Multi-domain strategy collaborative dynamic decision optimization method, system, equipment and medium

Abstract

The invention relates to the technical field of telecommunication. By providing the multi-domain strategy collaborative dynamic decision optimization method, system, equipment and medium, the method comprises the following steps: extracting cross-domain common characteristics of network layer topology, equipment state and alarm information, retaining differential characteristics of a transmission network and a data network, and generating a uniform resource data set; on the basis of the unified data set, a global optimization strategy is generated through linkage decision making of a three-level collaboration module; matching the strategy into an atomic instruction set executable by a heterogeneous network manager by utilizing semantic ontology mapping and syntax tree matching; and according to the instruction execution feedback data, dynamically adjusting the strategy weight and carrying out iterative optimization, and generating an updated global strategy, so as to achieve the technical effects of improving the cross-domain resource perception precision, enhancing the hierarchical decision collaborative efficiency and reducing the multi-manufacturer instruction conversion error rate.

Description

Technical Field

[0001] This invention relates to the field of telecommunications technology, and in particular to dynamic decision optimization methods, systems, devices, and media for multi-domain strategy collaboration. Background Technology

[0002] Against the backdrop of the deep integration of smart grids and communication networks, collaborative management and control of multi-domain heterogeneous networks has become a core requirement for ensuring the reliable operation of critical infrastructure. Efficient scheduling and dynamic optimization of end-to-end cross-network services are directly related to the stability and service continuity of power communication systems, and are the underlying technological pillars supporting the digital transformation of smart grids.

[0003] However, the relevant multi-domain network collaborative management technologies have the following problems: lack of unified awareness of cross-domain resources, resulting in data silos; lack of hierarchical decision-making and dynamic linkage mechanism, resulting in an imbalance between resource utilization and business reliability targets; and heterogeneity of multi-vendor protocols, causing instruction conversion errors. Summary of the Invention

[0004] Therefore, it is necessary to provide dynamic decision optimization methods, systems, devices, and media for multi-domain strategy collaboration to address the aforementioned technical problems, so as to improve the accuracy of cross-domain resource perception, enhance the efficiency of hierarchical decision collaboration, and reduce the error rate of multi-vendor instruction conversion.

[0005] Firstly, this application provides a dynamic decision optimization method for multi-domain policy collaboration, the method comprising:

[0006] The network layer's topology information, device status information, and alarm information are dynamically perceived and processed based on cross-domain common feature extraction to generate a unified resource dataset; the cross-domain common feature extraction preserves the differentiated characteristics of the transmission network and the data network.

[0007] Based on a unified resource dataset, a three-level collaborative decision-making mechanism involving the transport layer collaboration module, the data layer collaboration module, and the cross-domain collaboration module is used to perform hierarchical collaborative decision-making and generate a global optimization strategy.

[0008] Based on a global optimization strategy, a dual adaptation mechanism of instruction semantic ontology mapping and protocol syntax tree matching is used to dynamically adapt instructions from multiple vendors, generating an atomic operation instruction set that can be executed by heterogeneous network management systems.

[0009] Based on the execution feedback data returned by the network layer after executing the atomic operation instruction set, a dynamic weight adjustment mechanism driven by parameter reliability evaluation is used to perform closed-loop verification and iterative optimization of the strategy, generating an updated global optimization strategy.

[0010] Furthermore, based on a global optimization strategy, a dual adaptation mechanism of instruction semantic ontology mapping and protocol syntax tree matching is used to dynamically adapt instructions from multiple vendors, generating a set of atomic operation instructions executable by heterogeneous network management systems, including:

[0011] Based on the semantic web rule reasoning engine, the global optimization strategy is processed by instruction semantic ontology mapping to generate device-independent atomic operation descriptors;

[0012] Using the following formula, based on a multi-vendor protocol syntax tree feature library, node matching is performed on device-independent atomic operation descriptors to generate intermediate expression for the protocol syntax:

[0013]

[0014] Among them, M(A) i ,T k ) represents the device-independent atomic operation descriptor A i With protocol syntax tree T k Match degree, V k Representation of syntax tree T k The set of nodes in ω l Let f represent the weight of the l-th feature, δ represent the feature matching function, and f l (A i ) represents the l-th feature of the atomic operation descriptor, f l (N j ) represents node N j The l-th feature, where L represents the total number of features;

[0015] The intermediate expressions of the protocol syntax are subjected to semantic-syntactic consistency verification to generate an atomic operation instruction set that can be executed by the heterogeneous network management system.

[0016] Furthermore, based on the semantic web rule inference engine, the global optimization strategy is processed by instruction semantic ontology mapping to generate device-independent atomic operation descriptors, including:

[0017] Based on the unified information model, the global optimization strategy is processed by business intent parsing to generate business operation metaphrases;

[0018] Based on the semantic web rule reasoning engine, network atomic function mapping is performed on business operation meta-languages ​​to generate atomic operation semantic graphs.

[0019] Based on the YANG data modeling language, a hierarchical node encapsulation process is performed on the atomic operation semantic graph to generate device-independent atomic operation descriptors.

[0020] Furthermore, semantic-syntactic consistency verification is performed on the intermediate expressions of the protocol syntax to generate a set of atomic operation instructions executable by the heterogeneous network management system, including:

[0021] Based on a multi-vendor semantic rule library, semantic integrity conflict detection is performed on intermediate expressions of protocol syntax to generate semantically compliant intermediate expressions.

