CN114462925A - Inventory abnormal asset identification method and device and terminal equipment - Google Patents

Abstract

The invention provides an inventory abnormal asset identification method, an inventory abnormal asset identification device and terminal equipment, wherein the method comprises the following steps: acquiring a circulation track and attribute data of a target asset; calculating asset abnormal probability corresponding to the asset circulation track, extracting characteristic values corresponding to target asset characteristics from the attribute data, and generating characteristic vectors corresponding to target assets on the basis of the asset abnormal probability and the characteristic values corresponding to the target asset characteristics; and inputting the characteristic vector into a preset abnormal asset identification model, and determining an inventory abnormal identification result of the target asset. The method, the device and the terminal equipment for identifying the abnormal inventory assets can improve the efficiency and the accuracy of identifying the abnormal inventory assets.

Description

Inventory abnormal asset identification method, device and terminal equipment

technical field

The invention belongs to the technical field of inventory abnormal asset identification, and more particularly relates to a method, device and terminal equipment for inventory abnormal asset identification.

Background technique

In the daily asset operation management of the provincial-level company of the Internet, it is found that the phenomenon of inconsistency in the measurement of assets and accounts, such as the unit and status of the inventory assets, is inconsistent. The measurement assets themselves have the characteristics of a large number, scattered locations, and complex processing links. At present, managers usually use traditional warehouse inventory methods such as on-site inspection and business inspection to predict inventory assets, which results in a large workload and low efficiency in identifying abnormal inventory assets.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a method, device and terminal equipment for identifying abnormal inventory assets, so as to solve the technical problem of low identification efficiency of inventory abnormal assets in the prior art.

A first aspect of the embodiments of the present invention provides a method for identifying abnormal inventory assets, including:

Obtain the circulation trajectory and attribute data of the target asset;

Calculate the asset abnormality probability corresponding to the asset circulation trajectory, extract the feature value corresponding to each target asset feature from the attribute data, and generate a feature vector corresponding to the target asset based on the asset abnormality probability and the feature value corresponding to each target asset feature ;

The feature vector is input into a preset abnormal asset identification model to determine the abnormal inventory identification result of the target asset.

In a possible implementation manner, the calculating the asset abnormality probability corresponding to the asset circulation trajectory includes:

judging whether the asset circulation trajectory matches the historical asset circulation trajectory in the preset database;

If there is a certain historical circulation trajectory that matches the asset circulation trajectory, the asset abnormality probability corresponding to the historical circulation trajectory is used as the asset abnormality probability corresponding to the asset circulation trajectory;

If there is no historical asset circulation trajectory matching the asset circulation trajectory in the preset database, the asset circulation trajectory is input into the preset trajectory detection model to obtain the asset abnormality probability corresponding to the asset circulation trajectory.

In a possible implementation manner, before inputting the asset circulation trajectory into a preset trajectory detection model, the method for identifying abnormal assets in inventory further includes:

Extracting each circulation node in the asset circulation trajectory and the circulation time corresponding to each circulation node, and generating a feature sequence based on each circulation node and the circulation time corresponding to each circulation node;

Correspondingly, the asset circulation trajectory is input into a preset trajectory detection model to obtain the asset abnormality probability corresponding to the asset circulation trajectory, including:

Inputting the feature sequence into a preset trajectory detection model to obtain the asset abnormality probability corresponding to the asset circulation trajectory.

In a possible implementation manner, the generating a feature sequence based on each flow node and the flow time corresponding to each flow node includes:

Represent each flow node with different values;

A preset time series with an empty value is obtained. At a certain moment in the preset time series, the circulation node where the target asset is located at that moment is determined based on the circulation time of each circulation node, and the circulation node where the target asset is located at that moment is determined. The value of is used as the corresponding value at this moment;

The corresponding values of each moment in the preset time series are determined, and the preset time series of the corresponding values of all the moments is determined as the characteristic sequence.

In a possible implementation manner, generating a feature vector corresponding to the target asset based on the asset abnormality probability and the feature value corresponding to each target asset feature includes:

Generate a eigenvector based on the asset abnormality probability, denoted as the first eigenvector; generate a eigenvector based on the eigenvalues corresponding to the characteristics of each target asset, denoted as the second eigenvector; synthesize the first eigenvector and the first eigenvector Two eigenvectors to obtain the eigenvectors corresponding to the target asset.

