Remote security management method for low-power-consumption intelligent lock cluster

Abstract

The invention discloses a remote security management method for a low-power-consumption smart lock cluster, and relates to the field of smart home. Classifying the intelligent lock cluster into at least one local cluster; carrying out emergency treatment on the intelligent lock with the fault; obtaining a fault triggering symptom and a fault triggering use frequency; taking the intelligent lock with the fault triggering symptom as a target intelligent lock; the residual use frequency of the target intelligent lock is obtained through prediction, the use habit of the target intelligent lock is formed, the fault time of the target intelligent lock is predicted based on the use habit, and the target intelligent lock is maintained within the fault time of the target intelligent lock. According to the method, the important coefficient is obtained, emergency processing is carried out on the intelligent lock with the fault, and the residual use frequency and the fault time of the target intelligent lock are obtained through prediction, so that fault processing and fault prediction are more fit with the actual condition of the intelligent lock, and the safety problem caused by the fault is avoided as much as possible.

Description

Technical Field

[0001] The present invention relates to the field of smart homes, and specifically to a remote security management method for a low-power smart lock cluster. Background Art

[0002] A smart door lock management system is a system that realizes remote control and management of door locks through the combination of smart hardware and software. The smart door lock management system can monitor the door lock status in real time, record the usage of the door lock, and improve the security and convenience of the door lock. The smart door lock management system is applicable to various scenarios such as homes, offices, and hotels, meeting the needs of different users. The management of smart locks is mainly reflected in the handling and prediction of faults when the smart locks may malfunction. However, in the prior art, when dealing with and predicting faults, the usage habits and application scenarios of smart locks are lacking, resulting in insufficient pertinence in handling and prediction, and thus the security needs to be improved. Summary of the Invention

[0003] To solve the above technical problems, a remote security management method for a low-power smart lock cluster is provided, and this technical solution solves the problems raised in the above background art.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0005] A remote security management method for a low-power smart lock cluster includes:

[0006] Obtain a smart lock cluster, which is composed of at least one smart lock, and obtain the application scenarios of the smart locks;

[0007] Evaluate the importance level of the application scenarios of the smart locks to obtain an importance coefficient;

[0008] Based on the importance coefficient, classify the smart lock cluster into at least one local cluster;

[0009] Set a multi-modal unlocking method, remotely monitor the smart locks in the local cluster, and during the monitoring process, obtain the fault characteristics of the malfunctioning smart locks and summarize them into a fault characteristic database;

[0010] Based on the importance coefficient, perform emergency processing on the malfunctioning smart locks;

[0011] Identify the symptoms of the fault characteristics of the smart locks to obtain the fault trigger symptoms and the fault trigger usage frequency;

[0012] Check all the fault trigger symptoms in the fault characteristic database among the smart locks in the smart lock cluster, and use the smart locks with the fault trigger symptoms as target smart locks;

[0013] Predict the remaining usage frequency of the target intelligent lock, form the usage habit of the target intelligent lock, and based on the usage habit, predict the failure time of the target intelligent lock. During the failure time of the target intelligent lock, repair the target intelligent lock.

[0014] Preferably, the steps for evaluating the importance degree of the application scenario of the intelligent lock to obtain the importance coefficient include the following:

[0015] Obtain the protected space of the application scenario of the intelligent lock. The protected space is the space entered through the door lock where the intelligent lock is installed;

[0016] Identify items within the protected space to obtain the value cost of the items;

[0017] Obtain the minimum value of the re-acquisition time of the items as the additional time;

[0018] Obtain the necessary usage time of the items. The necessary usage time is the time when the items are needed again after being lost or damaged;

[0019] Set an additional coefficient. When the additional time does not exceed the necessary usage time, the additional coefficient is equal to 1. Otherwise, the additional coefficient is equal to the sum of the absolute value of the difference between the additional time and the necessary usage time and 1;

[0020] Multiply the value cost of the items by the additional coefficient to obtain the additional cost, and accumulate the additional costs of the items within the protected space to obtain the importance coefficient.