[0022] Using the following formula, and through vendor-specific protocol syntax constraints, semantically compliant intermediate expressions are subjected to syntactic structure compliance transformation to generate a syntactically normalized instruction set:

[0023]

[0024] Among them, C compliance S represents the grammatical structure compliance assessment value, where n represents the total number of grammatical constraint rules, and S represents the grammatical structure compliance assessment value. m Represents the actual output of the m-th syntax structure, R m This represents the standard requirement for the m-th grammar rule, ∈ represents a small constant to prevent division by zero, and α m This represents the importance index of the m-th grammar rule;

[0025] The syntax-normalized instruction set is encapsulated into executable atomic operations to generate an executable atomic operation instruction set for heterogeneous network management systems.

[0026] Furthermore, based on a unified resource dataset, a hierarchical collaborative decision-making process is implemented through a three-level linkage decision-making mechanism involving the transport layer collaboration module, the data layer collaboration module, and the cross-domain collaboration module to generate a global optimization strategy, including:

[0027] Based on the unified resource dataset, the cross-domain end-to-end configuration policy generation process is performed through the transport layer collaboration module to generate the transport network end-to-end configuration policy.

[0028] Based on a unified resource dataset, risk level linkage determination and root cause tracing are performed through a data layer collaboration module to generate data network fault analysis results.

[0029] Based on the end-to-end configuration strategy of the transmission network and the fault analysis results of the data network, the cross-domain resource conflict resolution is carried out through the cross-domain collaboration module to generate a global optimization strategy.

[0030] Furthermore, based on the end-to-end configuration strategy of the transmission network and the fault analysis results of the data network, cross-domain resource conflict resolution is performed through the cross-domain collaboration module to generate a global optimization strategy, including:

[0031] Based on the service flow SLA constraints, fault-resource correlation analysis is performed on the end-to-end configuration strategy of the transmission network and the fault analysis results of the data network to generate fault-resource correlation analysis results.

[0032] The results of the fault-resource correlation analysis are quantitatively evaluated using link reliability indicators to generate a dynamic resource arbitration scheme.

[0033] Based on the dynamic resource arbitration scheme, a multi-objective Pareto front solution is performed to generate a global optimization strategy.

[0034] Furthermore, based on the execution feedback data returned by the network layer after executing the atomic operation instruction set, a dynamic weight adjustment mechanism driven by parameter reliability evaluation is used to perform closed-loop verification and iterative optimization of the strategy, generating an updated global optimization strategy, including:

[0035] Perform parameter reliability quantification and evaluation on the execution feedback data to generate link reliability indicators;

[0036] Based on the link reliability index, dynamic weight adjustment processing of the path calculation cost function is performed to generate an updated path calculation cost function.

[0037] Based on the updated path cost function, cross-domain resource rescheduling decision processing is performed to generate an updated global optimization strategy.

[0038] Secondly, this application also provides a dynamic decision optimization system for multi-domain strategy collaboration, the system comprising:

[0039] The resource awareness module is used to perform multi-domain resource dynamic awareness processing based on cross-domain common feature extraction on network layer topology information, device status information and alarm information, and generate a unified resource dataset; the cross-domain common feature extraction retains the differentiated individual characteristics of the transmission network and the data network.

[0040] The collaborative decision-making module is used to perform hierarchical collaborative decision-making based on a unified resource dataset, through a three-level linkage decision-making mechanism of the transport layer collaborative module, the data layer collaborative module, and the cross-domain collaborative module, to generate a global optimization strategy.

[0041] The instruction adaptation module is used to dynamically adapt instructions from multiple vendors based on a global optimization strategy and through a dual adaptation mechanism of instruction semantic ontology mapping and protocol syntax tree matching, generating an atomic operation instruction set that can be executed by heterogeneous network management systems.

[0042] The closed-loop optimization module is used to perform closed-loop verification and iterative optimization of the strategy based on the execution feedback data returned by the network layer after the execution of the atomic operation instruction set, through a dynamic weight adjustment mechanism driven by parameter reliability evaluation, and to generate an updated global optimization strategy.

[0043] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of any of the methods in the first aspect of this application.

[0044] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any of the methods in the first aspect of this application.

[0045] This application provides a dynamic decision-making optimization method, system, device, and medium for multi-domain policy collaboration. The method includes: performing multi-domain resource dynamic perception processing on network layer topology information, device status information, and alarm information based on cross-domain common feature extraction to generate a unified resource dataset; wherein the cross-domain common feature extraction preserves the differentiated individual characteristics of the transmission network and the data network; based on the unified resource dataset, performing hierarchical collaborative decision-making processing through a three-level linkage decision-making mechanism of the transmission layer collaboration module, the data layer collaboration module, and the cross-domain collaboration module to generate a global optimization strategy; based on the global optimization strategy, performing multi-vendor instruction dynamic adaptation processing through a dual adaptation mechanism of instruction semantic ontology mapping and protocol syntax tree matching to generate a set of atomic operation instructions executable by heterogeneous network management; based on the execution feedback data returned by the network layer after executing the atomic operation instruction set, performing policy closed-loop verification and iterative optimization processing through a dynamic weight adjustment mechanism driven by parameter reliability evaluation to generate an updated global optimization strategy, thereby achieving the technical effects of improving the accuracy of cross-domain resource perception, enhancing the efficiency of hierarchical decision-making collaboration, and reducing the error rate of multi-vendor instruction conversion. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 This is a flowchart of a dynamic decision optimization method for multi-domain strategy collaboration in one embodiment of the present invention;

[0048] Figure 2 In one embodiment of the present invention, a flowchart of a global optimization strategy is generated based on a unified resource dataset and a three-level linkage decision-making mechanism consisting of a transport layer collaboration module, a data layer collaboration module, and a cross-domain collaboration module.

