displacement curve: the correspondence between the peak current and the displacement progress reflects the mechanical position of the load.

[0098] In one embodiment, the environmental compensation feature may include, but is not limited to, at least one of the following: Temperature-compensated current characteristics: The current characteristics corrected for ambient temperature. At low temperatures, the viscosity of the lubricating grease increases, naturally causing the current to rise. Temperature compensation is needed to eliminate the influence of environmental factors.

[0099] Battery voltage compensated response time: The response time is adjusted based on the battery voltage. Under low voltage conditions, the motor speed decreases, naturally increasing the response time. Voltage compensation is needed to eliminate the influence of the power supply status.

[0100] Correction factor under low temperature, high humidity or low power conditions: A factor that corrects for health characteristics under extreme environmental conditions.

[0101] For example, a health feature vector can be represented as: F = [I_peak, I_avg, I_rms, I_integral, T_total, T_stage1, ..., S_total, V_avg, D_hall_grating, R_retry, Temp, ...]; Where I_peak is the peak current, I_avg is the average current, I_rms is the root mean square value of the current, I_integral is the current integral, T_total is the total action time, T_stage1 is the time consumed in the first stage, S_total is the total displacement, V_avg is the average voltage, D_hall_grating is the deviation between the Hall position and the end point of the grating displacement, R_retry is the number of retries, and so on.

[0102] In one embodiment, reference is made to Figure 3 , Figure 3 The flowchart of step S203 provided in the embodiment of this disclosure, the target abnormality score of each action stage obtained according to the deviation between the health characteristics of the lock body and the health baseline may include, but is not limited to, steps S301 to S304.

[0103] Step S301: For each action stage, calculate the normalized deviation value between the health characteristics of different types of lock bodies and the health baseline.

[0104] In this step, for each health characteristic at each action stage, the deviation between the current characteristic value and the normal average value of the corresponding characteristic in the health baseline is calculated, and the deviation is normalized by dividing by the normal fluctuation range to obtain the normalized deviation value.

[0105] For characteristic types where an increase in value indicates greater resistance or abnormal severity, such as current value, time consumption, current integral, and endpoint deviation, the normalized deviation value is calculated as follows: Z(k,j) = max(0, (F(k,j) - mu(k,j)) / sigma(k,j)); Where F(k,j) is the actual value of the j-th feature in the k-th stage of the current action, mu(k,j) is the normal average value of the corresponding feature in the healthy baseline, sigma(k,j) is the normal fluctuation range (standard deviation) of the corresponding feature in the healthy baseline, and Z(k,j) is the normalized deviation value. If the current feature value does not exceed the normal range, the deviation value is 0.

[0106] For characteristic types where a decrease in value indicates increased or abnormally aggravated resistance, such as displacement velocity, the normalized deviation value is calculated as follows: Z(k,j) = max(0, (mu(k,j) - F(k,j)) / sigma(k,j)); For example, suppose the average current mu during phase S4 (locking tongue engagement phase) in a healthy baseline is 500mA, and the standard deviation sigma is 50mA. If the average current during phase S4 in the current action is 580mA, then the normalized deviation is (580-500) / 50 = 1.6. It can be seen that the current characteristic value is 1.6 standard deviations higher than the normal value, indicating a certain degree of abnormality.

[0107] Step S302: Weighted summation of multiple normalized deviation values ​​to obtain the resistance anomaly score corresponding to the action stage. The maximum value among the multiple resistance anomaly scores and the normalized deviation value of the last action stage is taken as the instantaneous anomaly score corresponding to the current lock body action.

[0108] In this step, the normalized deviation values ​​of various features within each action phase are weighted and summed to obtain the resistance anomaly score A(k) for that phase. Then, the maximum value among the resistance anomaly scores of all phases and the normalized deviation value of the last action phase is taken as the instantaneous anomaly score A_now for the current action.

[0109] Taking the extraction of stage average current I_avg(k), stage peak current I_peak(k), stage current integral Q(k), stage time T(k), stage average displacement velocity V_s(k), and deviation D_end between Hall effect point and grating end point as lock body health features for each stage as an example, the normalized deviation value corresponding to the deviation D_end between Hall effect point and grating end point can be used as the normalized deviation value of the last action stage.

