CN108196986A - Unit exception detection method, device, computer equipment and storage medium - Google Patents

Abstract

The present invention proposes an equipment abnormality detection method, device, computer equipment, and storage medium, wherein the method includes: acquiring a time-domain monitoring signal obtained by monitoring the equipment; performing frequency-domain transformation on the time-domain monitoring signal to obtain a frequency-domain monitoring signal ;According to the dimensionless processing algorithm, adjust the amplitudes of frequency domain components at all levels in the frequency domain monitoring signal; input the frequency domain components at all levels after amplitude adjustment in the frequency domain monitoring signal into the pre-trained anomaly detection model, and get anomaly Detection results; wherein, the anomaly detection model has learned the corresponding relationship between the frequency domain components at all levels adjusted by the dimensionless processing algorithm and the anomaly detection results. This method can realize real-time and automatic anomaly detection for massive devices, and improve the efficiency and accuracy of result detection.

Description

Device abnormality detection method, device, computer device and storage medium

technical field

The present invention relates to the technical field of information processing, in particular to an equipment abnormality detection method, device, computer equipment and storage medium.

Background technique

With the continuous development and innovation of computer technology, Internet of Things technology has made great progress in many fields. At present, the number of devices connected to the Internet of Things is relatively large. How to manage and monitor the status of each device in real time and detect whether the device is abnormal has very far-reaching significance.

In the prior art, for a sensor whose output signal is a periodic function in the device, the existing abnormal causes are manually summarized, and then according to the frequency spectrum of the sensor output signal, it is judged whether the sensor is abnormal. In this way, by manually judging whether the sensor is abnormal, the efficiency and accuracy are low, and the workload is large, which is not suitable for the detection of massive equipment.

Contents of the invention

The present invention aims to solve one of the technical problems in the related art at least to a certain extent.

For this reason, the first purpose of the present invention is to propose a device anomaly detection method to realize real-time and automatic anomaly detection of massive equipment, improve the efficiency and accuracy of result detection, and solve the problem of existing manual judgment. Whether the sensor is abnormal, the efficiency and accuracy are low, and the workload is heavy, so it is not suitable for technical problems in the detection of massive equipment.

The second object of the present invention is to provide a device abnormality detection device.

A third object of the present invention is to propose a computer device.

A fourth object of the present invention is to provide a computer-readable storage medium.

A fifth object of the present invention is to provide a computer program product.

In order to achieve the above purpose, the embodiment of the first aspect of the present invention proposes a device abnormality detection method, including:

Obtain the time-domain monitoring signal obtained by monitoring the equipment;

performing frequency-domain transformation on the time-domain monitoring signal to obtain a frequency-domain monitoring signal;

According to the dimensionless processing algorithm, adjust the amplitudes of the frequency domain components at all levels in the frequency domain monitoring signal;

In the frequency domain monitoring signal, the amplitude-adjusted frequency domain components at all levels are input into the pre-trained abnormality detection model to obtain the abnormality detection result; wherein, the abnormality detection model has been learned and adopts the dimensionless processing algorithm Correspondence between adjusted frequency-domain components of each level and anomaly detection results.

The equipment anomaly detection method in the embodiment of the present invention converts the single-dimensional time-domain monitoring signal obtained by monitoring the equipment into a multi-dimensional frequency-domain monitoring signal, and then can obtain more amplitude-frequency features, and then use the dimensionless processing algorithm , adjusting the amplitudes of the frequency domain components at all levels in the frequency domain monitoring signal can amplify the amplitude characteristics of the frequency domain components corresponding to the abnormal data, and finally use the pre-trained anomaly detection model to detect the frequency domain monitoring signal, and get The detection results can realize real-time and automatic anomaly detection for massive devices. In addition, since the anomaly detection model has learned the corresponding relationship between the frequency domain components adjusted by the dimensionless processing algorithm and the anomaly detection results, the frequency domain monitoring signal can be detected according to the anomaly detection model, which can improve the detection results. It solves the technical problems in the prior art that manually judge whether the sensor is abnormal, the efficiency and accuracy are low, and the workload is large, and it is not suitable for the detection of massive equipment.

