Model training method and software-defined terminal operating system evaluation method

Abstract

The present disclosure relates to the technical field of electric power, in particular to a model training method and a software-defined terminal operating system evaluation method. The model training method comprises: obtaining container state information, terminal state information, application program state information in a first time period, and power distribution equipment state information in a second time period; obtaining power distribution terminal feature information according to the container state information, the terminal state information and the application program state information, and obtaining power distribution equipment feature information according to the power distribution equipment state information; and training a software-defined terminal operating system evaluation model according to the power distribution terminal feature information and the power distribution equipment feature information to obtain a target software-defined terminal operating system evaluation model. The scheme solves the problem of complex evaluation conditions required for evaluating power distribution equipment management, reduces the cost and improves the accuracy of the evaluation result.

Description

Technical Field

[0001] This disclosure relates to the field of power technology, specifically to a model training method and a software-defined terminal operating system evaluation method. Background Technology

[0002] As one of the most important energy sources in contemporary society, electricity has become increasingly crucial with the rapid development of power distribution networks and the growing demands of customers. In recent years, with the continuous expansion of power distribution networks, the number of power distribution devices connected to these networks has also increased. However, due to the diverse functions of these devices, the functions of the power distribution terminals used to manage them in the power distribution network have become quite complex. For different power distribution devices, corresponding software function modules need to be developed for the power distribution terminals, and the system programs need to be upgraded. Alternatively, different power distribution terminals may need to be set up for different power distribution devices, resulting in high costs for setting up power distribution terminals.

[0003] In related technologies, a software-defined power distribution terminal operating system can be used to enable the power distribution terminal to have customizable hardware and software functions, allowing it to adapt to the needs of different applications. In different usage scenarios, power distribution equipment can be flexibly deployed with applications for managing different power distribution devices, thereby reducing the cost of setting up the power distribution terminal. Specifically, to evaluate the management of power distribution equipment, different evaluation conditions can be constructed when the power distribution equipment is deployed in different application scenarios and with different applications. Based on the collected relevant parameters, an evaluation is performed according to the constructed evaluation conditions to determine whether the management of the power distribution equipment can meet the corresponding requirements.

[0004] While the above-mentioned solutions can assess the management of power distribution equipment to determine whether it can meet the corresponding requirements, the diverse usage scenarios of power distribution terminals and the variety of applications deployed on them make the assessment conditions required by the above solutions quite complex. This increases the cost of assessing the management of power distribution equipment and reduces the efficiency and accuracy of such assessments. Summary of the Invention

[0005] To address the problems in related technologies, this disclosure provides a model training method and a software-defined terminal operating system evaluation method.

[0006] In a first aspect, this disclosure provides a model training method, including:

[0007] The system acquires the container status information of the target container on the power distribution terminal in the first time period, the terminal status information of the power distribution terminal in the first time period, the application status information of the target application, and the power distribution equipment status information of the target power distribution equipment in the second time period. The target container is used to carry the target application, and the target application is used to manage the target power distribution equipment. The first time period is earlier than the second time period.

[0008] The power distribution terminal characteristic information is obtained based on the container status information, terminal status information, and application status information, and the power distribution equipment characteristic information is obtained based on the power distribution equipment status information.

[0009] A software-defined terminal operating system (SDOS) evaluation model is obtained. The SOS evaluation model is trained based on the characteristic information of the power distribution terminal and the characteristic information of the power distribution equipment to obtain the target SOS evaluation model.

[0010] In one implementation of this disclosure, before obtaining the power distribution terminal feature information based on container status information, terminal status information, and application status information, the method further includes:

[0011] Obtain the container state weight corresponding to the container state information, the terminal state weight corresponding to the terminal state information, and the application state weight corresponding to the application state information.

[0012] Based on container status information, terminal status information, and application status information, the characteristic information of the power distribution terminal is obtained, including:

[0013] The power distribution terminal characteristic information is obtained based on the container status information, container status weight, terminal status information, terminal status weight, application status information, and application status weight.

[0014] In one implementation of this disclosure, the container status information is used to indicate at least one of the following: the target container's creation time, startup time, number of times the target container has been started, startup speed, stop time, deletion time, version number, network address, network speed, container identifier, resource utilization, number of processes, number of threads, port used, interface used, and working log.

[0015] In one implementation of this disclosure, the terminal status information is used to indicate at least one of the following: the process management function of the target power distribution terminal, the file system type supported by the target power distribution terminal, the hardware module type supported by the target power distribution terminal, the network protocol supported by the target power distribution terminal, the log management function supported by the target power distribution terminal, the resource utilization rate of the target power distribution terminal, and the operating system version number of the target power distribution terminal.

[0016] In one implementation of this disclosure, the application status information is used to indicate at least one of the following: the version number of the target application, the number of processes corresponding to the target application, the process identifier corresponding to the target application, the installation time of the target application, the configuration time of the target application, the stop time of the target application, the restart time of the target application, the uninstallation time of the target application, the version number of the target application, the upgrade time of the target application, and the resource utilization rate of the target application.

[0017] In one implementation of this disclosure, before training the software-defined terminal operating system evaluation model based on the power distribution terminal feature information and the power distribution equipment feature information, the method further includes:

[0018] Receive the updated weight parameters sent by the edge server, and update the software-defined terminal operating system evaluation model according to the updated weight parameters;

[0019] The software-defined terminal operating system evaluation model is trained based on the characteristic information of the power distribution terminal and the characteristic information of the power distribution equipment to obtain the target software-defined terminal operating system evaluation model, including:

[0020] The updated software-defined terminal operating system evaluation model is trained by taking the characteristic information of the power distribution terminal as input and the characteristic information of the power distribution equipment as output.

[0021] In response to the convergence of the trained software-defined terminal operating system evaluation model, the trained software-defined terminal operating system evaluation model is stored as the target software-defined terminal operating system evaluation model.

[0022] In one implementation of this disclosure, the method further includes:

[0023] In response to the failure of the trained software-defined terminal operating system evaluation model to converge, the gradient update vector is obtained based on the trained software-defined terminal operating system evaluation model and sent to the edge server.

[0024] Secondly, this disclosure provides a method for evaluating a software-defined terminal operating system, including:

[0025] The container status information of the target container on the power distribution terminal in the third time period, the terminal status information of the power distribution terminal in the third time period, and the application status information of the target application are obtained. The target container is used to carry the target application, and the target application is used to manage the target power distribution equipment.

[0026] The power distribution terminal characteristic information is obtained based on the container status information, terminal status information, and application status information.

[0027] Obtain the target software-defined terminal operating system evaluation model, input the power distribution terminal feature information into the target software-defined terminal operating system evaluation model, and obtain the output power distribution equipment feature information;

[0028] The evaluation results of the software-defined terminal operating system are obtained based on the characteristic information of the power distribution equipment.

[0029] Thirdly, this disclosure provides a model training apparatus, including:

[0030] The first information acquisition module is configured to acquire the container status information of the target container on the power distribution terminal in the first time period, the terminal status information of the power distribution terminal in the first time period, the application status information of the target application, and the power distribution equipment status information of the target power distribution equipment in the second time period. The target container is used to carry the target application, and the target application is used to manage the target power distribution equipment. The first time period is earlier than the second time period.

[0031] The first feature acquisition module is configured to acquire power distribution terminal feature information based on container status information, terminal status information and application status information, and acquire power distribution equipment feature information based on power distribution equipment status information.

[0032] The model training module is configured to acquire a software-defined terminal operating system evaluation model, and train the software-defined terminal operating system evaluation model based on the characteristic information of the power distribution terminal and the characteristic information of the power distribution equipment to obtain a target software-defined terminal operating system evaluation model.

[0033] In one implementation of this disclosure, the first feature acquisition module is further configured as follows:

[0034] Obtain the container state weight corresponding to the container state information, the terminal state weight corresponding to the terminal state information, and the application state weight corresponding to the application state information.

[0035] The first feature acquisition module is specifically configured as follows:

[0036] The power distribution terminal characteristic information is obtained based on the container status information, container status weight, terminal status information, terminal status weight, application status information, and application status weight.

[0037] In one implementation of this disclosure, the container status information is used to indicate at least one of the following: the target container's creation time, startup time, number of times the target container has been started, startup speed, stop time, deletion time, version number, network address, network speed, container identifier, resource utilization, number of processes, number of threads, port used, interface used, and working log.

