Fault Self-Recovery Method Based on Smart Doorbell System

Abstract

The present invention relates to the technical field of doorbell fault management, and particularly to a fault self-recovery method based on an intelligent doorbell system. The present invention analyzes the non-direct fault repair requirements from two aspects: the version update of the target object and its historical usage performance, to determine whether the target object needs system repair, so as to reduce the fault risk of the target object. Through information feedback, fault feature prediction and repair matching analysis are carried out, that is, through the method of fault repair matching, the target object is accurately and quickly self-repaired for faults, so as to improve the fault self-repair efficiency of the target object. And through an in-depth method, a repair matching feasibility evaluation output analysis is carried out on the collected repair record data, that is, from the perspective of the repair feasibility of the preset repair plan, it is judged whether the repair feasibility of the preset repair plan can directly output and implement automatic repair, and then the preset repair plan is reasonably adopted to improve the self-repair effect of the target object.

Description

Technical Field

[0001] The present invention relates to the technical field of doorbell fault management, and particularly to a fault self-recovery method based on an intelligent doorbell system. Background Art

[0002] In recent years, with the continuous upgrading of the intelligent industry, smart home has gradually entered the daily life of ordinary people. To facilitate people's daily life and meet people's greater needs for more intelligent technologies in more homes, intelligent doorbells have emerged as the times require. Currently, intelligent doorbells on the market can already meet people's needs for remote control or observing the situation at home through the wake-up function;

[0003] However, in the prior art, the faults of the intelligent doorbell system cannot be effectively and reasonably self-repaired and managed, thus reducing the fault self-recovery efficiency of the intelligent doorbell system. Moreover, it is impossible to supervise the non-direct fault repair requirements from two aspects of version update and historical usage performance, which increases the difficulty of self-repair of the intelligent doorbell system. At the same time, it is impossible to analyze and output a judgment on the feasibility of the preset self-repair plan for faults, and it is impossible to accurately understand the maintenance situation of the repair effect of the preset self-repair plan for faults, reducing the fault self-recovery management efficiency of the intelligent doorbell system.

[0004] In view of the above technical defects, a solution is proposed. Summary of the Invention

[0005] The purpose of the present invention is to provide a fault self-recovery method based on an intelligent doorbell system to solve the above-mentioned technical defects. The present invention analyzes the non-direct fault repair requirements from two aspects of the version update and historical usage performance of the target object to determine whether the target object needs system repair, so as to reduce the fault risk of the target object. Through information feedback, fault feature prediction and repair matching analysis are carried out, that is, through the method of fault repair matching, the target object is accurately and quickly self-repaired for faults, so as to improve the fault self-recovery efficiency of the target object. And through an in-depth method, a repair matching feasibility evaluation output analysis is carried out on the collected repair record data, that is, from the perspective of the repair feasibility of the preset repair plan, it is judged whether the repair feasibility of the preset repair plan can be directly output and automatic repair can be implemented, and then the preset repair plan is reasonably adopted to improve the self-repair effect of the target object.

[0006] The purpose of the present invention can be achieved by the following technical solutions: A fault self-recovery method based on an intelligent doorbell system includes the following steps:

[0007] Step 1: Set the intelligent doorbell system as the target object, collect the current version information of the target object for version operation obstacle evaluation feedback analysis, and output and feedback the obtained version normal instruction or version risk instruction;

[0008] Step 2: Conduct control obstacle repair evaluation and supervision analysis on the collected control Internet of Things information from the perspective of the historical usage performance of the target object, discriminate and process the obtained control obstacle evaluation coefficient, and output and feedback the obtained fault risk control instruction or stability instruction;

[0009] Step 3: Based on the fault feature prediction and repair matching analysis of the target object under information feedback, discriminate and process the obtained revised evaluation coefficient. If an effective instruction is obtained, proceed to Step 4; if a correction instruction is obtained, output and feedback;

[0010] Step 4: Analyze the working condition information of the collected target object based on the effective instruction, discriminate and process the output result of the obtained fault prediction model. If a fault instruction is obtained, proceed to Step 5; if a compliance instruction is obtained, output and feedback;

[0011] Step 5: Based on the output analysis of the repair matching feasibility evaluation under information progression, discriminate and process the obtained repair adoption evaluation coefficient, and output and feedback the obtained repair feedback instruction or available instruction.