[0049] Figure 3 This is a structural diagram of a dynamic decision optimization system for multi-domain strategy collaboration in one embodiment of the present invention. Detailed Implementation

[0050] To make the above-mentioned objects, features, and advantages of this application more apparent and understandable, the specific implementation methods of this application will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a full understanding of this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0051] First, the application scenarios of the embodiments of this application are described. The embodiments of this application provide a dynamic decision optimization method, system, device, and medium for multi-domain strategy collaboration applicable to, but not limited to, such scenarios. For example, ensuring stable transmission of highly sensitive cross-network services such as power dispatching and financial transactions to avoid interruptions; achieving efficient and unified management of multi-vendor hybrid deployment equipment to reduce human configuration errors; and quickly rebuilding core business channels to reduce socio-economic losses when communication networks are paralyzed by disasters such as typhoons and earthquakes.

[0052] As an illustration, the dynamic decision optimization method, system, device and medium for multi-domain strategy collaboration provided in this application embodiment can also be applied to other application scenarios. This is only an example and does not limit the specific application scenarios.

[0053] In one exemplary embodiment, such as Figure 1 As shown, this application provides a dynamic decision optimization method for multi-domain policy collaboration, which includes:

[0054] S101: Perform multi-domain resource dynamic perception processing based on cross-domain common feature extraction on the topology information, device status information and alarm information of the network layer to generate a unified resource dataset; among which, the cross-domain common feature extraction retains the differentiated individual characteristics of the transmission network and the data network.

[0055] Specifically, network topology information, device status information, and alarm information are collected, including information related to the transmission network and data network. Common features are extracted from the collected information from a cross-domain perspective, while retaining the unique characteristics of the transmission network and data network. Then, based on the extracted cross-domain common features and the retained unique characteristics, multi-domain resources are dynamically sensed and processed, and relevant information is integrated to generate a unified resource dataset that includes both cross-domain common features and the unique characteristics of the transmission network and data network.

[0056] S102: Based on a unified resource dataset, a three-level collaborative decision-making mechanism involving the transport layer collaboration module, the data layer collaboration module, and the cross-domain collaboration module is used to generate a global optimization strategy.

[0057] Specifically, based on a unified resource dataset, the transport layer coordination module analyzes resource data within the transport network and generates local transport layer policies by combining transport network orchestration and fault analysis capabilities. Simultaneously, the data layer coordination module analyzes data network resource data and generates local data layer policies using data network orchestration and fault analysis capabilities.

[0058] Subsequently, the local strategies of the transport layer and data layer and the unified resource dataset are input into the cross-domain collaboration module. This module comprehensively considers cross-domain resource mapping and fault results through cross-domain joint fault analysis, multi-layer path orchestration calculation and parameter reliability evaluation, and finally generates a global optimization strategy.

[0059] S103: Based on a global optimization strategy, it performs dynamic adaptation of instructions from multiple vendors through a dual adaptation mechanism of instruction semantic ontology mapping and protocol syntax tree matching, generating an atomic operation instruction set that can be executed by heterogeneous network management systems.

[0060] Specifically, semantic parsing is performed on the business requirements and network operations involved in the global optimization strategy to construct an instruction semantic ontology. This maps the abstract operational concepts in the strategy into standardized semantic entities, clarifying the semantic connotation and logical relationships of each operation. For the instruction formats of different network management vendors, their syntactic structures are analyzed, protocol syntax trees are constructed, and the syntactic rules and format specifications of each vendor's instructions are identified.

[0061] The standardized semantic entities in the semantic ontology are matched with the protocol syntax tree to find the correspondence between semantics and syntax. Based on the heterogeneous command characteristics of multi-vendor network management systems, semantic-to-syntax conversion is performed. The converted commands are verified and adjusted to ensure that they meet the syntax requirements of each vendor's network management system, generating an atomic operation command set executable by the heterogeneous network management system.

[0062] S104: Based on the execution feedback data returned by the network layer after executing the atomic operation instruction set, the strategy is closed-loop checked and iteratively optimized through a dynamic weight adjustment mechanism driven by parameter reliability evaluation, and an updated global optimization strategy is generated.

[0063] Specifically, the system obtains the execution feedback data returned after the network layer executes the atomic operation instruction set, including device operation results, network status changes, alarm information, etc., analyzes it, extracts key parameters such as operation success rate, network performance indicators, fault location information, etc., and generates a structured feedback dataset.

[0064] Based on the feedback dataset, the parameter reliability evaluation module is used to evaluate the reliability of the network parameters involved in the global optimization strategy. Combined with the preset evaluation index system, the deviation between the actual performance of each parameter and the expected target is quantified, and the parameter reliability evaluation results are generated.

[0065] Based on the parameter reliability assessment results, a dynamic parameter adjustment mechanism is used to adaptively adjust the decision weights of each business module in the global optimization strategy. The adjusted weight parameters are input into the cross-domain collaboration module, and combined with the unified resource dataset and the latest network status feedback, cross-domain joint fault analysis, multi-layer path orchestration calculation, and parameter reliability verification are performed again. The path planning and resource scheduling schemes in the original global optimization strategy are checked in a closed loop, deviation terms are corrected, and an updated global optimization strategy with integrated execution feedback data, parameter reliability evaluation, and dynamic weight adjustment is generated.

