A big data-based solution recommendation system

Abstract

The application discloses a solution recommendation system based on big data, and comprises an equipment fault information library, a self-checking detection module, a normal starting detection module, an abnormal starting detection module and a solution matching module. The equipment fault information library is used for statistically storing equipment historical abnormal data information, abnormal symptom manifestation tables and historical solution information of each batch identification. One historical abnormal data information corresponds to one abnormal symptom manifestation table, one historical abnormal data information corresponds to multiple historical solution information, one historical solution information comprises symptom manifestation, historical solution content, fault occurrence rate and feedback success rate, and the abnormal symptom manifestation table is a table of the collection of symptom manifestations in the multiple historical solution information corresponding to the historical abnormal data information.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of big data, and particularly relates to a solution recommendation system based on big data. BACKGROUND

[0002] Due to friction, external force, stress, electrostatic force and the like, the equipment is difficult to avoid failure in the use process. In the traditional habit, when the equipment fails, a special equipment technical maintenance personnel is found to solve the equipment failure. However, if the equipment fails, the special equipment technical maintenance personnel is found to maintain the efficiency. Time is spent, and money is spent, so self-repairing equipment failure becomes a big trend. SUMMARY

[0003] The present application aims to provide a solution recommendation system based on big data to solve the problems in the prior art.

[0004] To achieve the above-mentioned purpose, the present application provides the following technical scheme:

[0005] A solution recommendation system based on big data, the recommendation system comprises: an equipment failure information library, a self-checking detection module, a normal start detection module, an abnormal start detection module and a solution matching module, the equipment failure information library is used for statistically storing the equipment historical abnormal data information, the abnormal symptom performance table and the historical solution information of each batch identification, wherein one historical abnormal data information corresponds to one abnormal symptom performance table, one historical abnormal data information corresponds to multiple historical solution information, one historical solution information includes symptom performance, historical solution content, failure occurrence rate and feedback success rate, the abnormal symptom performance table is a table of the set of symptom performances in the multiple historical solution information corresponding to the historical abnormal data information, the self-checking detection module performs self-checking when the equipment to be recommended starts, and when the equipment to be recommended detects an abnormal condition, the batch identification of the equipment to be recommended is collected, the current abnormal data information is taken as the first abnormal data information, the normal start detection module continuously monitors the running state of the equipment when the equipment can normally start, and when the equipment cannot run in the monitoring process, the abnormal data information when the equipment cannot run is obtained, and the abnormal data information to be matched is determined, the abnormal start detection module monitors the restart state of the equipment and the abnormal data information at each restart when the equipment cannot normally start, and determines the abnormal data information to be matched accordingly, the solution matching module selects one historical abnormal data information as the matching abnormal data information from the equipment failure information library for the abnormal data information to be matched, and selects the corresponding historical solution content from the historical abnormal data information.

[0006] Preferably, the normal start detection module comprises a running detection module, a second abnormal data information collection module, a similarity comparison module and a restart control module. The running detection module continuously monitors the running state of the device, and when the device cannot run during the monitoring process, the second abnormal data information collection module collects the current abnormal data information as second abnormal data information. The similarity comparison module is used to compare the similarity between the first abnormal data information and the second abnormal data information. When the similarity is greater than or equal to a first similarity threshold, the first abnormal data information is taken as the abnormal data information to be matched. When the similarity is less than the first similarity threshold, the restart control module controls the device to automatically restart, and the abnormal data information after the device restarts is taken as the abnormal data information to be matched.