[0012] Preferably, the version operation obstacle evaluation feedback analysis process is as follows:

[0013] Set the intelligent doorbell system as the target object, collect the operation period of the target object, and set it as the time threshold. Obtain the current version information of the target object within the time threshold. The version information includes the version number and the version upgrade date. Extract the text characters of the version information of the target object, and set the string composed of the extracted text characters of the version information of the target object as the preliminary inspection evaluation recognition string;

[0014] Obtain the latest version information of the target object within the time threshold, set the string composed of the extracted text characters of the latest version information of the target object as the standard evaluation recognition string, and conduct a one-by-one comparison and analysis between the preliminary inspection evaluation recognition string and the standard evaluation recognition string to obtain a version normal instruction or a version risk instruction.

[0015] Preferably, the control obstacle repair evaluation and supervision analysis process is as follows:

[0016] Obtain the control Internet of Things information of the target object within the time threshold. The control Internet of Things information includes the control evaluation coefficient and the measurement and control response index. Set the number of the control evaluation coefficient and the measurement and control response index that are greater than or equal to the preset control evaluation coefficient threshold and the preset measurement and control response index threshold as the control obstacle evaluation coefficient, and conduct discrimination and processing on the control obstacle evaluation coefficient to obtain a fault risk control instruction or a stability instruction.

[0017] Preferably, the control evaluation coefficient represents the mean of the interval durations corresponding to the IoT defect features of the target object, and the IoT defect features include interruptions and lags; the analysis process of the measurement and control response index is as follows: Obtain the response performance values corresponding to the historical control times of the target object, obtain the number of times the response performance value is less than the preset response performance value threshold, and set it as the measurement and control response index. The response performance value represents the duration between sending an instruction to control the target object and the execution moment of the target object.

[0018] Preferably, the analysis process of the fault feature prediction and repair matching is as follows:

[0019] Obtain the historical operating condition information of multiple groups of target objects within the time threshold. The historical operating condition information includes temperature and abnormal sound values. Divide the historical fault information into a training set and a validation set, and perform preprocessing on the training set and the validation set. The preprocessing includes data cleaning and screening. Based on the preprocessed training set, construct a fault prediction model, and perform verification processing on the constructed fault prediction model through the validation set. Obtain the number of deviations between the output result of the fault prediction model and the known result, and set it as the revision evaluation coefficient;

[0020] Perform discrimination processing on the revision evaluation coefficient: If the revision evaluation coefficient is equal to zero, generate a valid instruction; if the revision evaluation coefficient is not equal to zero, generate a correction instruction.

[0021] Preferably, when a valid instruction is generated, obtain the operating condition information of the current target object within the time threshold, input the operating condition information into the fault prediction model, obtain the output result of the fault prediction model, and perform discrimination processing on the output result of the fault prediction model to obtain a qualified instruction or a fault instruction.

[0022] Preferably, the analysis process of the repair matching feasibility evaluation output is as follows:

[0023] Obtain the fault codes corresponding to the output results belonging to the preset fault list, obtain the preset repair solutions based on the fault codes, obtain the repair record data of the preset repair solutions. The repair record data includes repair feasibility and repair performance coefficients. Compare and analyze the repair feasibility and repair performance coefficients with the preset repair feasibility threshold and the preset repair performance coefficient threshold. Set the number of repair feasibility and repair performance coefficients that are less than the preset repair feasibility threshold and the preset repair performance coefficient threshold as the repair adoption evaluation coefficient, and perform discrimination processing on the repair adoption evaluation coefficient to obtain a repair feedback instruction or an available instruction.

[0024] Preferably, for each preset repair solution in the total number of repairs using the preset repair solution corresponding to the usage fault code, obtain the duration of continuous operation after repair. Set the ratio of the number of repairs of the preset repair solution corresponding to the fault code with a continuous operation duration greater than the preset continuous operation duration threshold to the total number of repairs as the repair feasibility. The continuous operation duration refers to the duration between the end of repair using the preset repair solution corresponding to the fault code and the moment when the same fault code appears again; obtain the ratio of the number of times the response performance value is lower than the preset response performance value within the preset duration after repair of the preset repair solution corresponding to the fault code to the total number of times, and set it as the recovery ability value. Set the ratio of the number of repairs of the preset repair solution corresponding to the recovery ability value greater than the preset recovery ability value threshold to the total number of repairs as the repair performance coefficient.