[0038] In one implementation of this disclosure, the terminal status information is used to indicate at least one of the following: the process management function of the target power distribution terminal, the file system type supported by the target power distribution terminal, the hardware module type supported by the target power distribution terminal, the network protocol supported by the target power distribution terminal, the log management function supported by the target power distribution terminal, the resource utilization rate of the target power distribution terminal, and the operating system version number of the target power distribution terminal.

[0039] In one implementation of this disclosure, the application status information is used to indicate at least one of the following: the version number of the target application, the number of processes corresponding to the target application, the process identifier corresponding to the target application, the installation time of the target application, the configuration time of the target application, the stop time of the target application, the restart time of the target application, the uninstallation time of the target application, the version number of the target application, the upgrade time of the target application, and the resource utilization rate of the target application.

[0040] In one implementation of this disclosure, the model training module is further configured as follows:

[0041] Receive the updated weight parameters sent by the edge server, and update the software-defined terminal operating system evaluation model according to the updated weight parameters;

[0042] The model training module is specifically configured as follows:

[0043] The updated software-defined terminal operating system evaluation model is trained by taking the characteristic information of the power distribution terminal as input and the characteristic information of the power distribution equipment as output.

[0044] In response to the convergence of the trained software-defined terminal operating system evaluation model, the trained software-defined terminal operating system evaluation model is stored as the target software-defined terminal operating system evaluation model.

[0045] In one implementation of this disclosure, the model training module is further configured as follows:

[0046] In response to the failure of the trained software-defined terminal operating system evaluation model to converge, the gradient update vector is obtained based on the trained software-defined terminal operating system evaluation model and sent to the edge server.

[0047] Fourthly, this disclosure provides a software-defined terminal operating system evaluation apparatus, comprising:

[0048] The second information acquisition module is configured to acquire the container status information of the target container on the power distribution terminal in the third time period, the terminal status information of the power distribution terminal in the third time period, and the application status information of the target application, wherein the target container is used to carry the target application, and the target application is used to manage the target power distribution equipment.

[0049] The second feature acquisition module is configured to acquire power distribution terminal feature information based on container status information, terminal status information and application status information.

[0050] The third feature acquisition module is configured to acquire the target software-defined terminal operating system evaluation model, input the power distribution terminal feature information into the target software-defined terminal operating system evaluation model, and obtain the output power distribution equipment feature information.

[0051] The evaluation result acquisition module is configured to acquire the evaluation results of the software-defined terminal operating system based on the characteristic information of the power distribution equipment.

[0052] Fifthly, embodiments of this disclosure provide an electronic device, including a memory and a processor, wherein the memory is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method as described in the first aspect, any implementation of the first aspect, or any one of the second aspects.

[0053] In a sixth aspect, embodiments of this disclosure provide a computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the method as described in any one of the first aspects, any implementation of the first aspect, and the second aspect.

[0054] According to the technical solution provided in this disclosure, a software-defined terminal operating system (SDS) evaluation model is obtained by acquiring container status information of the target container on the distribution terminal in the first time period, terminal status information of the distribution terminal in the first time period, application status information of the target application, and distribution equipment status information of the target distribution equipment in the second time period. Based on the container status information, terminal status information, and application status information, SDS feature information is obtained, and based on the distribution equipment status information, distribution equipment feature information is obtained. The SDS evaluation model is then trained based on the SDS feature information and distribution equipment feature information to obtain a target SDS evaluation model. Since the target container is used to host the target application, and the target application is used to manage the target distribution equipment, and the first time period is earlier than the second time period, the target SDS evaluation model can be understood as having learned the relationship between the status of the target application used to manage the target distribution equipment, the status of the target container used to host the target application, and the status of the distribution terminal where the target container is located in the earlier time period, and the status of the target distribution equipment managed by the target application in the later time period. Therefore, based on this target software-defined terminal operating system (SDOS) evaluation model, in different usage scenarios, it is relatively easy to estimate the power distribution equipment characteristic information corresponding to the power distribution equipment status information in later time periods based on the container status information, terminal status information, and application status information collected in earlier time periods. This allows for the acquisition of the corresponding SOS evaluation results based on the power distribution equipment characteristic information, thereby determining whether the power distribution equipment's operating status in later time periods, under the management of the target application, meets the corresponding requirements. The above solution improves the efficiency and accuracy of power distribution equipment management evaluation while ensuring low cost.

[0055] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0056] Other features, objects, and advantages of this disclosure will become more apparent from the following detailed description of non-limiting embodiments, taken in conjunction with the accompanying drawings. In the drawings:

[0057] Figure 1 A flowchart illustrating a model training method according to an embodiment of the present disclosure is shown.

[0058] Figure 2 A flowchart illustrating a software-defined terminal operating system evaluation method according to an embodiment of the present disclosure is shown.

[0059] Figure 3 A structural block diagram of a model training apparatus according to an embodiment of the present disclosure is shown.

[0060] Figure 4 A structural block diagram of a software-defined terminal operating system evaluation apparatus according to an embodiment of the present disclosure is shown.

[0061] Figure 5 A structural block diagram of an electronic device according to an embodiment of the present disclosure is shown.

[0062] Figure 6 A schematic diagram of the structure of a computer system suitable for implementing the method according to embodiments of the present disclosure is shown. Detailed Implementation

[0063] In the following, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings to enable those skilled in the art to readily implement them. Furthermore, for clarity, portions unrelated to the description of exemplary embodiments have been omitted from the drawings.

[0064] In this disclosure, it should be understood that terms such as “comprising” or “having” are intended to indicate the presence of features, figures, steps, behaviors, components, parts or combinations thereof disclosed in this specification, and are not intended to exclude the possibility of the presence or addition of one or more other features, figures, steps, behaviors, components, parts or combinations thereof.

[0065] It should also be noted that, unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other. This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0066] In this disclosure, any operation involving the acquisition of user information or user data, or the display of user information or user data to others, is an operation authorized or confirmed by the user, or actively selected by the user.

[0067] In related technologies, a software-defined power distribution terminal operating system can be used to enable the power distribution terminal to have customizable hardware and software functions, allowing it to adapt to the needs of different applications. In different usage scenarios, power distribution equipment can be flexibly deployed with applications for managing different power distribution devices, thereby reducing the cost of setting up the power distribution terminal. Specifically, to evaluate the management of power distribution equipment, different evaluation conditions can be constructed when the power distribution equipment is deployed in different application scenarios and with different applications. Based on the collected relevant parameters, an evaluation is performed according to the constructed evaluation conditions to determine whether the management of the power distribution equipment can meet the corresponding requirements.

[0068] While the above-mentioned approach can assess the management of power distribution equipment to determine whether it can meet the corresponding requirements, the assessment conditions required are also quite complex, considering the diverse usage scenarios of power distribution terminals and the variety of applications deployed on them. This increases the cost of assessing the management of power distribution equipment and reduces its efficiency and accuracy.

[0069] In the technical solution provided by this disclosure, a software-defined terminal operating system (SDS) evaluation model is obtained by acquiring the container status information of the target container on the distribution terminal in the first time period, the terminal status information of the distribution terminal in the first time period, the application status information of the target application, and the distribution equipment status information of the target distribution equipment in the second time period. Based on the container status information, terminal status information, and application status information, SDS feature information is obtained, and based on the distribution equipment status information, distribution equipment feature information is obtained. This model is then trained using the SDS feature information and the distribution equipment feature information to obtain a target SDS evaluation model. Since the target container is used to host the target application, and the target application is used to manage the target distribution equipment, and the first time period is earlier than the second time period, the target SDS evaluation model can be understood as having learned the relationship between the status of the target application used to manage the target distribution equipment, the status of the target container used to host the target application, and the status of the distribution terminal where the target container is located in the earlier time period, and the status of the target distribution equipment managed by the target application in the later time period. Therefore, based on this target software-defined terminal operating system (SDOS) evaluation model, in different usage scenarios, it is relatively easy to estimate the power distribution equipment characteristic information corresponding to the power distribution equipment status information in later time periods based on the container status information, terminal status information, and application status information collected in earlier time periods. This allows for the acquisition of the corresponding SOS evaluation results based on the power distribution equipment characteristic information, thereby determining whether the power distribution equipment's operating status in later time periods, under the management of the target application, meets the corresponding requirements. The above solution improves the efficiency and accuracy of power distribution equipment management evaluation while ensuring low cost.

