Test link management method based on automatic verification and version control
By constructing a dynamic graph structure and a hybrid verification mechanism, combined with quantum state encoding and digital twin models, the problem of incomplete dependency verification in test link management of power dispatching systems was solved, achieving full coverage and real-time optimization of complex systems and avoiding power grid failures.
Patent Information
- Application Number
- CN202511305162.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-12
- Publication Date
- 2026-02-03
AI Technical Summary
The lack of unified semantic specifications in existing software testing link management leads to incomplete verification of dependencies in complex systems. In particular, in power dispatching systems, configuration errors of relay protection devices may be overlooked, resulting in the false triggering of backup protection actions and the escalation of cascading failures.
A hybrid verification mechanism based on quantum annealing machines, graph attention networks, and quantum gates is adopted to construct a dynamic graph structure. Test link configuration data is processed through quantum state encoding to generate verification result reports. Real-time simulation and optimization are performed using a digital twin model, the timing constraint threshold is dynamically adjusted, and rule patches are generated for online repair.
It achieves full-coverage verification of complex dependencies, avoids omissions in manual verification, ensures the stability of the power grid system, predicts cascading failure paths through quantum filtering and digital twin models, dynamically adjusts constraint thresholds, forms a self-learning closed loop, and realizes real-time optimization and repair of rules.
Smart Images
Figure CN121455809A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of data processing, and in particular to a test link management method based on automated verification and version control. BACKGROUND
[0002] The existing software field test link is a series of structured test stages in the software development life cycle, starting from the unit test stage, in which the developer verifies whether the function logic of a single code unit is correct; then enters the integration test stage, in which the test team combines multiple unit modules to check whether the interface interaction and data flow are consistent; then advances to the system test stage, in which the quality assurance engineer simulates the real environment to evaluate whether the entire software system meets the requirement specifications and design goals; and finally completes the acceptance test stage, in which the user or customer operates in the actual scene to confirm that the software meets the business requirements and expected effects.
[0003] The existing test link management based on automated verification and version control has the following technical pain points, specifically, the lack of unified semantic specifications in the management process makes it difficult to fully cover complex dependency relationships during manual verification. In the power dispatching system, when the test link configuration does not clearly define the strict timing logic of "relay protection action" and "circuit breaker tripping", manual checking may overlook the configuration error that the protection signal is ahead of the tripping instruction. Such errors are not discovered in time, which will cause the relay protection device to trigger the backup protection action, leading to the expansion of the chain failure, and finally causing the regional power grid to split. SUMMARY
[0004] In view of the deficiencies of the prior art, the present application provides a test link management method based on automated verification and version control, which solves the technical problem of incomplete verification of complex system dependency relationships due to the lack of unified semantic specifications in the test link.
[0005] To solve the above technical problems, the specific content of the present application is as follows: The test link management method based on automated verification and version control provided by the present application comprises: Step 1, obtaining test link configuration data and historical version data, based on the test link configuration data and the historical version data, converting into a dynamic graph structure including test link nodes, the dynamic graph structure defining the constraint relationship between nodes, the constraint relationship including timing constraints; Step 2, the preset quantum annealing machine is applied to verify whether the action sequence in the dynamic graph structure meets the timing constraint, and node information violating the timing constraint is generated; the graph attention network is applied to verify whether the node mutual exclusion relationship in the dynamic graph structure meets the mutual exclusion constraint in the constraint relationship, and mutual exclusion conflict information is generated; the quantum gate circuit is applied to verify whether the interface parameter transmission in the dynamic graph structure meets the preset rule, and parameter difference information is generated; the node information violating the timing constraint, the mutual exclusion conflict information and the parameter difference information are summarized to generate a verification result; Step 3, according to the node information violating the timing constraint, the mutual exclusion conflict information and the parameter difference information, a verification result report is generated, and based on the verification result, a semantic rule identifier is extracted to generate a version snapshot including the semantic rule identifier; Step 4, based on the version snapshot generated in step 3, a digital twin model of the test link is established, real-time running data of the test link is collected, and the real-time running data is input to the digital twin model; Step 5, the quantum filtering is used to process the noise in the real-time running data to generate denoised running data, the denoised running data is used to predict the chain failure path, according to the prediction result, the timing constraint threshold in the dynamic graph structure is dynamically adjusted, and the constraint relationship between nodes in the dynamic graph structure is updated according to the adjustment of the timing constraint threshold to generate an optimized dynamic graph structure.
[0006] Further, the test link management method based on automatic verification and version control of the application comprises the following steps: Synchronously reading the test link configuration file and the historical version record stored in a distributed storage; Based on the test link configuration file, the quantum state encoding processing is performed on the relay protection parameters in the test link configuration file to form a superposition state data unit; Wherein, the synchronous reading of the test link configuration file and the historical version record stored in the distributed storage flows to the quantum state encoding processing on the relay protection parameters in the test link configuration file.
[0007] Further, the test link management method based on automatic verification and version control of the application comprises the following steps: Taking the relay protection device and the circuit breaker as the graph nodes; Defining the constraint edges between the nodes, the constraint edges including the trigger edges representing state transition, the timing constraint edges representing the maximum allowed time difference, and the mutual exclusion edges representing operation mutual exclusion; Converting the power grid power frequency characteristics into the energy parameters of the timing constraint edges; Wherein, the synchronous reading of the test link configuration file and the historical version record stored in the distributed storage flows to the quantum state encoding processing on the relay protection parameters in the test link configuration file.