[0021] Preferably, the steps for classifying the intelligent lock cluster into at least one local cluster based on the importance coefficient include the following:

[0022] Take the interval length of the value range of the importance coefficient as the feature distance;

[0023] Evenly divide the value range of the importance coefficient into at least one feature interval, and the length of the feature interval is equal to the feature distance;

[0024] Merge the intelligent locks whose importance coefficients belong to the local interval into the same feature cluster, calculate the difference degree of the application scenarios of any two intelligent locks in the feature cluster. When the difference degree of the application scenarios of any two intelligent locks in the feature cluster is less than the preset value, take the feature distance as the segmentation distance. Otherwise, update the feature distance to half of the original value and repeat the previous step, where the preset value is the allowable error set based on experience;

[0025] Evenly divide the value range of the importance coefficient into at least one local interval, and the length of the local interval is equal to the segmentation distance;

[0026] Merge the intelligent locks whose importance coefficients belong to the local interval into the same local cluster.

[0027] Preferably, the difference degree between the application scenarios of any two intelligent locks in the calculation feature cluster includes the following steps:

[0028] Arbitrarily select two intelligent locks in the feature cluster, and respectively use them as the first intelligent lock and the second intelligent lock;

[0029] Arrange the additional costs of the items in the protection space of the first intelligent lock from largest to smallest to form a first vector;

[0030] Arrange the additional costs of the items in the protection space of the second intelligent lock from largest to smallest to form a second vector;

[0031] When the number of components in the first vector and the second vector is different, then supplement components at the ends of the first vector and the second vector, and the supplemented components are all 0, so that the number of components in the first vector and the second vector after component supplementation is the same;

[0032] Subtract the second vector from the first vector to obtain a difference vector;

[0033] Divide the modulus length of the difference vector by the sum of the modulus lengths of the first vector and the second vector to obtain the difference degree.

[0034] Preferably, the steps for setting multi-modal unlocking methods and remotely monitoring the intelligent locks in the local cluster include the following:

[0035] The multi-modal unlocking methods are fingerprint recognition, face recognition, vein recognition, and voiceprint recognition. The user selects one of them as the preferred recognition method according to the demand;

[0036] Based on the Internet of Things technology, the intelligent lock is connected to a smart phone or other intelligent devices. The user can remotely monitor the door lock status and receive real-time alerts through the smart phone or other intelligent devices.

[0037] Preferably, the steps for performing emergency processing on the intelligent lock with a fault based on the importance coefficient include the following:

[0038] Identify the fault. When it is a fault that cannot be remotely processed, contact the maintenance personnel for processing;

[0039] When it is a fault that can be remotely processed, take the minimum value of the importance coefficient as the target coefficient, divide the average value of all the importance coefficients of the local cluster where the intelligent lock is located by the target coefficient and round up to obtain the number of passed verification questions. The verification questions are the characteristics of the items in the protection space of the application scenario of the intelligent lock;

[0040] When the number of verification questions answered correctly by the user reaches the number of passed verification questions, the user passes the verification; otherwise, the user fails the verification;

[0041] When the user passes the verification, the lock is unlocked remotely; otherwise, contact the maintenance personnel for handling.

[0042] Preferably, the steps of identifying the fault symptoms of the intelligent lock to obtain the fault trigger symptoms and the fault trigger usage frequency include the following:

[0043] Obtain the historical data of the parameter indicators of the intelligent lock with fault characteristics, where the parameter indicators are the unlocking response speed, the possibility of the lock being cracked, the battery life, the system compatibility, and the implementation speed of the remote management function;

[0044] Fit the historical data of the parameter indicators to obtain a parameter fitting function, where time is the independent variable and the historical data of the parameter indicators is the dependent variable;

[0045] Calculate the first derivative of the parameter fitting function. Based on the first derivative, divide the value range of the independent variable of the parameter fitting function into at least one continuous interval, satisfying that the first derivative is always greater than 0 or always less than 0 or always equal to 0 within the continuous interval, and the first derivatives in adjacent continuous intervals are different from each other;

[0046] Sort the continuous intervals in the order of time occurrence. When the length of the latter of two adjacent continuous intervals exceeds twice the length of the former, then take the latter of the two adjacent continuous intervals as the symptom interval;

[0047] Substitute the midpoint of the symptom interval into the parameter fitting function to obtain a critical value. The fault trigger symptom is that the value of the parameter indicator exceeds the critical value;

[0048] Count the number of times the intelligent lock with the fault trigger symptom first shows the fault characteristics as the fault trigger usage frequency.