In order to achieve the above purpose, the embodiment of the second aspect of the present invention proposes an equipment abnormality detection device, including:

An acquisition module, configured to acquire a time-domain monitoring signal obtained by monitoring the equipment;

A transformation module, configured to perform frequency-domain transformation on the time-domain monitoring signal to obtain a frequency-domain monitoring signal;

An adjustment module, configured to adjust the amplitudes of frequency domain components at all levels in the frequency domain monitoring signal according to a dimensionless processing algorithm;

The detection module is used to input the amplitude-adjusted frequency domain components of each level in the frequency domain monitoring signal into the pre-trained abnormality detection model to obtain the abnormality detection result; wherein, the abnormality detection model has been learned and adopted the The corresponding relationship between the adjusted frequency domain components of each level and the anomaly detection results after the dimensionless processing algorithm is described.

The equipment anomaly detection device of the embodiment of the present invention converts the single-dimensional time-domain monitoring signal obtained by monitoring the equipment into a multi-dimensional frequency-domain monitoring signal, and then can obtain more amplitude-frequency characteristics, and then use the dimensionless processing algorithm , adjusting the amplitudes of the frequency domain components at all levels in the frequency domain monitoring signal can amplify the amplitude characteristics of the frequency domain components corresponding to the abnormal data, and finally use the pre-trained anomaly detection model to detect the frequency domain monitoring signal, and get The detection results can realize real-time and automatic anomaly detection for massive devices. In addition, since the anomaly detection model has learned the corresponding relationship between the frequency domain components adjusted by the dimensionless processing algorithm and the anomaly detection results, the frequency domain monitoring signal can be detected according to the anomaly detection model, which can improve the detection results. It solves the technical problems in the prior art that manually judge whether the sensor is abnormal, the efficiency and accuracy are low, and the workload is large, and it is not suitable for the detection of massive equipment.

To achieve the above object, the embodiment of the third aspect of the present invention proposes a computer device, including: a memory, a processor, and a computer program stored in the memory and operable on the processor, when the processor executes the program , implementing the device abnormality detection method described in the embodiment of the first aspect of the present invention.

In order to achieve the above object, the embodiment of the fourth aspect of the present invention provides a computer-readable storage medium on which a computer program is stored, and the feature is that when the program is executed by a processor, the computer program described in the embodiment of the first aspect of the present invention is implemented. The above-mentioned equipment anomaly detection method.

In order to achieve the above purpose, the embodiment of the fifth aspect of the present invention provides a computer program product, when the instructions in the computer program product are executed by the processor, the device abnormality detection as described in the embodiment of the first aspect of the present invention is performed method.

Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.

Description of drawings

The above and/or additional aspects and advantages of the present invention will become apparent and easy to understand from the following description of the embodiments in conjunction with the accompanying drawings, wherein:

FIG. 1 is a schematic flow chart of a first method for detecting equipment abnormality provided by an embodiment of the present invention;

Fig. 2 is a schematic diagram of Fourier series expansion of the time domain monitoring signal in the embodiment of the present invention;

FIG. 3 is a schematic flowchart of a second method for detecting an abnormality in equipment provided by an embodiment of the present invention;

FIG. 4 is a schematic structural diagram of an equipment abnormality detection device provided by an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of another equipment abnormality detection device provided by an embodiment of the present invention;

Figure 6 shows a block diagram of an exemplary computer device suitable for implementing embodiments of the invention.

Detailed ways

Embodiments of the present invention are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals designate the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary and are intended to explain the present invention and should not be construed as limiting the present invention.

Aiming at the technical problems in the prior art that manually judge whether the sensor is abnormal, the efficiency and accuracy are low, and the workload is large, and it is not suitable for the detection of massive equipment. In the embodiment of the present invention, the equipment is monitored to obtain The single-dimensional time-domain monitoring signal is converted into a multi-dimensional frequency-domain monitoring signal, and then more amplitude-frequency characteristics can be obtained, and then the dimensionless processing algorithm is used to adjust the amplitudes of frequency-domain components at all levels in the frequency-domain monitoring signal, which can be Amplify the amplitude characteristics of the frequency domain component corresponding to the abnormal data, and finally use the pre-trained abnormal detection model to detect the frequency domain monitoring signal and obtain the detection result, which can realize real-time and automatic abnormal detection for massive equipment. In addition, since the anomaly detection model has learned the corresponding relationship between the frequency domain components adjusted by the dimensionless processing algorithm and the anomaly detection results, the frequency domain monitoring signal can be detected according to the anomaly detection model, which can improve the detection results. efficiency and accuracy.

The equipment anomaly detection method, device, computer equipment and storage medium in the embodiments of the present invention will be described below with reference to the accompanying drawings.

FIG. 1 is a schematic flowchart of a first method for detecting device abnormality provided by an embodiment of the present invention.