[0110] Where k represents the stage number, taking values ​​from S1 to S5. The stage current integral Q(k) can be calculated as follows: Q(k) = sum(I_i * delta_t); Where I_i is the motor current at the i-th sampling point in this stage, and delta_t is the sampling period, which can be 10ms in this embodiment.

[0111] The resistance anomaly score A(k) for each stage is calculated using a weighted method: A(k)=0.30×Z_I_avg(k)+0.2×Z_Q(k)+0.20×Z_T(k)+0.15×Z_I_peak(k)+0.10×Z_V_s(k); Where Z_I_avg(k) is the normalized deviation of the stage average current, Z_Q(k) is the normalized deviation of the stage current integral, Z_T(k) is the normalized deviation of the stage time, Z_I_peak(k) is the normalized deviation of the stage peak current, and Z_V_s(k) is the normalized deviation of the stage average displacement velocity. The above weight values ​​are only examples and can be adaptively adjusted according to different lock body types and scenario requirements in actual applications.

[0112] The instantaneous anomaly score A_now for a single action is the maximum value among the resistance anomaly scores for each stage and the normalized deviation value for the last action stage: A_now=max(A(S1),A(S2),A(S3),A(S4),A(S5),Z_D_end); Where Z_D_end is the normalized deviation value corresponding to the deviation between the Hall effect point and the grating end point.

[0113] Step S303: For each action stage, calculate the average value of the second historical health features of the most recent Q valid second historical lock body actions, obtain the window deviation value based on the deviation between the average value of the second historical health features and the health baseline, and take the maximum value of multiple window deviation values ​​as the long-term drift score corresponding to the current lock body action.

[0114] Where Q is a positive integer, and Q can be 20.

[0115] In this step, to eliminate the influence of incidental factors on anomaly assessment, a long-term drift score A_trend is also calculated. Specifically, the average value of the characteristics of each stage in the most recent Q valid actions is first calculated, and then these average values ​​are compared with the healthy baseline to obtain the window deviation value. The maximum window deviation in each stage is then taken as the long-term drift score.

[0116] For example, regarding the characteristic of average current in phase S4, the average current value in phase S4 during the most recent 20 valid locking actions is obtained, the average value of these 20 values ​​F_avg_S4_I_avg is calculated, and then the normalized deviation between this average value and the normal average value of average current in phase S4 in the healthy baseline is calculated: Z_window_S4_I_avg=max(0,(F_avg_S4_I_avg-mu(S4,I_avg)) / sigma(S4,I_avg)); Similarly, calculate the window deviation values ​​for various features at each stage, then calculate the window resistance anomaly score for each stage, and finally take the maximum value among all the window resistance anomaly scores as the long-term drift score A_trend.

[0117] It should be noted that the long-term drift score reflects the state change trend of the lock body over a longer time scale and has a good indicative effect on continuous and gradual degradation processes. By combining the immediate anomaly score and the long-term drift score, it is possible to effectively distinguish between occasional anomalies and long-term degradation.

[0118] In one embodiment, the value of Q can be adjusted according to the actual application scenario. The larger the Q value, the more obvious the smoothing effect and the stronger the ability to suppress short-term fluctuations.

[0119] In one embodiment, when calculating the long-term drift score, it is also checked whether the number of valid actions has reached Q. If the cumulative number of valid actions is less than Q, the average value is calculated using the currently available valid actions, and the confidence level of the long-term drift score is marked as low for reference in subsequent decision-making.

[0120] Step S304: Determine the immediate anomaly score and the long-term drift score as the target anomaly score.

[0121] In one embodiment, the abnormality type and maintenance reminder of the smart lock are determined based on the target abnormality score. Specifically, the abnormality type of the smart lock can be determined based on the immediate abnormality score, the long-term drift score, and the abnormal action stage, and the maintenance reminder is determined based on the immediate abnormality score, the long-term drift score, and the abnormality type.

[0122] In this step, the anomaly type can be determined according to the distribution pattern of the anomaly score, the stage of the anomaly occurrence, and the trend of feature changes, based on preset judgment rules.

[0123] Specifically, when the long-term drift scores corresponding to the second, third, and fourth stages are all greater than the first score threshold, and the average current integral and stage consumption time in X consecutive effective lock body actions show an upward trend, and there is no sudden increase in the average current integral and stage consumption time of a single action stage, the abnormal type of the smart door lock is determined to be lubrication degradation trend.