[0066] One embodiment of this application provides a dynamic decision optimization method for multi-domain policy collaboration, comprising: performing multi-domain resource dynamic perception processing on network layer topology information, device status information, and alarm information based on cross-domain common feature extraction to generate a unified resource dataset; wherein the cross-domain common feature extraction preserves the differentiated individual characteristics of the transmission network and the data network; based on the unified resource dataset, performing hierarchical collaborative decision processing through a three-level linkage decision mechanism of the transmission layer collaboration module, the data layer collaboration module, and the cross-domain collaboration module to generate a global optimization strategy; based on the global optimization strategy, performing multi-vendor instruction dynamic adaptation processing through a dual adaptation mechanism of instruction semantic ontology mapping and protocol syntax tree matching to generate a set of atomic operation instructions executable by heterogeneous network management; based on the execution feedback data returned by the network layer after executing the atomic operation instruction set, performing policy closed-loop verification and iterative optimization processing through a dynamic weight adjustment mechanism driven by parameter reliability evaluation to generate an updated global optimization strategy, thereby achieving the technical effects of improving the accuracy of cross-domain resource perception, enhancing the efficiency of hierarchical decision collaboration, and reducing the error rate of multi-vendor instruction conversion.

[0067] Furthermore, based on a global optimization strategy, a dual adaptation mechanism of instruction semantic ontology mapping and protocol syntax tree matching is used to dynamically adapt instructions from multiple vendors, generating a set of atomic operation instructions executable by heterogeneous network management systems, including:

[0068] Based on the semantic web rule reasoning engine, the global optimization strategy is processed by instruction semantic ontology mapping to generate device-independent atomic operation descriptors;

[0069] Using the following formula, based on a multi-vendor protocol syntax tree feature library, node matching is performed on device-independent atomic operation descriptors to generate intermediate expression for the protocol syntax:

[0070]

[0071] Among them, M(A) i ,T k ) represents the device-independent atomic operation descriptor A i With protocol syntax tree T k Match degree, V k Representation of syntax tree Tk The set of nodes in ω l Let f represent the weight of the l-th feature, δ represent the feature matching function, and f l (A i ) represents the l-th feature of the atomic operation descriptor, f l (N j ) represents node N j The l-th feature, where L represents the total number of features;

[0072] The intermediate expressions of the protocol syntax are subjected to semantic-syntactic consistency verification to generate an atomic operation instruction set that can be executed by the heterogeneous network management system.

[0073] Specifically, a semantic web rule inference engine is used to semantically parse the business operation requirements and network control logic contained in the global optimization strategy, mapping the abstract strategy content to a pre-built instruction semantic ontology. Through the concept hierarchy and relationship definition in the ontology, device-specific attributes are stripped away to generate atomic operation descriptors that are independent of specific vendor devices. These descriptors present basic operation units in the form of standardized semantic entities, such as configuring ports and querying topologies.

[0074] The feature library of multi-vendor protocol syntax trees is invoked to match the feature parameters of atomic operation descriptors with syntax tree nodes one by one. The degree of fit between descriptor features and syntax tree node features is quantified by a feature matching function to generate an intermediate expression of protocol syntax that represents the semantic-syntax mapping relationship. This expression integrates the commonalities and differences of syntax rules from multiple vendors.

[0075] The intermediate expressions of the protocol syntax are bidirectionally validated. On the one hand, this verifies whether the syntax structure conforms to the instruction format specifications of the target vendor's network management system. On the other hand, it confirms whether the semantic logic completely preserves the business intent of the atomic operation descriptors. By eliminating ambiguities and correcting format conflicts, an atomic operation instruction set that meets the syntax requirements of heterogeneous network management systems and is semantically accurate is generated. This atomic operation instruction set can be directly used for the control and execution of devices from multiple vendors.

[0076] Furthermore, based on the semantic web rule inference engine, the global optimization strategy is processed by instruction semantic ontology mapping to generate device-independent atomic operation descriptors, including:

[0077] Based on the unified information model, the global optimization strategy is processed by business intent parsing to generate business operation metaphrases;

[0078] Based on the semantic web rule reasoning engine, network atomic function mapping is performed on business operation meta-languages ​​to generate atomic operation semantic graphs.

[0079] Based on the YANG data modeling language, a hierarchical node encapsulation process is performed on the atomic operation semantic graph to generate device-independent atomic operation descriptors.

[0080] Specifically, based on a unified information model, the business requirements involved in the global optimization strategy are deeply analyzed, the specific technical implementation details are stripped away, and abstract business objectives such as resource orchestration and fault diagnosis are extracted and transformed into standardized business operation metaphrases to clarify the objects, actions and expected results of business operations.

[0081] Using the Semantic Web rule reasoning engine, business operation metaphrases are semantically associated and mapped with network atomic functions. Based on the reasoning logic built into the rule engine, an atomic operation semantic graph including operation flow and parameter passing relationship is constructed to achieve a structured transformation from business intent to network function.

[0082] Using the YANG data modeling language (Yet Another Next Generation), the functional nodes in the atomic operation semantic graph are hierarchically encapsulated. Following the structural specifications of YANG model, such as container nodes and leaf list nodes, the semantic graph is converted into standardized, device-independent atomic operation descriptors, giving it cross-vendor device universality and scalability.

[0083] Furthermore, semantic-syntactic consistency verification is performed on the intermediate expressions of the protocol syntax to generate a set of atomic operation instructions executable by the heterogeneous network management system, including:

[0084] Based on a multi-vendor semantic rule library, semantic integrity conflict detection is performed on intermediate expressions of protocol syntax to generate semantically compliant intermediate expressions.

