A method for evaluating the capabilities of network entities based on digital twins

Abstract

This invention relates to a method for evaluating the capabilities of network entities based on digital twins, belonging to the field of network capability evaluation technology, and solves the problem of poor accuracy in existing network entity capability evaluation results. The method includes: mapping network entities to network digital twins; obtaining all task flows from the task start node to the task end node in the network digital twin and their equivalent transfer functions; determining the capability of each task flow based on its equivalent transfer function; determining the importance of each task flow based on its equivalent transfer function and the equivalent transfer functions of all task flows; and obtaining the capability evaluation result of the network entity based on the capability and importance of all task flows.

Description

Technical Field

[0001] This invention relates to the field of network capability evaluation technology, and in particular to a method for evaluating the capabilities of network entities based on digital twins. Background Technology

[0002] In recent years, the internet and mobile internet industries have flourished, reaching a stage of technological maturity and solidified business models. Currently, various industries have greatly expanded online work methods and business operations, making the integration of the physical and virtual worlds, and digital services based on digital twins, a key development focus. Furthermore, the emergence of the metaverse concept, geared towards future application scenarios, is gradually becoming a future integrated solution based on existing mobile internet development, comprehensively integrating various application models, and utilizing various key technologies.

[0003] In the future, the metaverse will become a parallel space to the physical world in which people work and live. During its synchronous operation and interaction with the physical world, a quantitative evaluation method is needed to assess how digital twins in the virtual world simulate physical entities and the quality of their simulation. However, physical entities are complex and dynamically changing, making it difficult to evaluate their capabilities in a fixed way. Therefore, exploring a capability evaluation method for digital twin systems used in metaverse applications is particularly important.

[0004] Since the 1960s, researchers in this field have been studying system capability evaluation, and after many years of development, they have achieved considerable research results. Commonly used capability evaluation methods can be divided into four categories. First, expert evaluation methods, which construct a system capability evaluation index system based on extensive expert opinions and obtain index weights through multi-expert comprehensive evaluation. This method can evaluate and predict indicators that are difficult to quantify due to a lack of data, but it inevitably leads to a certain degree of subjectivity and bias in the evaluation and prediction results. Second, analytical methods, which are effective in overcoming estimation errors caused by coarse quantification, compare the performance of systems with a uniform quantification resolution, and can also utilize the analytic hierarchy process (AHP) and its extensions to evaluate system capability. This method essentially calculates the evaluation values ​​of capability indicators by establishing mathematical models; it is simple to calculate and easy to understand and apply. Its disadvantage is that the factors considered in the model construction process are limited, the results are too abstract, and the system lacks objectivity. Third, simulation methods, representative of which include capability evaluation methods based on system dynamics models or artificial neural networks, as well as evaluation methods based on multi-agent and automatic control system theories. This evaluation method simulates the actual operation or development process of the system, but it suffers from high system complexity and weak representation of the connection relationships between nodes. Fourthly, network analysis, which compensates for the shortcomings of agent-based modeling and other methods in modeling system information interaction, abstracts various entities in complex systems as nodes and various connection relationships as links. It has become an important method for studying system characteristics in recent years. Characterizing system capabilities from the perspective of complex networks is an evaluation method that closely reflects the structural characteristics of the system and can better represent the interaction relationships between nodes.

[0005] The above analysis shows that, in the process of effectively evaluating system capabilities, it is necessary to perform network-based characterization of the system and study its capabilities on this basis.

[0006] Research on the capability evaluation of digital twin systems for metaverse applications is limited and still in its early stages. Relevant studies, based on networked representation techniques, model the connection relationships between nodes and links to evaluate the transmission capabilities of communication networks serving as the framework for the virtual world. Modeling nodes, receiving and transmitting queues allows for the generation of expressions and lower bounds for performance parameters through analysis. Furthermore, some studies equate digital twin systems to a topology composed of nodes, utilizing corresponding topological features (such as degree distribution and clustering coefficients) to evaluate system capabilities. Additionally, the states within the digital twin system can be partitioned, and the network system capability can be solved by analyzing the transition processes between these states.

[0007] The above analysis shows that, in order to measure the capabilities of a digital twin system, it is necessary to characterize its structure and, based on this, further study the modeling and evaluation methods of its capabilities.

