Automatic correction method and system for test process of pumped storage power station

Abstract

The embodiment of the invention discloses an automatic correction method and system for a test process of a pumped storage power station. The method comprises the steps that a digital power station equipment model is constructed, and an equipment relation topological graph and a performance parameter set are obtained; establishing a control logic rule base based on the equipment relation topological graph and the performance parameter set; constructing a preset security boundary system based on the control logic rule base and the security operation standard; testing steps of the testing process document are converted into structured testing sequence data, and standardized testing process description is obtained; aiming at the standardized test process description, the control logic rule base and the preset security boundary system, performing logic conflict identification to obtain a conflict detection result set; and generating multiple sets of candidate correction schemes based on the conflict detection result set and the test coverage rate requirement. The method can solve the problem that in the prior art, a test process lacks dynamic adaptive capacity, intelligent conflict recognition and automatic correction capacity.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power system automation testing, in particular to a method and system for automatically correcting the test process of a pumped storage power station. BACKGROUND

[0002] As an important peak regulation and frequency modulation facility in the power system, the accuracy and effectiveness of the test process of a pumped storage power station are directly related to the safe operation and performance of the power station. With the development of the complexity and intelligence of the power system, higher requirements are placed on the management and optimization of the test process of a pumped storage power station.

[0003] Currently, the industry generally adopts the method of manually preparing and maintaining the test process, mainly relying on experienced engineers to develop standardized test sequences according to the characteristics of the equipment and the operation specifications. Another common practice is to use static test process templates to guide test execution through pre-set checklists and operation manuals.

[0004] The most relevant in the prior art is a test process management system based on a rule engine, which logically verifies and executes control over test steps through a pre-defined rule set. The technical principle is to decompose the test process into discrete operation units, each unit containing preconditions, execution actions and post-state, and the system verifies the rationality of the process through a rule matching mechanism. When a violation of the pre-set rules is detected, the system will issue an alarm or prevent execution.

[0005] However, the existing technology has significant technical defects: first, it lacks dynamic adaptability and cannot automatically adjust the test process according to the upgrade of the power station system or changes in operating conditions, resulting in a disconnection between the test scheme and the actual operating conditions; second, it lacks intelligent conflict recognition and automatic correction capabilities, still requiring a large amount of manual intervention to handle complex logical conflicts and redundant operations, which is inefficient and prone to human error. SUMMARY

[0006] The purpose of the present application is to provide a method and system for automatically correcting the test process of a pumped storage power station, aiming to solve the problem of lack of dynamic adaptability, intelligent conflict recognition and automatic correction capability in the prior art.

[0007] To achieve the above-mentioned purpose, the present application provides a method for automatically correcting the test process of a pumped storage power station, comprising the following steps: Based on the configuration parameters and operating data of the power station, a digital power station equipment model containing the hydro-generator unit, the pump, and the transformer is constructed, obtaining the equipment relationship topology graph and the performance parameter set; Based on the equipment relationship topology graph and the performance parameter set, the control dependency relationship and the operation constraint conditions between the digital power station equipment of the hydro-generator unit, the pump, and the transformer are analyzed, and a control logic rule library is established; determine the operation parameter boundary values of each device and system based on the control logic rule base and the safe operation standard, and construct a preset safe boundary system; digitally analyze and process the existing test procedure document, convert the test steps of the test procedure document into structured test sequence data, and obtain a standardized test procedure description; For the standardized test procedure description, the control logic rule base and the preset safe boundary system, logical conflict identification is performed to detect contradictions and redundant operations between test steps, and a conflict detection result set is obtained. Based on the conflict detection result set and the test coverage requirement, multiple sets of candidate correction schemes are generated.

[0008] The application also provides an automatic correction system for a pumped storage power station test procedure, comprising: A device modeling module is used to construct a digital power station device model containing a hydro-generator unit, a pumping pump and a transformer based on power station configuration parameters and operation data, to obtain a device relationship topology graph and a performance parameter set. A rule base establishment module is used to analyze the control dependency relationship and operation constraint conditions between the digital power station devices of the hydro-generator unit, the pumping pump and the transformer based on the device relationship topology graph and the performance parameter set, and to establish a control logic rule base. A safe boundary construction module is used to determine the operation parameter boundary values of each device and system based on the control logic rule base and the safe operation standard, and to construct a preset safe boundary system. A procedure analysis module is used to digitally analyze and process the existing test procedure document, convert the test steps of the test procedure document into structured test sequence data, and obtain a standardized test procedure description. A conflict identification module is used to perform logical conflict identification for the standardized test procedure description, the control logic rule base and the preset safe boundary system, to detect contradictions and redundant operations between test steps, and to obtain a conflict detection result set. A correction scheme generation module is used to generate multiple sets of candidate correction schemes based on the conflict detection result set and the test coverage requirement.