[0025] The beneficial effects of the present invention are as follows:

[0026] (1) The present invention analyzes the non-direct fault repair requirements from two aspects: the version update of the target object and the historical usage performance, to determine whether the target object needs system repair, so as to reduce the fault risk of the target object. That is, conduct a version operation obstacle evaluation and feedback analysis on the version information to determine whether the current version of the target object affects the operation, so as to update the version in a timely manner. And judge the fault risk and the degree of repair requirements of the target object from the perspective of historical usage performance, so as to conduct timely repair management on the target object in a timely manner;

[0027] (2) The present invention conducts fault feature prediction and repair matching analysis through information feedback, that is, accurately performs rapid fault self-repair on the target object through the method of fault repair matching, so as to improve the fault self-repair efficiency of the target object. And conduct a repair matching feasibility evaluation and output analysis on the collected repair record data in an in-depth manner, that is, analyze from the perspective of the repair feasibility of the preset repair solution to judge whether the repair feasibility of the preset repair solution can directly output and implement automatic repair, and then reasonably adopt the preset repair solution to improve the self-repair effect of the target object. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The present invention will be further described below with reference to the accompanying drawings;

[0029] Figure 1 is the reference analysis diagram of the method of the present invention;

[0030] Figure 2 is the partial analysis diagram of the second embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0031] Next, in combination with the accompanying drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0032] Embodiment 1:

[0033] Please refer to Figures 1 to 2 As shown, the present invention is a fault self-recovery method based on an intelligent doorbell system, including the following steps:

[0034] Step 1: Set the intelligent doorbell system as the target object, collect the current version information of the target object for version operation obstacle evaluation and feedback analysis, and output and feedback the obtained version normal instruction or version risk instruction;

[0035] Step 2: Conduct control obstacle repair evaluation and supervision analysis on the collected control IoT information from the perspective of the historical usage performance of the target object, perform discrimination processing on the obtained control obstacle evaluation coefficient, and output and feedback the obtained fault risk control instruction or stable instruction;

[0036] Step 3: Based on the fault feature prediction and repair matching analysis of the target object under information feedback, perform discrimination processing on the obtained revised evaluation coefficient. If an effective instruction is obtained, enter Step 4; if a correction instruction is obtained, output and feedback;

[0037] Step 4: Analyze the working condition information of the collected target object based on the effective instruction, perform discrimination processing on the output result of the obtained fault prediction model. If a fault instruction is obtained, enter Step 5; if a compliance instruction is obtained, output and feedback;

[0038] Step 5: Based on the repair matching feasibility evaluation output analysis under information progression, perform discrimination processing on the obtained repair adoption evaluation coefficient, and output and feedback the obtained repair feedback instruction or available instruction;

[0039] The present invention analyzes the repair requirements from two aspects: the version update of the target object and the historical usage performance, to determine whether the target object needs system repair, that is, collect the current version information of the target object, and conduct version operation obstacle evaluation and feedback analysis on the version information to determine whether the current version of the target object affects the operation, so as to update the version in a timely manner to reduce the fault risk of the target object. The specific process of version operation obstacle evaluation and feedback analysis is as follows:

[0040] The smart doorbell system is set as the target object, the running time period of the target object is collected and set as the time threshold, the current version information of the target object within the time threshold is obtained, the version information includes the version number, version upgrade date, etc., the version information of the target object is extracted with text characters, and the string composed of the text character extraction of the version information of the target object is set as the initial inspection and evaluation recognition string;

[0041] Get the latest version information of the target object within the time threshold, extract the text characters from the latest version information of the target object and set the string as the standard evaluation identification string, and compare and analyze the initial evaluation identification string with the standard evaluation identification string one by one:

[0042] If the initial check evaluation identification string and the standard evaluation identification string are in one-to-one correspondence, a normal version instruction is generated;

[0043] If the initial check evaluation identification string and the standard evaluation identification string are in one-to-one correspondence, a version risk instruction is generated, and a normal version instruction or a version risk instruction is output as feedback, so as to update the version information of the target object according to the information feedback to reduce the failure risk of the target object;

[0044] In the embodiment of the present invention, the collected control IoT information is subjected to control obstacle repair evaluation and supervision analysis from the perspective of historical usage performance, that is, the failure risk and repair demand degree of the target object are judged from the perspective of historical usage performance, so as to timely perform repair management on the target object. The specific control obstacle repair evaluation and supervision analysis process is as follows:

[0045] The control IoT information of the target object within the time threshold is obtained, and the control IoT information includes the control evaluation coefficient and the measurement and control response index. The number of control evaluation coefficients and measurement and control response indexes that are greater than or equal to the preset control evaluation coefficient threshold and the preset measurement and control response index threshold is set as the control obstruction evaluation coefficient, and the control obstruction evaluation coefficient is discriminated:

[0046] If the control obstruction evaluation coefficient = 0 or the control obstruction evaluation coefficient = 1, a fault risk control instruction is generated;

[0047] If the control obstacle evaluation coefficient = 2, a stable instruction is generated, and the fault risk control instruction or stable instruction is output and fed back, so as to conduct reasonable and targeted management of the target object based on the information feedback, so as to reduce the impact of long-term use on the operation of the target object, and at the same time help to ensure the continuous use and stable performance of the target object;

[0048] In the embodiment of the present invention, the control evaluation coefficient represents the average value of the interval duration corresponding to the IoT defect characteristics of the target object. The IoT defect characteristics include interruption, freeze, etc. It should be noted that the larger the value of the control evaluation coefficient, the smaller the control abnormality risk of the target object;

[0049] In the embodiment of the present invention, the analysis process of the measurement and control response index is as follows: Obtain the response performance value corresponding to the historical control times of the target object, obtain the number of times when the response performance value is less than the preset response performance value threshold, and set it as the measurement and control response index. The response performance value represents the duration between sending the instruction to control the target object and the execution moment of the target object. It should be noted that the measurement and control response index is an influence parameter reflecting the control performance of the target object. The larger the value of the measurement and control response index, the smaller the control anomaly risk of the target object.

[0050] Embodiment Two:

[0051] When generating a version risk instruction or a fault risk control instruction, based on the version update of the target object or the fault feature prediction and repair matching analysis under the abnormal historical usage performance, that is, through the way of fault repair matching, accurately perform rapid self-repair of faults on the target object to improve the fault self-repair efficiency of the target object. The specific fault feature prediction and repair matching analysis process is as follows:

[0052] Obtain multiple groups of historical working condition information of the target object within the time threshold. The historical working condition information includes temperature, abnormal sound value, etc. Divide the historical fault information into a training set and a verification set, and perform preprocessing on the training set and the verification set. The preprocessing includes data cleaning, screening, etc. Build a fault prediction model based on the preprocessed training set, and perform verification processing on the built fault prediction model through the verification set. Obtain the number of deviations between the output result of the fault prediction model and the known result, and set it as the revision evaluation coefficient;

[0053] In the embodiment of the present invention, abnormal sound means that the sound decibel of the target object deviates from the preset range;

[0054] Perform discriminant processing on the revision evaluation coefficient:

[0055] If the revision evaluation coefficient is equal to zero, generate a valid instruction;

[0056] If the revision evaluation coefficient is not equal to zero, generate a correction instruction. When generating a correction instruction, immediately perform the preset warning operation corresponding to the correction instruction to perform targeted correction on the fault prediction model to improve the extraction accuracy of the fault prediction model;

[0057] When generating a valid instruction, obtain the working condition information of the current target object within the time threshold, input the working condition information into the fault prediction model, obtain the output result of the fault prediction model, and perform discriminant processing on the output result of the fault prediction model:

[0058] If the output result does not belong to the preset fault list, generate a compliance instruction and output feedback;

[0059] If the output result belongs to a preset fault list, a fault instruction is generated;

[0060] When a fault instruction is generated, the present invention analyzes from the perspective of the repair feasibility of a preset repair plan, that is, performs a repair matching feasibility evaluation output analysis on the collected repair record data, that is, determines whether the repair feasibility of the preset repair plan can directly output and implement automatic repair, and then reasonably adopts the preset repair plan to improve the self-repair effect of the target object. The specific repair matching feasibility evaluation output analysis process is as follows:

[0061] Obtain the fault code corresponding to the output result belonging to the preset fault list, obtain the preset repair plan based on the fault code, obtain the repair record data of the preset repair plan, the repair record data includes the repair feasibility and the repair performance coefficient, compare and analyze the repair feasibility and the repair performance coefficient with the preset repair feasibility threshold and the preset repair performance coefficient threshold, set the number of the repair feasibility and the repair performance coefficient that are less than the preset repair feasibility threshold and the preset repair performance coefficient threshold as the repair adoption evaluation coefficient, and perform a discrimination process on the repair adoption evaluation coefficient:

[0062] If the repair adoption evaluation coefficient = 1 or the repair adoption evaluation coefficient = 2, a repair feedback instruction is generated. When the repair feedback instruction is generated, immediately perform the preset warning operation corresponding to the repair feedback instruction to remind the operation and management personnel to use other methods to perform self-repair processing on the target object, thereby improving the self-repair management efficiency of the target object;