[0049] Preferably, the steps of predicting the remaining usage frequency of the target intelligent lock include the following:

[0050] Obtain at least one fault trigger symptom of the target intelligent lock as the target fault trigger symptom;

[0051] Take the reciprocal of the fault trigger usage frequency of the target fault trigger symptom as the fault trigger probability, and subtract the fault trigger probability from 1 to get the non-fault trigger probability;

[0052] Multiply the non-fault trigger probabilities of the target fault trigger symptom together to get the comprehensive non-fault trigger probability, and subtract the comprehensive non-fault trigger probability from 1 to get the comprehensive trigger probability;

[0053] Round up the reciprocal of the comprehensive trigger probability to get the remaining usage frequency of the target intelligent lock.

[0054] Preferably, the steps of forming the usage habit of the target intelligent lock include the following:

[0055] Divide a year into at least one time period based on the hours;

[0056] During the time period, the number of times the target smart lock is used is counted as the target number, and the average of the target number is taken to obtain the average number of times used;

[0057] When the difference between the target number of times and the average number of times accounts for a ratio of the average number of times of use that is less than the allowable error, the time period is used as the usage period of the target smart lock. Otherwise, the time period is updated to twice the original time period and the previous step is repeated.

[0058] Count the number of times the target smart lock is used during its usage period as the characteristic number;

[0059] The feature count is divided by the usage cycle to obtain the usage frequency of the target smart lock.

[0060] Preferably, the predicting the failure time of the target smart lock based on usage habits comprises the following steps:

[0061] The remaining usage frequency is divided by the usage frequency of the target smart lock to obtain the failure time of the target smart lock.

[0062] Compared with the prior art, the present invention has the following beneficial effects:

[0063] By obtaining the importance coefficient, performing emergency treatment on faulty smart locks, and predicting the remaining usage frequency and failure time of the target smart lock, we can perform smart lock fault handling and fault prediction in combination with the application scenarios and usage habits of the smart locks, so that the fault handling and fault prediction are more in line with the actual situation of the smart locks, and thus use higher-level processing solutions for smart locks that protect more important items. At the same time, we can more accurately predict the time when a fault will occur, and then perform maintenance in advance to avoid safety problems caused by faults as much as possible. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 A schematic diagram of a remote security management method for a low-power smart lock cluster of the present invention;

[0065] Figure 2 A schematic diagram of a process for evaluating the importance of application scenarios of smart locks of the present invention to obtain an important coefficient;

[0066] Figure 3 A schematic diagram of a process of classifying a smart lock cluster into at least one local cluster based on an important coefficient of the present invention;

[0067] Figure 4 A schematic diagram of a process for calculating the difference between application scenarios of any two smart locks in a feature cluster of the present invention;

[0068] Figure 5 This is a schematic flow chart for setting up a multi-modal unlocking method in the present invention and remotely monitoring intelligent locks in a local cluster.

[0069] Figure 6 This is a schematic flow chart for emergently handling faulty intelligent locks based on importance coefficients in the present invention.

[0070] Figure 7 This is a schematic flow chart for identifying symptoms of the fault characteristics of intelligent locks in the present invention to obtain fault-triggering symptoms and fault-triggering usage frequencies.

[0071] Figure 8 This is a schematic flow chart for predicting the remaining usage frequency of a target intelligent lock in the present invention.

[0072] Figure 9 This is a schematic flow chart for forming the usage habits of a target intelligent lock in the present invention. Detailed implementation manners

[0073] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments in the following description are only examples, and those skilled in the art can think of other obvious variations.

[0074] Referring to Figure 1 As shown, a remote security management method for a low-power intelligent lock cluster includes:

[0075] Obtain an intelligent lock cluster, which consists of at least one intelligent lock, and obtain the application scenarios of the intelligent locks.

[0076] Evaluate the importance level of the application scenarios of the intelligent locks to obtain importance coefficients.

[0077] Based on the importance coefficients, classify the intelligent lock cluster into at least one local cluster.