[0070] Figure 1 A flowchart illustrating a model training method according to an embodiment of the present disclosure is shown. Figure 1 As shown, the model training method includes the following steps S101-S103:

[0071] In step S101, the container status information of the target container on the power distribution terminal in the first time period, the terminal status information of the power distribution terminal in the first time period, the application status information of the target application, and the power distribution equipment status information of the target power distribution equipment in the second time period are obtained.

[0072] The target container is used to host the target application, which is used to manage the target power distribution equipment. The first time period is earlier than the second time period.

[0073] In one embodiment of this disclosure, a distribution terminal can be understood as a software-defined terminal located at the edge of a distribution network. The distribution terminal can construct containers, and these containers can host corresponding applications (APPs) in a micro-application manner, with the applications implementing specific distribution service functions. For example, to implement a distribution transformer monitoring service function, the distribution terminal can construct container A, which hosts the distribution transformer monitoring APP, and the APP implements the specific distribution transformer monitoring service function. Similarly, to implement a smart meter information collection service function, the distribution terminal can construct container B, which hosts the smart meter information collection APP, and the APP implements the specific smart meter information collection service function. The APPs hosted in container A and container B can be isolated from each other to prevent interference. Furthermore, if new service functions are required, the distribution terminal can continue to construct new containers, each container hosting at least one APP.

[0074] In one embodiment of this disclosure, container status information can be understood as information used to indicate the working status of the corresponding container and the container attributes of the corresponding container.

[0075] In one embodiment of this disclosure, terminal status information can be understood as information used to indicate the working status of the corresponding power distribution terminal, the working parameters of the corresponding power distribution terminal, and the hardware and software configuration of the corresponding power distribution terminal.

[0076] In one embodiment of this disclosure, application status information can be understood as information used to indicate the working status of the corresponding application, the configuration of the corresponding application, etc.

[0077] In one embodiment of this disclosure, the power distribution equipment status information can be understood as information used to indicate the operating status, operating parameters, and environmental parameters of the corresponding power distribution equipment. For example, when the power distribution equipment is a smart meter, the status information may include the smart meter's voltage, current, power consumption, and power; when the power distribution equipment is a distribution transformer, the status information may include the transformer's ambient temperature, ambient humidity, oil temperature, oil level, and three-phase current.

[0078] In step S102, the power distribution terminal feature information is obtained based on the container status information, terminal status information and application status information, and the power distribution equipment feature information is obtained based on the power distribution equipment status information.

[0079] In one embodiment of this disclosure, obtaining power distribution terminal feature information based on container status information, terminal status information, and application status information can be understood as performing calculations by substituting the container status information, terminal status information, and application status information into a pre-set algorithm to obtain the power distribution terminal feature information. Alternatively, it can also involve querying a pre-acquired power distribution terminal feature database based on the container status information, terminal status information, and application status information to obtain the corresponding power distribution terminal feature information.

[0080] For example, scores can be obtained based on container status information, terminal status information, and application status information, and these scores can be accumulated to obtain power distribution terminal feature information.

[0081] In one embodiment of this disclosure, obtaining power distribution equipment feature information based on power distribution equipment status information can be understood as performing calculations by substituting the power distribution equipment status information into a pre-set algorithm to obtain power distribution terminal feature information. Alternatively, it can involve querying a pre-acquired power distribution equipment feature database based on the power distribution equipment status information to obtain the corresponding power distribution equipment feature information.

[0082] In step S103, a software-defined terminal operating system evaluation model is obtained, and the software-defined terminal operating system evaluation model is trained based on the power distribution terminal feature information and the power distribution equipment feature information to obtain a target software-defined terminal operating system evaluation model.

[0083] In one embodiment of this disclosure, the software-defined terminal operating system evaluation model can be pre-acquired or obtained from other devices or systems. The software-defined terminal operating system evaluation model can be a neural network (NN) model, a convolutional neural network (CNN) model, or a long short-term memory (LSTM) network model, etc. This disclosure does not specifically limit the implementation method of the software-defined terminal operating system evaluation model.

[0084] In the technical solution provided by this disclosure, a software-defined terminal operating system (SDS) evaluation model is obtained by acquiring the container status information of the target container on the distribution terminal in the first time period, the terminal status information of the distribution terminal in the first time period, the application status information of the target application, and the distribution equipment status information of the target distribution equipment in the second time period. Based on the container status information, terminal status information, and application status information, SDS feature information is obtained, and based on the distribution equipment status information, distribution equipment feature information is obtained. This model is then trained using the SDS feature information and the distribution equipment feature information to obtain a target SDS evaluation model. Since the target container is used to host the target application, and the target application is used to manage the target distribution equipment, and the first time period is earlier than the second time period, the target SDS evaluation model can be understood as having learned the relationship between the status of the target application used to manage the target distribution equipment, the status of the target container used to host the target application, and the status of the distribution terminal where the target container is located in the earlier time period, and the status of the target distribution equipment managed by the target application in the later time period. Therefore, based on this target software-defined terminal operating system (SDOS) evaluation model, in different usage scenarios, it is relatively easy to estimate the power distribution equipment characteristic information corresponding to the power distribution equipment status information in later time periods based on the container status information, terminal status information, and application status information collected in earlier time periods. This allows for the acquisition of the corresponding SOS evaluation results based on the power distribution equipment characteristic information, thereby determining whether the power distribution equipment's operating status in later time periods, under the management of the target application, meets the corresponding requirements. The above solution improves the efficiency and accuracy of power distribution equipment management evaluation while ensuring low cost.

[0085] In one implementation of this disclosure, before obtaining the power distribution terminal feature information based on container status information, terminal status information, and application status information, the method further includes:

[0086] Obtain the container state weight corresponding to the container state information, the terminal state weight corresponding to the terminal state information, and the application state weight corresponding to the application state information.

[0087] Based on container status information, terminal status information, and application status information, the characteristic information of the power distribution terminal is obtained, including:

[0088] The power distribution terminal characteristic information is obtained based on the container status information, container status weight, terminal status information, terminal status weight, application status information, and application status weight.

[0089] In one embodiment of this disclosure, the container state weight can be understood as the magnitude of the influence of container state information on the process of acquiring power distribution terminal feature information; the terminal state weight can be understood as the magnitude of the influence of terminal state information on the process of acquiring power distribution terminal feature information; and the application state weight can be understood as the magnitude of the influence of application state information on the process of acquiring power distribution terminal feature information.

[0090] For example, container status information, terminal status information, and application status information can be understood as primary indicator information, and each of these can include multiple corresponding secondary indicator information. Specifically, the container status weight can include the corresponding secondary indicator weight for each secondary indicator in the container status information, and the sum of all secondary indicator weights in the container status weight equals 1. Similarly, the terminal status weight can include the corresponding secondary indicator weight for each secondary indicator in the terminal status information, and the sum of all secondary indicator weights in the terminal status weight equals 1; the application status weight can include the corresponding secondary indicator weight for each secondary indicator in the application status information, and the sum of all secondary indicator weights in the application status weight equals 1.

[0091] In one embodiment of this disclosure, obtaining the state weight, terminal state weight, and program state weight can be done by reading pre-stored state weights, terminal state weights, and program state weights, or by receiving state weights, terminal state weights, and program state weights sent by other devices or systems.

[0092] In the technical solution provided by this disclosure, by obtaining the container state weight corresponding to the container state information, the terminal state weight corresponding to the terminal state information, and the program state weight corresponding to the application state information, and obtaining the power distribution terminal feature information based on the container state information, container state weight, terminal state information, terminal state weight, application state information, and program state weight, it is convenient to adjust the magnitude of the influence of the container state information, terminal state information, and application state information on the process of obtaining the power distribution terminal feature information, thereby improving the user experience.

[0093] In one implementation of this disclosure, the container status information is used to indicate at least one of the following: the target container's creation time, startup time, number of times the target container has been started, startup speed, stop time, deletion time, version number, network address, network speed, container identifier, resource utilization, number of processes, number of threads, port used, interface used, and working log.

[0094] In one embodiment of this disclosure, a container identifier can be understood as an identifier used to indicate the corresponding container and distinguish it from other containers. A container's resource utilization rate can be understood as the container's utilization rate of at least one of the following resources: Central Processing Unit (CPU), memory, and hard disk. The number of processes in a container can be understood as the number of processes created by the container. The number of threads in a container can be understood as an indicator of the number of threads created by the container.