[0008] In summary, research on the evaluation of the capabilities (corresponding network entity capabilities) of digital twin systems for metaverse applications, both domestically and internationally, is significantly insufficient, mainly due to the following problems: 1) Existing networked modeling for capability evaluation focuses more on the topological properties of the network model, neglecting the practical significance of the connections between entities within the digital twin system; 2) Existing capability-solving models for networked systems often only analyze the importance of each link, with insufficient research on tasks and activities that significantly impact system capabilities. Furthermore, the determination of link importance primarily relies on analyzing the impact of the presence or absence of a particular link on information transmission capabilities, resulting in a rather one-sided evaluation. Summary of the Invention

[0009] Based on the above analysis, the embodiments of the present invention aim to provide a network entity capability evaluation method based on digital twins, in order to solve the problem of poor accuracy of existing network entity capability evaluation results.

[0010] This invention discloses a method for evaluating the capabilities of network entities based on digital twins, comprising:

[0011] Map network entities to network digital twins, and obtain all task flows and their equivalent transfer functions from the task start node to the task end node in the network digital twin;

[0012] The capabilities of each task flow are determined based on its equivalent transfer function.

[0013] The importance of each task flow is determined based on the equivalent transfer function of each task flow and all task flows.

[0014] Based on the capabilities and importance of all task flows, the capability evaluation results of the network entities are obtained.

[0015] Based on the above solution, the present invention also makes the following improvements:

[0016] Furthermore, the task flow represents the nodes and directed links that are passed sequentially during the transmission of task activities from the task start node to the task end node;

[0017] The equivalent transfer function for obtaining all task flows includes:

[0018] Take any two adjacent nodes in the task flow and the directed link between them as a basic capability unit; in the basic capability unit, the node pointed to by the directed link is the receiving node, and the other node is the executing node.

[0019] For each basic capability unit, a transfer function is constructed based on the capability distribution of the execution node's task activities and the activity probability of transmitting task activities from the execution node to the receiving node.

[0020] The product of the transfer functions of all the basic units of capability in each task flow is taken as the equivalent transfer function of the corresponding task flow.

[0021] Furthermore, the transfer function of the basic capability unit:

[0022]

[0023] Where, p ij This represents the probability of transmitting task activities from execution node i to receiving node j. The characteristic function represents the distribution of the execution node's ability to perform tasks, t∈R, where R represents the real number field and t takes any real number.

[0024] Furthermore, the sum of the activity probabilities of transmitting task activities from the same execution node to all its target receiving nodes is 1.

[0025] Furthermore, the importance of the l-th task flow, im(l), satisfies:

[0026]

[0027] in, l ranges from 1 to L, where L represents the total number of all task flows, and t is any real number.

[0028] Furthermore, the ability to determine the corresponding task flow based on the equivalent transfer function of each task flow includes:

[0029] According to the equivalent transfer function W of the l-th task flow l (t), determine the equivalent characteristic function of the l-th task flow.

[0030]

[0031] Based on the equivalent characteristic function of the l-th task flow The ability to determine the l-th task flow.

[0032] Furthermore, the capability of the task flow refers to the average capability of the task flow; the capability evaluation result of the network entity is the average capability evaluation result.

[0033] The average capacity E(l) of the l-th task flow satisfies:

[0034]

[0035] Furthermore, the average ability evaluation result E is:

[0036]

[0037] Furthermore, the capability of the task flow refers to the variance capability of the task flow, and the capability evaluation result of the network entity is the variance capability evaluation result;

[0038] The variance capability D(l) of the l-th task flow satisfies:

[0039]

[0040] Furthermore, the variance capability evaluation result D is:

[0041]

[0042] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0043] The network entity capability evaluation method based on digital twins provided by this invention fully considers the practical significance of the connection relationships between entities in the digital twin system to form basic capability units and equivalent transfer functions of each task flow. At the same time, it analyzes the importance of task flows from the perspective of task activities. The network capability evaluation results obtained are more comprehensive and accurate, and have extremely important practical significance.

[0044] In this invention, the above-described technical solutions can be combined with each other to achieve more preferred combinations. Other features and advantages of this invention will be set forth in the following description, and some advantages may become apparent from the description or be learned by practicing the invention. The objects and other advantages of this invention can be realized and obtained from what is particularly pointed out in the description and drawings. Attached Figure Description

[0045] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts.