[0078] Set up a multi-modal unlocking method, remotely monitor the intelligent locks in the local cluster, and during the monitoring process, obtain the fault characteristics of the faulty intelligent locks and summarize them into a fault characteristic database.

[0079] Based on the importance coefficients, emergently handle the faulty intelligent locks.

[0080] Identify symptoms of the fault characteristics of the intelligent locks to obtain fault-triggering symptoms and fault-triggering usage frequencies.

[0081] Check all the fault-triggering symptoms in the fault characteristic database among the intelligent locks in the intelligent lock cluster, and regard the intelligent locks with fault-triggering symptoms as target intelligent locks.

[0082] Predict the remaining usage frequency of the target intelligent lock, form the usage habit of the target intelligent lock, and based on the usage habit, predict the failure time of the target intelligent lock. During the failure time of the target intelligent lock, repair the target intelligent lock.

[0083] In this solution, due to the different application scenarios of intelligent locks, their requirements for security are also different. For users who can directly unlock through the intelligent lock, their security does not need to be overly managed. The main problem is that when the intelligent lock fails, whether the user can enter the house remotely. Since it is uncertain whether the user is a security personnel, corresponding algorithms are set up in the follow-up for verification. At the same time, to avoid this security problem caused by failures, the failures are also identified in advance and then repaired in advance, and corresponding algorithms are also set up in the follow-up to solve it.

[0084] Refer to Figure 2 As shown, to evaluate the importance degree of the application scenario of the intelligent lock and obtain the importance coefficient, the following steps are included:

[0085] Obtain the protected space of the application scenario of the intelligent lock, where the protected space is the space entered through the door lock where the intelligent lock is installed;

[0086] Identify the items in the protected space to obtain the value cost of the items;

[0087] Obtain the minimum value of the re-acquisition time of the items as the additional time;

[0088] Obtain the necessary usage time of the items, where the necessary usage time is the time when the items are needed again after being lost or damaged;

[0089] Set an additional coefficient. When the additional time does not exceed the necessary usage time, the additional coefficient is equal to 1. Otherwise, the additional coefficient is equal to the sum of the absolute value of the difference between the additional time and the necessary usage time and 1;

[0090] Multiply the value cost of the items by the additional coefficient to obtain the additional cost, and accumulate the additional costs of the items in the protected space to obtain the importance coefficient.

[0091] The significance of the necessary usage time is reflected in scenarios such as hospitals. Because the drugs or instruments stored here are necessary and may be used at a certain time in the future. Therefore, once damaged or stolen, the loss is not only the cost itself but also a series of other problems. Therefore, this situation must be distinguished, so the necessary usage time is set and the additional coefficient is formed accordingly.

[0092] Refer to Figure 3 As shown, based on the importance coefficient, classifying the intelligent lock cluster into at least one local cluster includes the following steps:

[0093] Take the interval length of the value range of the important coefficient as the characteristic distance;

[0094] Evenly divide the value range of the important coefficient into at least one characteristic interval, and the length of the characteristic interval is equal to the characteristic distance;

[0095] Merge the intelligent locks whose important coefficients belong to the local interval into the same characteristic cluster, calculate the difference degree of the application scenarios of any two intelligent locks in the characteristic cluster. When the difference degree of the application scenarios of any two intelligent locks in the characteristic cluster is less than the preset value, then take the characteristic distance as the segmentation distance. Otherwise, update the characteristic distance to half of the original value and repeat the previous step, where the preset value is the allowable error set based on experience;

[0096] Evenly divide the value range of the important coefficient into at least one local interval, and the length of the local interval is equal to the segmentation distance;

[0097] Merge the intelligent locks whose important coefficients belong to the local interval into the same local cluster.

[0098] The local cluster is to summarize the intelligent locks with similar application scenarios. When making importance judgments later, since there may be certain errors in the important coefficients of individual intelligent locks, in order to control the errors, use the average value of the important coefficients of all intelligent locks in the local cluster for substitution, then the overall error can be controlled within a relatively small range, which can improve the accuracy of subsequent overall management.