[0095] In the technical solution provided by the embodiments of this disclosure, by limiting the container status information to indicate at least one of the following: the creation time of the target container, the start time of the target container, the number of times the target container is started, the start speed of the target container, the stop time of the target container, the deletion time of the target container, the version number of the target container, the network address of the target container, the network speed of the target container, the container identifier of the target container, the resource utilization rate of the target container, the number of processes of the target container, the number of threads of the target container, the port used by the target container, the interface used by the target container, and the working log of the target container, the container status information can more comprehensively reflect the working status of the target container and its own container attributes.

[0096] In one implementation of this disclosure, the terminal status information is used to indicate at least one of the following: the process management function of the target power distribution terminal, the file system type supported by the target power distribution terminal, the hardware module type supported by the target power distribution terminal, the network protocol supported by the target power distribution terminal, the log management function supported by the target power distribution terminal, the resource utilization rate of the target power distribution terminal, and the operating system version number of the target power distribution terminal.

[0097] In one embodiment of this disclosure, the hardware module type can be understood as indicating the type of hardware module installed in the power distribution terminal to implement the corresponding function. The resource utilization rate of the power distribution terminal can be understood as the percentage of at least one of the following resources of the power distribution terminal: CPU, memory, and hard disk.

[0098] In the technical solution provided by the embodiments of this disclosure, by limiting the terminal status information to indicate at least one of the following: the process management function of the target power distribution terminal, the file system type supported by the target power distribution terminal, the hardware module type supported by the target power distribution terminal, the network protocol supported by the target power distribution terminal, the log management function supported by the target power distribution terminal, the resource utilization rate of the target power distribution terminal, and the operating system version number of the target power distribution terminal, the terminal status information can more comprehensively reflect the working status, working parameters, and hardware and software configuration of the target power distribution terminal.

[0099] In one implementation of this disclosure, the application status information is used to indicate at least one of the following: the version number of the target application, the number of processes corresponding to the target application, the process identifier corresponding to the target application, the installation time of the target application, the configuration time of the target application, the stop time of the target application, the restart time of the target application, the uninstallation time of the target application, the version number of the target application, the upgrade time of the target application, and the resource utilization rate of the target application.

[0100] In one embodiment of this disclosure, the process corresponding to the application can be understood as the process created by the application. The application upgrade time can be understood as the time when the application was upgraded to the current version.

[0101] In the technical solution provided by the embodiments of this disclosure, by limiting the application status information to indicate at least one of the following: the version number of the target application, the number of processes corresponding to the target application, the process identifier corresponding to the target application, the installation time of the target application, the configuration time of the target application, the stop time of the target application, the restart time of the target application, the uninstallation time of the target application, the version number of the target application, the upgrade time of the target application, and the resource utilization rate of the target application, the application status information can more comprehensively reflect the working status and configuration of the target application.

[0102] In one implementation of this disclosure, before training the software-defined terminal operating system evaluation model based on the power distribution terminal feature information and the power distribution equipment feature information, the method further includes:

[0103] Receive the updated weight parameters sent by the edge server, and update the software-defined terminal operating system evaluation model according to the updated weight parameters;

[0104] The software-defined terminal operating system evaluation model is trained based on the characteristic information of the power distribution terminal and the characteristic information of the power distribution equipment to obtain the target software-defined terminal operating system evaluation model, including:

[0105] The updated software-defined terminal operating system evaluation model is trained by taking the characteristic information of the power distribution terminal as input and the characteristic information of the power distribution equipment as output.

[0106] In response to the convergence of the trained software-defined terminal operating system evaluation model, the trained software-defined terminal operating system evaluation model is stored as the target software-defined terminal operating system evaluation model.

[0107] In one implementation of this disclosure, the method further includes:

[0108] In response to the failure of the trained software-defined terminal operating system evaluation model to converge, the gradient update vector is obtained based on the trained software-defined terminal operating system evaluation model and sent to the edge server.

[0109] In one implementation of this disclosure, an edge server is used to aggregate gradient update vectors and update the weight parameters of the software-defined terminal operating system evaluation model on the edge server according to the aggregated gradient update vectors to obtain updated weight parameters. The edge server can be a cloud server or a server provided by a software-defined terminal operating system evaluation service provider. It should be noted that the model training method can be executed by edge devices located in the power distribution network, close to the power distribution equipment and power distribution terminals. One edge server can correspond to one or more edge devices. For example, the software-defined terminal operating system evaluation service provider can divide its managed area into multiple blocks, and multiple edge devices in each block can correspond to one edge server.

[0110] The evaluation model for software-defined terminal operating systems on edge servers can be a neural network model, a convolutional neural network model, or a long short-term memory network model, etc.

[0111] In one implementation of this disclosure, the updated weight parameters received by the edge device from the edge server are obtained by the edge server aggregating gradient update vectors sent by multiple edge devices and updating the weight parameters of the software-defined terminal operating system evaluation model on the edge server according to the aggregated gradient update vectors. Therefore, the updated software-defined terminal operating system evaluation model on the edge device can reflect the common patterns of power distribution terminal feature information and power distribution equipment feature information learned by the software-defined terminal operating system evaluation model on the edge server in the previous training round. Then, using distribution terminal feature information as input and distribution equipment feature information as output, the updated software-defined terminal operating system evaluation model is trained. This allows the updated software-defined terminal operating system evaluation model on the edge device to learn not only common patterns but also personalized patterns specific to the distribution terminal and distribution equipment feature information acquired by the edge device itself. This enables the trained software-defined terminal operating system evaluation model on the edge device to learn the private patterns of the distribution terminal and distribution equipment feature information acquired by the edge device. If the trained software-defined terminal operating system evaluation model on the edge device fails to converge, it indicates that the model still needs further training. The software-defined terminal operating system (SDOS) evaluation model on the edge device acquires and sends gradient update vectors, enabling the edge server to continue obtaining corresponding update weight parameters based on the gradient update vectors uploaded by multiple edge devices. This allows for continued training of the SOS evaluation model on each edge device. When the trained SOS evaluation model on the edge device converges, it can be considered that the converged SOS evaluation model can accurately obtain the corresponding power distribution equipment feature information based on the power distribution equipment status information. The converged SOS evaluation model can be stored as the target SOS evaluation model, which is a model that can be used to obtain power distribution equipment feature information with high accuracy.

[0112] In the above technical solution, on the one hand, the final target software-defined terminal operating system evaluation model can be a model that learns both common rules and private rules, and the accuracy of the power distribution equipment feature information obtained based on this model is relatively high; on the other hand, since the process of continuing to train the software-defined terminal operating system evaluation model on each edge device is jointly executed by the edge device and the edge server, compared with training the software-defined terminal operating system evaluation model by the edge device or the server alone, less processing resources are required and the training speed is faster.

[0113] Figure 2A flowchart illustrating a software-defined terminal operating system evaluation method according to embodiments of the present disclosure is shown. Figure 2 As shown, the software-defined terminal operating system evaluation method includes the following steps S201–S203:

[0114] In step S201, the container status information of the target container on the power distribution terminal in the third time period, the terminal status information of the power distribution terminal in the third time period, and the application status information of the target application are obtained.

[0115] The target container is used to host the target application, which in turn is used to manage the target power distribution equipment.

[0116] The target container is used to host the target application, which is used to manage the target power distribution equipment. The first time period is earlier than the second time period.

[0117] In one embodiment of this disclosure, a distribution terminal can be understood as a software-defined terminal located at the edge of a distribution network. The distribution terminal can construct containers, and these containers can host corresponding applications (APPs) in a micro-application manner, with the applications implementing specific distribution service functions. For example, to implement a distribution transformer monitoring service function, the distribution terminal can construct container A, which hosts the distribution transformer monitoring APP, and the APP implements the specific distribution transformer monitoring service function. Similarly, to implement a smart meter information collection service function, the distribution terminal can construct container B, which hosts the smart meter information collection APP, and the APP implements the specific smart meter information collection service function. The APPs hosted in container A and container B can be isolated from each other to prevent interference. Furthermore, if new service functions are required, the distribution terminal can continue to construct new containers, each container hosting at least one APP.

[0118] In one embodiment of this disclosure, container status information can be understood as information used to indicate the working status of the corresponding container and the container attributes of the corresponding container.

[0119] In one embodiment of this disclosure, terminal status information can be understood as information used to indicate the working status of the corresponding power distribution terminal, the working parameters of the corresponding power distribution terminal, and the hardware and software configuration of the corresponding power distribution terminal.