[0008] Further, the test link management method based on automated verification and version control of the present application, step 2 includes: Using a quantum annealing machine to verify whether the action sequence in the dynamic graph structure meets the timing constraint, and generating node information that violates the timing constraint; Using a graph attention network to verify whether the node mutual exclusion relationship in the dynamic graph structure meets the mutual exclusion constraint, and generating mutual exclusion conflict information; Using a quantum gate circuit to verify whether the interface parameter transmission in the dynamic graph structure meets the preset rule, and generating parameter difference information; Constructing a state space projection of the fault scenario, the state space projection being based on the node information that violates the timing constraint, the mutual exclusion conflict information, and the parameter difference information; Summarizing the node information that violates the timing constraint, the mutual exclusion conflict information, the parameter difference information, and the state space projection data to generate a verification result.
[0009] Further, the test link management method based on automated verification and version control of the present application, step 2 further includes: Converting the timing constraint into an Ising model coupling coefficient; Minimizing an Ising model energy function to solve an optimal action sequence; Comparing the actual action sequence in the dynamic graph structure with the optimal action sequence; Identifying nodes in the actual action sequence that deviate from the optimal action sequence to generate node information that violates the timing constraint; Among them, converting the timing constraint into an Ising model coupling coefficient flows to minimizing an Ising model energy function to solve an optimal action sequence; minimizing an Ising model energy function to solve an optimal action sequence flows to comparing the actual action sequence in the dynamic graph structure with the optimal action sequence; comparing the actual action sequence in the dynamic graph structure with the optimal action sequence flows to identifying nodes in the actual action sequence that deviate from the optimal action sequence to generate node information that violates the timing constraint.
[0010] Further, the test link management method based on automated verification and version control of the present application, step 3 includes: Based on the node information that violates the timing constraint generated in step 2, marking the nodes that violate the timing constraint in the three-dimensional power grid model; Based on the mutual exclusion conflict information generated in step 2, visualizing and displaying the mutual exclusion conflict in the three-dimensional power grid model; Based on the node information that violates the timing constraint and the mutual exclusion conflict information, generating a repair instruction; Wherein, based on the node information of violating timing constraints generated in step 2, the flow direction of the node violating timing constraints is marked in the three-dimensional power grid model; based on the mutual exclusion conflict information generated in step 2, the mutual exclusion conflict is visualized and displayed in the three-dimensional power grid model; based on the mutual exclusion conflict information generated in step 2, the mutual exclusion conflict flow direction is visualized and displayed in the three-dimensional power grid model; based on the node information of violating timing constraints and the mutual exclusion conflict information, repair instructions are generated.
[0011] Further, the test link management method based on automated verification and version control of the application, step 3 comprises: Based on the verification result generated in step 2, the semantic rule identifier is extracted, the hash value of the semantic rule identifier is calculated, the hash value is used as a leaf node, a quantum Merkle tree is constructed, version change records are obtained, and the version change records are cryptographically signed; Wherein, based on the verification result generated in step 2, the semantic rule identifier is extracted; the hash value of the semantic rule identifier is calculated; the hash value is used as a leaf node to construct a quantum Merkle tree; the hash value is used as a leaf node to construct a quantum Merkle tree; the version change records are obtained; and the version change records are cryptographically signed.
[0012] Further, the test link management method based on automated verification and version control of the application, step 4 comprises: Based on the version snapshot generated in step 3, the digital twin model is initialized; In the digital twin model, quantum mechanics equations are used to describe the state evolution of the relay protection device; Wherein, based on the version snapshot generated in step 3, the digital twin model is initialized; in the digital twin model, quantum mechanics equations are used to describe the state evolution of the relay protection device.
[0013] Further, the test link management method based on automated verification and version control of the application, step 5 comprises: Quantum filtering is used to process noise in real-time operation data collected in step 4 to generate denoised operation data; Based on the denoised operation data, the cascading failure path is predicted; According to the cascading failure path prediction result, the timing constraint threshold in the dynamic graph structure is dynamically adjusted; According to the dynamically adjusted timing constraint threshold, the weight of the dynamic graph structure is updated; The noise in the real-time operation data collected in step 4 is processed using quantum filtering to generate denoised operation data, and the denoised operation data is used to predict a cascading failure path; based on the denoised operation data, the cascading failure path is predicted; according to the cascading failure path prediction result, the time constraint threshold in the dynamic graph structure is dynamically adjusted; according to the cascading failure path prediction result, the time constraint threshold in the dynamic graph structure is dynamically adjusted; according to the dynamically adjusted time constraint threshold, the weight of the dynamic graph structure is updated.