[0099] Refer to Figure 4 As shown, calculating the difference degree of the application scenarios of any two intelligent locks in the characteristic cluster includes the following steps:

[0100] Randomly select two intelligent locks in the characteristic cluster and use them as the first intelligent lock and the second intelligent lock respectively;

[0101] Arrange the additional costs of the items in the protection space of the first intelligent lock from largest to smallest to form a first vector;

[0102] Arrange the additional costs of the items in the protection space of the second intelligent lock from largest to smallest to form a second vector;

[0103] When the number of components in the first vector and the second vector is different, then supplement components at the ends of the first vector and the second vector, and the supplemented components are all 0. After supplementing the components, the number of components in the first vector and the second vector is the same;

[0104] Subtract the second vector from the first vector to get a difference vector;

[0105] Divide the modulus length of the difference vector by the sum of the modulus lengths of the first vector and the second vector to get the difference degree.

[0106] When making comparisons, vectors are used for processing. When subtracting vectors, it is necessary to ensure that their dimensions are the same, that is, the number of components is the same. However, since the first vector and the second vector are respectively composed of the additional costs of the items in the protection space of the first smart lock and the additional costs of the items in the protection space of the second smart lock, the number of their additional costs may be different. Therefore, it is necessary to supplement the first vector and the second vector so that the dimensions of the supplemented first vector and second vector are the same. At the same time, it is necessary to ensure that the supplement has no impact on the result. Therefore, the supplemented component is 0.

[0107] Refer to Figure 5 As shown, setting up a multi-modal unlocking method and remotely monitoring the smart locks in a local cluster includes the following steps:

[0108] The multi-modal unlocking methods are fingerprint recognition, face recognition, vein recognition, and voiceprint recognition. Users can select one of them as the preferred recognition method according to their needs;

[0109] Based on Internet of Things technology, the smart lock is connected to a smartphone or other smart devices. Users can remotely monitor the door lock status and receive real-time alerts through the smartphone or other smart devices.

[0110] Refer to Figure 6 As shown, based on the importance coefficient, the emergency treatment for a malfunctioning smart lock includes the following steps:

[0111] Identify the fault. When it is a fault that cannot be remotely processed, contact the maintenance personnel for handling;

[0112] When it is a fault that can be remotely processed, take the minimum value of the importance coefficient as the target coefficient. Divide the average value of all the importance coefficients of the local cluster where the smart lock is located by the target coefficient and round down to obtain the number of passed verification questions. The verification questions are the characteristics of the items in the protection space of the application scenario of the smart lock;

[0113] When the number of verification questions answered correctly by the user reaches the number of passed verification questions, the user passes the verification; otherwise, the user fails the verification;

[0114] When the user passes the verification, remotely unlock the door; otherwise, contact the maintenance personnel for handling.

[0115] For different smart locks, since the importance of the items they protect is different, when the user cannot unlock the door, it is necessary to conduct different levels of verification according to the different importance of the items protected by the smart lock to ensure that the user is a reliable person. Furthermore, when the user enters the indoor area, their behavior is controllable, reducing the possible risks.

[0116] Refer to Figure 7As shown in the figure, the steps for identifying the symptoms of the failure characteristics of the smart lock and obtaining the failure trigger symptoms and the failure trigger usage frequency are as follows:

[0117] Obtain the historical data of the parameter indicators of the smart lock with failure characteristics, where the parameter indicators are the unlocking response speed, the possibility of the lock being cracked, the battery life, the system compatibility, and the implementation speed of the remote management function;

[0118] Fit the historical data of the parameter indicators to obtain a parameter fitting function, where time is the independent variable and the historical data of the parameter indicators is the dependent variable;

[0119] Calculate the first derivative of the parameter fitting function. Based on the first derivative, divide the value range of the independent variable of the parameter fitting function into at least one continuous interval, satisfying that the first derivative is always greater than 0 or always less than 0 or always equal to 0 within the continuous interval, and the first derivatives in adjacent continuous intervals are different from each other;

[0120] Sort the continuous intervals in the order of time occurrence. When the length of the latter of two adjacent continuous intervals exceeds twice the length of the former, then take the latter of the two adjacent continuous intervals as the symptom interval;

[0121] Substitute the midpoint of the symptom interval into the parameter fitting function to obtain a critical value. The failure trigger symptom is that the value of the parameter indicator exceeds the critical value;

[0122] Count the number of times the smart lock with the failure trigger symptom first shows the failure characteristics as the failure trigger usage frequency.