[0046] Figure 1 Here is a flowchart of the network entity capability evaluation method based on digital twins provided in Embodiment 1 of the present invention;

[0047] Figure 2 This is a model architecture diagram of a capability-oriented network digital twin provided in Embodiment 1 of the present invention;

[0048] Figure 3 The structure of the basic capability unit provided in Embodiment 1 of the present invention;

[0049] Figure 4The topology of the network entity provided in Embodiment 2 of the present invention;

[0050] Figure 5 This is a model architecture diagram of the network digital twin provided in Embodiment 2 of the present invention. Detailed Implementation

[0051] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0052] The capabilities of network entities and network digital twins refer to the effectiveness with which the physical world entities simulated by the network digital twin complete tasks. The capabilities of a network digital twin can be used as a measure of the capabilities of the network entities it simulates. The evaluation of system capabilities is based on the effective evaluation of the capabilities of each entity in the system. It uses the performance of each entity as a foundation and modeling and simulation as a means to comprehensively evaluate the system capabilities.

[0053] Example 1

[0054] A specific embodiment of the present invention discloses a method for evaluating the capabilities of network entities based on digital twins, the flowchart of which is shown below. Figure 1 As shown, it includes the following steps:

[0055] Step S1: Map the network entity into a network digital twin, and obtain all task flows from the task start node to the task end node in the network digital twin and their equivalent transfer functions;

[0056] In this embodiment, the network digital twin is obtained by mapping network entities. Exemplarily, the network entities simulated by the network digital twin mainly include computer communication networks. Common computer communication networks are composed of terminal devices, routers, switches, and network equipment connected together. Depending on the network size, they can be divided into local area networks (LANs), metropolitan area networks (MANs), or wide area networks (WANs). Different topologies can be formed based on the different topologies of the computer communication network. In a computer communication network, each device is considered a node, and the connection between devices is considered a link, thus forming a corresponding topology. There are many types of computer communication networks, ranging from small communication networks formed between multiple hosts and printers in an office to large-scale urban rail transit network communication systems. Based on the topology of the computer communication network, nodes and links in the network digital twin can be mapped. Specifically, the nodes and links in the network entity are mapped to the nodes and links in the network digital twin, respectively. The distribution of the ability of nodes in the network digital twin to perform tasks is consistent with the distribution of the ability of corresponding nodes in the network entity to perform tasks. The probability of transmitting tasks from one node to another in the network digital twin is consistent with the probability of transmitting tasks between corresponding nodes in the network entity.

[0057] Nodes in a network digital twin are divided into three categories: task initiation nodes, transmission nodes, and task termination nodes. The process of a network digital twin executing a task can be understood as the application generating and sending information through the task initiation node, transmitting it through one or more transmission nodes, and finally reaching the task termination node. The model architecture diagram of a capability-oriented network digital twin is shown below. Figure 2 As shown. In this embodiment, the task flow represents the nodes and directed links sequentially traversed during the transmission of task activity from the task start node to the task end node. For example, in... Figure 2 In this context, S1→R1→T1 represents a task flow. Within a network digital twin, depending on the different task activities and their transmission paths, a network digital twin may contain several task flows. These task flows form the basis for evaluating the capabilities of the network digital twin.

[0058] To accurately measure network capabilities, it is necessary to establish equivalent transfer functions for each task flow within the network digital twin. The capabilities of each task flow during its transmission through each link are random variables following different probability distributions. Therefore, the equivalent transfer function of the network digital twin is a probability distribution of multiple random variables, making it difficult to solve. Thus, in this embodiment, we consider establishing and solving mathematical models for the basic capability units within the network digital twin.