[0063] If the repair adoption evaluation coefficient = 0, an available instruction is generated. When the available instruction is generated, output and feedback the preset repair plan corresponding to the fault code, and then perform self-repair processing on the target object according to the preset repair plan to improve the fault self-repair efficiency of the target object;

[0064] In the embodiment of the present invention, obtain the continuous operation duration after each preset repair plan is repaired in the total number of repairs using the preset repair plan corresponding to the fault code, and set the ratio of the number of repairs of the preset repair plan corresponding to the fault code with the continuous operation duration greater than the preset continuous operation duration threshold to the total number of repairs as the repair feasibility. The continuous operation duration represents the duration between the end of the repair by the preset repair plan corresponding to the fault code and the moment when the same fault code appears again. It should be noted that the repair feasibility is an influence parameter reflecting the repair efficiency of the preset repair plan corresponding to the fault code. The larger the value of the repair feasibility, the higher the feasibility of the repair method through the set preset repair plan;

[0065] In the embodiment of the present invention, the ratio of the number of times that the response performance value within the preset duration after repair corresponding to the preset repair solution for the fault code is lower than the preset response performance value to the total number of times is obtained, and it is set as the recovery ability value. The ratio of the number of times of repair corresponding to the preset repair solution with the recovery ability value greater than the preset recovery ability value threshold to the total number of repairs is set as the repair performance coefficient. It should be noted that the larger the value of the repair performance coefficient, the higher the feasibility of the repair method by setting the preset repair solution;

[0066] In summary, the present invention analyzes the non-direct fault repair requirements from two aspects: the version update of the target object and the historical usage performance, so as to judge whether the target object needs system repair, reduce the fault risk of the target object, that is, conduct a version operation obstacle evaluation feedback analysis on the version information to judge whether the current version of the target object affects the operation, so as to update the version in time, and judge the fault risk and the degree of repair requirements of the target object from the perspective of historical usage performance, so as to conduct timely repair management on the target object in time;

[0067] Fault feature prediction and repair matching analysis are carried out through the way of information feedback, that is, the target object is accurately and quickly self-repaired from faults through the way of fault repair matching, so as to improve the fault self-repair efficiency of the target object. And the repair matching feasibility evaluation output analysis is carried out on the collected repair record data in an in-depth manner, that is, analyzed from the perspective of the repair feasibility of the preset repair solution to judge whether the repair feasibility of the preset repair solution can directly output and implement automatic repair, and then reasonably adopt the preset repair solution to improve the self-repair effect of the target object.

[0068] The setting of the size of the threshold is for the convenience of comparison. Regarding the size of the threshold, it depends on the amount of sample data and the number of base numbers set by those skilled in the art for each group of sample data; as long as the proportional relationship between the parameter and the quantified value is not affected;

[0069] The size of the coefficient is a specific value obtained by quantifying each parameter for the convenience of subsequent comparison. Regarding the size of the coefficient, it depends on the amount of sample data and the corresponding operation coefficient initially set by those skilled in the art for each group of sample data; as long as the proportional relationship between the parameter and the quantified value is not affected.