[0124] Where X is a positive integer, for example, X can be 10, 15 or 20. The first score threshold can be 1.5.

[0125] Lubrication degradation is one of the most common degradation modes in smart locks. As usage time increases, the grease inside the lock body gradually ages, dries out, or becomes contaminated with dust, leading to an overall increase in frictional resistance in the transmission system. This degradation mode is characterized by a simultaneous increase in resistance across multiple stages, but the rate of increase is relatively slow and uniform, without any abrupt changes at any particular stage.

[0126] For example, suppose a door lock has been used for more than a year. The average current integral and stage time of the S2, S3 and S4 stages of the last 20 locking actions all show a continuous upward trend. The long-term drift scores of the three stages are all greater than 1.5, and there is no situation where any stage is significantly worse than other stages. This is judged as a lubrication degradation trend.

[0127] When the instantaneous abnormal score corresponding to the fourth stage is greater than the second score threshold, and the stage consumption time of the fourth stage rises more than the first preset ratio relative to the baseline, and the stage average current rises more than the second preset ratio relative to the baseline, while the second and third stages do not reach the same abnormal level, the abnormal type of the smart door lock is determined to be strike box offset, door frame deformation, or increased lock tongue engagement resistance.

[0128] The second score threshold can be set to 2.0. The first preset ratio can be set to 20%. The second preset ratio can be set to 25%.

[0129] When the door frame settles, the strike box position shifts, or the fit between the latch and the strike box changes, the resistance during the latch engagement phase increases significantly, while the resistance changes little during the transmission and idle travel phases. This abnormal pattern is characterized by the anomaly being concentrated in phase S4, while phases S2 and S3 remain largely normal.

[0130] For example, suppose that in the recent locking actions of a door lock, the instantaneous abnormality score of the S4 stage reaches 2.3, the S4 stage time increases from 200ms at the baseline to 260ms, the average current of the S4 stage increases from 500mA at the baseline to 640mA, while the abnormality scores of the S2 and S3 stages are both less than 1.0, which is determined to be an increase in the locking resistance of the bolt.

[0131] When the instantaneous anomaly score corresponding to the second or third stage is greater than the third score threshold, and the grating displacement shows zero speed or short-term retreat in the second or third stage, while the current curve shows a sudden peak increase, it is determined that the abnormality type of the smart door lock is that there is an abnormality in the gear, linkage or transmission mechanism.

[0132] The third score threshold can be set to 3.0.

[0133] Abnormalities in the transmission mechanism typically manifest as jamming, slippage, or increased backlash during transmission. When gears are worn, connecting rods are deformed, or foreign objects are present in the transmission path, the transmission speed may drop to zero or briefly regress, and a spike may appear on the current curve.

[0134] For example, suppose that during a locking action, the instantaneous anomaly score of a door lock in stage S3 reaches 3.5. The grating displacement data shows that the velocity is zero for about 50ms in the middle of stage S3. Then the current suddenly jumps from 500mA to 900mA and then falls back, which is determined to be an anomaly in the transmission mechanism.

[0135] When the displacement of the latch grating has reached more than the third preset proportion of the total stroke pulses of the grating, and the Hall position signal has not stabilized within the preset confirmation time, or the deviation between the Hall position and the grating end point exceeds the preset multiple of the baseline deviation for Y consecutive times, the abnormal type of the smart door lock is determined to be a deviation in the Hall position, magnet position, or latch end point position.

[0136] The third preset ratio can be 95%. Y is a positive integer, for example, Y can be 5. The preset multiplier can be 2.

[0137] When the Hall sensor position shifts, the magnet becomes loose, or the end position of the latch changes, a deviation will occur between the triggering timing of the Hall position signal and the end position of the grating displacement.

[0138] For example, suppose the total travel of the grating of a door lock is 1000 pulses. When the bolt reaches the mechanical end point, the count is 1000 pulses. However, the Hall position signal is triggered when the count reaches 980 pulses. In five consecutive actions, the deviation exceeds twice the baseline deviation, which is determined to be a position detection deviation.

[0139] If the current does not show obvious abnormalities in the first, second, third, fourth, or fifth stage, but the stage duration increases by more than the fourth preset ratio, and the battery voltage drops by more than the voltage threshold during the operation, the abnormality type of the smart door lock is determined to be increased battery internal resistance or decreased power supply capacity.