As shown in Figure 1, the device anomaly detection method includes the following steps:

In step 101, a time-domain monitoring signal obtained by monitoring a device is obtained.

The device abnormality detection method of the embodiment of the present invention can be used to detect whether the sensor whose output signal is a periodic function in the device is abnormal. When abnormality detection is performed on this type of sensor, the output signal of this type of sensor can be obtained, which is recorded as a time-domain monitoring signal in the embodiment of the present invention. Optionally, the time-domain monitoring signal is marked as f(x).

Step 102, performing frequency domain transformation on the time domain monitoring signal to obtain the frequency domain monitoring signal.

In the embodiment of the present invention, since the acquired time-domain monitoring signal is a periodic function, Fourier series expansion can be performed on the time-domain monitoring signal to obtain a frequency-domain monitoring signal, and the frequency-domain monitoring signal is an infinite number of sine waves The combination.

Specifically, the time-domain monitoring signal f(x) is expanded by Fourier series, and the obtained frequency-domain monitoring signal is:

Among them, k is the frequency domain component The series, A _k is the amplitude.

As an example, refer to FIG. 2 , which is a schematic diagram of Fourier series expansion of the time-domain monitoring signal in an embodiment of the present invention. Waveform 1 represents the time-domain monitoring signal, and waveforms 2, 3, 4, etc. represent the frequency-domain components of each level in the frequency-domain monitoring signal obtained after Fourier series expansion of the time-domain monitoring signal.

Further, since f(x) is a periodic function, the natural frequency is relatively obvious, and the natural frequency is in a lower frequency band. Therefore, in the embodiment of the present invention, in order to reduce the workload of the system and improve the processing efficiency, the frequency domain components of all levels of the frequency domain monitoring signal can be screened, and the frequency domain components of the preset levels are reserved, wherein the reserved preset The frequency component of the series includes the frequency domain component of the natural frequency, for example, mark the preset series as M. Further, since the high-frequency components in the frequency domain components are noise, in the embodiment of the present invention, denoising can be performed on the frequency domain components of all levels of the frequency domain monitoring signal, so as to filter out frequency components whose frequency is higher than the preset threshold. domain components, thereby improving the accuracy of detection results. Wherein, the preset threshold can be set according to the application scenario of the device.

Step 103, according to the dimensionless processing algorithm, adjust the amplitudes of the frequency domain components of each level in the frequency domain monitoring signal.

Wherein, the dimensionless processing algorithm is used to make the adjusted amplitudes of the frequency domain components at all levels become scalar.

In the embodiment of the present invention, according to the dimensionless processing algorithm, after adjusting the amplitudes of the frequency domain components at all levels in the frequency domain monitoring signal, the amplitude characteristics of the frequency domain components of the natural frequency can be weakened, and at the same time, the frequency domain component amplitudes corresponding to the abnormal data can be amplified. value feature. For example, the dimensionless processing algorithm can be a normalization algorithm, or the dimensionless processing algorithm can be any other algorithm that can weaken the amplitude characteristics of the frequency domain components of the natural frequency, and at the same time amplify the amplitude characteristics of the frequency domain components corresponding to the abnormal data. The embodiments of the invention do not limit this.

In the embodiment of the present invention, after the frequency-domain monitoring signal is obtained, the amplitudes of frequency-domain components at all levels in the frequency-domain monitoring signal can be adjusted according to a dimensionless processing algorithm to obtain the adjusted frequency-domain monitoring signal. Optionally, mark the adjusted frequency-domain monitoring signal as f'(x).

As a possible implementation, the normalization algorithm can be used to adjust the amplitudes of the frequency domain components at all levels in the frequency domain monitoring signal, so that the amplitudes of the frequency domain components of the adjusted natural frequency are weakened, and the abnormal data corresponding to The amplitude characteristics of the frequency domain components are amplified.

Optionally, when the dimensionless processing algorithm is a normalization algorithm, assuming that there is no abnormality in the sensor in the device, the frequency domain monitoring signal is:

Among them, since f ₁ (x) is a periodic function, the natural frequency is more obvious, and because the high-frequency component is noise, therefore, the natural frequency is in the lower frequency band, and the amplitude of the frequency domain component marking the natural frequency is a _f , then a _f >>a _k (k≠f).

Using the normalization algorithm, normalize each level of frequency domain components in f ₁ (x) so that the coefficient of each level of frequency domain components is 1, and record so that the coefficient of each level of frequency domain components is The normalization parameter a _k used when 1.