[0014] Further, the test link management method based on automated verification and version control of the present application further comprises: Output real-time state evolution data from the digital twin model; Convert the real-time state evolution data into a quantum state feature vector through a quantum harmonic oscillator layer; Based on the quantum state feature vector, apply the Pauli operator to match the abnormal evolution mode to generate a rule patch; Inject the rule patch into the hybrid verification mechanism of step 2 to update the time constraint, mutual exclusion constraint and preset rules; Wherein, the real-time state evolution data output from the digital twin model flows through the quantum harmonic oscillator layer to convert the real-time state evolution data into a quantum state feature vector; the real-time state evolution data is converted into a quantum state feature vector through the quantum harmonic oscillator layer; based on the quantum state feature vector, the Pauli operator is applied to match the abnormal evolution mode to generate a rule patch; based on the quantum state feature vector, the Pauli operator is applied to match the abnormal evolution mode to generate a rule patch; the rule patch is injected into the hybrid verification mechanism of step 2 to update the time constraint, mutual exclusion constraint and preset rules.
[0015] The present application has the following advantages: The beneficial effects of the present application are embodied in the following aspects: dynamic graph structure unified semantic modeling, mixed verification mechanism multi-dimensional coverage, closed-loop optimization real-time rule updating to solve the problem of incomplete dependency relationship verification: the dynamic graph structure converts the test link configuration data and historical version data into node relationships defining timing constraints, the constraint edge encapsulates the interaction logic of relay protection action and circuit breaker tripping, and eliminates semantic ambiguity; the mixed verification mechanism includes quantum annealing machine to verify the timing compliance of action sequence and identify the violation nodes, graph attention network to detect mutual exclusion operation conflicts, quantum gate circuit to check interface parameter consistency, and three modules to cover the blind area of timing, mutual exclusion and interface dependency relationship verification in parallel; the digital twin model loads semantic rules based on version snapshots, drives state evolution simulation with real-time running data, predicts cascading failure paths after quantum filtering denoising and dynamically adjusts the timing constraint threshold, optimizes the dynamic graph structure weight to form a self-learning closed loop; the rule patch mechanism converts the state evolution data output by the digital twin model into quantum state feature vectors, generates incremental rule patches by matching abnormal patterns with Pauli operators, and updates the timing constraint boundary and mutual exclusion relationship definition by injecting the verification mechanism, to realize online hot repair of verification rules. The whole process establishes a technical system of "unified semantic modeling, multi-dimensional verification, real-time closed-loop optimization and dynamic evolution of rules", avoiding the risk of regional power grid splitting caused by missing complex dependency relationships in manual verification. BRIEF DESCRIPTION OF DRAWINGS
[0016] In order to more clearly illustrate the technical solutions of the present application, the following will briefly introduce the drawings needed in the embodiments. Obviously, for those skilled in the art, other drawings can also be obtained without creative labor based on the drawings.
[0017] Figure 1 The flowchart of the test link management method based on automatic verification and version control provided by the embodiments of the present application. DETAILED DESCRIPTION
[0018] In order to make the technical solutions of the present application clearer, the following will combine the specific embodiments of the present application and the corresponding drawings to clearly and completely describe the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application. The following will combine the drawings to specifically describe the present application provided by the embodiments of the present application. In order to better understand the purpose of the present application, the following will further describe the present application in detail.
[0019] Please refer to Figure 1 The test link management method based on automatic verification and version control provided by the present application comprises the following steps: Step 1, obtain test link configuration data and historical version data, and convert the test link configuration data and the historical version data into a dynamic graph structure including test link nodes, the dynamic graph structure defining constraint relationships between the nodes, the constraint relationships including timing constraints; Step 2, applying a preset quantum annealing machine to verify whether the action sequence in the dynamic graph structure meets the timing constraints, generating node information violating the timing constraints; applying a graph attention network to verify whether the node mutual exclusion relationship in the dynamic graph structure meets the mutual exclusion constraint in the constraint relationship, generating mutual exclusion conflict information; applying a quantum gate circuit to verify whether the interface parameter transmission in the dynamic graph structure meets the preset rule, generating parameter difference information; and aggregating the node information violating the timing constraints, the mutual exclusion conflict information, and the parameter difference information to generate a verification result; Step 3, generating a verification result report according to the node information violating the timing constraints, the mutual exclusion conflict information, and the parameter difference information, and extracting semantic rule identifiers based on the verification result to generate a version snapshot including the semantic rule identifiers; Step 4, based on the version snapshot generated in step 3, establishing a digital twin model of the test link, collecting real-time running data of the test link, and inputting the real-time running data to the digital twin model; Step 5, using quantum filtering to process noise in the real-time running data to generate de-noised running data, using the de-noised running data to predict a cascading failure path, dynamically adjusting a timing constraint threshold in the dynamic graph structure according to the prediction result, updating the constraint relationships between the nodes in the dynamic graph structure according to the adjustment of the timing constraint threshold, and generating an optimized dynamic graph structure.
[0020] Obtain test link configuration data and historical version data, and convert the test link configuration data and the historical version data into a dynamic graph structure including test link nodes. The dynamic graph structure defines constraint relationships between the nodes, and the constraint relationships include timing constraints. The historical version data provides a configuration change trajectory, and the test link configuration data is parsed into graph nodes and associated relationships. The nodes in the dynamic graph structure represent entities such as relay protection devices and circuit breakers, the constraint relationships describe action sequence dependency rules, and the timing constraints specify the order and time window limits of action execution.
