CN-122020517-A - Equipment fault early warning method, equipment, storage medium and program product

CN122020517ACN 122020517 ACN122020517 ACN 122020517ACN-122020517-A

Abstract

The application discloses a device fault early warning method, device, storage medium and program product, and relates to the technical field of intelligent home/intelligent families, wherein the device fault early warning method comprises the steps of determining fault information of intelligent home devices through a mixed fault model based on multi-source operation data by acquiring the multi-source operation data of the intelligent home devices, and outputting early warning information based on the fault information; the hybrid fault model is used for determining abnormal information of at least two dimensions of the intelligent home equipment based on the multi-source operation data and carrying out fusion processing on the abnormal information of the at least two dimensions. The method avoids the limitation of fault detection in a single dimension, can identify progressive deterioration and sudden abnormality of the equipment performance in advance, reduces the false alarm rate of faults, ensures the reliability of early warning results and improves the use experience of users.

Inventors

DONG HAITAO

Assignees

海尔优家智能科技（北京）有限公司

Dates

Publication Date: 20260512
Application Date: 20251230

Claims (10)

1. An equipment fault early warning method, which is characterized by comprising the following steps: Acquiring multi-source operation data of intelligent household equipment, wherein the multi-source operation data at least comprises two of performance data, sensor data, log data and user behavior data; Determining fault information of the intelligent home equipment through a mixed fault model based on the multi-source operation data, wherein the mixed fault model is used for determining abnormal information of at least two dimensions of the intelligent home equipment based on the multi-source operation data and carrying out fusion processing on the abnormal information of the at least two dimensions; and outputting early warning information based on the fault information, wherein the early warning information is used for prompting the fault information of the intelligent household equipment.
2. The method of claim 1, wherein the hybrid fault model includes a trend anomaly branch, a transient anomaly branch, and a joint inference branch, wherein the fault information includes a fault type and a fault probability, wherein the determining fault information for the smart home device by the hybrid fault model based on the multi-source operational data includes: Extracting characteristics of the multi-source operation data to obtain time sequence characteristics and statistical characteristics of the multi-source operation data; inputting the time sequence characteristics into the trend abnormal branch to obtain abnormal trend information of the intelligent household equipment, wherein the abnormal trend information is used for reflecting the performance decline trend of the intelligent household equipment; Inputting the statistical characteristics into the transient abnormal branch to obtain burst abnormal information of the intelligent household equipment, wherein the burst abnormal information is used for reflecting the burst fault of the intelligent household equipment; and carrying out fusion processing on the abnormal trend information and the burst abnormal information through the combined reasoning branch, and determining the fault type and the fault probability of the intelligent household equipment.
3. The method of claim 1, wherein outputting the pre-warning information based on the fault information comprises: and determining a fault early warning strategy of the intelligent household equipment based on the fault information, and outputting early warning information according to the fault early warning strategy.
4. The method of claim 3, wherein the fault information comprises a fault type and a fault probability, and wherein the determining a fault early warning strategy for the smart home device based on the fault information comprises: Determining a fault influence range of the intelligent household equipment based on the fault type and the fault probability, wherein the fault influence range is used for representing a specific object and an associated range of the specific object, wherein the specific object is abnormal in equipment function, performance is reduced or service is interrupted due to the fault; And calling a preset early warning relation based on the fault type, the fault probability and the fault influence range, determining a fault early warning level of the intelligent household equipment, and determining the fault early warning strategy corresponding to the fault early warning level.
5. The method of claim 4, wherein the fault early warning level comprises a first early warning level, a second early warning level, a third early warning level, and a fourth early warning level, wherein the fault early warning policy comprises a first early warning policy, a second early warning policy, a third early warning policy, and a fourth early warning policy, and wherein the method further comprises: Under the condition that the fault early warning level is the first early warning level, determining that the fault early warning policy is a first early warning policy, wherein the first early warning policy comprises generating fault prompt information and sending the fault prompt information to a terminal of a first target user; Under the condition that the fault early-warning level is the second early-warning level, determining that the fault early-warning policy is a second early-warning policy, wherein the second early-warning policy comprises generating a fault analysis task and sending the fault analysis task to a terminal of a second target user; determining that the fault early warning strategy is a third early warning strategy under the condition that the fault early warning level is the third early warning level, wherein the third early warning strategy comprises generation of fault early warning information and a fault pre-detection task, sending of the fault early warning information to a terminal of the first target user and sending of the fault pre-detection task to a terminal of the second target user; And under the condition that the fault early warning level is the fourth early warning level, determining that the fault early warning policy is the fourth early warning policy, wherein the fourth early warning policy comprises generating a fault maintenance task and fault maintenance information, sending the fault maintenance information to the terminal of the first target user, and sending the fault maintenance task to the terminal of the second target user.
6. The method of claim 5, wherein the method further comprises: based on the fault influence range, constructing a fault propagation chain model of the intelligent home equipment, wherein the fault propagation chain model is used for quantifying influence paths and influence weights of single equipment faults on other equipment; Simulating a fault propagation path of the intelligent home equipment through the fault propagation chain model based on the multi-source operation data, determining a plurality of fault influencing equipment corresponding to the intelligent home equipment, and respectively determining fault early warning grades of the plurality of fault influencing equipment; And determining a combined early warning strategy of the plurality of fault influencing devices and the intelligent household device based on the fault early warning level of the plurality of fault influencing devices and the fault early warning level of the intelligent household device, wherein the combined early warning strategy is used for indicating the plurality of fault influencing devices and the intelligent household device to cooperatively perform fault early warning.
7. The method according to claim 1, wherein the method further comprises: Determining high-frequency user behaviors of the intelligent home equipment based on the user behavior data; Based on the high-frequency user behavior, adjusting an early warning threshold range corresponding to the multi-source operation data to obtain a target threshold range, wherein the early warning threshold range is used for distinguishing whether the multi-source operation data is abnormal or not; based on the multi-source operation data, determining the fault information of the intelligent home equipment through a hybrid fault model comprises the following steps: And determining fault information of the intelligent household equipment through a hybrid fault model based on the multi-source operation data and the target threshold range.
8. An electronic device comprising a memory and a processor, characterized in that the memory has stored therein a computer program, the processor being arranged to execute the method according to any of claims 1 to 7 by means of the computer program.
9. A computer-readable storage medium, characterized in that the computer-readable storage medium comprises a stored program, wherein the program when run performs the method of any one of claims 1 to 7.
10. A computer program product comprising a computer program which, when executed by a processor, implements the method of any one of claims 1 to 7.