[0009] Compared with the prior art, the application has the following beneficial effects: The application realizes the conversion from static information to dynamic model by constructing a digital power station equipment model, and provides a comprehensive and accurate data basis for subsequent analysis. A control logic rule library is established based on the equipment relationship topology graph and the performance parameter set, and the control dependency relationship and operation constraint conditions between the equipment are systematically extracted and formalized, so that the rules have explainability and operability. A multi-dimensional preset safety boundary system is constructed, and the safety constraints of the equipment level and the system level are considered at the same time, so as to provide comprehensive safety criteria for test flow correction. Intelligent conflict detection is realized through the logical conflict identification of the test flow. Based on the conflict detection result set and the test coverage requirement, the test efficiency, safety and coverage rate are balanced, and diversified test flow correction schemes are generated.

[0010] The above description is only a summary of the technical scheme of the present disclosure, in order to more clearly understand the technical means of the present disclosure, and can be implemented according to the content of the specification, and in order to make the above and other purposes, characteristics and advantages of the present disclosure more obvious and easy to understand, the following preferred embodiments are described in detail below, and the accompanying drawings are described as follows. BRIEF DESCRIPTION OF DRAWINGS

[0011] In order to more clearly illustrate the technical scheme of the embodiments of the present disclosure, the following will briefly introduce the drawings needed to be used in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present disclosure, and those skilled in the art can obtain other drawings according to these drawings without creating any creative labor.

[0012] Figure 1 The flowchart of the automatic correction method of the pumped storage power station test flow of the present application; Figure 2 The working flowchart of the equipment modeling module of the present application; Figure 3 The process chart of the establishment of the control logic rule library of the present application; Figure 4 The hierarchical structure chart of the construction of the safety boundary of the present application; Figure 5 The digital analysis flowchart of the test flow of the present application; Figure 6 The schematic diagram of the maximization principle of the sub-module of the reconnection sequence greedy algorithm of the present application; Figure 7 The application diagram of the fixed parameter approximation principle of the Multiwinner rule in the multi-objective optimization of the present application; Figure 8 The structural block diagram of the automatic correction system of the pumped storage power station test flow of the present application. DETAILED DESCRIPTION

[0013] The embodiments of the present disclosure will be described in detail below with reference to the drawings.

[0014] It should be apparent that the following describes the embodiments of the present disclosure through specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the disclosure. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, not all. The present disclosure can also be implemented or applied by other different specific embodiments, and various modifications or changes can be made to the details in the specification without departing from the spirit of the present disclosure. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict. Based on the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present disclosure.

[0015] As shown in Figure 1 The present application provides an automatic correction method for a pumped storage power station test process, comprising the following steps: S1: Based on the power station configuration parameters and operation data, a digital power station equipment model containing the hydroelectric generator set, the water pumping pump and the transformer is constructed, and the equipment relationship topology graph and the performance parameter set are obtained; S2: Based on the equipment relationship topology graph and the performance parameter set, the control dependency relationship and the operation constraint condition between the digital power station equipment of the hydroelectric generator set, the water pumping pump and the transformer are analyzed, and a control logic rule library is established; S3: Based on the control logic rule library and the safe operation standard, the operation parameter boundary value of each device and system is determined, and a preset safety boundary system is constructed; S4: The existing test process document is digitally analyzed and processed, the test steps of the test process document are converted into structured test sequence data, and a standardized test process description is obtained; S5: For the standardized test process description, the control logic rule library and the preset safety boundary system, logical conflict identification is performed, and the contradictions and redundant operations between the test steps are detected, and a conflict detection result set is obtained; S6: Based on the conflict detection result set and the test coverage requirement, multiple sets of candidate correction schemes are generated.