[0021] The quantum annealer is used to verify whether the action sequence in the dynamic graph structure meets the timing constraint, and to generate node information that violates the timing constraint. The quantum annealer converts the timing constraint into an optimization problem to solve the optimal path and identify the deviation nodes of the action sequence. The graph attention network is used to verify whether the mutual exclusion relationship meets the mutual exclusion constraint, and to generate mutual exclusion conflict information; the graph attention network analyzes the node association weight and detects the mutual exclusion operation conflict. The quantum gate circuit is used to verify whether the interface parameter transmission meets the preset rule, and to generate parameter difference information; the quantum gate circuit checks the consistency of the interface data stream. The timing constraint violation node information, mutual exclusion conflict information and parameter difference information are summarized to generate the verification result.
[0022] According to the timing constraint violation node information, the mutual exclusion conflict information and the parameter difference information, a three-dimensional visual verification result report is generated. The verification result report marks the abnormal node position and the conflict type. Based on the semantic rule identification of the verification result, the semantic rule identification encapsulates the constraint relationship logical features. A version snapshot including the semantic rule identification is generated, and the version snapshot records the semantic rule hash value and the version change signature.
[0023] Based on the version snapshot, the digital twin model of the test link is initialized, and the version snapshot provides a model initialization rule library. Real-time running data of the test link is collected, and the real-time running data is input into the digital twin model. The digital twin model maps the physical system state, and the real-time running data drives the dynamic simulation of the model.
[0024] The quantum filtering is used to process the noise in the real-time running data to generate the de-noised running data. The de-noised running data is used to predict the cascading failure path, and the prediction result identifies the potential failure propagation chain. According to the cascading failure path prediction result, the timing constraint threshold in the dynamic graph structure is dynamically adjusted. According to the timing constraint threshold adjustment result, the constraint relationship between the nodes in the dynamic graph structure is updated, and an optimized dynamic graph structure is generated. The optimized dynamic graph structure is fed back to step 1 to form an iterative enhancement closed loop.
[0025] Specifically, the test link management method based on automatic verification and version control of the application comprises the following steps: Synchronously reading the test link configuration file and the historical version record stored in the distributed storage; Based on the test link configuration file, the quantum state encoding processing is performed on the relay protection parameters in the test link configuration file to form a superposition state data unit; Wherein, the synchronous reading of the test link configuration file and the historical version record stored in the distributed storage flows to the quantum state encoding processing of the relay protection parameters in the test link configuration file.
[0026] Synchronous reading of the test link configuration file and historical version record of the distributed storage obtains original configuration data and historical change trajectory. Based on the test link configuration file, quantum state encoding processing is performed on the relay protection parameters, continuous parameters such as overcurrent threshold and action time delay are mapped into superposition states of quantum bits, and superposition state data units are formed. The quantum state encoding processing realizes the conversion of classical parameter vector into quantum information, and the superposition state data unit retains the quantum correlation between parameters, providing a quantumized input basis for dynamic graph construction.
[0027] Specifically, the test link management method based on automatic verification and version control of the application comprises the following steps: The relay protection device and the circuit breaker are used as graph nodes. The constraint edges between nodes are defined, including trigger edges representing state transition, timing constraint edges representing maximum allowed time difference, and mutual exclusion edges representing operation mutual exclusion. The power grid power frequency characteristics are converted into energy parameters of the timing constraint edges. The constraint edges between nodes are defined by using the relay protection device and the circuit breaker as graph nodes. The power grid power frequency characteristics are converted into energy parameters of the timing constraint edges.
[0028] The relay protection device and the circuit breaker are used as graph nodes of the dynamic graph structure to establish a power grid entity topology framework. The constraint edges between nodes are defined, including trigger edges describing state transition logic (such as “protection start→trip instruction”), timing constraint edges setting maximum allowed time difference, and mutual exclusion edges declaring operation mutual exclusion relationship (such as “reclosing and locking mutual exclusion”). The power grid 50Hz power frequency characteristics are converted into energy parameters of the timing constraint edges. The energy parameters quantify the physical constraints of the time window, so that the timing rules are bound with the power grid frequency characteristics.
[0029] Specifically, the test link management method based on automatic verification and version control of the application comprises the following steps: The quantum annealing machine is used to verify whether the action sequence in the dynamic graph structure conforms to the timing constraint, and node information violating the timing constraint is generated. The graph attention network is used to verify whether the node mutual exclusion relationship in the dynamic graph structure conforms to the mutual exclusion constraint, and mutual exclusion conflict information is generated. The quantum gate circuit is used to verify whether the interface parameter transmission in the dynamic graph structure conforms to the preset rule, and parameter difference information is generated. The state space projection of the fault scene is constructed, which is based on the node information violating the timing constraint, the mutual exclusion conflict information and the parameter difference information. The node information violating the timing constraint, the mutual exclusion conflict information, the parameter difference information and the state space projection data are summarized to generate the verification result.
[0030] Verifying action sequence timing compliance using quantum annealing machine: converting timing constraints into a combinatorial optimization problem, generating node position information that violates timing constraints. Verifying node mutual exclusion using graph attention network: analyzing node connection weights to detect conflicting operations, outputting mutual exclusion conflict information. Verifying interface parameters using quantum gate circuit: comparing interface definitions with actual parameters through quantum controlled NOT gate, generating parameter difference values. Projecting fault scenario state space: based on timing violation nodes, mutual exclusion conflicts and parameter differences, simulating fault evolution paths in Hilbert space. Generating verification results by summarizing all verification information, covering timing, mutual exclusion and interface consistency.