[0120] In one embodiment of this disclosure, application status information can be understood as information used to indicate the working status of the corresponding application, the configuration of the corresponding application, etc.

[0121] In step S202, the power distribution terminal characteristic information is obtained based on the container status information, terminal status information, and application status information.

[0122] In one embodiment of this disclosure, obtaining power distribution terminal feature information based on container status information, terminal status information, and application status information can be understood as performing calculations by substituting the container status information, terminal status information, and application status information into a pre-set algorithm to obtain the power distribution terminal feature information. Alternatively, it can also involve querying a pre-acquired power distribution terminal feature database based on the container status information, terminal status information, and application status information to obtain the corresponding power distribution terminal feature information.

[0123] For example, scores can be obtained based on container status information, terminal status information, and application status information, and these scores can be accumulated to obtain power distribution terminal feature information.

[0124] In one embodiment of this disclosure, obtaining power distribution equipment feature information based on power distribution equipment status information can be understood as performing calculations by substituting the power distribution equipment status information into a pre-set algorithm to obtain power distribution terminal feature information. Alternatively, it can involve querying a pre-acquired power distribution equipment feature database based on the power distribution equipment status information to obtain the corresponding power distribution equipment feature information.

[0125] In step S203, the target software-defined terminal operating system evaluation model is obtained, and the power distribution terminal feature information is input into the target software-defined terminal operating system evaluation model to obtain the output power distribution equipment feature information.

[0126] In one embodiment of this disclosure, the target software-defined terminal operating system evaluation model can be pre-acquired or obtained from other devices or systems. The target software-defined terminal operating system evaluation model can be a neural network (NN) model, a convolutional neural network (CNN) model, or a long short-term memory (LSTM) network model, etc. This disclosure does not specifically limit the implementation method of the software-defined terminal operating system evaluation model.

[0127] In step S204, the evaluation results of the software-defined terminal operating system are obtained based on the power distribution equipment characteristic information.

[0128] In the technical solution provided by this disclosure, the container status information of the target container on the power distribution terminal in the third time period, the terminal status information of the power distribution terminal in the third time period, and the application status information of the target application are obtained; power distribution terminal feature information is obtained based on the container status information, terminal status information, and application status information; a target software-defined terminal operating system evaluation model is obtained, and the power distribution terminal feature information is input into the target software-defined terminal operating system evaluation model to obtain the output power distribution equipment feature information; and the software-defined terminal operating system evaluation result is obtained based on the power distribution equipment feature information. In this model, the target container is used to host the target application, which in turn manages the target power distribution equipment. The target software-defined terminal operating system evaluation model can be understood as learning the patterns between the status of the target application managing the target power distribution equipment, the status of the target container hosting the target application, and the status of the power distribution terminal where the target container is located in an earlier time period, and the status of the target power distribution equipment managed by the target application in a later time period. Therefore, based on the software-defined terminal operating system evaluation results obtained from the power distribution equipment characteristic information output by the target software-defined terminal operating system evaluation model, it can be determined whether the working status of the power distribution equipment under the management of the target application meets the corresponding requirements after the third time period. This improves the efficiency and accuracy of power distribution equipment management evaluation while ensuring that the cost of evaluating power distribution equipment management is low.

[0129] Figure 3 A structural block diagram of a model training apparatus according to an embodiment of the present disclosure is shown. This apparatus can be implemented as part or all of an electronic device through software, hardware, or a combination of both.

[0130] like Figure 3 As shown, the model training device 300 includes:

[0131] The first information acquisition module 310 is configured to acquire the container status information of the target container on the power distribution terminal in the first time period, the terminal status information of the power distribution terminal in the first time period, the application status information of the target application, and the power distribution equipment status information of the target power distribution equipment in the second time period. The target container is used to carry the target application, and the target application is used to manage the target power distribution equipment. The first time period is earlier than the second time period.

[0132] The first feature acquisition module 320 is configured to acquire power distribution terminal feature information based on container status information, terminal status information and application status information, and acquire power distribution equipment feature information based on power distribution equipment status information.

[0133] The model training module 330 is configured to acquire a software-defined terminal operating system evaluation model, and to train the software-defined terminal operating system evaluation model based on the power distribution terminal feature information and the power distribution equipment feature information to obtain a target software-defined terminal operating system evaluation model.

[0134] According to the technical solution provided in this disclosure, a software-defined terminal operating system (SDS) evaluation model is obtained by acquiring container status information of the target container on the distribution terminal in the first time period, terminal status information of the distribution terminal in the first time period, application status information of the target application, and distribution equipment status information of the target distribution equipment in the second time period. Based on the container status information, terminal status information, and application status information, SDS feature information is obtained, and based on the distribution equipment status information, distribution equipment feature information is obtained. The SDS evaluation model is then trained based on the SDS feature information and distribution equipment feature information to obtain a target SDS evaluation model. Since the target container is used to host the target application, and the target application is used to manage the target distribution equipment, and the first time period is earlier than the second time period, the target SDS evaluation model can be understood as having learned the relationship between the status of the target application used to manage the target distribution equipment, the status of the target container used to host the target application, and the status of the distribution terminal where the target container is located in the earlier time period, and the status of the target distribution equipment managed by the target application in the later time period. Therefore, based on this target software-defined terminal operating system (SDOS) evaluation model, in different usage scenarios, it is relatively easy to estimate the power distribution equipment characteristic information corresponding to the power distribution equipment status information in later time periods based on the container status information, terminal status information, and application status information collected in earlier time periods. This allows for the acquisition of the corresponding SOS evaluation results based on the power distribution equipment characteristic information, thereby determining whether the power distribution equipment's operating status in later time periods, under the management of the target application, meets the corresponding requirements. The above solution improves the efficiency and accuracy of power distribution equipment management evaluation while ensuring low cost.

[0135] In one implementation of this disclosure, the first feature acquisition module is further configured as follows:

[0136] Obtain the container state weight corresponding to the container state information, the terminal state weight corresponding to the terminal state information, and the application state weight corresponding to the application state information.

[0137] The first feature acquisition module is specifically configured as follows:

[0138] The power distribution terminal characteristic information is obtained based on the container status information, container status weight, terminal status information, terminal status weight, application status information, and application status weight.

[0139] In one embodiment of this disclosure, the container state weight can be understood as the magnitude of the influence of container state information on the process of acquiring power distribution terminal feature information; the terminal state weight can be understood as the magnitude of the influence of terminal state information on the process of acquiring power distribution terminal feature information; and the application state weight can be understood as the magnitude of the influence of application state information on the process of acquiring power distribution terminal feature information.

[0140] For example, container status information, terminal status information, and application status information can be understood as primary indicator information, and each of these can include multiple corresponding secondary indicator information. Specifically, the container status weight can include the corresponding secondary indicator weight for each secondary indicator in the container status information, and the sum of all secondary indicator weights in the container status weight equals 1. Similarly, the terminal status weight can include the corresponding secondary indicator weight for each secondary indicator in the terminal status information, and the sum of all secondary indicator weights in the terminal status weight equals 1; the application status weight can include the corresponding secondary indicator weight for each secondary indicator in the application status information, and the sum of all secondary indicator weights in the application status weight equals 1.

[0141] In one embodiment of this disclosure, obtaining the state weight, terminal state weight, and program state weight can be done by reading pre-stored state weights, terminal state weights, and program state weights, or by receiving state weights, terminal state weights, and program state weights sent by other devices or systems.

[0142] In the technical solution provided by this disclosure, by obtaining the container state weight corresponding to the container state information, the terminal state weight corresponding to the terminal state information, and the program state weight corresponding to the application state information, and obtaining the power distribution terminal feature information based on the container state information, container state weight, terminal state information, terminal state weight, application state information, and program state weight, it is convenient to adjust the magnitude of the influence of the container state information, terminal state information, and application state information on the process of obtaining the power distribution terminal feature information, thereby improving the user experience.

[0143] In one implementation of this disclosure, the container status information is used to indicate at least one of the following: the target container's creation time, startup time, number of times the target container has been started, startup speed, stop time, deletion time, version number, network address, network speed, container identifier, resource utilization, number of processes, number of threads, port used, interface used, and working log.

[0144] In one embodiment of this disclosure, a container identifier can be understood as an identifier used to indicate the corresponding container and distinguish it from other containers. A container's resource utilization rate can be understood as the container's utilization rate of at least one of the following resources: Central Processing Unit (CPU), memory, and hard disk. The number of processes in a container can be understood as the number of processes created by the container. The number of threads in a container can be understood as an indicator of the number of threads created by the container.