In a possible implementation manner, the method for determining the characteristics of each target asset is:

All asset characteristics of the target asset are acquired, and principal component analysis is performed on all the asset characteristics to obtain the target asset characteristics of the target asset.

In a possible implementation manner, the abnormal asset identification model is an XGBoost model.

In a second aspect of the embodiments of the present invention, a device for identifying abnormal inventory assets is provided, including:

The data acquisition module is used to acquire the circulation trajectory and attribute data of the target asset;

The feature extraction module is used to calculate the asset abnormality probability corresponding to the asset circulation trajectory, extract the feature value corresponding to each target asset feature from the attribute data, and generate the feature value based on the asset abnormality probability and the feature value corresponding to each target asset feature The feature vector corresponding to the target asset;

The state identification module is used for inputting the feature vector into a preset abnormal asset identification model to determine the abnormal inventory identification result of the target asset.

In a third aspect of the embodiments of the present invention, a terminal device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program When implementing the steps of the above-mentioned method for identifying abnormal inventory assets.

In a fourth aspect of the embodiments of the present invention, a computer-readable storage medium is provided, where a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the above-mentioned method for identifying abnormal inventory assets are implemented .

The beneficial effects of the method, device and terminal equipment for identifying abnormal inventory assets provided by the embodiments of the present invention are:

Different from the manual verification scheme in the prior art, the present invention provides a method for identifying abnormal inventory assets based on big data, which can realize automatic verification and improve the efficiency of identifying abnormal inventory assets. Specifically, the present invention comprehensively considers the asset circulation trajectory and attribute data of the target asset. In this process, considering the uncertainty of the asset circulation trajectory (the trajectory length is uncertain, and the trajectory itself has randomness), the present invention is based on the asset circulation. The trajectory determines the asset abnormal probability, that is, the asset circulation trajectory is converted into the asset abnormal probability, and the asset circulation trajectory is added to the feature vector in the form of the asset abnormal probability, which realizes the comprehensive consideration of the asset circulation trajectory and attribute data, thus improving the inventory of abnormal assets Based on the identification efficiency, the identification accuracy of abnormal assets is improved.

Description of drawings

In order to illustrate the technical solutions in the embodiments of the present invention more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only for the present invention. In some embodiments, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without any creative effort.

1 is a schematic flowchart of a method for identifying abnormal inventory assets provided by an embodiment of the present invention;

2 is a structural block diagram of a device for identifying abnormal inventory assets provided by an embodiment of the present invention;

FIG. 3 is a schematic block diagram of a terminal device according to an embodiment of the present invention.

Detailed ways

In the following description, for the purpose of illustration rather than limitation, specific details such as specific system structures and technologies are set forth in order to provide a thorough understanding of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the present invention may be practiced in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.

In order to make the objectives, technical solutions and advantages of the present invention clearer, the following descriptions will be given through specific embodiments in conjunction with the accompanying drawings.

Please refer to FIG. 1. FIG. 1 is a schematic flowchart of a method for identifying abnormal inventory assets provided by an embodiment of the present invention. The method includes:

S101: Obtain the circulation track and attribute data of the target asset.

In this embodiment, the flow trajectory and attribute data of the target asset can be directly collected from the production scheduling platform MDS system and the marketing system of the metering center in the power grid. Among them, the attribute data of the target asset includes but is not limited to the equipment identification of the target asset, the unit where it is located, the equipment category, the specification, the wiring method, the voltage, the calibration current, the active power accuracy level, the filing date, etc. The circulation track of the target asset may include the date of each verification, the date of each meter installation, the date of each meter disassembly, the number of times of installation and disassembly, and the date of scrapping.

S102: Calculate the asset abnormality probability corresponding to the asset circulation trajectory, extract the feature value corresponding to each target asset feature from the attribute data, and generate a feature vector corresponding to the target asset based on the asset abnormality probability and the feature value corresponding to each target asset feature.

The asset circulation trajectory is the same, which means the use environment of the target asset is the same. Considering that when the target asset inventory is abnormal, the probability of the same problem occurring in the same application batch or in the same use environment is high. In this embodiment of the present invention, the asset circulation trajectory is used as the Part of the identification feature to improve the identification accuracy of anomalous assets in inventory.