[0085] Using the following formula, and through vendor-specific protocol syntax constraints, semantically compliant intermediate expressions are subjected to syntactic structure compliance transformation to generate a syntactically normalized instruction set:

[0086]

[0087] Among them, C compliance S represents the grammatical structure compliance assessment value, where n represents the total number of grammatical constraint rules, and S represents the grammatical structure compliance assessment value. m Represents the actual output of the m-th syntax structure, R m This represents the standard requirement for the m-th grammar rule, ∈ represents a small constant to prevent division by zero, and α m This represents the importance index of the m-th grammar rule;

[0088] The syntax-normalized instruction set is encapsulated into executable atomic operations to generate an executable atomic operation instruction set for heterogeneous network management systems.

[0089] Specifically, it invokes multi-vendor semantic rule bases to perform semantic integrity verification on the intermediate expressions of the protocol syntax. By comparing the operational semantics in the intermediate expressions with the standard semantic models defined in the rule bases, it detects the existence of semantic ambiguity, missing parameters, or logical conflicts. After eliminating conflicting items, it generates semantically compliant intermediate expressions to ensure that the instruction semantics are consistent with the business meaning. Figure 1 To.

[0090] Based on the grammatical constraints of vendor-specific protocols, semantically compliant intermediate expressions are substituted into the grammatical structure compliance evaluation formula. By calculating the deviation between the actual output of each grammatical structure and the standard requirements, and combining the deviation with a rule importance index, the deviation is weighted and processed. Parts that do not conform to the target vendor's grammatical specifications are automatically converted and corrected, generating a grammatical standardization instruction set that conforms to multi-vendor grammatical rules.

[0091] The standardized instruction set is structurally encapsulated and converted into atomic operation units recognizable by devices from various vendors, according to the interface protocol requirements of heterogeneous network management systems. By adding vendor-specific protocol headers, verification parameters, and execution logic, an atomic operation instruction set that can be directly issued to heterogeneous network management systems is generated to achieve unified control of devices across vendors.

[0092] like Figure 2 As shown, based on a unified resource dataset, a three-tiered collaborative decision-making mechanism involving the transport layer collaboration module, data layer collaboration module, and cross-domain collaboration module is used to generate a global optimization strategy, including:

[0093] S201: Based on the unified resource dataset, cross-domain end-to-end configuration policy generation is performed through the transport layer collaboration module to generate the transport network end-to-end configuration policy.

[0094] S202: Based on the unified resource dataset, risk level linkage judgment and root cause tracing are performed through the data layer collaboration module to generate data network fault analysis results;

[0095] S203: Based on the end-to-end configuration strategy of the transmission network and the fault analysis results of the data network, the cross-domain resource conflict resolution is carried out through the cross-domain collaboration module to generate a global optimization strategy.

[0096] Specifically, the transport layer coordination module extracts transport network topology information, device status data, link bandwidth parameters, and cross-domain configuration requirements from the unified resource dataset. Combined with the atomic capability management and resource management modules in the transport network orchestration function, it analyzes the resource availability of end-to-end paths within the transport network. At the same time, it calls the risk assessment module in the fault analysis function to pre-judge potential fault points. Based on the open requirements of the transport layer cross-domain configuration capability, it generates an end-to-end configuration strategy for the transport network, including path planning, bandwidth reservation, and fault switching mechanisms.

[0097] The data layer collaboration module synchronously acquires the data network topology, device configuration information, and alarm data from the unified resource dataset. It uses the risk level assessment model in the data network fault analysis function to classify network anomalies. Combined with the inter-layer correlation analysis capability of the root cause tracing module, it locates the specific nodes or configuration conflict points of the data network fault and generates data network fault analysis results including fault location, scope of impact, and preliminary handling suggestions.

[0098] The cross-domain collaboration module receives the end-to-end configuration strategy of the transmission network and the fault analysis results of the data network. It identifies potential resource conflicts through cross-domain resource mapping relationships, recalculates the cross-domain service path using the multi-layer path orchestration and routing module, and verifies the compatibility of configuration parameters in conjunction with the parameter reliability evaluation module. Then, it resolves cross-domain resource conflicts and generates a global optimization strategy that integrates the transmission configuration strategy, the data network fault handling plan, and the cross-domain path optimization.

[0099] Furthermore, based on the end-to-end configuration strategy of the transmission network and the fault analysis results of the data network, cross-domain resource conflict resolution is performed through the cross-domain collaboration module to generate a global optimization strategy, including:

[0100] Based on the service flow SLA (Service Level Agreement) constraints, fault-resource correlation analysis is performed on the end-to-end configuration strategy of the transmission network and the fault analysis results of the data network to generate fault-resource correlation analysis results;

[0101] The results of the fault-resource correlation analysis are quantitatively evaluated using link reliability indicators to generate a dynamic resource arbitration scheme.

[0102] Based on the dynamic resource arbitration scheme, a multi-objective Pareto front solution is performed to generate a global optimization strategy.

[0103] Specifically, based on the service flow SLA constraints, the path resources in the end-to-end configuration strategy of the transmission network are correlated and mapped with the fault nodes and impact range in the data network fault analysis results. The degree of impact of the fault on the transmission path resources is analyzed, the resource nodes and links affected by the fault are identified, and fault-resource correlation analysis results are generated to clarify the interaction relationship between faults and resources.

[0104] For affected links identified in the fault-resource correlation analysis, real-time status parameters of the links are extracted from the unified resource dataset. Combined with a pre-defined reliability evaluation index system, the reliability of the links is quantitatively assessed, and the reliability index value for each link is calculated. Based on the quantitative results, and considering both business priority and resource availability, a dynamic resource arbitration scheme is generated to determine the priority and strategy for resource reallocation, thereby achieving rational resource scheduling.