[0031] Specifically, the test link management method based on automated verification and version control of the present application, step 2 further comprises: Converting timing constraints into Ising model coupling coefficients; Minimizing Ising model energy function to solve optimal action sequence; Comparing actual action sequence in dynamic graph structure with optimal action sequence; Identifying nodes in actual action sequence that deviate from optimal action sequence, generating node information that violates timing constraints; Among them, converting timing constraints into Ising model coupling coefficients flows to minimizing Ising model energy function to solve optimal action sequence; minimizing Ising model energy function to solve optimal action sequence flows to comparing actual action sequence in dynamic graph structure with optimal action sequence; comparing actual action sequence in dynamic graph structure with optimal action sequence flows to identifying nodes in actual action sequence that deviate from optimal action sequence, generating node information that violates timing constraints.
[0032] Converting timing constraints into Ising model coupling coefficients: time difference threshold is mapped to the interaction strength between spin particles. Minimizing Ising model energy function: solving the optimal action sequence path. Comparing actual action sequence in dynamic graph structure with optimal sequence: aligning node execution order and timestamp. Identifying nodes deviating from optimal sequence: locating leading / lagging trigger nodes, generating node information that violates timing constraints. This process realizes mathematical modeling and automated diagnosis of timing rules.
[0033] Specifically, the test link management method based on automated verification and version control of the present application, step 3 comprises: Based on the node information that violates timing constraints generated in step 2, marking the nodes that violate timing constraints in the three-dimensional power grid model; Based on the mutual exclusion conflict information generated in step 2, visualizing and displaying mutual exclusion conflicts in the three-dimensional power grid model; Based on the node information that violates timing constraints and the mutual exclusion conflict information, generating repair instructions; Wherein, based on the node information of violating timing constraints generated in step 2, the flow direction of the node violating timing constraints is marked in the three-dimensional power grid model; based on the mutual exclusion conflict information generated in step 2, the mutual exclusion conflict is visualized and displayed in the three-dimensional power grid model; based on the node information of violating timing constraints and the mutual exclusion conflict information, repair instructions are generated.
[0034] Based on the node information of violating timing constraints, the action timeout / out-of-order node position is calibrated in the three-dimensional power grid model. Based on the mutual exclusion conflict information, the conflict operation area (such as the overlap area of reclosing and backup protection) is visualized in the three-dimensional model. According to the node violation and conflict information, repair instructions are generated, for example, "add a time delay calibration module for the circuit breaker node". The visual report realizes the integrated output of fault positioning and repair scheme.
[0035] Specifically, the test link management method based on automated verification and version control of the present application, step 3 includes: Based on the verification result generated in step 2, the semantic rule identifier is extracted, the hash value of the semantic rule identifier is calculated, the hash value is used as a leaf node, a quantum Merkle tree is constructed, version change records are obtained, and the version change records are cryptographically signed; Wherein, based on the verification result generated in step 2, the semantic rule identifier is extracted, the hash value of the semantic rule identifier is calculated, the hash value is used as a leaf node, a quantum Merkle tree is constructed, version change records are obtained, and the version change records are cryptographically signed;
[0036] The semantic rule identifier is extracted from the verification result: the logical features of the constraint relationship are coded as unique identifiers. The hash value of the semantic rule identifier is calculated: an irreversible rule digest is generated. A quantum Merkle tree is constructed with the hash value as a leaf node: the branch nodes use quantum superposition state to store the topological relationship. The historical version change records are obtained, and the change records are signed using lattice cryptography. The combination of quantum Merkle tree and cryptographic signature ensures the tamper-proofing and traceability of version snapshots.
[0037] Specifically, the test link management method based on automated verification and version control of the present application, step 4 includes: Based on the version snapshot generated in step 3, the digital twin model is initialized; In the digital twin model, quantum mechanics equations are used to describe the state evolution of the relay protection device; Wherein, based on the version snapshot generated in step 3, the digital twin model is initialized, and in the digital twin model, quantum mechanics equations are used to describe the state evolution of the relay protection device.
[0038] Initializing the digital twin model based on version snapshots: the version snapshot loads semantic rules into the twin body rule base. In the digital twin model, the Schrodinger equation is used to describe the state evolution of the relay protection device: the Hamiltonian of the equation contains the grid frequency parameter, and the quantum wave function simulates the state transition of the device. This step establishes a physically accurate virtual mapping system.
[0039] Specifically, the test link management method based on automated verification and version control of the application comprises the following steps: Using quantum filtering to process the noise in the real-time operation data collected in step 4 to generate denoised operation data; Based on the denoised operation data, the cascading failure path is predicted; According to the prediction result of the cascading failure path, the time constraint threshold in the dynamic graph structure is dynamically adjusted; According to the dynamically adjusted time constraint threshold, the weight of the dynamic graph structure is updated; Among them, using quantum filtering to process the noise in the real-time operation data collected in step 4 to generate denoised operation data flow based on denoised operation data, predict cascading failure path; based on the denoised operation data, the cascading failure path is predicted; according to the prediction result of the cascading failure path, the time constraint threshold in the dynamic graph structure is dynamically adjusted; according to the prediction result of the cascading failure path, the time constraint threshold in the dynamic graph structure is dynamically adjusted; according to the dynamically adjusted time constraint threshold, the weight of the dynamic graph structure is updated.