It should be noted that the normalization parameter a _k , k=1, 2, . . . , that is, each level of frequency domain component has a corresponding normalization parameter.

After normalization, the frequency domain monitoring signal is:

It can be seen that, relative to the natural frequency, before normalization, the ratio of the amplitude of the frequency domain component of the non-natural frequency to the amplitude of the frequency domain component of the natural frequency is Since a _f >>a _k (k≠f), the amplitude characteristics of the frequency domain components of the natural frequency are more obvious, and after normalization, the frequency domain component amplitude of the non-natural frequency and the frequency domain component amplitude of the natural frequency The ratio of the values becomes 1 (the amplitude of the natural frequency is 1), and the amplitude characteristic of the frequency domain component of the natural frequency is weakened.

When the sensor in the equipment is abnormal, it is assumed that the frequency domain monitoring signal is:

where b _k ≈ a _k , for abnormal data.

Using the normalization algorithm, each level of frequency domain component in f ₂ (x) is divided by the corresponding normalization parameter a _k , and the normalized frequency domain monitoring signal is obtained as:

in, Relative to the natural frequency, before normalization, the ratio of the amplitude of the frequency domain component corresponding to the abnormal data to the amplitude of the frequency domain component of the natural frequency is After normalization, the ratio of the amplitude of the frequency domain component corresponding to the abnormal data to the amplitude of the frequency domain component of the natural frequency becomes (The magnitude of the natural frequency is 1). Since a _f >>a _k (k≠f), we have Therefore, after normalization, the ratio of the amplitude of the frequency-domain component of the abnormal data to the amplitude of the frequency-domain component of the natural frequency increases compared with that before adjustment, that is, the amplitude characteristic of the frequency-domain component corresponding to the abnormal data is amplified.

In step 104, the amplitude-adjusted frequency domain components of each level in the frequency domain monitoring signal are input into the pre-trained anomaly detection model to obtain an anomaly detection result.

Among them, the anomaly detection model has learned the corresponding relationship between the adjusted frequency domain components at all levels and the anomaly detection results using the dimensionless processing algorithm.

It can be seen from step 103 that when the sensor in the device is abnormal, the acquired time domain monitoring signal carries abnormal data, and after adjusting the amplitudes of frequency domain components at all levels in the frequency domain monitoring signal according to the dimensionless processing algorithm, the abnormal data corresponds to The amplitude characteristics of the frequency domain components of the frequency domain are amplified, and at the same time, the amplitude characteristics of the frequency domain components of the natural frequency are also weakened. Therefore, in the frequency domain monitoring signal, the frequency domain components of all levels after amplitude adjustment are input into the pre-trained anomaly detection In the model, anomaly detection results can be obtained. Since the anomaly detection model has learned the corresponding relationship between the frequency domain components adjusted by the dimensionless processing algorithm and the anomaly detection results, the frequency domain monitoring signal can be detected according to the anomaly detection model, which can improve the efficiency of result detection and accuracy.

The equipment anomaly detection method of this embodiment converts the single-dimensional time-domain monitoring signal obtained by monitoring the equipment into a multi-dimensional frequency-domain monitoring signal, thereby obtaining more amplitude-frequency features, and then using a dimensionless processing algorithm, Adjusting the amplitudes of the frequency domain components at all levels in the frequency domain monitoring signal can amplify the amplitude characteristics of the frequency domain components corresponding to the abnormal data, and finally use the pre-trained abnormal detection model to detect the frequency domain monitoring signal and obtain the detection As a result, real-time, automatic anomaly detection for massive devices can be realized. In addition, since the anomaly detection model has learned the corresponding relationship between the frequency domain components adjusted by the dimensionless processing algorithm and the anomaly detection results, the frequency domain monitoring signal can be detected according to the anomaly detection model, which can improve the detection results. efficiency and accuracy.

In the embodiment of the present invention, before the detection, the abnormality detection model may be trained in advance. The above process will be described in detail below with reference to FIG. 3 .

FIG. 3 is a schematic flow chart of a second method for detecting device abnormality provided by an embodiment of the present invention.

As shown in Figure 3, before step 104, the device abnormality detection method may also include the following steps:

Step 201, acquiring the time-domain historical signals monitored during the historical operation of the equipment.

In the embodiment of the present invention, during the operation of the system monitoring equipment, the signal output by the sensor whose output signal is a periodic function in the equipment can be saved, so that when training the abnormality detection model, the time domain monitored during the historical operation of the equipment can be obtained historical signal.