[0105] Based on a dynamic resource arbitration scheme, an optimization model is constructed that includes multiple objectives such as service continuity, resource utilization, and fault recovery rate. Using a multi-objective Pareto front algorithm, a Pareto optimal solution set is obtained while satisfying service SLA constraints. The optimal solution is selected from this set to generate a global optimization strategy. This strategy integrates transmission network configuration, data network fault handling, and resource optimization scheduling to achieve comprehensive optimization of multiple objectives.

[0106] Furthermore, based on the execution feedback data returned by the network layer after executing the atomic operation instruction set, a dynamic weight adjustment mechanism driven by parameter reliability evaluation is used to perform closed-loop verification and iterative optimization of the strategy, generating an updated global optimization strategy, including:

[0107] Perform parameter reliability quantification and evaluation on the execution feedback data to generate link reliability indicators;

[0108] Based on the link reliability index, dynamic weight adjustment processing of the path calculation cost function is performed to generate an updated path calculation cost function.

[0109] Based on the updated path cost function, cross-domain resource rescheduling decision processing is performed to generate an updated global optimization strategy.

[0110] Specifically, link status parameters are extracted from execution feedback data, and the reliability of the links is quantitatively evaluated based on a preset reliability evaluation index system. This generates a link reliability index that includes the reliability scores of each link, thus clarifying the trustworthiness of different links under the current network conditions.

[0111] Based on the link reliability index, the weights of each influencing factor in the path calculation cost function are dynamically adjusted. Links with low reliability are given increased weights in the cost function to improve their probability of being avoided by the path selection algorithm; links with high reliability are given decreased weights to make them more likely to be selected as the optimal path, thus generating an updated path calculation cost function.

[0112] By utilizing the updated path cost function and combining it with the real-time resource status in the unified resource dataset, cross-domain service paths are recalculated, and resource allocation schemes are optimized. By comparing the resource utilization and service SLA satisfaction of the old and new paths, cross-domain resource rescheduling decisions are made, generating an updated global optimization strategy that includes path optimization and resource reallocation to achieve continuous improvement in network performance.

[0113] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0114] In one embodiment, such as Figure 3 As shown, this application also provides a dynamic decision optimization system 300 for multi-domain strategy collaboration, the system 300 including:

[0115] The resource awareness module 301 is used to perform multi-domain resource dynamic awareness processing based on cross-domain common feature extraction on network layer topology information, device status information and alarm information to generate a unified resource dataset; wherein the cross-domain common feature extraction retains the differentiated individual characteristics of the transmission network and the data network;

[0116] The collaborative decision-making module 302 is used to perform hierarchical collaborative decision-making processing based on a unified resource dataset through a three-level linkage decision-making mechanism of the transport layer collaborative module, the data layer collaborative module, and the cross-domain collaborative module, and to generate a global optimization strategy.

[0117] The instruction adaptation module 303 is used to perform dynamic adaptation of instructions from multiple vendors based on a global optimization strategy and through a dual adaptation mechanism of instruction semantic ontology mapping and protocol syntax tree matching, to generate an atomic operation instruction set that can be executed by heterogeneous network management.

[0118] The closed-loop optimization module 304 is used to perform closed-loop verification and iterative optimization of the strategy based on the execution feedback data returned by the network layer after the execution of the atomic operation instruction set, through a dynamic weight adjustment mechanism driven by parameter reliability evaluation, and to generate an updated global optimization strategy.

[0119] Specifically, the resource awareness module 301 collects network layer topology information, device status information, and alarm information, including relevant data from the transmission network and data network. Then, it extracts the common features of the above information from a cross-domain perspective, while retaining the unique characteristics of the transmission network and data network. Based on this, it performs dynamic awareness processing on multi-domain resources, ultimately integrating and generating a unified resource dataset that includes both cross-domain common features and the unique characteristics of the transmission network and data network.

[0120] The collaborative decision-making module 302 is based on a unified resource dataset and uses a three-level linkage decision-making mechanism for processing. This includes: the transport layer collaborative module analyzes resource data within the transport network and generates an end-to-end configuration strategy for the transport network by combining the transport network orchestration and fault analysis functions; the data layer collaborative module synchronously analyzes data network resource data and generates data network fault analysis results using data network orchestration and fault analysis capabilities; and the cross-domain collaborative module receives the above strategies and results, performs cross-domain resource conflict resolution processing, and finally generates a global optimization strategy.

[0121] The instruction adaptation module 303 processes the instruction using a dual adaptation mechanism based on the global optimization strategy. This includes: mapping the strategy to the instruction semantic ontology through the semantic web rule reasoning engine to generate device-independent atomic operation descriptors; performing node matching processing on the descriptors based on a multi-vendor protocol syntax tree feature library to generate intermediate protocol syntax expressions; and performing semantic-syntax consistency verification on the intermediate expressions to generate a set of atomic operation instructions executable by the heterogeneous network management system.

[0122] The closed-loop optimization module 304 obtains the execution feedback data returned by the network layer after executing the atomic operation instruction set, and processes it through a dynamic weight adjustment mechanism. This includes: performing parameter reliability quantification and evaluation on the feedback data to generate link reliability indicators; dynamically adjusting the path cost function based on the above indicators to generate an updated path cost function; and performing cross-domain resource rescheduling decision processing based on the updated function to generate an updated global optimization strategy, so as to realize the closed-loop verification and iterative optimization of the strategy.