[0040] Using quantum filtering to process the noise in the real-time operation data: filtering out sensor signal interference to generate denoised operation data. Based on the denoised data, the cascading failure path is predicted: identifying the cascading failure chain caused by protection misoperation. According to the prediction result of the failure path, the time constraint threshold is dynamically adjusted: for example, shortening the protection action time window. According to the adjusted threshold, the weight of the dynamic graph structure is updated: strengthening the constraint relationship strength of the key node. The weight update realizes the adaptive optimization of the graph structure.
[0041] Specifically, the test link management method based on automated verification and version control of the application further comprises: Outputting real-time state evolution data from the digital twin model; Converting the real-time state evolution data into a quantum state feature vector through a quantum harmonic oscillator layer; Based on the quantum state feature vector, applying the Pauli operator to match the abnormal evolution mode to generate a rule patch; Injecting the rule patch into the hybrid verification mechanism of step 2 to update the time constraint, mutual exclusion constraint and preset rules; Wherein, the real-time state evolution data stream output from the digital twin model is converted into a quantum state feature vector through the quantum harmonic oscillator layer; the real-time state evolution data is converted into a quantum state feature vector through the quantum harmonic oscillator layer, and based on the quantum state feature vector, a rule patch is generated by applying the Pauli operator to match the abnormal evolution mode; based on the quantum state feature vector, a rule patch is generated by applying the Pauli operator to match the abnormal evolution mode, and the rule patch is injected into the mixed verification mechanism of step 2 to update the timing constraints, mutual exclusion constraints and preset rules.
[0042] Real-time state evolution data output from the digital twin model: capture the quantum state evolution trajectory of the relay protection device. Convert the evolution data through the quantum harmonic oscillator layer: map the time domain state sequence to the frequency domain quantum state feature vector. Based on the quantum state feature vector, apply the Pauli operator to match the abnormal mode: for example, the σ_z operator detects that the state offset exceeds the standard. Generate a rule patch: quantify the abnormal correction parameters (such as timing threshold increment). Inject the rule patch into the mixed verification mechanism: update the timing constraint boundary, mutual exclusion relationship definition and interface rule. This process forms a closed-loop optimization mechanism of "digital twin monitoring to online rule hot update".
[0043] The technical scheme of the present application solves the problem of incomplete test link dependency relationship verification by constructing a unified semantic framework and a closed-loop verification optimization mechanism. First, the test link configuration data and historical version data are converted into a dynamic graph structure, which defines the constraint relationship between nodes including timing constraints. The topology entity is established by taking the relay protection device and the circuit breaker as the graph nodes. The constraint edges encapsulate the state transition logic, time window constraints and operation mutual exclusion rules. The power grid power frequency characteristics are converted into the energy parameters of the timing constraint edges, realizing the semantic unified mapping of physical rules to mathematical models. A hybrid verification mechanism is applied to cover multi-dimensional dependency relationships: quantum annealing machine verifies whether the action sequence conforms to the timing constraints and identifies the violation nodes, graph attention network analyzes the node connection weight to detect mutual exclusion conflicts, quantum gate circuit checks the consistency of interface parameter transmission, and the verification results are summarized to generate timing constraint violation node information, mutual exclusion conflict information and parameter difference information. Based on the verification results, semantic rule identifiers are extracted, which abstract the core features of the constraint relationship, and version snapshots including semantic rule identifiers are generated to realize rule traceable binding. Through the digital twin model, the version snapshot is loaded to initialize the rule library, the real-time running data is input to drive the state evolution simulation, the quantum filter is used to predict the cascading failure path after denoising, the timing constraint threshold is dynamically adjusted and the dynamic graph structure weight is updated, forming a closed loop of modeling, verification, execution and optimization. The rule patch mechanism enhances the dynamic correction capability: the real-time state evolution data from the digital twin model is converted into a quantum state feature vector through the quantum harmonic oscillator layer, the Pauli operator is applied to match the abnormal mode to generate a rule patch, and the hybrid verification mechanism is used to update the timing constraints, mutual exclusion constraints and preset rules. The whole process realizes semantic specification unification, verification dimension fullness and rule correction automation, and completely solves the technical defect that manual verification cannot cover complex dependency relationships.
[0044] The specific implementation of the test link management method based on automatic verification and version control is carried out in the application scenario of the power dispatching system. In the background art, when the test link configuration does not clearly define the strict timing logic of the relay protection action and the circuit breaker tripping, manual verification may ignore the configuration error that the protection signal is ahead of the tripping instruction. To solve this technical problem, the implementation process is as follows: First, the test link configuration file and the historical version record stored in a distributed manner are read synchronously. Quantum state encoding processing is performed on the relay protection parameters in the configuration file, and the overcurrent threshold, action time delay and other parameters are mapped to the superposition state of quantum bits to form a superposition state data unit. Taking the relay protection device and the circuit breaker as dynamic graph nodes, three types of constraint edges are defined: trigger edges describe the state transition logic of "protection starting to tripping instruction", timing constraint edges set the maximum allowed time difference, and mutual exclusion edges declare the mutual exclusion operation of "reclosing and locking out". The power grid 50Hz power frequency characteristics are converted into the energy parameters of the timing constraint edges, so that the time window constraints are bound to the physical characteristics.