[0007] Preferably, the abnormal start detection module comprises a restart state detection module and a human restart abnormality detection module. The restart state detection module monitors the restart state of the device within a preset time period when the device cannot start normally, and judges that the device restarts as a human restart when the number of times that the time interval between two restarts within the preset time period is not equal is greater than or equal to a first number threshold. The human restart abnormality detection module works when the number of times that the time interval between two restarts within the preset time period is equal is less than the first number threshold. The abnormal data information at the first restart is taken as the abnormal data information to be matched. The human restart abnormality detection module comprises a restart data information comparison module and a power state detection module. The restart data information comparison module is used to obtain the abnormal data information at each restart when the device restarts is judged as a human restart. If the abnormal data information at each restart is the same, the abnormal data information at the restart is taken as the abnormal data information to be matched. If the abnormal data information at each restart is not the same, the power state detection module works. The power state detection module monitors the time length during which the power state of the device remains unchanged after each restart, and when the time length during which the power state of the device remains unchanged after a certain restart is greater than or equal to a time length threshold, the restart is regarded as the last restart, and the abnormal data information at the last restart is taken as the abnormal data information to be matched.

[0008] Preferably, the solution matching module comprises a batch identification selection module, a similarity comparison module, a matching abnormal data information and a solution selection module. The batch identification selection module is configured to obtain each historical abnormal data information of the device with the same batch identification as the device to be recommended in the device fault information library. The similarity comparison module is configured to compare the similarity of each historical abnormal data information obtained by the batch identification selection module and the matching abnormal data information of the device to be recommended. The matching abnormal data information sorts the comparison results of the similarity comparison module in descending order, and selects the historical abnormal data information with the first order as the matching abnormal data information. The solution selection module selects the solution content of the abnormality of the device to be recommended from the historical solution information corresponding to the matching abnormal data information.

[0009] Preferably, the solution selection module comprises a symptom manifestation return module, a symptom manifestation similarity comparison module, a matching degree calculation module and a solution content recommendation module. The symptom manifestation return module is configured to receive the abnormal symptom manifestation of the device to be recommended returned by the user. The symptom manifestation similarity comparison module is configured to compare the similarity of the abnormal symptom manifestation returned by the symptom manifestation return module and the symptom manifestation in each historical solution information corresponding to the matching abnormal data information. The matching degree calculation module calculates the matching degree of each historical solution information according to the symptom manifestation similarity, the fault occurrence rate and the feedback success rate of each historical solution information. The solution content recommendation module sorts each historical solution information in descending order of matching degree, and selects the historical solution content corresponding to the historical solution information with the first order to recommend to the user.

[0010] A solution recommendation method based on big data, the recommendation method comprising the following:

[0011] A device fault information library is pre-set, which statistically stores historical abnormal data information, abnormal symptom manifestation tables and historical solution information of each batch identification. One historical abnormal data information corresponds to one abnormal symptom manifestation table, one historical abnormal data information corresponds to multiple historical solution information, one historical solution information includes symptom manifestation, historical solution content, fault occurrence rate and feedback success rate, and the abnormal symptom manifestation table is a table of the set of symptom manifestations corresponding to multiple historical solution information of the historical abnormal data information.

[0012] When the device to be recommended starts, self-checking is performed. If the device to be recommended detects abnormal conditions, the batch identification of the device to be recommended is collected, and the current abnormal data information is taken as the first abnormal data information.

[0013] If the device can be normally started, the running state of the device is continuously monitored, if the device cannot run in the monitoring process, the abnormal data information when the device cannot run is acquired, and the abnormal data information to be matched is determined;

[0014] If the device cannot be normally started, the restart state of the device and the abnormal data information at each restart are monitored, and the abnormal data information to be matched is determined according to the abnormal data information.

[0015] A historical abnormal data information is selected as the matching abnormal data information from the device fault information library for the abnormal data information to be matched, and the corresponding historical solution content is selected from the historical abnormal data information.

[0016] Preferably, the acquisition of the abnormal data information when the device cannot run and the determination of the abnormal data information to be matched include:

[0017] The running state of the device is continuously monitored, if the device cannot run in the monitoring process, the current abnormal data information is collected as the second abnormal data information, the similarity of the first abnormal data information and the second abnormal data information is the first similarity,

[0018] If the first similarity is greater than or equal to the first similarity threshold, the first abnormal data information is taken as the abnormal data information to be matched,

[0019] If the first similarity is less than the first similarity threshold, the device is automatically restarted, and the abnormal data information after the device is restarted is taken as the abnormal data information to be matched.