[0140] The fourth preset ratio can be 20%. The voltage threshold can be set to 15% of the battery's open-circuit voltage or a fixed voltage value, such as 0.5V.

[0141] Increased battery internal resistance can cause a significant drop in battery voltage during motor startup or operation. Although the current value may be normal, the reduced motor speed leads to a longer response time. This abnormal mode is characterized by an extended response time but a generally normal current curve.

[0142] For example, suppose a door lock performs a locking action when the battery is low. The no-load battery voltage is 6.0V, and the voltage drops to 4.8V when the motor starts. The total action time increases from the baseline of 800ms to 1000ms, but the current value at each stage is within the normal range, which is judged as a decrease in power supply capacity.

[0143] In some embodiments, when the long-term drift score is greater than the first score threshold and there are X consecutive effective lock body actions, or when there are Y consecutive slight deviations in the same period, the maintenance reminder is determined to be a level one reminder. Specifically, it can prompt the user to check whether the door is closed properly, whether there are foreign objects on the lock tongue, and whether the latch box is rubbing.

[0144] Where X can be 20 and Y can be 5. Slight bias can be defined as a normalized bias value between 1.0 and 2.0.

[0145] A Level 1 alert is a warning level, indicating that a slight but persistent abnormality has been detected, and users are advised to perform a basic check.

[0146] When the long-term drift score exceeds the first score threshold and the abnormality type is a trend of lubrication degradation or an increase in lock tongue engagement resistance, the maintenance reminder is determined to be a level two reminder. Specifically, it can prompt the user or maintenance personnel to perform cleaning, lubrication, lock box calibration, or door frame inspection.

[0147] Level 2 alerts are maintenance alerts, indicating that a continuous degradation trend of the lock body has been confirmed, and users or maintenance personnel are advised to perform proactive maintenance.

[0148] For example, when the anomaly type is increased resistance to latch engagement, a secondary alert is output: "A continuous increase in resistance at the end of the locking mechanism has been detected, which may indicate that the latch box is misaligned or the latch friction is increased. It is recommended to check the latch box position and clean and lubricate the latch."

[0149] When the instantaneous anomaly score exceeds the third score threshold, or the anomaly type is transmission anomaly, Hall position deviation, frequent retries, or near stall, the maintenance reminder is determined to be a level three reminder. Specifically, it can prompt for maintenance as soon as possible and limit the number of automatic retries or enter protection mode.

[0150] Level 3 alert is an emergency level, indicating that the door lock has a high risk of malfunction and immediate maintenance is recommended.

[0151] The maintenance reminder can include the abnormal stage, the abnormal type, and the recommended action. For example, when the S4 bolt engagement stage triggers a secondary reminder, the output maintenance reminder could be: "A continuous increase in resistance at the end of the locking phase has been detected, which may indicate bolt box misalignment or increased bolt friction. It is recommended to check the bolt box position and clean and lubricate the bolt."

[0152] While outputting maintenance reminders, protection strategies can also be implemented. These strategies could include reducing the number of automatic retries to avoid repeated stalling and damage to the motor; reducing motor drive force or using segmented drive during abnormal phases; pausing forced locking when a door is detected not closing properly; providing advance notice of battery replacement when battery is low and resistance is high; adding abnormal door locks to the maintenance queue and generating work orders for centralized management scenarios; or marking abnormal data to prevent it from incorrectly updating the normal health baseline.

[0153] In one embodiment, before acquiring the motor drive current signal, position detection signal, and displacement detection signal of the smart lock during the current lock body action, the smart lock can also be instructed to perform multiple lock body actions in response to a baseline establishment command, and the motor drive current signal, position detection signal, and displacement detection signal of each lock body action can be acquired.

[0154] After the smart lock is installed, it enters the baseline learning phase. Baseline learning can be triggered manually by the installer or automatically after the lock is first powered on. It instructs the lock to perform multiple complete lock body actions, including locking and unlocking, to collect the sample data needed to establish a healthy baseline.

[0155] The baseline learning phase requires the door lock to operate under normal installation conditions and with proper door frame fit. If the door is not closed properly, the battery is too low, or other abnormalities occur during the baseline learning process, the user can be prompted to resolve the issue and restart the baseline learning process.

[0156] Based on the motor drive current signal, position detection signal, and displacement detection signal, each lock body action in multiple lock body actions is divided into multiple action stages. The lock body health characteristics of each action stage of each lock body action are extracted to obtain the first historical health characteristics.