Considering the randomness of the trajectory length and the trajectory of the asset circulation trajectory, in this embodiment, when identifying abnormal assets in inventory based on the asset circulation trajectory, it is inconvenient to directly generate a feature vector corresponding to the target asset according to the asset circulation trajectory. Moreover, although the asset circulation trajectory is random, in view of the huge number of target assets and the high degree of coincidence of asset circulation trajectories, on this basis, using the entire asset circulation trajectory as a feature vector to calculate every time will consume a lot of unnecessary. computing resources. Therefore, the embodiment of the present invention first calculates the asset abnormality probability corresponding to the asset circulation trajectory, that is, the asset abnormality probability is added to the calculation of the target asset feature vector as the characterization quantity of the asset circulation trajectory, so as to comprehensively consider the attribute information of the target asset and the asset circulation trajectory. . Among them, the asset abnormality probability corresponding to the asset circulation trajectory refers to the abnormality probability of the target asset's inventory calculated on the basis of the asset circulation trajectory as the influence parameter of the target asset inventory abnormality.

S103: Input the feature vector into the preset abnormal asset identification model, and determine the abnormal inventory identification result of the target asset.

In this embodiment, the abnormal inventory identification result of the target asset is that the inventory of the target asset is abnormal and the inventory of the target asset is normal. The abnormal inventory of the target asset indicates that there is an error in the inventory information of the target asset, such as the target asset has been lost, the inventory location of the target asset is wrong, and the use state of the target asset is wrong. Among them, the target assets can be electric energy meters, circuit breakers, etc.

In a possible implementation manner of this embodiment, the abnormal asset identification model may be an XGBoost model. The XGBoost model first trains a base learner from the initial training set, and then adjusts the training sample distribution according to the performance of the base learner, so that the previously misclassified samples will receive more attention in the follow-up, and then train the next base learner based on the adjusted sample distribution. Learner, giving higher weight to base learners with higher accuracy. This is repeated until the number of base learners reaches a pre-specified value. The combined goal is to minimize the exponential loss function through a combination of learners. XGBoost adds regular terms such as leaf node weights and tree depths to the cost function, which can control the complexity of the model on the one hand, and prevent overfitting on the other hand. At the same time, it uses the second-order Taylor expansion approximation to the cost function, which makes the approximate optimization of the objective function closer to the actual value, thereby improving the prediction accuracy. XGBoost has good convergence effect, stable process, fast calculation speed, and strong ability to solve practical problems. Therefore, the XGBoost model is used in this embodiment to identify abnormal inventory assets.

It can be concluded from the above that, different from the manual verification scheme in the prior art, the embodiment of the present invention provides a method for identifying abnormal inventory assets based on big data, which can realize automatic verification and improve the efficiency of identifying abnormal inventory assets. Specifically, the embodiment of the present invention comprehensively considers the asset circulation trajectory and attribute data of the target asset. In this process, considering the uncertainty of the asset circulation trajectory (the trajectory length is uncertain, and the trajectory itself has randomness), the present invention implements For example, the asset abnormality probability is determined according to the asset circulation trajectory, that is, the asset circulation trajectory is converted into the asset abnormality probability, and the asset circulation trajectory is added to the feature vector in the form of the asset abnormality probability, which realizes the comprehensive consideration of the asset circulation trajectory and attribute data. On the basis of improving the efficiency of identifying abnormal assets in inventory, the accuracy of identifying abnormal assets is improved.

In a possible implementation, calculating the asset anomaly probability corresponding to the asset circulation trajectory, including:

Determine whether the asset circulation trajectory matches the historical asset circulation trajectory in the preset database.

If there is a certain historical circulation trajectory that matches the asset circulation trajectory, the asset abnormality probability corresponding to the historical circulation trajectory is used as the asset abnormality probability corresponding to the asset circulation trajectory.

If there is no historical asset circulation trajectory matching the asset circulation trajectory in the preset database, input the asset circulation trajectory into the preset trajectory detection model to obtain the asset abnormality probability corresponding to the asset circulation trajectory.

In this embodiment, judging whether the asset circulation trajectory matches the historical asset circulation trajectory in the preset database is to detect whether there is a historical asset circulation trajectory in the preset database that is the same as the asset circulation trajectory of the target asset. If the asset circulation trajectory is the same as the historical asset circulation trajectory, it means that there is a match between the historical circulation trajectory and the asset circulation trajectory. If there is no historical asset circulation trajectory that is the same as the asset circulation trajectory of the target asset, it means that there is no historical circulation trajectory matching the asset circulation trajectory.

In this embodiment, the preset trajectory detection model may be a neural network model obtained by pre-training, and the neural network model takes the asset circulation trajectory as an input and takes the asset abnormality probability as an output.