[0020] Preferably, the monitoring of the restart state of the device and the abnormal data information at each restart and the determination of the abnormal data information to be matched include:

[0021] If the device cannot be normally started, the restart state of the device to be recommended within a preset time period is monitored, if the number of times that the time interval between two restarts within the preset time period is not equal is greater than or equal to a first number threshold, the abnormal data information at each restart is acquired, if the abnormal data information at each restart is the same, the abnormal data information at the restart is taken as the abnormal data information to be matched, if the abnormal data information at each restart is not the same, the time length that the power state of the device remains unchanged after each restart is monitored, if there is a time length that the power state of the device remains unchanged after a restart is greater than or equal to a time length threshold, the restart is the last restart, and the abnormal data information at the last restart is taken as the abnormal data information to be matched; if the number of times that the time interval between two restarts within the preset time period is equal is less than a number threshold, the abnormal data information at the first restart is taken as the abnormal data information to be matched.

[0022] More preferably, the selecting a historical abnormal data information as the matching abnormal data information from the equipment fault information library comprises:

[0023] Obtaining each historical abnormal data information of the same equipment as the to-be-recommended equipment batch identifier in the equipment fault information library, and comparing the similarity of each historical abnormal data information and the to-be-matched abnormal data information of the to-be-recommended equipment, sorting each similarity in descending order, and selecting the first historical abnormal data information as the matching abnormal data information.

[0024] More preferably, the selecting a historical abnormal data information as the matching abnormal data information from the equipment fault information library comprises

[0025] Transmitting the abnormal symptom manifestation table corresponding to the historical abnormal data information to the user, and the user selects the corresponding abnormal symptom manifestation of the to-be-recommended equipment;

[0026] Comparing the similarity m of the symptom manifestation in each historical solution information and the symptom manifestation selected by the user,

[0027] Then the matching degree of each historical solution information

[0028] Z=0.6*m+0.3*n+0.1*q, wherein n is the fault occurrence rate of the historical solution information, and q is the feedback success rate of the historical solution information;

[0029] Sorting each historical solution information in descending order of matching degree, and selecting the historical solution content corresponding to the first historical solution information to recommend to the user.

[0030] Compared with the prior art, the beneficial effects of the present application are: the present application distinguishes between the artificial restart of the equipment and the self-restart of the equipment by monitoring the restart state of the equipment within a preset time period, so that the judgment of the abnormal data information of the equipment is more accurate, and the recommended solution according to the abnormal data information is also more accurate. BRIEF DESCRIPTION OF DRAWINGS

[0031] Figure 1 A module schematic diagram of the solution recommendation system based on big data. DETAILED DESCRIPTION

[0032] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0033] Please refer to Figure 1 In the embodiment of the present application, a solution recommendation system based on big data includes a device fault information library, a self-checking detection module, a normal start detection module, an abnormal start detection module, and a solution matching module. The device fault information library is used to statistically store device historical abnormal data information, abnormal symptom performance tables, and historical solution information of each batch identification. One historical abnormal data information corresponds to one abnormal symptom performance table, one historical abnormal data information corresponds to multiple historical solution information, one historical solution information includes symptom performance, historical solution content, fault occurrence rate, and feedback success rate, and the abnormal symptom performance table is a table of the set of symptom performances of the multiple historical solution information corresponding to the historical abnormal data information. The self-checking detection module performs self-checking when the device to be recommended starts, and collects the batch identification of the device to be recommended when the device to be recommended detects an abnormal condition, and takes the current abnormal data information as the first abnormal data information. The normal start detection module continuously monitors the running state of the device when the device can start normally, and obtains abnormal data information when the device cannot run during the monitoring process, and determines the abnormal data information to be matched. The abnormal start detection module monitors the restart state of the device and the abnormal data information at each restart when the device cannot start normally, and determines the abnormal data information to be matched accordingly. The solution matching module selects one historical abnormal data information from the device fault information library as the matching abnormal data information for the abnormal data information to be matched, and selects the corresponding historical solution content from the historical abnormal data information.