Step 202, performing frequency domain transformation on the time domain historical signal to obtain the frequency domain historical signal.

In the embodiment of the present invention, Fourier series expansion may be performed on the time-domain historical signal to obtain the frequency-domain historical signal, and the frequency-domain historical signal is a combination of infinitely many sine waves.

It can be understood that by transforming the single-dimensional time-domain historical signal into a multi-dimensional frequency-domain historical signal, more amplitude-frequency features can be obtained, so that when the abnormality detection model is used for abnormality detection, the accuracy of the detection result can be improved.

Step 203, according to the dimensionless processing algorithm, adjust the amplitudes of the frequency domain components of each level in the frequency domain history signal.

In the embodiment of the present invention, according to the dimensionless processing algorithm, after adjusting the amplitudes of the frequency domain components at all levels in the frequency domain historical signal, the amplitude characteristics of the frequency domain components of the natural frequency can be weakened, and at the same time, the frequency domain component amplitudes corresponding to the abnormal data can be amplified. value feature. For example, the dimensionless processing algorithm can be a normalization algorithm, or the dimensionless processing algorithm can be any other algorithm that can weaken the amplitude characteristics of the frequency domain components of the natural frequency, and at the same time amplify the amplitude characteristics of the frequency domain components corresponding to the abnormal data. The embodiments of the invention do not limit this.

In the embodiment of the present invention, after the frequency domain historical signal is obtained, the amplitudes of the frequency domain components at all levels in the frequency domain historical signal can be adjusted according to the dimensionless processing algorithm to obtain the adjusted frequency domain historical signal.

As a possible implementation, the normalization algorithm can be used to adjust the amplitudes of the frequency domain components at all levels in the frequency domain historical signal, so that the amplitude characteristics of the frequency domain components of the adjusted natural frequency are weakened, and the abnormal data corresponding to The amplitude characteristics of the frequency domain components are amplified.

In step 204, an abnormality detection model is trained according to the amplitude-adjusted frequency domain components of each level in the frequency domain historical signal and the corresponding historical operating status of the equipment.

Wherein, the historical running state includes normal state and abnormal state.

Specifically, according to the adjusted frequency domain components at all levels corresponding to the historical signal amplitudes corresponding to the normal state, a positive sample for training the anomaly detection model is generated, and the positive sample is used to train the anomaly detection model.

In the embodiment of the present invention, the anomaly detection model is trained by adjusting the frequency domain components at all levels adjusted according to the amplitudes in the frequency domain historical signal and the corresponding historical operating status of the equipment, so that the anomaly detection model can be learned and obtained using dimensionless processing. The corresponding relationship between the frequency domain components at all levels and the abnormality detection results adjusted by the algorithm can improve the efficiency and accuracy of the result detection when detecting the frequency domain monitoring signals according to the abnormality detection model.

As a possible implementation, regression prediction can be used to train the anomaly detection model using the Long Short Term Memory (LSTM) neural network, so that the learning of the anomaly detection model using the LSTM neural network can be obtained using dimensionless The corresponding relationship between the frequency domain components at all levels adjusted by the processing algorithm and the abnormality detection results.

The equipment anomaly detection method of this embodiment can obtain more amplitude-frequency features by transforming the single-dimensional time-domain historical signal into a multi-dimensional frequency-domain historical signal, so that when using the anomaly detection model for anomaly detection, the detection result can be improved. the accuracy. According to the dimensionless processing algorithm, adjusting the amplitudes of the frequency domain components at all levels in the frequency domain historical signal can amplify the amplitude characteristics of the frequency domain components corresponding to the abnormal data, and finally according to the adjusted amplitudes of the frequency domain historical signals. Level frequency domain components, and the corresponding historical operating status of the equipment, to train the anomaly detection model, so that the anomaly detection model can learn to obtain the correspondence between the frequency domain components of each level adjusted by the dimensionless processing algorithm and the abnormality detection results relationship, so that the efficiency and accuracy of the result detection can be improved when detecting the frequency domain monitoring signal by using the anomaly detection model.

In order to realize the above embodiments, the present invention further proposes a device abnormality detection device.

Fig. 4 is a schematic structural diagram of a device abnormality detection device provided by an embodiment of the present invention.

As shown in FIG. 4 , the device anomaly detection device 400 includes: an acquisition module 410 , a transformation module 420 , an adjustment module 430 , and a detection module 440 . in,

The obtaining module 410 is configured to obtain a time-domain monitoring signal obtained by monitoring the device.