[0059] Specifically, for each task flow, any two adjacent nodes in the task flow and the directed link between them are taken as a basic unit of capability, such as... Figure 3 As shown, this basic capability unit is the fundamental computational unit in the process of evaluating the capabilities of a digital twin. In this basic capability unit, node j, pointed to by the directed link, serves as the receiving node, and another node i serves as the executing node; wherein, the directed link (i,j) carries the task activities between i and j, and p... ij E represents the probability of transmitting task activities from execution node i to receiving node j. ij This represents the capability of the task activity carried on the directed link (i,j). The activity probability of the task activity represents the likelihood of its execution. It should be noted that the sum of the activity probabilities of transmitting task activities from the same execution node to all its target receiving nodes is 1. There is also a flow of capability between nodes. The capability of a node to transmit task activities through the link is represented by the transfer function of the capability basic unit. For each capability basic unit, the transfer function of the capability basic unit is constructed based on the capability distribution of the execution node to execute task activities and the activity probability of transmitting task activities from the execution node to the receiving node; the transfer function of the capability basic unit is shown in formula (1):

[0060]

[0061] Where, p ij This represents the probability of transmitting task activities from execution node i to receiving node j. The characteristic function represents the distribution of the execution node's ability to perform tasks, t∈R, where R represents the real number field and t takes any real number.

[0062] In probability theory, closed-form expressions for probabilistic features can be obtained through moment generating functions or characteristic functions. In practical modeling, considering that some random distributions may not have moment generating functions, and that characteristic functions are the Fourier transforms of probability density functions and exist for any distribution, they are introduced into the process of solving for the equivalent transfer function of a network digital twin.

[0063] In the basic unit of capability of a network digital twin, the capability E of the task activities carried on the directed link (i,j) between nodes. ij It is a continuous random variable with probability density function f(E) ij The capability E of each basic capability unit can be calculated. ij The characteristic function is:

[0064]

[0065] Where t∈R. According to relevant knowledge of probability theory, the characteristic function... The calculation follows these rules:

[0066] First, if E1, E2, ..., Ek ,…,E K Let K be K independent random variables, each with a corresponding characteristic function, denoted as . Then the characteristic function of the sum of K independent random variables Y is the product of the characteristic functions of each random variable. Therefore, we have...

[0067] Second, the characteristic function E ij The value of the nth derivative of the random variable E at t=0 is the value of the random variable E. ij The nth order origin moment. This law will be used in steps S2 and S3 to calculate the task flow capability.

[0068] If a task flow contains K basic capability units connected in series, then the probability of that task flow occurring is: p k This represents the probability that the k-th basic capability unit successfully executes the task activity. Combining this with Rule 1, we know that when finding the equivalent transfer function of a task flow, the product of the transfer functions of all basic capability units in each task flow can be used as the equivalent transfer function of that task flow. That is, assuming the l-th task flow includes K cascaded basic capability units, then the equivalent transfer function of the l-th task flow is:

[0069]

[0070] Among them, W k (t) is the transfer function of the k-th basic capability unit in the l-th task flow.

[0071] Furthermore, considering that different task flows are connected in parallel, the sum of the equivalent transfer functions of the parallel task flows can be obtained according to the probability addition formula, which serves as the equivalent transfer function W(t) of the network digital twin, as shown in formula (4).

[0072]

[0073] Step S2: Determine the capabilities of the corresponding task flow based on the equivalent transfer function of each task flow;

[0074] A network digital twin comprises multiple task flows, each with a capability that is a random function. Evaluating the network's capabilities requires calculating the capability of each task flow separately.

[0075] Specifically, W l (t) can be expressed as Besides the form , it can also be expressed in the following forms:

[0076]

[0077] Where, p lThis represents the product of the probabilities that all basic capability units in the l-th task flow successfully execute the task activity. This represents the product of the characteristic functions of the capability distribution of all basic capability units in the l-th task flow. According to formula (4):

[0078]

[0079] That is, the equivalent transitive probability p l =W l (0).

[0080] Therefore, we can use the equivalent transfer function W of the l-th task flow. l (t), determine the equivalent characteristic function of the l-th task flow.

[0081]

[0082] Based on the equivalent characteristic function of the l-th task flow The ability to determine the l-th task flow;

[0083] Where l ranges from 1 to L, and L represents the total number of all task flows.

[0084] According to Rule 2 described above, the value of the nth derivative of the characteristic function at t=0 is the nth raw moment of the random variable. It should be noted that raw moments of different orders have different meanings and corresponding capability representations; in this embodiment, the following two capability representation methods are given:

[0085] (1) The capability of a task flow refers to its average capability. In this case, by calculating the first-order moment at the origin, it can be known that the average capability E(l) of the l-th task flow satisfies:

[0086]

[0087] Here, the first-order moment at the origin represents the average value of the capability.