[0034] The normal start detection module includes a running detection module, a second abnormal data information collection module, a similarity comparison module, and a restart control module. The running detection module continuously monitors the running state of the device, and causes the second abnormal data information collection module to collect the current abnormal data information as the second abnormal data information when the device cannot run during the monitoring process. The similarity comparison module is used to compare the similarity between the first abnormal data information and the second abnormal data information. When the similarity is greater than or equal to a first similarity threshold, the first abnormal data information is taken as the abnormal data information to be matched. When the similarity is less than the first similarity threshold, the restart control module controls the device to automatically restart, and the abnormal data information after the device restarts is taken as the abnormal data information to be matched.

[0035] The abnormal start detection module comprises a restart state detection module and a manual restart anomaly detection module. The restart state detection module monitors the restart state of the device to be recommended within a preset time period when the device cannot start normally, and determines that the device restarts manually when the number of time intervals between two restarts within the preset time period is not equal to a first number threshold. The manual restart anomaly detection module works when the number of time intervals between two restarts within the preset time period is equal to the first number threshold. The device restarts automatically when the number of time intervals between two restarts within the preset time period is less than the first number threshold. The abnormal data information at the first restart is used as the abnormal data information to be matched. The manual restart anomaly detection module comprises a restart data information comparison module and a power state detection module. The restart data information comparison module is used to obtain abnormal data information at each restart when the device restarts manually. If the abnormal data information at each restart is the same, the abnormal data information at the restart is used as the abnormal data information to be matched. If the abnormal data information at each restart is not the same, the power state detection module is used. The power state detection module monitors the time length during which the power state of the device remains unchanged after each restart, and determines that the last restart is the last restart when the time length during which the power state of the device remains unchanged after a certain restart is greater than or equal to a time length threshold. The abnormal data information at the last restart is used as the abnormal data information to be matched.

[0036] The solution matching module comprises a batch identification selection module, a similarity comparison module, a matching abnormal data information, and a solution selection module. The batch identification selection module is used to obtain each historical abnormal data information of the device with the same batch identification as the device to be recommended in the device fault information library. The similarity comparison module is used to compare the similarity between each historical abnormal data information obtained by the batch identification selection module and the abnormal data information to be matched of the device to be recommended. The matching abnormal data information sorts the comparison results of the similarity comparison module in descending order, and selects the historical abnormal data information ranked first as the matching abnormal data information. The solution selection module selects the solution content of the abnormal device to be recommended from the historical solution information corresponding to the matching abnormal data information.

[0037] The solution selection module includes a symptom manifestation returning module, a symptom manifestation similarity comparison module, a matching degree calculation module, and a solution content recommendation module. The symptom manifestation returning module is configured to receive abnormal symptom manifestations of the to-be-recommended device returned by the user. The symptom manifestation similarity comparison module is configured to compare the similarity of the abnormal symptom manifestations returned by the symptom manifestation returning module and the symptom manifestations in each historical solution information corresponding to the matching abnormal data information. The matching degree calculation module is configured to calculate the matching degree of each historical solution information according to the symptom manifestation similarity, the failure occurrence rate, and the feedback success rate of each historical solution information. The solution content recommendation module is configured to sort each historical solution information in descending order of the matching degree, and recommend the historical solution content corresponding to the historical solution information with the highest matching degree to the user.

[0038] A solution recommendation method based on big data, the recommendation method comprising the following:

[0039] A device failure information library is pre-set, which statistically stores device historical abnormal data information, abnormal symptom manifestation tables, and historical solution information of each batch identifier. One historical abnormal data information corresponds to one abnormal symptom manifestation table, one historical abnormal data information corresponds to multiple historical solution information, one historical solution information includes symptom manifestations, historical solution content, failure occurrence rate, and feedback success rate, and the abnormal symptom manifestation table is a table of the set of symptom manifestations corresponding to multiple historical solution information of historical abnormal data information.