[0045] Then the hybrid verification mechanism is applied: the timing constraints are converted into the coupling coefficients of the Ising model, and the optimal action sequence is solved by minimizing the energy function of the quantum annealing machine. By comparing the actual action sequence in the dynamic graph structure with the optimal sequence, the rule-violating nodes that are ahead of time or lagging behind are identified. The graph attention network analyzes the node connection weights to detect mutually exclusive operation conflicts, such as the overlapping area of the reclosing and backup protection operation conflicts. The quantum gate circuit checks the consistency of the interface parameter transmission through the controlled NOT gate and generates parameter difference information such as the error of the current transformer ratio. The above verification results are summarized to project the state space of the fault scenario, and the cascading failure path is predicted.
[0046] Based on the verification results, the rule-violating node positions are marked in the three-dimensional power grid model, and the mutually exclusive conflict areas are visualized. The semantic rule identifiers are extracted to calculate the hash values, which are used as leaf nodes to build a quantum Merkle tree, and the version change records are signed using lattice cryptography to generate tamper-proof version snapshots. When initializing the digital twin model, the quantum mechanical equations are used to describe the state evolution of the relay protection device, and the equation Hamiltonian contains the power grid frequency parameter.
[0047] After inputting the real-time running data into the digital twin model, the quantum filter filters out sensor noise. Based on the denoised data, the cascading failure path is predicted, and the timing constraint threshold is dynamically adjusted: when the prediction shows that the protection action may trigger a cascading failure, the action time window threshold is shortened. According to the new threshold, the dynamic graph structure weight is updated, and the constraint strength of the key nodes is strengthened. The state evolution data output by the digital twin model is converted into a quantum state feature vector through the quantum harmonic oscillator layer, and the Pauli operator is used to match the abnormal pattern to generate rule patches, such as expanding the definition of mutually exclusive relationships or correcting the interface parameter tolerance. The hybrid verification mechanism is injected to realize rule hot updating.
Claims
1. A test pipeline management method based on automated verification and version control, characterized in that, include: Step 1: Obtain test link configuration data and historical version data. Based on the test link configuration data and historical version data, convert them into a dynamic graph structure that includes test link nodes. The dynamic graph structure defines the constraint relationships between nodes, including time constraints. Step 2: Apply a preset quantum annealing machine to verify whether the action sequence in the dynamic graph structure conforms to the temporal constraints, and generate node information that violates the temporal constraints; apply a graph attention network to verify whether the mutual exclusion relationship between nodes in the dynamic graph structure conforms to the mutual exclusion constraint in the constraint relationship, and generate mutual exclusion conflict information. Quantum gate circuits are used to verify whether the interface parameter passing in the dynamic graph structure conforms to the preset rules, and parameter difference information is generated; the node information, mutual exclusion conflict information and parameter difference information that violate timing constraints are summarized to generate the verification results; Step 3: Based on the node information that violates the timing constraints, the mutual exclusion conflict information, and the parameter difference information, generate a verification result report, and based on the verification results, extract semantic rule identifiers and generate a version snapshot including semantic rule identifiers; Step 4: Based on the version snapshot generated in Step 3, establish a digital twin model of the test link, collect real-time running data of the test link, and input the real-time running data into the digital twin model; Step 5: Use quantum filtering to process the noise in the real-time running data to generate denoised running data. Use the denoised running data to predict cascading failure paths. Based on the prediction results, dynamically adjust the timing constraint thresholds in the dynamic graph structure. Based on the adjustment of the timing constraint thresholds, update the constraint relationships between nodes in the dynamic graph structure to generate an optimized dynamic graph structure.
2. The test link management method based on automated verification and version control according to claim 1, characterized in that, Step 1 includes: Synchronously read the test link configuration file and historical version records from the distributed storage; Based on the test link configuration file, the relay protection parameters in the test link configuration file are quantum state encoded to form superposition state data units; Among them, the synchronous reading of the test link configuration file and historical version record flow of the distributed storage is based on the test link configuration file, and the relay protection parameters in the test link configuration file are processed by quantum state encoding.
3. The test link management method based on automated verification and version control according to claim 2, characterized in that, Step 1 includes: Relay protection devices and circuit breakers are used as nodes in the diagram; Define constraint edges between nodes. Constraint edges include trigger edges representing state transitions, timing constraint edges representing the maximum allowable time difference, and mutual exclusion edges representing mutually exclusive operations. Transform the power frequency characteristics of the power grid into energy parameters of time-constrained edges; Among them, relay protection devices and circuit breakers are used as graph nodes to define the constraint edges between nodes; defining the constraint edge flow between nodes transforms the power grid frequency characteristics into the energy parameters of the time-series constraint edges.
4. The test link management method based on automated verification and version control according to claim 3, characterized in that, Step 2 includes: The quantum annealing machine is used to verify whether the action sequence in the dynamic graph structure conforms to the temporal constraints, and to generate node information that violates the temporal constraints. Graph attention network is used to verify whether the mutual exclusion relationship between nodes in a dynamic graph structure conforms to mutual exclusion constraints, and mutual exclusion conflict information is generated. Quantum gate circuits are used to verify whether the interface parameter passing in the dynamic graph structure conforms to the preset rules, and parameter difference information is generated. Construct a state space projection of the fault scenario. The state space projection is based on node information that violates timing constraints, mutual exclusion conflict information, and parameter difference information. The system summarizes node information that violates timing constraints, mutual exclusion conflict information, parameter difference information, and state space projection data to generate verification results.