[0088] (2) The capability of a task flow refers to its variance capability. In this case, by calculating the second-order raw moments, it can be known that the variance capability D(l) of the l-th task flow satisfies:

[0089]

[0090] Here, the second-order origin moment characterizes the volatility of energy.

[0091] It should be noted that other higher-order origin moments also have corresponding capability representation angles. In practical applications, the desired higher-order origin moments can be selected according to the different needs of relevant technical personnel for capability evaluation, so as to obtain capability calculation results from the corresponding angles and use them for subsequent capability evaluation.

[0092] Step S3: Determine the importance of each task flow based on the equivalent transfer function of each task flow and all task flows;

[0093] Based on the above steps, the capability equivalence transfer function of the network digital twin and the average / fluctuating capability of each task flow have been obtained. The next step is to study the distinction in importance among each task flow during the capability evaluation process of the digital twin system.

[0094] Based on research methods in related fields, the importance of a model entity is quantified as the impact of the generation or termination of task flows during task execution on the rate of change of relevant system parameters. From the perspective of task flow importance in a network digital twin system, this refers to the degree to which changes in task flow capabilities affect the overall capability of the network digital twin. To analyze task flow importance, the derivative function of the capability equivalent transfer function of the digital twin network model with respect to a specific task flow capability parameter can be obtained. Combining this with the properties of the characteristic function, the evaluation value of task flow importance can be derived.

[0095] Specifically, the equivalent transfer function W(t) of the network digital twin is used to represent the equivalent transfer function W of the l-th task flow. l (t) Taking the derivative, we obtain the derivative function of the network capability parameters of the l-th task flow. Its first-order raw moment represents the degree of influence of the l-th task flow on the network digital twin capability. Let the proportion of the influence of the l-th task flow on the network digital twin capability to the influence of all task flows on the network model capability be the importance of the l-th task flow, denoted as im(l). That is, the importance im(l) of the l-th task flow satisfies:

[0096]

[0097] Step S4: Based on the capabilities and importance of all task flows, obtain the capability evaluation results of the network entities.

[0098] In this embodiment, the capability of a network entity is equal to the capability of its simulated network digital twin. Therefore, the capability of the solved network digital twin can be used as the capability of the corresponding network entity. The overall capability of the network digital twin is characterized by a weighted sum of the products of the importance and capability of each task flow. Specifically,

[0099] When the capability of the task flow refers to the average capability of the task flow, the capability evaluation result of the generated network digital twin is the average capability evaluation result of the network digital twin;

[0100] The average capability assessment result of the network digital twin (i.e., the average capability assessment result of the network entities) E is:

[0101]

[0102] When the capability of the task flow refers to the variance capability of the task flow, the capability evaluation result of the generated network digital twin is the variance capability evaluation result of the network digital twin.

[0103] The variance capability evaluation result of the network digital twin (i.e., the variance capability evaluation result of the network entity) D is:

[0104]

[0105] Example 2

[0106] Specific embodiment 2 of the present invention discloses a method for evaluating the capabilities of network entities based on digital twins, to verify the effectiveness of the method in the embodiment. The specific process is described as follows:

[0107] The task flow chain relationships and work sequence of the network digital twin are determined by the functional task planning of the physical world network entity system simulated by the network digital twin, and generated according to business requirements. Based on this, to verify the correctness and effectiveness of the algorithm, capability calculations are performed using user data generated by the software.

[0108] Assuming the network entity topology is as follows: Figure 4 As shown, in the network entity, terminal node S is the node that issues the task, terminal node T is the node that receives the task, and R1-R3 are the transmission nodes between the two terminal nodes. The physical device entities corresponding to the transmission nodes are routers or switches. The model architecture diagram of the network digital twin corresponding to this network entity is as follows. Figure 5 As shown.

[0109] Based on the capability performance data of each node, the probability of each task activity and the capability distribution of each node are shown in Table 1. It is assumed that the capability of each task activity follows a normal distribution, and the overall capability of the network digital twin needs to be evaluated. In practical applications, the above probability and capability distribution can be obtained through experiments on network entities or simulations of the corresponding network digital twin.

[0110] Table 1. Probability of each task activity and capability distribution of each node.