[0123] Instruction adapter module 303 is also used for:

[0124] Based on the semantic web rule reasoning engine, the global optimization strategy is processed by instruction semantic ontology mapping to generate device-independent atomic operation descriptors;

[0125] Using the following formula, based on a multi-vendor protocol syntax tree feature library, node matching is performed on device-independent atomic operation descriptors to generate intermediate expression for the protocol syntax:

[0126]

[0127] Among them, M(A) i ,T k ) represents the device-independent atomic operation descriptor A i With protocol syntax tree T k Match degree, V k Representation of syntax tree T k The set of nodes in ω l Let f represent the weight of the l-th feature, δ represent the feature matching function, and f l (A i) represents the l-th feature of the atomic operation descriptor, f l (N j ) represents node N j The l-th feature, where L represents the total number of features;

[0128] The intermediate expressions of the protocol syntax are subjected to semantic-syntactic consistency verification to generate an atomic operation instruction set that can be executed by the heterogeneous network management system.

[0129] Instruction adapter module 303 is also used for:

[0130] Based on the unified information model, the global optimization strategy is processed by business intent parsing to generate business operation metaphrases;

[0131] Based on the semantic web rule reasoning engine, network atomic function mapping is performed on business operation meta-languages ​​to generate atomic operation semantic graphs.

[0132] Based on the YANG data modeling language, a hierarchical node encapsulation process is performed on the atomic operation semantic graph to generate device-independent atomic operation descriptors.

[0133] Instruction adapter module 303 is also used for:

[0134] Based on a multi-vendor semantic rule library, semantic integrity conflict detection is performed on intermediate expressions of protocol syntax to generate semantically compliant intermediate expressions.

[0135] Using the following formula, and through vendor-specific protocol syntax constraints, semantically compliant intermediate expressions are subjected to syntactic structure compliance transformation to generate a syntactically normalized instruction set:

[0136]

[0137] Among them, C compliance S represents the grammatical structure compliance assessment value, where n represents the total number of grammatical constraint rules, and S represents the grammatical structure compliance assessment value. m Represents the actual output of the m-th syntax structure, R m This represents the standard requirement for the m-th grammar rule, ∈ represents a small constant to prevent division by zero, and α m This represents the importance index of the m-th grammar rule;

[0138] The syntax-normalized instruction set is encapsulated into executable atomic operations to generate an executable atomic operation instruction set for heterogeneous network management systems.

[0139] The collaborative decision-making module 302 is also used for:

[0140] Based on the unified resource dataset, the cross-domain end-to-end configuration policy generation process is performed through the transport layer collaboration module to generate the transport network end-to-end configuration policy.

[0141] Based on a unified resource dataset, risk level linkage determination and root cause tracing are performed through a data layer collaboration module to generate data network fault analysis results.

[0142] Based on the end-to-end configuration strategy of the transmission network and the fault analysis results of the data network, the cross-domain resource conflict resolution is carried out through the cross-domain collaboration module to generate a global optimization strategy.

[0143] The collaborative decision-making module 302 is also used for:

[0144] Based on the service flow SLA constraints, fault-resource correlation analysis is performed on the end-to-end configuration strategy of the transmission network and the fault analysis results of the data network to generate fault-resource correlation analysis results.

[0145] The results of the fault-resource correlation analysis are quantitatively evaluated using link reliability indicators to generate a dynamic resource arbitration scheme.

[0146] Based on the dynamic resource arbitration scheme, a multi-objective Pareto front solution is performed to generate a global optimization strategy.

[0147] The closed-loop optimization module 304 is also used for:

[0148] Perform parameter reliability quantification and evaluation on the execution feedback data to generate link reliability indicators;

[0149] Based on the link reliability index, dynamic weight adjustment processing of the path calculation cost function is performed to generate an updated path calculation cost function.

[0150] Based on the updated path cost function, cross-domain resource rescheduling decision processing is performed to generate an updated global optimization strategy.

[0151] In one embodiment, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0152] In one embodiment, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the above-described method embodiments.

[0153] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0154] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. A dynamic decision optimization method for multi-domain strategy collaboration, characterized in that, The method includes: The network layer's topology information, device status information, and alarm information are subjected to multi-domain resource dynamic sensing processing based on cross-domain common feature extraction to generate a unified resource dataset; wherein the cross-domain common feature extraction preserves the differentiated individual characteristics of the transmission network and the data network; Based on the unified resource dataset, a three-level linkage decision-making mechanism involving the transport layer collaboration module, the data layer collaboration module, and the cross-domain collaboration module is used to perform hierarchical collaborative decision-making processing and generate a global optimization strategy. Based on the global optimization strategy, a dual adaptation mechanism of instruction semantic ontology mapping and protocol syntax tree matching is used to dynamically adapt instructions from multiple vendors, generating an atomic operation instruction set that can be executed by heterogeneous network management systems. Based on the execution feedback data returned by the network layer after executing the atomic operation instruction set, a dynamic weight adjustment mechanism driven by parameter reliability evaluation is used to perform closed-loop verification and iterative optimization of the strategy, generating an updated global optimization strategy.

2. The dynamic decision optimization method for multi-domain strategy collaboration according to claim 1, characterized in that, Based on the global optimization strategy, a dual adaptation mechanism of instruction semantic ontology mapping and protocol syntax tree matching is used to dynamically adapt instructions from multiple vendors, generating a set of atomic operation instructions executable by heterogeneous network management systems, including: Based on the semantic web rule reasoning engine, the global optimization strategy is processed by instruction semantic ontology mapping to generate device-independent atomic operation descriptors. Using the following formula, based on a multi-vendor protocol syntax tree feature library, node matching processing is performed on the device-independent atomic operation descriptors to generate intermediate expression of the protocol syntax: Among them, M(A) i ,T k ) represents the device-independent atomic operation descriptor A i With protocol syntax tree T k Match degree, V k Representation of syntax tree T k The set of nodes in ω l Let f represent the weight of the l-th feature, δ represent the feature matching function, and f l (A i ) represents the l-th feature of the atomic operation descriptor, f l (N j ) represents node N j The l-th feature, where L represents the total number of features; The intermediate expression of the protocol syntax is subjected to semantic-syntactic consistency verification to generate an atomic operation instruction set that can be executed by the heterogeneous network management system.