In a possible implementation manner, before the asset circulation trajectory is input into the preset trajectory detection model, the method for identifying abnormal assets in inventory further includes:

Extract each transfer node and the corresponding transfer time of each transfer node in the asset transfer trajectory, and generate a feature sequence based on each transfer node and the corresponding transfer time of each transfer node.

Correspondingly, input the asset circulation trajectory into the preset trajectory detection model to obtain the asset abnormality probability corresponding to the asset circulation trajectory, including:

Input the feature sequence into the preset trajectory detection model to obtain the asset anomaly probability corresponding to the asset circulation trajectory.

In this embodiment, the length of the generated feature sequence is fixed, wherein the asset circulation track has a shorter length, and when the feature sequence is generated, the part of the insufficient track length is supplemented with "0".

In a possible implementation manner, a feature sequence is generated based on each flow node and the flow time corresponding to each flow node, including:

Each flow node is represented by a different value.

A preset time series with an empty value is obtained. At a certain moment in the preset time series, the circulation node where the target asset is located at that moment is determined based on the circulation time of each circulation node, and the circulation node where the target asset is located at that moment is determined. value as the corresponding value at this moment.

The corresponding values of each moment in the preset time series are determined, and the preset time series of the corresponding values of all the moments is determined as the characteristic sequence.

In this embodiment, the feature sequence represents the flow node where the target asset is located at each preset moment. For example, the transfer nodes are a, b, c, the target asset is at a at the first moment, the flow from a to b at the third moment, and the flow from b to c at the seventh moment, the preset time series contains ten moments, then this example corresponds to The feature sequence of is [aabbbbcccc].

In a possible implementation manner, a feature vector corresponding to the target asset is generated based on the abnormal probability of the asset and the feature value corresponding to each target asset feature, including:

A feature vector is generated based on the asset abnormality probability, which is recorded as the first feature vector. A eigenvector is generated based on the eigenvalues corresponding to each target asset feature, which is recorded as the second eigenvector. Synthesize the first feature vector and the second feature vector to obtain the feature vector corresponding to the target asset.

In this embodiment, a feature vector can be generated directly by using the asset abnormality probability as a feature value, a feature vector can be generated based on the feature values corresponding to each target asset feature, and finally the feature vector corresponding to the target asset can be obtained by synthesizing the feature vector.

In a possible implementation manner, the method for determining the characteristics of each target asset is:

Obtain all asset characteristics of the target asset, perform principal component analysis on all asset characteristics, and obtain the target asset characteristics of the target asset.

In this embodiment, the target asset feature refers to an asset feature whose degree of influence on the abnormality of the asset inventory is greater than a preset threshold. In order to effectively select such features, the embodiment of the present invention adopts a principal component analysis or a correlation analysis method to realize the screening of target asset features.

Corresponding to the method for identifying abnormal inventory assets in the above embodiment, FIG. 2 is a structural block diagram of a device for identifying abnormal inventory assets provided by an embodiment of the present invention. For the convenience of description, only the parts related to the embodiments of the present invention are shown. Referring to FIG. 2 , the abnormal inventory asset identification device 20 includes: a data acquisition module 21 , a feature extraction module 22 , and a state identification module 23 .

Among them, the data acquisition module 21 is used to acquire the circulation track and attribute data of the target asset.

The feature extraction module 22 is used to calculate the asset abnormality probability corresponding to the asset circulation trajectory, extract the feature value corresponding to each target asset feature from the attribute data, and generate the target asset corresponding feature based on the asset abnormality probability and the feature value corresponding to each target asset feature vector.

The state identification module 23 is configured to input the feature vector into a preset abnormal asset identification model to determine the abnormal inventory identification result of the target asset.

In a possible implementation manner, the feature extraction module 22 is specifically used for:

Determine whether the asset circulation trajectory matches the historical asset circulation trajectory in the preset database.

If there is a certain historical circulation trajectory that matches the asset circulation trajectory, the asset abnormality probability corresponding to the historical circulation trajectory is used as the asset abnormality probability corresponding to the asset circulation trajectory.

If there is no historical asset circulation trajectory matching the asset circulation trajectory in the preset database, input the asset circulation trajectory into the preset trajectory detection model to obtain the asset abnormality probability corresponding to the asset circulation trajectory.

In a possible implementation manner, the feature extraction module 22 is further configured to extract each circulation node in the asset circulation trajectory and the corresponding circulation time of each circulation node before inputting the asset circulation trajectory into the preset trajectory detection model, A feature sequence is generated based on each flow node and the flow time corresponding to each flow node.