[0111]

[0112] Based on Table 1, the task flows and their equivalent transfer functions can be obtained, as shown in Table 2:

[0113] Table 2 Task Flow and its Equivalent Transfer Function

[0114]

[0115] Correspondingly, the equivalent transfer function of a network digital twin:

[0116]

[0117] Based on the obtained equivalent transfer function, the average capability of each task flow can be obtained, as shown in Table 3.

[0118] Table 3 Average Capabilities of Each Task Flow

[0119]

[0120]

[0121] By combining the equivalent transfer functions of each task flow, the importance of each task flow can be obtained, as shown in Table 4.

[0122] Table 4 Importance of each task flow

[0123]

[0124] The overall average capability of the network digital twin is calculated by weighting and summing the products of the importance of each task flow and the average capability of that task flow, i.e.:

[0125]

[0126] Therefore, by performing the above calculations, the capability distribution of node task activities and the overall average capability of the network digital twin under the condition of a determined activity probability can be calculated. Correspondingly, the capability evaluation result of the network entity simulated by the network digital twin is also obtained, which is 2.192. It should be emphasized that this capability evaluation result is based on the probability of nodes transmitting tasks and the capability distribution of nodes executing task activities during the task activity transmission process of the network entity. Therefore, when the above probability and capability distribution parameters in the network entity change, the capability evaluation result will also change accordingly, but it can still be obtained according to the above calculation method.

[0127] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0128] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for evaluating capabilities of network entities based on digital twins, characterized in that, The method comprises the following steps: mapping a network entity into a network digital twin, obtaining all task flows from a task start node to a task end node in the network digital twin and an equivalent transfer function of each task flow; determining the capability of each task flow based on the equivalent transfer function of each task flow; determining the importance of each task flow based on the equivalent transfer function of each task flow and all task flows; obtaining the capability evaluation result of the network entity based on the capability and the importance of all task flows; the task flow represents the nodes and directed links that are sequentially passed through in the process of transmitting task activities from the task start node to the task end node; the equivalent transfer function of all task flows comprises the following steps: taking any two adjacent nodes and the directed link therebetween in one task flow as a capability basic unit; in the capability basic unit, the node to which the directed link points is an accept node, and the other node is an execution node; for each capability basic unit, constructing the transfer function of the capability basic unit based on the capability distribution of the execution node in executing task activities and the activity probability of transmitting task activities from the execution node to the accept node; taking the product of the transfer functions of all capability basic units in each task flow as the equivalent transfer function of the corresponding task flow; the transfer function of the capability basic unit comprises the following steps: where p ij denotes the activity probability of transmitting a task activity from the performing node i to the accepting node j, denotes the characteristic function of the capability distribution of the performing node to perform a task activity, t ∈ R, R denotes the field of real numbers, t takes any real number.

2. The digital twin based network entity capability evaluation method of claim 1, wherein, the sum of the activity probabilities of transmitting task activities from the same execution node to all accept nodes to which the execution node points is 1.

3. The digital twin based network entity capability evaluation method according to claim 1 or 2, c h a r a c t e r i z e d by, the importance im(l) of the lth task flow satisfies: wherein, l takes values from 1 to L, L denotes the total number of task flows, and t takes any real number.

4. The digital twin based network entity capability evaluation method of claim 3, wherein, the capability of each task flow comprises the following steps: According to the equivalent transfer function W of the lth task flow l (t), determine the equivalent characteristic function of the lth task flow Equivalent characteristic function according to the lth task flow Determine the capacity of the lth task flow.

5. The digital twin based network entity capability evaluation method of claim 4, wherein, the capability of the task flow refers to the average capability of the task flow; and the capability evaluation result of the network entity is an average capability evaluation result; the average capability E(l) of the lth task flow satisfies:

6. The digital twin based network entity capability evaluation method of claim 5, wherein, the average capability evaluation result E satisfies:

7. The digital twin based network entity capability evaluation method of claim 4, wherein, the capability of the task flow refers to the variance capability of the task flow; and the capability evaluation result of the network entity is a variance capability evaluation result; the variance capability D(l) of the lth task flow satisfies:

8. The digital twin based network entity capability evaluation method of claim 7, wherein, the variance capability evaluation result D satisfies:

Images