[0040] When the to-be-recommended device starts, self-checking is performed. If the to-be-recommended device detects an abnormal condition, the batch identifier of the to-be-recommended device is collected, and the current abnormal data information is taken as the first abnormal data information.

[0041] If the device can start normally, the running state of the device is continuously monitored. If the device cannot run during the monitoring process, the abnormal data information when the device cannot run is obtained, and the to-be-matched abnormal data information is determined.

[0042] The abnormal data information when the device cannot run is obtained, and the to-be-matched abnormal data information is determined.

[0043] The running state of the device is continuously monitored. If the device cannot run during the monitoring process, the current abnormal data information is collected as the second abnormal data information. The similarity of the first abnormal data information and the second abnormal data information is the first similarity.

[0044] If the first similarity is greater than or equal to the first similarity threshold, the first abnormal data information is taken as the to-be-matched abnormal data information.

[0045] If the first similarity is less than the first similarity threshold, the control device automatically restarts, and the abnormal data information after the device is restarted is taken as the abnormal data information to be matched.

[0046] If the device cannot be normally started, the restart state of the device and the abnormal data information at each restart are monitored, and the abnormal data information to be matched is determined according to the restart state and the abnormal data information at each restart;

[0047] The monitoring of the restart state of the device and the abnormal data information at each restart and the determination of the abnormal data information to be matched according to the restart state and the abnormal data information at each restart include:

[0048] If the device cannot be normally started, the restart state of the device within a preset time period is monitored, if the number of times that the time interval between two restarts within the preset time period is not equal is greater than or equal to a first number threshold, the abnormal data information at each restart is obtained, if the abnormal data information at each restart is the same, the abnormal data information at the restart is taken as the abnormal data information to be matched, if the abnormal data information at each restart is not the same, the length of time that the power state of the device remains unchanged after each restart is monitored, if there is a length of time that the power state of the device remains unchanged after a restart is greater than or equal to a length threshold, the restart is the last restart, and the abnormal data information at the last restart is taken as the abnormal data information to be matched; if the number of times that the time interval between two restarts within the preset time period is equal is less than a number threshold, the abnormal data information at the first restart is taken as the abnormal data information to be matched.

[0049] For the abnormal data information to be matched, a historical abnormal data information is selected from a device fault information library as a matching abnormal data information, and a corresponding historical solution content is selected from the historical abnormal data information:

[0050] The selection of the historical abnormal data information from the device fault information library as the matching abnormal data information includes:

[0051] The historical abnormal data information of the device with the same batch identifier as the device to be recommended in the device fault information library is obtained, and the similarity of each historical abnormal data information to the abnormal data information to be matched of the device to be recommended is compared, each similarity compared is sorted in descending order, and the historical abnormal data information ranked first is selected as the matching abnormal data information.

[0052] The selection of the corresponding historical solution content from the historical abnormal data information includes:

[0053] The abnormal symptom manifestation table corresponding to the historical abnormal data information is transmitted to the user, and the user selects the corresponding abnormal symptom manifestation of the device to be recommended;

[0054] Comparing the similarity m of the symptom performance in each historical solution information with the symptom performance selected by the user,

[0055] Then the matching degree of each historical solution information

[0056] Z=0.6*m+0.3*n+0.1*q, wherein n is the failure occurrence rate of the historical solution information, and q is the feedback success rate of the historical solution information;

[0057] Ranking each historical solution information according to the matching degree from large to small, and recommending the historical solution content corresponding to the historical solution information ranked first to the user

[0058] It is apparent for those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, but can be implemented in other concrete forms without departing from the spirit or essential characteristics of the present application. Therefore, the embodiments should be considered in all aspects as illustrative and not restrictive, and the scope of the present application is defined by the appended claims rather than the above description, and it is intended to embrace all changes falling within the meaning and range of equivalents of the elements of the claims. Any reference signs in the claims should not be considered as limiting the claims involved.