The transformation module 420 is configured to perform frequency domain transformation on the time domain monitoring signal to obtain the frequency domain monitoring signal.

As a possible implementation, the transformation module 420 is specifically used to perform Fourier series expansion on the time domain monitoring signal to obtain the frequency domain monitoring signal Among them, k is the frequency domain component The series, A _k is the amplitude.

The adjustment module 430 is configured to adjust the amplitudes of frequency domain components at various levels in the frequency domain monitoring signal according to a dimensionless processing algorithm.

As a possible implementation manner, the adjustment module 430 is specifically configured to use a normalization algorithm to adjust the amplitudes of frequency domain components at various levels in the frequency domain monitoring signal.

The detection module 440 is used to input the amplitude-adjusted frequency domain components of all levels in the frequency domain monitoring signal into the pre-trained anomaly detection model to obtain anomaly detection results; wherein, the anomaly detection model has been learned and adopts a dimensionless processing algorithm Correspondence between adjusted frequency-domain components of each level and anomaly detection results.

Further, in a possible implementation of the embodiment of the present invention, referring to FIG. 5, on the basis of the embodiment shown in FIG. 460.

The screening and denoising module 450 is used to perform frequency domain transformation on the time domain monitoring signal to obtain the frequency domain monitoring signal, and to filter the frequency domain components of the frequency domain monitoring signal at all levels, and retain the frequency domain components of the preset series; And/or, denoising is performed on frequency domain components of various levels of the frequency domain monitoring signal, so as to filter out frequency domain components whose frequency is higher than a preset threshold.

The training module 460 is used to obtain the time-domain historical signals monitored during the historical operation of the equipment; perform frequency-domain transformation on the time-domain historical signals to obtain the frequency-domain historical signals; according to the dimensionless processing algorithm, adjust The amplitude of the frequency domain component; according to the adjusted frequency domain components at all levels in the frequency domain historical signal and the corresponding historical operating status of the equipment, the abnormal detection model is trained; the historical operating status includes normal status and abnormal status.

It should be noted that the foregoing explanations on the embodiment of the device abnormality detection method are also applicable to the device anomaly detection apparatus 400 of this embodiment, and will not be repeated here.

The equipment anomaly detection device in this embodiment converts the single-dimensional time-domain monitoring signal obtained by monitoring the equipment into a multi-dimensional frequency-domain monitoring signal, thereby obtaining more amplitude-frequency characteristics, and then using a dimensionless processing algorithm, Adjusting the amplitudes of the frequency domain components at all levels in the frequency domain monitoring signal can amplify the amplitude characteristics of the frequency domain components corresponding to the abnormal data, and finally use the pre-trained abnormal detection model to detect the frequency domain monitoring signal and obtain the detection As a result, real-time, automatic anomaly detection for massive devices can be realized. In addition, since the anomaly detection model has learned the corresponding relationship between the frequency domain components adjusted by the dimensionless processing algorithm and the anomaly detection results, the frequency domain monitoring signal can be detected according to the anomaly detection model, which can improve the detection results. efficiency and accuracy.

In order to realize the above embodiments, the present invention also proposes a computer device, including: a memory and a processor, wherein the processor runs the executable program code by reading the executable program code stored in the memory The corresponding program is used to execute the device abnormality detection method described in the foregoing embodiments.

In order to clearly illustrate the specific structure of the foregoing computer devices, FIG. 6 shows a block diagram of an exemplary computer device 12 suitable for implementing the embodiments of the present invention. The computer device 12 shown in FIG. 6 is only an example, and should not limit the functions and scope of use of this embodiment of the present invention.

As shown in FIG. 6, computer device 12 takes the form of a general-purpose computer device. Components of computer device 12 may include, but are not limited to: one or more processors or processing units 16 , system memory 28 , bus 18 connecting various system components including system memory 28 and processing unit 16 .

Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus structures. For example, these architectures include but are not limited to Industry Standard Architecture (Industry Standard Architecture, referred to as ISA) bus, Micro Channel Architecture (Micro Channel Architecture, referred to as MAC) bus, enhanced ISA bus, Video Electronic Standard (Vedio Electronic Standard Association (referred to as VESA) local bus and peripheral component interconnect (Peripheral Component Interconnect, referred to as PCI) bus.

Computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by computer device 12 and include both volatile and nonvolatile media, removable and non-removable media.