[0123] When there is no failure, its parameters fluctuate regularly, that is, they fluctuate up and down within a range, and the length of the interval where its first derivative is continuously greater than 0 or less than 0 or equal to 0 also fluctuates within a range, and there will be no sudden significant increase. However, when a failure is about to occur, its parameters will inevitably show a different movement from the regular fluctuation, that is, they jump out of the fluctuation range and start to continuously decline or increase. Therefore, the length of the interval where its first derivative is continuously greater than 0 or less than 0 will be much longer than the previous interval. Therefore, the failure trigger symptom is obtained based on this principle.

[0124] Refer to Figure 8 As shown in the figure, the steps for predicting the remaining usage frequency of the target smart lock are as follows:

[0125] Obtain at least one failure trigger symptom of the target smart lock as the target failure trigger symptom;

[0126] Take the reciprocal of the failure trigger usage frequency of the target failure trigger symptom as the failure trigger probability, and 1 minus the failure trigger probability to obtain the failure non-trigger probability;

[0127] Multiply the probabilities of non-triggering for each potential fault trigger symptom of the target fault to obtain the comprehensive non-triggering probability. Subtract the comprehensive non-triggering probability from 1 to get the comprehensive triggering probability.

[0128] Take the integer part of the reciprocal of the comprehensive triggering probability to obtain the remaining usage frequency of the target smart lock.

[0129] Since there may be more than one potential fault trigger symptom for the target smart lock, when there are multiple potential fault trigger symptoms, its remaining usage frequency will necessarily decrease accordingly. However, it is not reasonable to estimate the remaining usage frequency by directly adding the triggering frequencies of the faults. Therefore, calculate the triggering probability. When the triggering frequency of the fault is larger, according to the calculation process of the triggering probability, the triggering probability is smaller, which is consistent with common sense. Therefore, in the following, it is only necessary to synthesize all the triggering probabilities of the faults. To this end, first calculate the non-triggering probabilities caused by all the triggering probabilities of the faults, which must be that each potential fault trigger symptom does not trigger, that is, multiply the probabilities of non-triggering of the faults. Then, subtract the comprehensive non-triggering probability from 1 to get the comprehensive triggering probability. According to the acquisition of the triggering probability, the remaining usage frequency of the target smart lock can be deduced from the comprehensive triggering probability. However, since the reciprocal of the comprehensive triggering probability may not be an integer, rounding is required.

[0130] Refer to Figure 9 As shown, the steps to form the usage habit of the target smart lock are as follows:

[0131] Divide the time of one year into at least one time period by hour.

[0132] During the time period, count the number of times the target smart lock is used as the target number, and take the average value of the target number to get the average usage number.

[0133] When the ratio of the difference between the target number and the average usage number to the average usage number is less than the allowable error, then take the time period as the usage cycle of the target smart lock; otherwise, update the time period to twice the original and repeat the previous step.

[0134] Count the number of times the target smart lock is used during the usage cycle as the characteristic number.

[0135] Divide the characteristic number by the usage cycle to get the usage frequency of the target smart lock.

[0136] When obtaining the usage habit of the target smart lock, it is necessary to obtain its usage cycle, that is, during the usage cycle, the number of times the smart lock is used is almost the same. Thus, the calculated usage frequency of the target smart lock is more credible. Therefore, obtain the usage cycle by gradually doubling the time period.

[0137] Based on usage habits, predicting the failure time of the target smart lock includes the following steps:

[0138] The remaining usage frequency is divided by the usage frequency of the target smart lock to obtain the failure time of the target smart lock.

[0139] Furthermore, the present solution also proposes a storage medium on which a computer-readable program is stored, and when the computer-readable program is called, the above-mentioned remote security management method of the low-power smart lock cluster is executed.

[0140] It is understandable that the storage medium may be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid state drive (SSD).