Claims

1. A big data based solution recommendation system, characterized in that: The recommendation system comprises a device fault information library, a self-checking detection module, a normal start detection module, an abnormal start detection module and a solution matching module. The device fault information library is used for statistically storing device historical abnormal data information, abnormal symptom manifestation tables and historical solution information of each batch identification. One historical abnormal data information corresponds to one abnormal symptom manifestation table, one historical abnormal data information corresponds to multiple historical solution information, one historical solution information comprises symptom manifestation, historical solution content, fault occurrence rate and feedback success rate, and the abnormal symptom manifestation table is a table of the set of symptom manifestations of the multiple historical solution information corresponding to the historical abnormal data information. The self-checking detection module performs self-checking when the device to be recommended starts, and collects the batch identification of the device to be recommended when the device to be recommended detects an abnormal condition, and takes the current abnormal data information as the first abnormal data information. The normal start detection module continuously monitors the running state of the device when the device can start normally, and obtains abnormal data information when the device cannot run during the monitoring process, and determines the abnormal data information to be matched. The abnormal start detection module monitors the restart state of the device and the abnormal data information at each restart when the device cannot start normally, and determines the abnormal data information to be matched accordingly. The solution matching module selects one historical abnormal data information from the device fault information library as the matched abnormal data information, and selects the corresponding historical solution content from the historical abnormal data information. The abnormal start detection module comprises a restart state detection module and a human restart abnormality detection module. The restart state detection module monitors the restart state of the device to be recommended within a preset time period when the device cannot start normally, and determines that the device restarts as human restart when the number of time intervals between two restarts within the preset time period is not equal to or greater than a first number threshold. The human restart abnormality detection module works when the number of time intervals between two restarts within the preset time period is equal to or less than the first number threshold, determines that the device restarts as automatic restart of the device, takes the abnormal data information at the first restart as the abnormal data information to be matched. The human restart abnormality detection module comprises a restart data information comparison module and a power state detection module. The restart data information comparison module is used for obtaining the abnormal data information at each restart when it is determined that the device restarts as human restart. If the abnormal data information at each restart is the same, the abnormal data information at the restart is taken as the abnormal data information to be matched. If the abnormal data information at each restart is not the same, the power state detection module is enabled. The power state detection module monitors the time length during which the power state of the device remains unchanged after each restart, and takes the last restart as the last restart when the time length during which the power state of the device remains unchanged after a certain restart is greater than or equal to a time length threshold, and takes the abnormal data information at the last restart as the abnormal data information to be matched.

2. The big data based solution recommendation system according to claim 1, characterized in that: The solution matching module comprises a batch identification selection module, an abnormal similarity comparison module, a matching abnormal data information and a solution selection module, the batch identification selection module is used for obtaining each historical abnormal data information of the same equipment as the to-be-recommended equipment batch identification in the equipment fault information library, the abnormal similarity comparison module is used for comparing the similarity of each historical abnormal data information obtained by the batch identification selection module and the to-be-matched abnormal data information of the to-be-recommended equipment, the matching abnormal data information sorts the comparison results of the similarity comparison module in descending order, and selects the historical abnormal data information ranked first as the matching abnormal data information, and the solution selection module selects the solution content of the abnormality of the to-be-recommended equipment from the historical solution information corresponding to the matching abnormal data information.

3. The big data based solution recommendation system according to claim 2, characterized in that: The solution selection module comprises a symptom manifestation return module, a symptom manifestation similarity comparison module, a matching degree calculation module and a solution content recommendation module, the symptom manifestation return module is used for receiving the abnormal symptom manifestation of the to-be-recommended equipment returned by the user, the symptom manifestation similarity comparison module is used for comparing the similarity of the abnormal symptom manifestation returned by the symptom manifestation return module and the symptom manifestation in each historical solution information corresponding to the matching abnormal data information, the matching degree calculation module calculates the matching degree of each historical solution information according to the symptom manifestation similarity, the fault occurrence rate and the feedback success rate of each historical solution information, and the solution content recommendation module sorts each historical solution information in descending order of matching degree, and selects the historical solution content corresponding to the historical solution information ranked first to recommend to the user.

Images