[0145] In the technical solution provided by the embodiments of this disclosure, by limiting the container status information to indicate at least one of the following: the creation time of the target container, the start time of the target container, the number of times the target container is started, the start speed of the target container, the stop time of the target container, the deletion time of the target container, the version number of the target container, the network address of the target container, the network speed of the target container, the container identifier of the target container, the resource utilization rate of the target container, the number of processes of the target container, the number of threads of the target container, the port used by the target container, the interface used by the target container, and the working log of the target container, the container status information can more comprehensively reflect the working status of the target container and its own container attributes.

[0146] In one implementation of this disclosure, the terminal status information is used to indicate at least one of the following: the process management function of the target power distribution terminal, the file system type supported by the target power distribution terminal, the hardware module type supported by the target power distribution terminal, the network protocol supported by the target power distribution terminal, the log management function supported by the target power distribution terminal, the resource utilization rate of the target power distribution terminal, and the operating system version number of the target power distribution terminal.

[0147] In one embodiment of this disclosure, the hardware module type can be understood as indicating the type of hardware module installed in the power distribution terminal to implement the corresponding function. The resource utilization rate of the power distribution terminal can be understood as the percentage of at least one of the following resources of the power distribution terminal: CPU, memory, and hard disk.

[0148] In the technical solution provided by the embodiments of this disclosure, by limiting the terminal status information to indicate at least one of the following: the process management function of the target power distribution terminal, the file system type supported by the target power distribution terminal, the hardware module type supported by the target power distribution terminal, the network protocol supported by the target power distribution terminal, the log management function supported by the target power distribution terminal, the resource utilization rate of the target power distribution terminal, and the operating system version number of the target power distribution terminal, the terminal status information can more comprehensively reflect the working status, working parameters, and hardware and software configuration of the target power distribution terminal.

[0149] In one implementation of this disclosure, the application status information is used to indicate at least one of the following: the version number of the target application, the number of processes corresponding to the target application, the process identifier corresponding to the target application, the installation time of the target application, the configuration time of the target application, the stop time of the target application, the restart time of the target application, the uninstallation time of the target application, the version number of the target application, the upgrade time of the target application, and the resource utilization rate of the target application.

[0150] In one embodiment of this disclosure, the process corresponding to the application can be understood as the process created by the application. The application upgrade time can be understood as the time when the application was upgraded to the current version.

[0151] In the technical solution provided by the embodiments of this disclosure, by limiting the application status information to indicate at least one of the following: the version number of the target application, the number of processes corresponding to the target application, the process identifier corresponding to the target application, the installation time of the target application, the configuration time of the target application, the stop time of the target application, the restart time of the target application, the uninstallation time of the target application, the version number of the target application, the upgrade time of the target application, and the resource utilization rate of the target application, the application status information can more comprehensively reflect the working status and configuration of the target application.

[0152] In one implementation of this disclosure, the model training module is further configured as follows:

[0153] Receive the updated weight parameters sent by the edge server, and update the software-defined terminal operating system evaluation model according to the updated weight parameters;

[0154] The model training module is specifically configured as follows:

[0155] The updated software-defined terminal operating system evaluation model is trained by taking the characteristic information of the power distribution terminal as input and the characteristic information of the power distribution equipment as output.

[0156] In response to the convergence of the trained software-defined terminal operating system evaluation model, the trained software-defined terminal operating system evaluation model is stored as the target software-defined terminal operating system evaluation model.

[0157] In one implementation of this disclosure, the model training module is further configured as follows:

[0158] In response to the failure of the trained software-defined terminal operating system evaluation model to converge, the gradient update vector is obtained based on the trained software-defined terminal operating system evaluation model and sent to the edge server.

[0159] In one implementation of this disclosure, an edge server is used to aggregate gradient update vectors and update the weight parameters of the software-defined terminal operating system evaluation model on the edge server according to the aggregated gradient update vectors to obtain updated weight parameters. The edge server can be a cloud server or a server provided by a software-defined terminal operating system evaluation service provider. It should be noted that the model training method can be executed by edge devices located in the power distribution network, close to the power distribution equipment and power distribution terminals. One edge server can correspond to one or more edge devices. For example, the software-defined terminal operating system evaluation service provider can divide its managed area into multiple blocks, and multiple edge devices in each block can correspond to one edge server.

[0160] The evaluation model for software-defined terminal operating systems on edge servers can be a neural network model, a convolutional neural network model, or a long short-term memory network model, etc.

[0161] In one implementation of this disclosure, the updated weight parameters received by the edge device from the edge server are obtained by the edge server aggregating gradient update vectors sent by multiple edge devices and updating the weight parameters of the software-defined terminal operating system evaluation model on the edge server according to the aggregated gradient update vectors. Therefore, the updated software-defined terminal operating system evaluation model on the edge device can reflect the common patterns of power distribution terminal feature information and power distribution equipment feature information learned by the software-defined terminal operating system evaluation model on the edge server in the previous training round. Then, using distribution terminal feature information as input and distribution equipment feature information as output, the updated software-defined terminal operating system evaluation model is trained. This allows the updated software-defined terminal operating system evaluation model on the edge device to learn not only common patterns but also personalized patterns specific to the distribution terminal and distribution equipment feature information acquired by the edge device itself. This enables the trained software-defined terminal operating system evaluation model on the edge device to learn the private patterns of the distribution terminal and distribution equipment feature information acquired by the edge device. If the trained software-defined terminal operating system evaluation model on the edge device fails to converge, it indicates that the model still needs further training. The software-defined terminal operating system (SDOS) evaluation model on the edge device acquires and sends gradient update vectors, enabling the edge server to continue obtaining corresponding update weight parameters based on the gradient update vectors uploaded by multiple edge devices. This allows for continued training of the SOS evaluation model on each edge device. When the trained SOS evaluation model on the edge device converges, it can be considered that the converged SOS evaluation model can accurately obtain the corresponding power distribution equipment feature information based on the power distribution equipment status information. The converged SOS evaluation model can be stored as the target SOS evaluation model, which is a model that can be used to obtain power distribution equipment feature information with high accuracy.

[0162] In the above technical solution, on the one hand, the final target software-defined terminal operating system evaluation model can be a model that learns both common rules and private rules, and the accuracy of the power distribution equipment feature information obtained based on this model is relatively high; on the other hand, since the process of continuing to train the software-defined terminal operating system evaluation model on each edge device is jointly executed by the edge device and the edge server, compared with training the software-defined terminal operating system evaluation model by the edge device or the server alone, less processing resources are required and the training speed is faster.

[0163] Figure 4A structural block diagram of a software-defined terminal operating system evaluation apparatus according to an embodiment of the present disclosure is shown. This apparatus can be implemented as part or all of an electronic device through software, hardware, or a combination of both.

[0164] like Figure 4 As shown, the software-defined terminal operating system evaluation device 400 includes:

[0165] The second information acquisition module 410 is configured to acquire the container status information of the target container on the power distribution terminal in the third time period, the terminal status information of the power distribution terminal in the third time period, and the application status information of the target application, wherein the target container is used to carry the target application and the target application is used to manage the target power distribution equipment.

[0166] The second feature acquisition module 420 is configured to acquire power distribution terminal feature information based on container status information, terminal status information and application status information.

[0167] The third feature acquisition module 430 is configured to acquire the target software-defined terminal operating system evaluation model, input the power distribution terminal feature information into the target software-defined terminal operating system evaluation model, and obtain the output power distribution equipment feature information.

[0168] The evaluation result acquisition module 440 is configured to acquire the evaluation results of the software-defined terminal operating system based on the characteristic information of the power distribution equipment.

[0169] In the technical solution provided by this disclosure, the container status information of the target container on the power distribution terminal in the third time period, the terminal status information of the power distribution terminal in the third time period, and the application status information of the target application are obtained; power distribution terminal feature information is obtained based on the container status information, terminal status information, and application status information; a target software-defined terminal operating system evaluation model is obtained, and the power distribution terminal feature information is input into the target software-defined terminal operating system evaluation model to obtain the output power distribution equipment feature information; and the software-defined terminal operating system evaluation result is obtained based on the power distribution equipment feature information. In this model, the target container is used to host the target application, which in turn manages the target power distribution equipment. The target software-defined terminal operating system evaluation model can be understood as learning the patterns between the status of the target application managing the target power distribution equipment, the status of the target container hosting the target application, and the status of the power distribution terminal where the target container is located in an earlier time period, and the status of the target power distribution equipment managed by the target application in a later time period. Therefore, based on the software-defined terminal operating system evaluation results obtained from the power distribution equipment characteristic information output by the target software-defined terminal operating system evaluation model, it can be determined whether the working status of the power distribution equipment under the management of the target application meets the corresponding requirements after the third time period. This improves the efficiency and accuracy of power distribution equipment management evaluation while ensuring that the cost of evaluating power distribution equipment management is low.