Correspondingly, the feature extraction module 22 is specifically used for:

Input the feature sequence into the preset trajectory detection model to obtain the asset anomaly probability corresponding to the asset circulation trajectory.

In a possible implementation manner, the feature extraction module 22 is specifically used for:

Each flow node is represented by a different value.

A preset time series with an empty value is obtained. At a certain moment in the preset time series, the circulation node where the target asset is located at that moment is determined based on the circulation time of each circulation node, and the circulation node where the target asset is located at that moment is determined. value as the corresponding value at this moment.

The corresponding values of each moment in the preset time series are determined, and the preset time series of the corresponding values of all the moments is determined as the characteristic sequence.

In a possible implementation manner, the feature extraction module 22 is specifically used for:

A feature vector is generated based on the asset abnormality probability, which is recorded as the first feature vector. A eigenvector is generated based on the eigenvalues corresponding to each target asset feature, which is recorded as the second eigenvector. Synthesize the first feature vector and the second feature vector to obtain the feature vector corresponding to the target asset.

In a possible implementation manner, the feature extraction module 22 is also used for:

Obtain all asset characteristics of the target asset, perform principal component analysis on all asset characteristics, and obtain the target asset characteristics of the target asset.

In a possible implementation, the abnormal asset identification model is an XGBoost model.

Referring to FIG. 3, FIG. 3 is a schematic block diagram of a terminal device according to an embodiment of the present invention. As shown in FIG. 3 , the terminal 300 in this embodiment may include: one or more processors 301 , one or more input devices 302 , one or more output devices 303 , and one or more memories 304 . The processor 301 , the input device 302 , the output device 303 and the memory 304 communicate with each other through the communication bus 305 . Memory 304 is used to store computer programs including program instructions. The processor 301 is used to execute program instructions stored in the memory 304 . The processor 301 is configured to invoke the program instructions to perform the following operations to perform the functions of the modules/units in the above device embodiments, for example, the functions of the modules 21 to 23 shown in FIG. 2 .

It should be understood that, in this embodiment of the present invention, the so-called processor 301 may be a central processing unit (Central Processing Unit, CPU), and the processor may also be other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), dedicated Integrated circuit (Application Specific Integrated Circuit, ASIC), off-the-shelf Programmable Gate Array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint direction information), a microphone, and the like, and the output device 303 may include a display (LCD, etc.), a speaker, and the like.

The memory 304 may include read only memory and random access memory, and provides instructions and data to the processor 301 . A portion of memory 304 may also include non-volatile random access memory. For example, memory 304 may also store device type information.

In specific implementation, the processor 301 , the input device 302 , and the output device 303 described in the embodiments of the present invention can execute the methods described in the first embodiment and the second embodiment of the method for identifying abnormal inventory assets provided by the embodiments of the present invention. For the implementation manner, the implementation manner of the terminal described in the embodiments of the present invention can also be implemented, and details are not described herein again.

In another embodiment of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program, and the computer program includes program instructions. When the program instructions are executed by a processor, all or all of the methods in the foregoing embodiments are implemented. Part of the process can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of the above method embodiments. Wherein, the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file or some intermediate form. The computer readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, removable hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access Memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in computer-readable media may be appropriately increased or decreased in accordance with the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media does not include It is an electrical carrier signal and a telecommunication signal.

The computer-readable storage medium may be an internal storage unit of the terminal in any of the foregoing embodiments, such as a hard disk or a memory of the terminal. The computer-readable storage medium may also be an external storage device of the terminal, such as a plug-in hard disk equipped on the terminal, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, and a flash memory card (Flash Card). )Wait. Further, the computer-readable storage medium may also include both an internal storage unit of the terminal and an external storage device. The computer-readable storage medium is used to store computer programs and other programs and data required by the terminal. The computer-readable storage medium can also be used to temporarily store data that has been or will be output.

Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. Interchangeability, the above description has generally described the components and steps of each example in terms of function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each particular application, but such implementations should not be considered beyond the scope of the present invention.

Those skilled in the art can clearly understand that, for the convenience and brevity of description, for the specific working process of the terminal and unit described above, reference may be made to the corresponding process in the foregoing method embodiments, which will not be repeated here.

In the several embodiments provided in this application, it should be understood that the disclosed terminal and method may be implemented in other manners. For example, the apparatus embodiments described above are only illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, for example, multiple units or components may be combined or integrated. to another system, or some features can be ignored, or not implemented. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces or units, and may also be electrical, mechanical or other forms of connection.