The system memory 28 may include a computer system readable medium in the form of a volatile memory, such as a random access memory (Random Access Memory, RAM for short) 30 and/or a cache memory 32 . Computer device 12 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 34 may be used to read and write to non-removable, non-volatile magnetic media (not shown in FIG. 6, commonly referred to as a "hard drive"). Although not shown in FIG. 6, a disk drive for reading and writing to removable non-volatile disks (such as "floppy disks") may be provided, as well as for removable non-volatile optical disks (such as CD-ROM, DVD-ROM or other optical media) CD-ROM drive. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. Memory 28 may include at least one program product having a set (eg, at least one) of program modules configured to perform the functions of various embodiments of the present invention.

A program/utility 40 having a set (at least one) of program modules 42 may be stored, for example, in memory 28, such program modules 42 including but not limited to an operating system, one or more application programs, other program modules, and program data , each or some combination of these examples may include implementations of network environments. Program modules 42 generally perform the functions and/or methodologies of the described embodiments of the invention.

The computer device 12 may also communicate with one or more external devices 14 (e.g., a keyboard, pointing device, display 24, etc.), and with one or more devices that enable a user to interact with the computer device 12, and/or with Any device (eg, network card, modem, etc.) that enables the computing device 12 to communicate with one or more other computing devices. Such communication may occur through input/output (I/O) interface 22 . Also, the computer device 12 can also communicate with one or more networks (eg, a local area network, a wide area network, and/or a public network, such as the Internet) through the network adapter 20 . As shown, network adapter 20 communicates with other modules of computer device 12 via bus 18 . It should be appreciated that although not shown, other hardware and/or software modules may be used in conjunction with computer device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, redundant array of independent disks (RedundantArray of Independent Disks, referred to as RAID) systems, tape drives, and data backup storage systems.

The processing unit 16 executes various functional applications and data processing by running the programs stored in the system memory 28 to realize the above-mentioned device abnormality detection method.

To achieve the above purpose, the present invention also proposes a computer program product. When the instructions in the computer program product are executed by the processor, the device abnormality detection method described in the foregoing embodiments is executed.

In order to realize the above-mentioned embodiments, the present invention also proposes a non-transitory computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, it can realize the device abnormality detection method as described in the above-mentioned embodiments .

In the description of this specification, descriptions referring to the terms "one embodiment", "some embodiments", "example", "specific examples", or "some examples" mean that specific features described in connection with the embodiment or example , structure, material or characteristic is included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the described specific features, structures, materials or characteristics may be combined in any suitable manner in any one or more embodiments or examples. In addition, those skilled in the art can combine and combine different embodiments or examples and features of different embodiments or examples described in this specification without conflicting with each other.

In addition, the terms "first" and "second" are used for descriptive purposes only, and cannot be interpreted as indicating or implying relative importance or implicitly specifying the quantity of indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include at least one of these features. In the description of the present invention, "plurality" means at least two, such as two, three, etc., unless otherwise specifically defined.

Any process or method descriptions in flowcharts or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing custom logical functions or steps of a process , and the scope of preferred embodiments of the invention includes alternative implementations in which functions may be performed out of the order shown or discussed, including substantially concurrently or in reverse order depending on the functions involved, which shall It is understood by those skilled in the art to which the embodiments of the present invention pertain.

The logic and/or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium, For use with instruction execution systems, devices, or devices (such as computer-based systems, systems including processors, or other systems that can fetch instructions from instruction execution systems, devices, or devices and execute instructions), or in conjunction with these instruction execution systems, devices or equipment used. For the purposes of this specification, a "computer-readable medium" may be any device that can contain, store, communicate, propagate or transmit a program for use in or in conjunction with an instruction execution system, device, or device. More specific examples (non-exhaustive list) of computer-readable media include the following: electrical connection with one or more wires (electronic device), portable computer disk case (magnetic device), random access memory (RAM), Read Only Memory (ROM), Erasable and Editable Read Only Memory (EPROM or Flash Memory), Fiber Optic Devices, and Portable Compact Disc Read Only Memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program can be printed, since the program can be read, for example, by optically scanning the paper or other medium, followed by editing, interpretation or other suitable processing if necessary. The program is processed electronically and stored in computer memory.

It should be understood that various parts of the present invention can be realized by hardware, software, firmware or their combination. In the embodiments described above, various steps or methods may be implemented by software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented by any one or a combination of the following techniques known in the art: a discrete Logic circuits, ASICs with suitable combinational logic gates, Programmable Gate Arrays (PGA), Field Programmable Gate Arrays (FPGA), etc.