[0157] For each action data collected during the baseline learning phase, the phase division and feature extraction are performed using the same method as in step S202 to obtain the first historical health feature. The first historical health feature will be used as the raw data for establishing the health baseline, and the health baseline will be established based on the first historical health feature.

[0158] In one embodiment, reference is made to Figure 4 , Figure 4 A flowchart for establishing a health baseline provided in this embodiment of the disclosure may include, but is not limited to, steps S401 to S403.

[0159] Step S401: After the smart lock is installed, obtain the first historical health features corresponding to each action stage of the first historical lock body action N consecutive times.

[0160] Where N is a positive integer.

[0161] In this step, the historical health characteristics corresponding to N consecutive successful lock actions are obtained. The value of N can be preset, for example, N can be 20, 30 or 50. Preferably, N is 30, which strikes a balance between ensuring statistical significance and shortening the learning time.

[0162] For example, after the door lock is installed, it automatically starts recording the health characteristics corresponding to 30 consecutive successful locking actions and 30 consecutive successful unlocking actions, which are used to establish locking baseline and unlocking baseline, respectively.

[0163] Step S402: If the first historical lock body action has at least one of the following conditions: low battery, door not closed tightly, user external force interference, motor retry, action timeout, Hall effect not in position, or stall protection, the corresponding first historical health feature will be removed.

[0164] In this step, to ensure that the health baseline reflects the health status of the lock body under normal conditions, quality control is performed on the first historical lock body operation. If any abnormality occurs during an operation, the corresponding health characteristic of that operation will not be included in the baseline calculation.

[0165] For example, check the recorded flags for each action, including battery level flag, door status flag, retry flag, timeout flag, position flag, and stall flag. If any flag indicates an abnormality, the sample is discarded.

[0166] Step S403: When the number of remaining first historical health features reaches the preset number, calculate the average and standard deviation of each first historical health feature in each action stage, obtain the health baseline based on the average and standard deviation, and save the health baseline.

[0167] In this step, once the number of valid samples reaches the preset number, the mean and standard deviation of the values ​​of each feature at each stage are calculated to form a healthy baseline.

[0168] A healthy baseline can be represented as: B={mu(k,j),sigma(k,j),P_total,H_end}; Where mu(k,j) represents the normal average value of the j-th feature in the k-th stage, sigma(k,j) represents the normal fluctuation range of the corresponding feature, P_total represents the total number of grating travel pulses, and H_end represents the normal Hall state in position.

[0169] In one embodiment, after step S403, in the lock body health baseline, if the standard deviation corresponding to the first historical health feature of the same type is lower than a preset proportion of the corresponding average value, then the lower limit of the standard deviation of the first historical health feature is set to the preset proportion of the corresponding average value.

[0170] To prevent excessively small standard deviations of certain features from causing overly sensitive anomaly detection and resulting in false alarms, a lower limit is set for the standard deviation. Specifically, if the standard deviation of a feature is less than 5% of its mean, then the lower limit for that standard deviation is set to 5% of the mean.

[0171] sigma(k,j)=max(real_sigma(k,j),0.05×mu(k,j)); Where real_sigma(k,j) is the actual calculated standard deviation, and 0.05 is the preset ratio, which can be adjusted according to actual needs.

[0172] For example, suppose the measured standard deviation of the average current in stage S3 is 5 mA, while the average current in this stage is 500 mA. The measured standard deviation is only 1% of the average, which is far less than the lower limit of 5%. Setting the lower limit of standard deviation to 25 mA keeps the sensitivity of subsequent anomaly detection within a reasonable range.

[0173] In one embodiment, after step S403, the first historical health features corresponding to each action stage of the M first historical lock body actions can be saved. The M first historical lock body actions are used for subsequent trend analysis, where M is a positive integer.

[0174] In this step, after the health baseline is established, feature data from the most recent M valid movements are saved for subsequent long-term drift score calculation. The value of M can be set according to actual needs. The larger the value of M, the better the smoothing effect of long-term trend analysis, but the slower the response to changes. M can be set to 20.

[0175] For example, after the healthy baseline is established, a sliding window containing the features of the 20 most recent valid locking actions is always maintained. Each time a new valid action is completed, the features of the new action are added to the window, and the oldest feature is removed, keeping the window size at 20 actions. Action data after an anomaly alert is issued is only used for anomaly recording and not for updating the healthy baseline, thus avoiding anomaly data contaminating the normal baseline.