Units described as separate components may or may not be physically separated, and components shown as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solutions in the embodiments of the present invention.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.

The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or modifications within the technical scope disclosed by the present invention. Replacement, these modifications or replacements should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims (10)

1. A method for identifying abnormal inventory assets, comprising: Obtain the circulation trajectory and attribute data of the target asset; Calculate the asset abnormality probability corresponding to the asset circulation trajectory, extract the feature value corresponding to each target asset feature from the attribute data, and generate a feature vector corresponding to the target asset based on the asset abnormality probability and the feature value corresponding to each target asset feature ; The feature vector is input into a preset abnormal asset identification model to determine the abnormal inventory identification result of the target asset. 2 . The method for identifying abnormal inventory assets according to claim 1 , wherein the calculating the abnormal asset probability corresponding to the asset circulation trajectory comprises: 2 . judging whether the asset circulation trajectory matches the historical asset circulation trajectory in the preset database; If there is a certain historical circulation trajectory that matches the asset circulation trajectory, the asset abnormality probability corresponding to the historical circulation trajectory is used as the asset abnormality probability corresponding to the asset circulation trajectory; If there is no historical asset circulation trajectory matching the asset circulation trajectory in the preset database, the asset circulation trajectory is input into the preset trajectory detection model to obtain the asset abnormality probability corresponding to the asset circulation trajectory. 3. The method for identifying abnormal inventory assets according to claim 2, wherein before the asset circulation trajectory is input into a preset trajectory detection model, the method for identifying abnormal inventory assets further comprises: Extracting each circulation node in the asset circulation trajectory and the circulation time corresponding to each circulation node, and generating a feature sequence based on each circulation node and the circulation time corresponding to each circulation node; Correspondingly, the asset circulation trajectory is input into a preset trajectory detection model to obtain the asset abnormality probability corresponding to the asset circulation trajectory, including: Inputting the feature sequence into a preset trajectory detection model to obtain the asset abnormality probability corresponding to the asset circulation trajectory. 4. The method for identifying abnormal inventory assets according to claim 3, wherein the generating a characteristic sequence based on each circulation node and the circulation time corresponding to each circulation node, comprising: Represent each flow node with different values; A preset time series with an empty value is obtained. At a certain moment in the preset time series, the circulation node where the target asset is located at that moment is determined based on the circulation time of each circulation node, and the circulation node where the target asset is located at that moment is determined. The value of is used as the corresponding value at this moment; The corresponding values of each moment in the preset time series are determined, and the preset time series of the corresponding values of all the moments is determined as the characteristic sequence. 5 . The method for identifying abnormal assets in inventory according to claim 1 , wherein the generating a feature vector corresponding to the target asset based on the abnormal probability of the asset and the feature value corresponding to each target asset feature, comprising: 6 . Generate a eigenvector based on the asset abnormality probability, denoted as the first eigenvector; generate a eigenvector based on the eigenvalues corresponding to the characteristics of each target asset, denoted as the second eigenvector; synthesize the first eigenvector and the first eigenvector Two eigenvectors to obtain the eigenvectors corresponding to the target asset. 6. The method for identifying abnormal assets in inventory as claimed in claim 1, wherein the method for determining the characteristics of each target asset is: All asset characteristics of the target asset are acquired, and principal component analysis is performed on all the asset characteristics to obtain the target asset characteristics of the target asset. 7. The method for identifying abnormal assets in inventory according to any one of claims 1 to 6, wherein the abnormal asset identification model is an XGBoost model. 8. A device for identifying abnormal assets in inventory, comprising: The data acquisition module is used to acquire the circulation trajectory and attribute data of the target asset; The feature extraction module is used to calculate the asset abnormality probability corresponding to the asset circulation trajectory, extract the feature value corresponding to each target asset feature from the attribute data, and generate the feature value based on the asset abnormality probability and the feature value corresponding to each target asset feature The feature vector corresponding to the target asset; The state identification module is used for inputting the feature vector into a preset abnormal asset identification model to determine the abnormal inventory identification result of the target asset. 9. A terminal device, comprising a memory, a processor and a computer program stored in the memory and running on the processor, characterized in that, when the processor executes the computer program, the implementation as claimed in the claims The steps of any one of 1 to 7 of the method. 10. A computer-readable storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented .

Images