[0170] This disclosure also discloses an electronic device. Figure 5 A structural block diagram of an electronic device according to an embodiment of the present disclosure is shown.

[0171] like Figure 5 As shown, the electronic device includes a memory and a processor, wherein the memory is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method according to embodiments of the present disclosure.

[0172] In a first aspect, this disclosure provides a model training method, including:

[0173] The system acquires the container status information of the target container on the power distribution terminal in the first time period, the terminal status information of the power distribution terminal in the first time period, the application status information of the target application, and the power distribution equipment status information of the target power distribution equipment in the second time period. The target container is used to carry the target application, and the target application is used to manage the target power distribution equipment. The first time period is earlier than the second time period.

[0174] The power distribution terminal characteristic information is obtained based on the container status information, terminal status information, and application status information, and the power distribution equipment characteristic information is obtained based on the power distribution equipment status information.

[0175] A software-defined terminal operating system (SDOS) evaluation model is obtained. The SOS evaluation model is trained based on the characteristic information of the power distribution terminal and the characteristic information of the power distribution equipment to obtain the target SOS evaluation model.

[0176] In one implementation of this disclosure, before obtaining the power distribution terminal feature information based on container status information, terminal status information, and application status information, the method further includes:

[0177] Obtain the container state weight corresponding to the container state information, the terminal state weight corresponding to the terminal state information, and the application state weight corresponding to the application state information.

[0178] Based on container status information, terminal status information, and application status information, the characteristic information of the power distribution terminal is obtained, including:

[0179] The power distribution terminal characteristic information is obtained based on the container status information, container status weight, terminal status information, terminal status weight, application status information, and application status weight.

[0180] In one implementation of this disclosure, the container status information is used to indicate at least one of the following: the target container's creation time, startup time, number of times the target container has been started, startup speed, stop time, deletion time, version number, network address, network speed, container identifier, resource utilization, number of processes, number of threads, port used, interface used, and working log.

[0181] In one implementation of this disclosure, the terminal status information is used to indicate at least one of the following: the process management function of the target power distribution terminal, the file system type supported by the target power distribution terminal, the hardware module type supported by the target power distribution terminal, the network protocol supported by the target power distribution terminal, the log management function supported by the target power distribution terminal, the resource utilization rate of the target power distribution terminal, and the operating system version number of the target power distribution terminal.

[0182] In one implementation of this disclosure, the application status information is used to indicate at least one of the following: the version number of the target application, the number of processes corresponding to the target application, the process identifier corresponding to the target application, the installation time of the target application, the configuration time of the target application, the stop time of the target application, the restart time of the target application, the uninstallation time of the target application, the version number of the target application, the upgrade time of the target application, and the resource utilization rate of the target application.

[0183] In one implementation of this disclosure, before training the software-defined terminal operating system evaluation model based on the power distribution terminal feature information and the power distribution equipment feature information, the method further includes:

[0184] Receive the updated weight parameters sent by the edge server, and update the software-defined terminal operating system evaluation model according to the updated weight parameters;

[0185] The software-defined terminal operating system evaluation model is trained based on the characteristic information of the power distribution terminal and the characteristic information of the power distribution equipment to obtain the target software-defined terminal operating system evaluation model, including:

[0186] The updated software-defined terminal operating system evaluation model is trained by taking the characteristic information of the power distribution terminal as input and the characteristic information of the power distribution equipment as output.

[0187] In response to the convergence of the trained software-defined terminal operating system evaluation model, the trained software-defined terminal operating system evaluation model is stored as the target software-defined terminal operating system evaluation model.

[0188] In one implementation of this disclosure, the method further includes:

[0189] In response to the failure of the trained software-defined terminal operating system evaluation model to converge, the gradient update vector is obtained based on the trained software-defined terminal operating system evaluation model and sent to the edge server.

[0190] Secondly, this disclosure provides a method for evaluating a software-defined terminal operating system, including:

[0191] The container status information of the target container on the power distribution terminal in the third time period, the terminal status information of the power distribution terminal in the third time period, and the application status information of the target application are obtained. The target container is used to carry the target application, and the target application is used to manage the target power distribution equipment.

[0192] The power distribution terminal characteristic information is obtained based on the container status information, terminal status information, and application status information.

[0193] Obtain the target software-defined terminal operating system evaluation model, input the power distribution terminal feature information into the target software-defined terminal operating system evaluation model, and obtain the output power distribution equipment feature information;

[0194] The evaluation results of the software-defined terminal operating system are obtained based on the characteristic information of the power distribution equipment.

[0195] Figure 6 A schematic diagram of the structure of a computer system suitable for implementing the method according to embodiments of the present disclosure is shown.

[0196] like Figure 6 As shown, the computer system includes a processing unit that can execute various methods described above based on a program stored in a read-only memory (ROM) or a program loaded from a storage portion into a random access memory (RAM). The RAM also stores various programs and data required for the operation of the computer system. The processing unit, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0197] The following components are connected to the I / O interface: input sections including keyboards, mice, etc.; output sections including cathode ray tubes (CRTs), liquid crystal displays (LCDs), and speakers; storage sections including hard disks, etc.; and communication sections including network interface cards such as LAN cards and modems. The communication section performs communication processes via a network such as the Internet. Drives are also connected to the I / O interface as needed. Removable media, such as disks, optical disks, magneto-optical disks, semiconductor memories, etc., are installed on the drive as needed so that computer programs read from them can be installed into the storage section as needed. The processing unit can be implemented as a CPU, GPU, TPU, FPGA, NPU, etc.

[0198] In particular, according to embodiments of this disclosure, the methods described above can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program tangibly embodied on a machine-readable medium, the computer program containing program code for performing the methods described above. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium.

[0199] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0200] The units or modules described in the embodiments of this disclosure can be implemented in software or programmable hardware. The described units or modules can also be located in a processor, and the names of these units or modules do not necessarily constitute a limitation on the unit or module itself.

[0201] In another aspect, this disclosure also provides a computer-readable storage medium, which may be a computer-readable storage medium included in the electronic device or computer system described above; or it may be a standalone computer-readable storage medium not assembled into a device. The computer-readable storage medium stores one or more programs, which are used by one or more processors to perform the methods described in this disclosure.

[0202] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

Claims

1. A model training method, characterized in that, The method comprises the following steps: obtaining container state information of a target container on a power distribution terminal in a first period, terminal state information of the power distribution terminal in the first period, application state information of a target application, and power distribution equipment state information of a target power distribution equipment in a second period, wherein the target container is used to carry the target application, the target application is used to manage the target power distribution equipment, the first period is earlier than the second period, the power distribution terminal is a terminal with software-defined function, and the target power distribution equipment is an equipment managed by the target application carried by the target container on the power distribution terminal; obtaining power distribution terminal feature information according to the container state information, the terminal state information, and the application state information, and obtaining power distribution equipment feature information according to the power distribution equipment state information; obtaining a software-defined terminal operating system evaluation model, training the software-defined terminal operating system evaluation model according to the power distribution terminal feature information and the power distribution equipment feature information to obtain a target software-defined terminal operating system evaluation model, and using the target software-defined terminal operating system evaluation model to predict power distribution equipment feature information corresponding to power distribution equipment state information collected in a later period according to container state information, terminal state information, and application state information collected in an earlier period, so as to obtain corresponding software-defined terminal operating system evaluation results according to the power distribution equipment feature information, and determine whether the working condition of the power distribution equipment in a later period meets corresponding requirements under the management of the application.

2. The model training method of claim 1, wherein, Before the step of obtaining the power distribution terminal feature information according to the container state information, the terminal state information, and the application state information, the method further comprises the following steps: obtaining a container state weight corresponding to the container state information, a terminal state weight corresponding to the terminal state information, and an application state weight corresponding to the application state information; the step of obtaining the power distribution terminal feature information according to the container state information, the terminal state information, and the application state information comprises the following step: obtaining the power distribution terminal feature information according to the container state information, the container state weight, the terminal state information, the terminal state weight, the application state information, and the application state weight.