[0176] Reference Figure 5 , Figure 5 The present disclosure provides an overall flowchart, which first collects current, position, displacement and environmental signals during the operation of the lock body to obtain full raw operation data of the lock body's actions.

[0177] The acquired raw signal is filtered and denoised, and the complete unlocking / locking process is divided into different action stages. Health feature vectors are extracted from the processed data to form feature indicators that can characterize the operating status of the lock body.

[0178] Determine if a valid health baseline exists. If no valid health baseline exists, proceed to the step of establishing a factory / installation / self-learning baseline. Construct a health benchmark for the lock body based on the current normal operation data. After completion, save the baseline and the current normal sample, storing the baseline data and the current normal sample.

[0179] If a valid health baseline exists, the extracted health features are compared with the health baseline. At the same time, the instantaneous deviation between the current operation and the baseline, as well as the long-term drift trend of the features over time, are calculated.

[0180] Determine if the maintenance reminder conditions have been met. If not, add the current operation data to the historical database and end the process after data iteration.

[0181] If the maintenance conditions are met, first identify the specific type of abnormality and the stage of action in which the abnormality occurred; then output a graded maintenance reminder according to the severity of the abnormality, and then execute the lock body protection strategy or generate a formal maintenance work order, and finally the process ends.

[0182] The following explanation uses an automatic smart door lock as an example. The lock body contains a DC motor, a reduction gear set, a main bolt, a Hall sensor, and a grating displacement detection structure. The door lock's main control MCU collects the motor circuit current via an ADC, reads the Hall sensor status via GPIO, and reads the grating pulses via a counting interface.

[0183] When the user performs the locking operation, the MCU records the start time of the action and samples the current and battery voltage at a 10ms sampling period, while also recording the Hall state and the grating pulse count. Based on the total stroke pulse count P_total in the installation baseline, the locking process is divided into four segments: S1 start segment, S2 idle stroke segment, S3 transmission load segment, S4 latch engagement segment, and S5 position confirmation segment. For each segment, the average current, peak current, current integral, stage duration, and average displacement speed are calculated.

[0184] After installation, 30 valid locking actions are used as self-learning samples. Samples involving low battery, incomplete door closure, failed actions, motor retries, Hall effect sensor malfunctions, and user-induced interference are removed. The average value mu(k,j) and standard deviation sigma(k,j) of each characteristic at each stage are calculated to generate an installation health baseline. Subsequently, each locking action compares the current stage characteristics with the installation health baseline, and calculates the immediate anomaly score A_now and the long-term drift score A_trend of the last 20 valid actions.

[0185] For example, if the total locking stroke of a door lock after installation is 1000 grating pulses, then S1 is 0 to 100 pulses, S2 is 100 to 400 pulses, S3 is 400 to 800 pulses, S4 is 800 to 1000 pulses, and S5 is the Hall effect confirmation time after the end point. If, in the last 20 valid locking actions, the average current of the latch engagement segment in S4 increases by more than 25% relative to the baseline, the time consumed in the S4 stage increases by more than 20% relative to the baseline, and A_trend is greater than 2.0, then it is determined that the latch engagement resistance has increased, and a maintenance reminder is output: "It is recommended to check the position of the latch box or clean and lubricate the latch."

[0186] If, during a certain operation, A_now of the S3 transmission load section is greater than 3.0, and the grating displacement experiences a brief pause during this stage, while the current curve shows a significant peak increase, but the locking can still be completed in the end, then it is determined that there is an early abnormality in the transmission mechanism. A level three maintenance reminder will be output, and the stage characteristics and abnormal curve of this operation will be recorded for after-sales analysis.

[0187] In this way, the door lock can identify mechanical degradation before the user perceives any obvious fault, thus enabling predictive maintenance.

[0188] This disclosure also provides a door lock, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described smart door lock health monitoring and maintenance method.

[0189] This disclosure also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described smart door lock health monitoring and maintenance method.

[0190] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0191] The embodiments described in this disclosure are for the purpose of more clearly illustrating the technical solutions of this disclosure and do not constitute a limitation on the technical solutions provided by this disclosure. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by this disclosure are also applicable to similar technical problems.