[0016] As shown in Figure 2 In the preferred embodiments of the present application, step S1 based on the actual configuration parameters and operation data of the power station, the digital power station equipment model containing the hydroelectric generator set, the water pumping pump and the transformer is constructed, and the equipment relationship topology graph and the performance parameter set are obtained, which specifically includes: S1.1: Obtain technical parameters of the hydroelectric generating set, the water pumping set, the transformer, the switching device, and the monitoring system based on the power station configuration document and the equipment list, and obtain an equipment basic information database; The power station configuration document and the equipment list containing actual configuration parameters of the power station can be engineering archives, equipment nameplate information, and technical manuals of the power station. First, basic parameters are extracted from the engineering archives, the equipment nameplate information, and the technical manuals of the power station. For the hydroelectric generating set, key parameters such as rated power, speed range, water head range, and starting time are collected; for the water pumping set, information such as head characteristic, flow characteristic, and power curve is collected; and for the transformer, basic data such as capacity specification, voltage grade, and impedance parameter are recorded. Then, basic configuration information of the switching device and the monitoring system is recorded, and a complete equipment basic information database is established.

[0017] S1.2: Based on the equipment basic information database, a topological analysis algorithm is used to identify the physical connection relationship and the electrical connection relationship between the devices, and an adjacency matrix and a connected graph between the devices are constructed, and a device relationship topological graph is obtained.

[0018] The execution of the topological analysis algorithm specifically includes the following steps: (1) Each record is mapped into a device node object in a unified format: wherein may be spatial coordinates , or a plant area code; contains key attributes such as rated power, voltage grade, and pipe diameter.

[0019] (2) The physical connection relationship is vectorized to obtain a node-edge sketch. In this process, adjacent, same medium, and same pipe diameter pipe sections can be merged into a logical edge. Then, according to a prior flow direction template of pump-valve-set, a direction is added to the undirected edge to obtain a physical adjacency submatrix: (3) A graph convolution network (GCN) is used for one-time node classification to distinguish between “functional nodes” (such as generators, transformers, and circuit breakers) and “non-functional nodes” (such as bus sections and short cables) in the electrical connection relationship. The non-functional nodes and the edges at both ends thereof are merged into a logical edge with a lower weight, and finally an electrical adjacency submatrix is obtained: (3) and are fused into a weighted adjacency matrix: wherein the coefficient ​According to engineering experience: if an edge appears in both the physical and electrical layers, the weight is set to 2, indicating stronger coupling in both domains; if it only appears in one layer, the weight is 1. Thus, we obtain which not only preserves the original coupling strength but can also be directly used in subsequent graph algorithms.

[0020] (4) Perform connected component detection. If there are isolated nodes, use the "shortest hanging path" method to complete the missing connections, and finally generate a device relationship topology graph. In the device relationship topology graph, nodes can be attached with all attributes, and edges can be attached with connection type, cable length, pipe diameter, etc. metadata, which can be directly called by control logic rule library, security boundary system and test flow correction module.

[0021] S1.3: Based on the device relationship topology graph and the operation data, a dynamic characteristic model of each device is established through a performance modeling algorithm to obtain the performance parameter set, wherein the dynamic characteristic model of the device includes power characteristics, efficiency curve, and response time.

[0022] The performance modeling algorithm is executed and specifically includes the following steps: S1.3 Based on the device relationship topology graph and the operation data, a dynamic characteristic model of each device is established through a performance modeling algorithm to obtain the performance parameter set, wherein the dynamic characteristic model of the device includes power characteristics, efficiency curve, and response time.

[0023] The performance modeling algorithm is executed and specifically includes the following steps: (1) Extract the static parameters of each device from the device relationship topology graph, including the basic technical parameters of the device (such as rated power, rated voltage, rated speed, etc.). Collect device operation data, including real-time monitoring data (such as power, speed, temperature, etc.) and historical operation data. Clean and preprocess these data to remove outliers and noise, and ensure the accuracy and consistency of the data.