3. The dynamic decision optimization method for multi-domain strategy collaboration according to claim 2, characterized in that, The semantic web rule-based reasoning engine performs instruction semantic ontology mapping processing on the global optimization strategy to generate device-independent atomic operation descriptors, including: Based on the unified information model, the global optimization strategy is processed by business intent parsing to generate business operation metaphrases; Based on the semantic web rule reasoning engine, network atomic function mapping is performed on the business operation meta-languages ​​to generate an atomic operation semantic graph. Based on the YANG data modeling language, the atomic operation semantic graph is subjected to hierarchical node encapsulation processing to generate the device-independent atomic operation descriptor.

4. The dynamic decision optimization method for multi-domain strategy collaboration according to claim 2, characterized in that, The step of performing semantic-syntactic consistency verification on the intermediate expression of the protocol syntax to generate an executable atomic operation instruction set for the heterogeneous network management system includes: Based on a multi-vendor semantic rule library, semantic integrity conflict detection processing is performed on the intermediate expression of the protocol syntax to generate a semantically compliant intermediate expression. Using the following formula, and through vendor-specific protocol syntax constraints, the semantically compliant intermediate expression is subjected to syntactic structure compliance transformation to generate a syntactic normalization instruction set: Among them, C compliance S represents the grammatical structure compliance assessment value, where n represents the total number of grammatical constraint rules, and S represents the grammatical structure compliance assessment value. m Represents the actual output of the m-th syntax structure, R m This represents the standard requirement for the m-th grammar rule, ∈ represents a small constant to prevent division by zero, and α m This represents the importance index of the m-th grammar rule; The syntax-normalized instruction set is encapsulated into executable atomic operations to generate the heterogeneous network management system's executable atomic operation instruction set.

5. The dynamic decision optimization method for multi-domain strategy collaboration according to claim 1, characterized in that, Based on the unified resource dataset, a three-level collaborative decision-making mechanism involving the transport layer collaboration module, data layer collaboration module, and cross-domain collaboration module is used to generate a global optimization strategy, including: Based on the unified resource dataset, cross-domain end-to-end configuration policy generation processing is performed through the transport layer collaboration module to generate the transport network end-to-end configuration policy. Based on the unified resource dataset, risk level linkage determination and root cause tracing are performed through the data layer collaboration module to generate data network fault analysis results. Based on the end-to-end configuration strategy of the transmission network and the fault analysis results of the data network, the cross-domain resource conflict resolution is carried out through the cross-domain collaboration module to generate the global optimization strategy.

6. The dynamic decision optimization method for multi-domain strategy collaboration according to claim 5, characterized in that, Based on the end-to-end configuration strategy of the transmission network and the fault analysis results of the data network, the cross-domain resource conflict resolution process is performed through the cross-domain collaboration module to generate the global optimization strategy, including: Based on the service flow SLA constraints, fault-resource correlation analysis is performed on the end-to-end configuration strategy of the transmission network and the fault analysis results of the data network to generate fault-resource correlation analysis results. The fault-resource correlation analysis results are subjected to link reliability index quantitative evaluation processing to generate a dynamic resource arbitration scheme; Based on the dynamic resource arbitration scheme, a multi-objective Pareto front solution is performed to generate the global optimization strategy.

7. The dynamic decision optimization method for multi-domain strategy collaboration according to claim 1, characterized in that, The process involves using the execution feedback data returned by the network layer after executing the atomic operation instruction set, and employing a dynamic weight adjustment mechanism driven by parameter reliability evaluation to perform closed-loop policy verification and iterative optimization, thereby generating an updated global optimization policy, including: The execution feedback data is subjected to parameter reliability quantification evaluation processing to generate link reliability indicators; Based on the link reliability index, dynamic weight adjustment processing of the path calculation cost function is performed to generate an updated path calculation cost function. Based on the updated path cost function, cross-domain resource rescheduling decision processing is performed to generate the updated global optimization strategy.

8. A dynamic decision optimization system with multi-domain strategy collaboration, characterized in that, The system includes: The resource awareness module is used to perform multi-domain resource dynamic awareness processing based on cross-domain common feature extraction on network layer topology information, device status information and alarm information to generate a unified resource dataset; wherein the cross-domain common feature extraction preserves the differentiated individual characteristics of the transmission network and the data network. The collaborative decision-making module is used to perform hierarchical collaborative decision-making processing based on the unified resource dataset through a three-level linkage decision-making mechanism of the transport layer collaborative module, the data layer collaborative module, and the cross-domain collaborative module, and generate a global optimization strategy. The instruction adaptation module is used to perform dynamic adaptation processing of multi-vendor instructions based on the global optimization strategy through a dual adaptation mechanism of instruction semantic ontology mapping and protocol syntax tree matching, and generate an atomic operation instruction set that can be executed by heterogeneous network management. The closed-loop optimization module is used to perform closed-loop verification and iterative optimization of the strategy based on the execution feedback data returned by the network layer after executing the atomic operation instruction set, through a dynamic weight adjustment mechanism driven by parameter reliability evaluation, and to generate an updated global optimization strategy.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the dynamic decision optimization method for multi-domain strategy collaboration as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the dynamic decision optimization method for multi-domain policy collaboration as described in any one of claims 1 to 7.

Images