[0192] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this disclosure, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0193] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of the embodiments disclosed herein, depending on actual needs.

[0194] Those skilled in the art will understand that all or some of the steps, apparatuses, or functional modules / units in the methods disclosed above can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0195] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in this disclosure and the foregoing drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0196] It should be understood that in this disclosure, "at least one item" means one or more, and "more than one" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0197] In the several embodiments provided in this disclosure, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0198] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of this disclosure, depending on actual needs.

[0199] Furthermore, the functional units in the various embodiments of this disclosure can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0200] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this disclosure, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this disclosure. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0201] The preferred embodiments of the present disclosure have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present disclosure. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and spirit of the present disclosure shall be within the scope of the claims of the present disclosure.

Claims

1. A method for health monitoring and maintenance of a smart door lock, characterized in that, include: Acquire the motor drive current signal, position detection signal, and displacement detection signal of the smart door lock during the current lock body action process; The current lock body action is divided into multiple action stages based on the motor drive current signal, the position detection signal, and the displacement detection signal, and the lock body health features of each action stage are extracted respectively. Obtain the health baseline of the smart lock after installation, and obtain the target abnormality score for each action stage based on the deviation between the lock body health characteristics and the health baseline. The anomaly type of the smart lock and maintenance reminders are determined based on the target anomaly score.

2. The method according to claim 1, characterized in that, The step of obtaining the target abnormality score for each of the action stages based on the deviation between the lock body health characteristics and the health baseline includes: For each of the aforementioned action stages, calculate the normalized deviation values ​​between the different types of lock body health characteristics and the health baseline; The multiple normalized deviation values ​​are weighted and summed to obtain the resistance anomaly score corresponding to the action stage. The maximum value among the multiple resistance anomaly scores and the normalized deviation value of the last action stage is taken as the instantaneous anomaly score corresponding to the current lock body action. For each of the aforementioned action stages, the average value of the second historical health features of the most recent Q valid second historical lock body actions is calculated. A window deviation value is obtained based on the deviation between the average value of the second historical health features and the health baseline. The maximum value of the multiple window deviation values ​​is taken as the long-term drift score corresponding to the current lock body action, where Q is a positive integer. The immediate anomaly score and the long-term drift score are determined as the target anomaly score.

3. The method according to claim 2, characterized in that, The step of determining the anomaly type and maintenance reminder of the smart lock based on the target anomaly score includes: The anomaly type of the smart lock is determined based on the instantaneous anomaly score, the long-term drift score, and the abnormal action phase. Maintenance reminders are determined based on the immediate anomaly score, the long-term drift score, and the anomaly type.

4. The method according to claim 3, characterized in that, The current lock body action includes locking or unlocking. The multiple action stages are divided into at least a first stage, a second stage, a third stage, a fourth stage, and a fifth stage. When the current lock body action is the locking action, the first stage, the second stage, the third stage, the fourth stage, and the fifth stage are, in sequence, the motor start stage, the idle travel stage, the transmission load stage, the lock tongue engagement stage, and the position confirmation stage. When the current lock body action is the unlocking action, the first stage, the second stage, the third stage, the fourth stage and the fifth stage are, in sequence, the motor start stage, the transmission engagement stage, the lock tongue retraction stage, the clutch reset stage and the position confirmation stage. The process of determining the anomaly type of the smart lock based on the immediate anomaly score, the long-term drift score, and the abnormal action phase includes: When the long-term drift scores corresponding to the second stage, the third stage, and the fourth stage are all greater than the first score threshold, and the average current integral and stage consumption time in X consecutive effective lock body actions show an upward trend, and there is no sudden increase in the average current integral and stage consumption time of a single action stage, the abnormal type of the smart door lock is determined to be lubrication degradation trend, where X is a positive integer; When the instantaneous abnormal score corresponding to the fourth stage is greater than the second score threshold, and the stage consumption time of the fourth stage rises more than the first preset ratio relative to the baseline, and the stage average current rises more than the second preset ratio relative to the baseline, while the second stage and the third stage do not reach the same abnormality level, the abnormality type of the smart door lock is determined to be strike box offset, door frame deformation, or increased lock tongue engagement resistance. When the instantaneous anomaly score corresponding to the second stage or the third stage is greater than the third score threshold, and the grating displacement shows zero speed or short-term retreat in the second stage or the third stage, and the current curve shows a peak surge, it is determined that the anomaly type of the smart door lock is an anomaly in the gear, linkage or transmission mechanism. When the displacement of the latch grating has reached more than the third preset proportion of the total stroke pulses of the grating, and the Hall position signal has not stabilized within the preset confirmation time, or the deviation between the Hall position and the grating end point exceeds the preset multiple of the baseline deviation for Y consecutive times, the abnormal type of the smart door lock is determined to be that there is a deviation in the Hall position, the magnet position, or the latch end point position, where Y is a positive integer. If the current in the first stage, second stage, third stage, fourth stage, or fifth stage is not abnormal, but the stage duration increases by more than a fourth preset ratio, and the battery voltage drops by more than a voltage threshold during the operation, the abnormality type of the smart door lock is determined to be an increase in battery internal resistance or a decrease in power supply capacity.