[0141] To sum up, the advantages of the present invention are: by obtaining important coefficients, performing emergency processing on faulty smart locks, and predicting the remaining usage frequency and fault time of the target smart lock, it is possible to combine the application scenarios and usage habits of the smart locks to perform fault processing and fault prediction of the smart locks, so that the fault processing and fault prediction are more in line with the actual situation of the smart locks, so that higher-level processing solutions can be used for smart locks that protect more important items. At the same time, it can more accurately predict the time when a fault will occur, and then perform maintenance in advance to avoid safety problems caused by the fault as much as possible.

[0142] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A remote security management method for a low-power smart lock cluster, characterized in that: include: Obtain a smart lock cluster, which consists of at least one smart lock, and obtain an application scenario of the smart lock; Evaluate the importance of the application scenarios of smart locks and obtain the importance coefficient; Based on the importance coefficient, the smart lock cluster is classified into at least one local cluster; Set up a multi-modal unlocking method to remotely monitor the smart locks in the local cluster. During the monitoring process, obtain the fault characteristics of the faulty smart locks and summarize them in the fault characteristic database; Based on the importance coefficient, emergency treatment is carried out for faulty smart locks; Identify the fault characteristics of the smart lock and obtain the fault triggering signs and the frequency of fault triggering. All fault triggering signs in the fault feature database are checked in the smart locks of the smart lock cluster, and the smart locks with fault triggering signs are used as target smart locks; The remaining usage frequency of the target smart lock is predicted to form usage habits of the target smart lock. Based on the usage habits, the failure time of the target smart lock is predicted, and the target smart lock is repaired within the failure time of the target smart lock.

2. According to claim 1, a remote security management method for a low-power smart lock cluster is characterized in that: The evaluation of the importance of the application scenario of the smart lock to obtain the importance coefficient includes the following steps: Obtain the protection space of the application scenario of the smart lock, where the protection space is the space entered through the door lock installed with the smart lock; Identify items in the protection space and obtain the value cost of the items; Get the minimum value of the item's re-retrieval time as additional time; The necessary usage time to obtain the item, which is the time it takes for the item to be needed again after it is lost or damaged; Set an additional coefficient. When the additional time does not exceed the necessary usage time, the additional coefficient is equal to 1. Otherwise, the additional coefficient is equal to the sum of the absolute value of the difference between the additional time and the necessary usage time and 1. The value cost of the item is multiplied by the additional coefficient to obtain the additional cost, and the additional costs of the items in the protection space are accumulated to obtain the important coefficient.

3. According to claim 2, a remote security management method for a low-power smart lock cluster is characterized in that: The method of classifying the smart lock cluster into at least one local cluster based on the importance coefficient comprises the following steps: The length of the interval of the value range of the important coefficient is used as the characteristic distance; The value range of the important coefficient is evenly divided into at least one characteristic interval, and the length of the characteristic interval is equal to the characteristic distance; The smart locks whose important coefficients belong to the local interval are merged into the same feature cluster, and the difference between the application scenarios of any two smart locks in the feature cluster is calculated. When the difference between the application scenarios of any two smart locks in the feature cluster is less than the preset value, the feature distance is used as the segmentation distance. Otherwise, the feature distance is updated to half of the original one, and the previous step is repeated, where the preset value is the allowable error set based on experience; The value range of the important coefficients is evenly divided into at least one local interval, and the length of the local interval is equal to the segmentation distance; The smart locks whose important coefficients belong to the local interval are merged into the same local cluster.

4. According to claim 3, a remote security management method for a low-power smart lock cluster is characterized in that: The calculation of the difference between the application scenarios of any two smart locks in the feature cluster comprises the following steps: Arbitrarily select two smart locks in the feature cluster as the first smart lock and the second smart lock respectively; Arrange the additional costs of the items in the protection space of the first smart lock from large to small to form a first vector; Arrange the additional costs of the items in the protection space of the second smart lock from large to small to form a second vector; When the number of components in the first vector and the second vector is different, components are supplemented at the ends of the first vector and the second vector, and the supplemented components are all 0. After the supplemented components, the number of components in the first vector and the second vector is the same; Subtract the first vector from the second vector to obtain a difference vector; The difference is obtained by dividing the modulus of the difference vector by the sum of the moduli of the first vector and the second vector.