[0070] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution of the present invention and its inventive concept, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. A fault self-recovery method based on an intelligent doorbell system, characterized in that, It includes the following steps: Step 1: Set the intelligent doorbell system as the target object, collect the current version information of the target object for version operation obstacle evaluation feedback analysis, and output and feedback the obtained version normal instruction or version risk instruction; The version information includes the version number and the version upgrade date. Extract the text characters of the version information of the target object, and set the string composed of the extracted text characters of the version information of the target object as the preliminary inspection evaluation recognition string; Step 2: Conduct control obstacle repair evaluation supervision analysis on the collected control and Internet of Things information from the perspective of the historical usage performance of the target object, perform discriminant processing on the obtained control obstacle evaluation coefficient, and output and feedback the obtained fault risk control instruction or stable instruction; Step 3: Based on the fault feature prediction and repair matching analysis of the target object under information feedback, perform discriminant processing on the obtained revised evaluation coefficient. If an effective instruction is obtained, enter Step 4; if a correction instruction is obtained, output and feedback; Obtain the number of deviations between the output result of the fault prediction model and the known result, and set it as the revised evaluation coefficient; Step 4: Analyze the working condition information of the collected target object based on the effective instruction, perform discriminant processing on the output result of the obtained fault prediction model. If a fault instruction is obtained, enter Step 5; if a compliance instruction is obtained, output and feedback; Step 5: Based on the repair matching feasibility evaluation output analysis under information progression, perform discriminant processing on the obtained repair adoption evaluation coefficient, and output and feedback the obtained repair feedback instruction or available instruction; The process of the repair matching feasibility evaluation output analysis is as follows: Compare and analyze the repair feasibility and repair performance coefficient with the preset repair feasibility threshold and the preset repair performance coefficient threshold. Set the number corresponding to the repair feasibility and repair performance coefficient that is less than the preset repair feasibility threshold and the preset repair performance coefficient threshold as the repair adoption evaluation coefficient, and perform discriminant processing on the repair adoption evaluation coefficient to obtain a repair feedback instruction or an available instruction; Obtain the fault code corresponding to the output result belonging to the preset fault list, obtain the preset repair plan based on the fault code, and obtain the repair record data of the preset repair plan. The repair record data includes the repair feasibility and the repair performance coefficient; Obtain the continuous operation duration after each preset repair plan is repaired in the total number of repairs using the preset repair plan corresponding to the fault code. Set the ratio of the number of repairs of the preset repair plan corresponding to the fault code with the continuous operation duration greater than the preset continuous operation duration threshold to the total number of repairs as the repair feasibility. The continuous operation duration refers to the duration from the end of the repair using the preset repair plan corresponding to the fault code to the moment when the same fault code appears again; Obtain the ratio of the number of times the response performance value is lower than the preset response performance value within the preset duration after the repair of the preset repair plan corresponding to the fault code to the total number of repairs, and set it as the recovery ability value. Set the ratio of the number of repairs of the preset repair plan corresponding to the recovery ability value greater than the preset recovery ability value threshold to the total number of repairs as the repair performance coefficient.

2. The fault self-recovery method based on the intelligent doorbell system according to claim 1, wherein, The process of the version operation obstacle evaluation feedback analysis is as follows: The smart doorbell system is set as the target object, the running time period of the target object is collected and set as the time threshold, and the current version information of the target object within the time threshold is obtained; The latest version information of the target object within the time threshold is obtained, and the string composed of text characters extracted from the latest version information of the target object is set as the standard evaluation identification string. The initial evaluation identification string is compared and analyzed with the standard evaluation identification string one by one to obtain normal version instructions or version risk instructions.

3. The fault self-recovery method based on the intelligent doorbell system according to claim 1, characterized in that The control obstacle remediation assessment supervision and analysis process is as follows: The control IoT information of the target object within the time threshold is obtained, and the control IoT information includes the control evaluation coefficient and the measurement and control response index. The number of control evaluation coefficients and measurement and control response indexes that are greater than or equal to the preset control evaluation coefficient threshold and the preset measurement and control response index threshold is set as the control obstacle evaluation coefficient, and the control obstacle evaluation coefficient is judged and processed to obtain a fault risk control instruction or a stability instruction.

4. The fault self-recovery method based on the intelligent doorbell system according to claim 3, wherein, The control evaluation coefficient represents the average value of the interval duration corresponding to the IoT defect characteristics of the target object, and the IoT defect characteristics include interruption and freeze; the response performance value represents the duration between sending the instruction to control the target object and the execution time of the target object; The analysis process of the measurement and control response index is as follows: obtaining the response performance value corresponding to the historical control times of the target object, obtaining the number of times the response performance value is less than the preset response performance value threshold, and setting it as the measurement and control response index.

5. The fault self-recovery method based on the intelligent doorbell system according to claim 1, characterized in that The fault feature prediction and repair matching analysis process is as follows: Obtain historical operating condition information of multiple groups of target objects within the time threshold, including temperature and abnormal sound value, divide the historical fault information into training set and verification set, preprocess the training set and verification set, preprocessing includes data cleaning and screening, build a fault prediction model based on the preprocessed training set, and verify the built fault prediction model through the verification set; The revised evaluation coefficient is judged: if the revised evaluation coefficient is equal to zero, a valid instruction is generated; if the revised evaluation coefficient is not equal to zero, a correction instruction is generated.

6. The fault self-recovery method based on the intelligent doorbell system according to claim 5, characterized in that, When a valid instruction is generated, the operating condition information of the current target object within the time threshold is obtained, the operating condition information is input into the fault prediction model, the output result of the fault prediction model is obtained, and the output result of the fault prediction model is discriminated and processed to obtain a standard instruction or a fault instruction.

Images