3. The model training method of claim 1, wherein, The container state information is used to indicate at least one of the following: creation time of the target container, start time of the target container, start number of the target container, start speed of the target container, stop time of the target container, deletion time of the target container, version number of the target container, network address of the target container, network speed of the target container, container identifier of the target container, resource occupancy rate of the target container, process number of the target container, thread number of the target container, port used by the target container, interface used by the target container, and working log of the target container.

4. The model training method of claim 1, wherein, The terminal state information is used to indicate at least one of a process management function of the target power distribution terminal, a file system type supported by the target power distribution terminal, a hardware module type of a supported driver of the target power distribution terminal, a network protocol supported by the target power distribution terminal, a log management function supported by the target power distribution terminal, a resource occupancy rate of the target power distribution terminal, and an operating system version number of the target power distribution terminal, and the target power distribution terminal is a first time period power distribution terminal.

5. The model training method of claim 1, wherein, The application program state information is used to indicate at least one of a version number of the target application program, a number of processes corresponding to the target application program, a process identifier corresponding to the target application program, an installation time of the target application program, a configuration time of the target application program, a stop time of the target application program, a restart time of the target application program, an uninstallation time of the target application program, a version number of the target application program, an upgrade time of the target application program, and a resource occupancy rate of the target application program.

6. The model training method of any one of claims 1-5, characterized in that, Before the software-defined terminal operating system evaluation model is trained according to the power distribution terminal feature information and the power distribution equipment feature information, the method further comprises: receiving an updated weight parameter sent by the edge server, and updating the software-defined terminal operating system evaluation model according to the updated weight parameter; The software-defined terminal operating system evaluation model is trained according to the power distribution terminal feature information and the power distribution equipment feature information to obtain a target software-defined terminal operating system evaluation model, comprising: training the updated software-defined terminal operating system evaluation model by taking the power distribution terminal feature information as input and taking the power distribution equipment feature information as output; in response to the trained software-defined terminal operating system evaluation model converging, storing the trained software-defined terminal operating system evaluation model as the target software-defined terminal operating system evaluation model.

7. The model training method of claim 6, wherein, The method further comprises: in response to the trained software-defined terminal operating system evaluation model not converging, obtaining a gradient update vector according to the trained software-defined terminal operating system evaluation model, and sending the gradient update vector to the edge server.

8. A software-defined terminal operating system evaluation method, characterized by, comprising: obtaining container state information of a target container on a third time period power distribution terminal, terminal state information of the third time period power distribution terminal, and application program state information of a target application program, wherein the target container is used to carry the target application program, the target application program is used to manage a target power distribution equipment, the power distribution terminal is a terminal with a software-defined function, and the target power distribution equipment is an equipment managed by the target application program carried by the target container on the power distribution terminal; obtaining power distribution terminal feature information according to the container state information, the terminal state information, and the application program state information; obtaining a target software-defined terminal operating system evaluation model, inputting the power distribution terminal feature information into the target software-defined terminal operating system evaluation model to obtain output power distribution equipment feature information; According to the software-defined terminal operating system evaluation result obtained according to the power distribution equipment feature information, it is determined whether the working condition of the power distribution equipment under the management of the application program in a third period later than the first period meets the corresponding requirement.

9. A model training apparatus characterized by comprising: The method comprises: a first information acquisition module configured to acquire container state information of a target container on a power distribution terminal, terminal state information of the power distribution terminal in a first period, application program state information of a target application program, and power distribution equipment state information of a target power distribution equipment in a second period, wherein the target container is used to carry the target application program, the target application program is used to manage the target power distribution equipment, the first period is earlier than the second period, the power distribution terminal is a terminal with software-defined function, and the target power distribution equipment is a device managed by the target application program carried by the target container on the power distribution terminal; a first feature acquisition module configured to acquire power distribution terminal feature information according to the container state information, the terminal state information, and the application program state information, and acquire power distribution equipment feature information according to the power distribution equipment state information; a model training module configured to acquire a software-defined terminal operating system evaluation model, train the software-defined terminal operating system evaluation model according to the power distribution terminal feature information and the power distribution equipment feature information, and acquire a target software-defined terminal operating system evaluation model, wherein the target software-defined terminal operating system evaluation model is used to estimate power distribution equipment feature information corresponding to power distribution equipment state information collected in a later period according to container state information, terminal state information, and application program state information collected in an earlier period, so as to acquire a corresponding software-defined terminal operating system evaluation result according to the power distribution equipment feature information, and determine whether the working condition of the power distribution equipment in a later period under the management of the application program meets the corresponding requirement. 10.The model training apparatus of claim 9, wherein, The first feature acquisition module is further configured to: acquire a container state weight corresponding to the container state information, a terminal state weight corresponding to the terminal state information, and a program state weight corresponding to the application program state information; The first feature acquisition module is specifically configured to: acquire the power distribution terminal feature information according to the container state information, the container state weight, the terminal state information, the terminal state weight, the application program state information, and the program state weight. 11.The model training apparatus of claim 9, wherein, The container state information is used to indicate at least one of the following: creation time of the target container, start time of the target container, start number of the target container, start speed of the target container, stop time of the target container, deletion time of the target container, version number of the target container, network address of the target container, network speed of the target container, container identifier of the target container, resource occupancy rate of the target container, process number of the target container, thread number of the target container, port used by the target container, interface used by the target container, and working log of the target container. 12.The model training apparatus of claim 9, wherein, The terminal state information is used to indicate at least one of a process management function of the target power distribution terminal, a file system type supported by the target power distribution terminal, a hardware module type of a supported driver of the target power distribution terminal, a network protocol supported by the target power distribution terminal, a log management function supported by the target power distribution terminal, a resource occupancy rate of the target power distribution terminal, and an operating system version number of the target power distribution terminal, and the target power distribution terminal is a first time period power distribution terminal. 13.The model training apparatus of claim 9, wherein, The application program state information is used to indicate at least one of a version number of the target application program, a number of processes corresponding to the target application program, a process identifier corresponding to the target application program, an installation time of the target application program, a configuration time of the target application program, a stop time of the target application program, a restart time of the target application program, an uninstallation time of the target application program, a version number of the target application program, an upgrade time of the target application program, and a resource occupancy rate of the target application program. 14.The model training apparatus according to any one of claims 9-13, characterized in that, The model training module is further configured to: receive an updated weight parameter sent by the edge server, and update the software-defined terminal operating system evaluation model according to the updated weight parameter; The model training module is further configured to: train the updated software-defined terminal operating system evaluation model by taking the power distribution terminal feature information as input and taking the power distribution equipment feature information as output; in response to the trained software-defined terminal operating system evaluation model converging, store the trained software-defined terminal operating system evaluation model as the target software-defined terminal operating system evaluation model. 15.The model training apparatus of claim 14, wherein, The model training module is further configured to: in response to the trained software-defined terminal operating system evaluation model not converging, obtain a gradient update vector according to the trained software-defined terminal operating system evaluation model, and send the gradient update vector to the edge server.

16. A software-defined terminal operating system evaluation apparatus, comprising: comprising: The second information acquisition module is configured to acquire container state information of a target container on a power distribution terminal in a third time period, terminal state information of the power distribution terminal in the third time period, and application program state information of a target application program, wherein the target container is used to carry the target application program, the target application program is used to manage a target power distribution equipment, the power distribution terminal is a terminal with a software-defined function, and the target power distribution equipment is an equipment managed by the target application program carried by the target container on the power distribution terminal; The second feature acquisition module is configured to acquire power distribution terminal feature information according to the container state information, the terminal state information, and the application program state information; The third feature acquisition module is configured to acquire a target software-defined terminal operating system evaluation model, input the power distribution terminal feature information into the target software-defined terminal operating system evaluation model, and acquire output power distribution equipment feature information. The evaluation result obtaining module is configured to obtain a software-defined terminal operating system evaluation result according to the power distribution equipment characteristic information, so as to determine whether the working condition of the power distribution equipment in the application management after the third time period meets the corresponding requirement.

17. An electronic device, comprising: A computer program product including a memory and a processor; wherein the memory is configured to store one or more computer instructions; wherein the one or more computer instructions are executed by the processor to implement the method steps of any one of claims 1-8.

18. A computer readable storage medium having stored thereon computer instructions, wherein, The computer instructions are executed by the processor to implement the method steps of any one of claims 1-8.

Images