Those of ordinary skill in the art can understand that all or part of the steps carried by the methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. During execution, one or a combination of the steps of the method embodiments is included.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing module, each unit may exist separately physically, or two or more units may be integrated into one module. The above-mentioned integrated modules can be implemented in the form of hardware or in the form of software function modules. If the integrated modules are realized in the form of software function modules and sold or used as independent products, they can also be stored in a computer-readable storage medium.

The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, and the like. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention, those skilled in the art can make the above-mentioned The embodiments are subject to changes, modifications, substitutions and variations.

Claims (10)

1. An apparatus abnormality detection method, characterized in that the method comprises the steps of:

acquiring a time domain monitoring signal obtained by monitoring equipment;

carrying out frequency domain transformation on the time domain monitoring signal to obtain a frequency domain monitoring signal;

adjusting the amplitude of each level of frequency domain component in the frequency domain monitoring signal according to a dimensionless processing algorithm;

inputting each level of frequency domain components with the adjusted amplitudes in the frequency domain monitoring signals into a pre-trained anomaly detection model to obtain anomaly detection results; and the anomaly detection model learns the corresponding relation between each level of frequency domain component adjusted by the dimensionless processing algorithm and the anomaly detection result.

2. The method for detecting device anomalies according to claim 1, wherein the adjusting the amplitudes of the frequency-domain components of each level in the frequency-domain monitoring signal according to a dimensionless processing algorithm includes:

and adjusting the amplitude of each level of frequency domain component in the frequency domain monitoring signal by adopting a normalization algorithm.

3. The method for detecting device anomaly according to claim 1, wherein after the frequency-domain transformation is performed on the time-domain monitoring signal to obtain a frequency-domain monitoring signal, the method further comprises:

screening all levels of frequency domain components of the frequency domain monitoring signal, and reserving frequency domain components of preset levels;

and/or denoising each level of frequency domain components of the frequency domain monitoring signal to filter out frequency domain components with frequencies higher than a preset threshold value.

4. The method according to claim 1, wherein before inputting the frequency domain components of each level, which have been amplitude-adjusted, in the frequency domain monitoring signal into a pre-trained anomaly detection model to obtain an anomaly detection result, the method further comprises:

acquiring a time domain historical signal monitored in the historical operation process of equipment;

performing frequency domain transformation on the time domain historical signal to obtain a frequency domain historical signal;

adjusting the amplitude of each level of frequency domain component in the frequency domain historical signal according to the dimensionless processing algorithm;

training the abnormal detection model according to each level of frequency domain components after amplitude adjustment in the frequency domain historical signal and the historical operating state corresponding to the equipment; the historical operating state comprises a normal state and an abnormal state.

5. The device anomaly detection method according to claim 4, wherein said training of said anomaly detection model comprises:

and training an abnormality detection model adopting an LSTM neural network by adopting a regression prediction mode.

6. The apparatus anomaly detection method according to claim 1, wherein said frequency-domain transforming the time-domain monitor signal to obtain a frequency-domain monitor signal comprises:

fourier series expansion is carried out on the time domain monitoring signal to obtain a frequency domain monitoring signalWhere k is the frequency domain componentNumber of stages of (A)_kIs the amplitude.

7. An apparatus for detecting abnormality of a device, characterized in that the apparatus comprises:

the acquisition module is used for acquiring a time domain monitoring signal obtained by monitoring equipment;

the transformation module is used for carrying out frequency domain transformation on the time domain monitoring signal to obtain a frequency domain monitoring signal;

the adjusting module is used for adjusting the amplitude of each level of frequency domain component in the frequency domain monitoring signal according to a dimensionless processing algorithm;

the detection module is used for inputting each level of frequency domain components with the adjusted amplitudes in the frequency domain monitoring signals into a pre-trained anomaly detection model to obtain anomaly detection results; and the anomaly detection model learns the corresponding relation between each level of frequency domain component adjusted by the dimensionless processing algorithm and the anomaly detection result.

8. A computer device, comprising: memory, processor and computer program stored on the memory and executable on the processor, which when executed by the processor implements the device anomaly detection method according to any one of claims 1 to 6.

9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the device abnormality detection method according to any one of claims 1 to 6.

10. A computer program product, wherein instructions, when executed by a processor, perform the device anomaly detection method of any one of claims 1-6.