Description

Equipment fault early warning method, equipment, storage medium and program product Technical Field The application relates to the field of intelligent home/smart home, in particular to a device fault early warning method, device, storage medium and program product. Background Along with the rapid development of the Internet of things technology, the intelligent household equipment is widely applied to household life scenes and comprises diversified equipment such as air conditioners, washing machines, water purifiers, intelligent door locks, lighting systems and the like, and the diversified intelligent household equipment realizes remote control and data interaction through the Internet, so that the life convenience of users is remarkably improved. While intelligent home devices are increasingly popular, fault detection and maintenance problems are also increasingly prominent. In the prior art, smart home device fault detection relies on single-dimensional "heartbeat packet" communication status monitoring (e.g., whether the device is online). However, the existing fault detection scheme only depends on communication state detection, so that fault identification is incomplete, false alarm and missing alarm are serious. Disclosure of Invention The application provides a device fault early warning method, device, storage medium and program product, which are used for solving the technical problems of incomplete fault identification, false alarm and serious missing report caused by the fact that the existing fault detection scheme only depends on communication state detection. In a first aspect, the present application provides an apparatus fault early warning method, including: Acquiring multi-source operation data of intelligent household equipment, wherein the multi-source operation data at least comprises two of performance data, sensor data, log data and user behavior data; Determining fault information of the intelligent home equipment through a mixed fault model based on the multi-source operation data, wherein the mixed fault model is used for determining abnormal information of at least two dimensions of the intelligent home equipment based on the multi-source operation data and carrying out fusion processing on the abnormal information of the at least two dimensions; and outputting early warning information based on the fault information, wherein the early warning information is used for prompting the fault information of the intelligent household equipment. Optionally, the hybrid fault model includes a trend anomaly branch, a transient anomaly branch and a joint inference branch, the fault information includes a fault type and a fault probability, and the determining, based on the multi-source operation data, the fault information of the smart home device through the hybrid fault model includes: Extracting characteristics of the multi-source operation data to obtain time sequence characteristics and statistical characteristics of the multi-source operation data; inputting the time sequence characteristics into the trend abnormal branch to obtain abnormal trend information of the intelligent household equipment, wherein the abnormal trend information is used for reflecting the performance decline trend of the intelligent household equipment; Inputting the statistical characteristics into the transient abnormal branch to obtain burst abnormal information of the intelligent household equipment, wherein the burst abnormal information is used for reflecting the burst fault of the intelligent household equipment; and carrying out fusion processing on the abnormal trend information and the burst abnormal information through the combined reasoning branch, and determining the fault type and the fault probability of the intelligent household equipment. Optionally, the outputting the early warning information based on the fault information includes: and determining a fault early warning strategy of the intelligent household equipment based on the fault information, and outputting early warning information according to the fault early warning strategy. Optionally, the fault information comprises a fault type and a fault probability, and the determining the fault early warning strategy of the intelligent home equipment based on the fault information comprises the following steps: Determining a fault influence range of the intelligent household equipment based on the fault type and the fault probability, wherein the fault influence range is used for representing a specific object and an associated range of the specific object, wherein the specific object is abnormal in equipment function, performance is reduced or service is interrupted due to the fault; And calling a preset early warning relation based on the fault type, the fault probability and the fault influence range, determining a fault early warning level of the intelligent household equipment, and determining the fault early warning strategy correspond