5. A remote security management method for a low-power smart lock cluster according to claim 4, characterized in that: The multi-modal unlocking method is set to remotely monitor the smart locks of the local cluster, including the following steps: The multi-modal unlocking methods include fingerprint recognition, face recognition, vein recognition and voiceprint recognition. Users can choose one of them as the priority recognition method according to their needs. Based on the Internet of Things technology, smart locks are connected to smartphones or other smart devices, and users can remotely monitor the door lock status and receive real-time alerts through smartphones or other smart devices.

6. A remote security management method for a low-power smart lock cluster according to claim 5, characterized in that: The emergency treatment of the faulty smart lock based on the importance coefficient includes the following steps: Identify the fault and if it cannot be handled remotely, contact maintenance personnel for processing; When it is a remotely processable fault, the minimum value of the important coefficient is used as the target coefficient, and the average value of all important coefficients of the local cluster where the smart lock is located is divided by the target coefficient and rounded to obtain the number of passes of the verification problem. The verification problem is the characteristics of the items in the protection space of the application scenario of the smart lock; If the number of verification questions answered correctly by the user reaches the number of verification questions passed, the user passes the verification; otherwise, the user fails the verification; When the user passes the verification, the door is unlocked remotely, otherwise, the maintenance personnel are contacted for processing.

7. A remote security management method for a low-power smart lock cluster according to claim 6, characterized in that: The step of identifying the fault characteristics of the smart lock and obtaining the fault triggering symptom and the fault triggering usage frequency comprises the following steps: Obtain historical data on the parameter indicators of smart locks with fault characteristics, including unlocking response speed, possibility of lock cracking, battery life, system compatibility, and speed of remote management function implementation; Fitting the historical data of the parameter index to obtain the parameter fitting function, in which time is the independent variable and the historical data of the parameter index is the dependent variable; Calculating the first-order derivative of the parameter fitting function, and dividing the value range of the independent variable of the parameter fitting function into at least one continuous interval based on the first-order derivative, satisfying that the first-order derivative in the continuous interval is always greater than 0, always less than 0, or always equal to 0, and the first-order derivatives in adjacent continuous intervals are different from each other; The continuous intervals are sorted in chronological order, and when the length of the interval after the adjacent continuous interval exceeds twice the length of the interval before the adjacent continuous interval, the adjacent continuous interval after the adjacent continuous interval is used as the sign interval; Substitute the midpoint of the symptom interval into the parameter fitting function to obtain the critical value. The fault triggering symptom is when the value of the parameter index exceeds the critical value. The number of times the smart lock with fault triggering signs is used for the first time and the fault characteristics are shown is counted as the fault triggering usage frequency.

8. A remote security management method for a low-power smart lock cluster according to claim 7, characterized in that: The predicting of the remaining usage frequency of the target smart lock comprises the following steps: Acquire at least one fault triggering symptom of the target smart lock as the target fault triggering symptom; The reciprocal of the fault trigger usage frequency of the target fault trigger symptom is used as the fault trigger possibility, and 1 minus the fault trigger possibility is used to obtain the fault non-trigger possibility. Multiply the fault non-triggering possibilities of the target fault triggering symptom to obtain the comprehensive non-triggering possibility, and subtract the comprehensive non-triggering possibility from 1 to obtain the comprehensive triggering possibility; The possible inverse of the comprehensive trigger is rounded to the integer to obtain the remaining usage frequency of the target smart lock.

9. A remote security management method for a low-power smart lock cluster according to claim 8, characterized in that: The forming of the usage habits of the target smart lock comprises the following steps: Divide a year into at least one time period based on the hours; During the time period, the number of times the target smart lock is used is counted as the target number, and the average of the target number is taken to obtain the average number of times used; When the difference between the target number of times and the average number of times accounts for a ratio of the average number of times of use that is less than the allowable error, the time period is used as the usage period of the target smart lock. Otherwise, the time period is updated to twice the original time period and the previous step is repeated. Count the number of times the target smart lock is used during its usage period as the characteristic number; The feature count is divided by the usage cycle to obtain the usage frequency of the target smart lock.

10. A remote security management method for a low-power smart lock cluster according to claim 9, characterized in that: The method of predicting the failure time of the target smart lock based on the usage habits includes the following steps: The remaining usage frequency is divided by the usage frequency of the target smart lock to obtain the failure time of the target smart lock.

Images