5. The method according to claim 4, characterized in that, The process of determining maintenance reminders based on the immediate anomaly score, the long-term drift score, and the anomaly type includes: When the long-term drift score is greater than the first score threshold and there are X consecutive effective lock body actions, or when there are Y consecutive slight deviations in the same period, the maintenance reminder is determined to prompt the user to check whether the door is closed properly, whether there are foreign objects on the lock tongue, and whether the strike box is rubbing. When the long-term drift score is greater than the first score threshold and the abnormality type is lubrication degradation trend or increased lock tongue engagement resistance, the maintenance reminder is determined to prompt the user or maintenance personnel to perform cleaning, lubrication, lock box calibration or door frame inspection. When the instantaneous anomaly score is greater than the third score threshold, or when the anomaly type is transmission anomaly, Hall position deviation, frequent retries, or near stall, the maintenance reminder is determined to prompt for maintenance as soon as possible, and the number of automatic retries is limited or the system enters protection mode.

6. The method according to claim 1, characterized in that, The lock body health characteristics include at least one of the following: current characteristics, time characteristics, position displacement characteristics, consistency characteristics, and environmental compensation characteristics; The current-related characteristics include at least one of the following: starting peak current, steady-state average current, maximum current, root mean square current value, current integral value, location or time of current peak occurrence, segmented current average value, current fluctuation amplitude, and current rise slope before stall protection. The time-related features include at least one of the following: total action duration, time consumed in each action stage, arrival confirmation delay, number of retries, and number of action timeouts. The position displacement features include at least one of the following: starting position, ending position, actual travel length, displacement velocity, displacement acceleration, whether the displacement curve has pauses, backtracking or jitter, and the deviation between the Hall position signal and the grating displacement endpoint. The consistency features include at least one of the following: curve similarity when the same action is performed multiple times, resistance difference during unlocking and locking processes, characteristic deviation under different door opening directions, installation directions, or different lock tongue combinations, and phase relationship between current curve and displacement curve. The environmental compensation characteristics include at least one of the following: current characteristics after temperature compensation, operating time after battery voltage compensation, and correction coefficients under low temperature, high humidity, or low battery conditions.

7. The method according to claim 1, characterized in that, Before obtaining the health baseline of the smart lock after installation, the method further includes: After the smart lock is installed, the first historical health features corresponding to each action stage of the first historical lock body action of the smart lock for N consecutive times are obtained, where N is a positive integer; If the first historical lock body action has at least one of the following conditions: low battery, door not closed tightly, user interference, motor retry, action timeout, Hall effect not in position, or stall protection, the corresponding first historical health feature will be removed. When the number of remaining first historical health features reaches a preset number, the average value and standard deviation of each first historical health feature in each action stage are calculated respectively, and the health baseline is obtained based on the average value and standard deviation, and the health baseline is saved.

8. The method according to claim 7, characterized in that, The method further includes: In the lock body health baseline, if the standard deviation corresponding to the first historical health feature of the same type is lower than a preset proportion of the corresponding average value, then the lower limit of the standard deviation of the first historical health feature is set to a preset proportion of the corresponding average value. After determining the health baseline, the first historical health features corresponding to each action stage of the first historical lock body action M times are saved. The M first historical lock body actions are used for subsequent trend analysis, where M is a positive integer.

9. A door lock, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the smart door lock health monitoring and maintenance method according to any one of claims 1 to 8.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the smart door lock health monitoring and maintenance method according to any one of claims 1 to 8.

Images