5. The test link management method based on automated verification and version control according to claim 4, characterized in that, Step 2 also includes: Transform the temporal constraints into Ising model coupling coefficients; Solving for the optimal action sequence by minimizing the energy function of the Ising model; Compare the actual action sequence with the optimal action sequence in the dynamic graph structure; Identify nodes in the actual action sequence that deviate from the optimal action sequence and generate node information that violates the timing constraints; Specifically, the temporal constraints are transformed into the flow of the coupling coefficient of the Ising model, and the energy function of the Ising model is minimized to solve for the optimal action sequence. The flow of the optimal action sequence obtained by minimizing the energy function of the Ising model is compared with the actual action sequence and the optimal action sequence in the dynamic graph structure. The flow of the actual action sequence and the optimal action sequence in the dynamic graph structure is compared to identify the nodes in the actual action sequence that deviate from the optimal action sequence, and the node information that violates the temporal constraints is generated.
6. The test link management method based on automated verification and version control according to claim 5, characterized in that, Step 3 includes: Based on the node information that violates the timing constraints generated in step 2, the nodes that violate the timing constraints are marked in the three-dimensional power grid model; Based on the mutual exclusion conflict information generated in step 2, the mutual exclusion conflict is visualized in the three-dimensional power grid model; Based on node information that violates timing constraints and mutual exclusion conflict information, a repair instruction is generated; Specifically, based on the node information that violates the timing constraints generated in step 2, the flow direction of the node that violates the timing constraints is marked in the three-dimensional power grid model; based on the mutual exclusion conflict information generated in step 2, the mutual exclusion conflict is visualized in the three-dimensional power grid model; based on the mutual exclusion conflict information generated in step 2, the flow direction of the mutual exclusion conflict is visualized in the three-dimensional power grid model; and based on the node information that violates the timing constraints and the mutual exclusion conflict information, a repair instruction is generated.
7. The test link management method based on automated verification and version control according to claim 6, characterized in that, Step 3 includes: Based on the verification results generated in step 2, semantic rule identifiers are extracted, hash values of semantic rule identifiers are calculated, and quantum Merkle trees are constructed using hash values as leaf nodes to obtain version change records and perform cryptographic signatures on the version change records. Specifically, based on the verification results generated in step 2, the semantic rule identifier flow is extracted and the hash value of the semantic rule identifier is calculated; the hash value of the semantic rule identifier is used as the leaf node to construct a quantum Merkle tree; the hash value is used as the leaf node to construct the quantum Merkle tree flow to obtain the version change record; the version change record flow is used to perform a cryptographic signature on the version change record.
8. The test link management method based on automated verification and version control according to claim 7, characterized in that, Step 4 includes: Initialize the digital twin model based on the version snapshot generated in step 3; In the digital twin model, quantum mechanical equations are used to describe the state evolution of relay protection devices; Among them, based on the version snapshot generated in step 3, the initialization of the digital twin model flow is carried out. In the digital twin model, quantum mechanical equations are used to describe the state evolution of the relay protection device.
9. The test link management method based on automated verification and version control according to claim 8, characterized in that, Step 5 includes: The noise in the real-time running data collected in step 4 is processed using quantum filtering to generate denoised running data. Based on the denoised operating data, predict the cascading failure path; Based on the cascading failure path prediction results, dynamically adjust the timing constraint thresholds in the dynamic graph structure; Update the weights of the dynamic graph structure based on the dynamically adjusted temporal constraint thresholds; Specifically, the noise in the real-time running data collected in step 4 is processed using quantum filtering to generate denoised running data flow. Based on the denoised running data, the cascading failure path is predicted. Based on the denoised running data, the cascading failure path flow is predicted. Based on the cascading failure path prediction results, the temporal constraint threshold in the dynamic graph structure is dynamically adjusted. Based on the cascading failure path prediction results, the temporal constraint threshold in the dynamic graph structure is dynamically adjusted. Based on the dynamically adjusted temporal constraint threshold, the weights of the dynamic graph structure are updated.
10. The test link management method based on automated verification and version control according to claim 9, characterized in that, Also includes: Output real-time state evolution data from the digital twin model; Real-time state evolution data is converted into quantum state feature vectors through a quantum harmonic oscillator layer; Based on quantum state feature vectors, Pauli operators are applied to match anomalous evolution patterns and generate rule patches. Inject the rule patch into the hybrid verification mechanism in step 2 to update the timing constraints, mutual exclusion constraints, and preset rules; Specifically, the flow of real-time state evolution data output from the digital twin model is as follows: the real-time state evolution data is converted into quantum state feature vectors through a quantum harmonic oscillator layer; the flow of the real-time state evolution data converted into quantum state feature vectors through the quantum harmonic oscillator layer is based on the quantum state feature vectors, and Pauli operators are applied to match abnormal evolution patterns to generate rule patches; the flow of the rule patches is then injected into the hybrid verification mechanism in step 2 to update the temporal constraints, mutual exclusion constraints, and preset rules.