[0024] (2) Dynamic characteristic model construction Hydroelectric generating set: Establish a power characteristic model to describe the relationship between power and water head and flow; establish an efficiency curve model to reflect the efficiency change of the device under different working conditions; and establish a response time model to describe the time delay of the device from receiving an instruction to actual response. Power characteristic model (describing the relationship between power P and water head H and flow Q): Where: P: output power (unit: watt, W) η: current efficiency (dimensionless) ρ: density of water (unit: kg / m³) g: acceleration due to gravity (unit: meters per second squared, m / s²) Q: flow rate (unit: cubic meters per second, m³ / s) H: water head (unit: meters, m) Efficiency curve model (reflecting the efficiency change of the device under different working conditions): Where: η: current efficiency η0: efficiency under optimal working conditions a and b: efficiency change coefficients (obtained through experiments or fitting) Q opt and H opt : flow rate and water head under optimal working conditions Response time model (describing the time delay from receiving an instruction to the actual response of the device): Where: τ: response time (unit: seconds, s) τ0: reference response time c and d: response time change coefficients (obtained through experiments or fitting) Q set and H set : set flow rate and water head Pump: Establish power characteristic model to describe the relationship between power and lift and flow rate; establish efficiency curve model to reflect the efficiency change of the device under different working conditions; establish response time model to describe the time delay from receiving an instruction to the actual response of the device.

[0025] Power characteristic model (describing the relationship between power P and lift H and flow rate Q): Where the parameter meanings are the same as those of the hydroelectric generator set.

[0026] Efficiency curve model (reflecting the efficiency change of the device under different working conditions): Where the parameter meanings are the same as those of the hydroelectric generator set.

[0027] Response time model (describing the time delay from receiving an instruction to the actual response of the device): Where the parameter meanings are the same as those of the hydroelectric generator set.

[0028] Transformer: Establish a power characteristic model to describe the power output of the transformer under different loads; establish an efficiency curve model to reflect the efficiency changes of the equipment under different loads; establish a response time model to describe the dynamic response characteristics of the equipment when the load changes.

[0029] Power characteristic model (describes the power output of a transformer under different loads): in: P: Output power (unit: watts, W) P rated Rated power S: Current load (unit: volt-amperes, VA) S rated Rated load Efficiency curve model (reflecting the efficiency changes of equipment under different loads): in: η: Current efficiency η max Maximum efficiency a: Efficiency variation coefficient (obtained through experimentation or fitting) S and S rated Current load and rated load Response time model (describes the dynamic response characteristics of a device under varying loads): in: τ: Response time (in seconds, s) τ0: Reference response time c: Response time variation coefficient (obtained through experimentation or fitting) S set : Set load (3) Model calibration and verification The established dynamic characteristic model is calibrated using operational data. Model parameters are adjusted through a fitting algorithm to ensure the model accurately reflects the actual operating characteristics of the equipment. The model is then validated by comparing its predicted values ​​with actual operational data to evaluate its accuracy and reliability. If there is a significant deviation between the model predictions and the actual data, the model is adjusted and optimized.

[0030] (4) Generation of performance parameter set Key parameters are extracted from the calibrated dynamic characteristic model to form a performance parameter set. This set includes parameters such as the power characteristics, efficiency curves, and response time of the equipment under different operating conditions. The performance parameter set is stored in a structured format for easy subsequent analysis and application.

[0031] As Figure 3 shown, in the preferred embodiment of the present application, step S2 analyzes the control dependency relationship and operation constraint condition between the digital power plant equipment of the hydroelectric generating set, the water pumping pump and the transformer based on the equipment relationship topology graph and the performance parameter set, and establishes a control logic rule library, specifically including: S2.1: Based on the equipment relationship topology graph and the performance parameter set, the equipment control logic relationship matrix is obtained by analyzing the equipment start-stop sequence, interlocking relationship and protection condition through a logic extraction algorithm; For example, before the hydroelectric generating set is started, it must ensure that the water inlet valve is opened, the main bearing oil pressure is normal, the speed regulation system is ready, and other preconditions; and analyze the interlocking relationship between the equipment, such as the main transformer can only be subjected to excitation test in the unit shutdown state. By analyzing the above operation conditions and constraint relationships, a complete control logic relationship matrix is established to represent the control dependency and sequence constraint between the equipment.

[0032] The execution of the above logic extraction algorithm specifically includes: first traversing all possible start-stop paths in a graph theory manner, then mapping the three constraints of “interlocking, protection, sequence” into Boolean functions using formal rules, and finally outputting a matrix readable by a machine and inferable, that is, the equipment control logic relationship matrix.

[0033] S2.2: Based on the equipment control logic relationship matrix, the control logic is converted into an IF-THEN format rule expression to obtain a standardized control rule set; The control logic relationship is converted into a standardized IF-THEN rule expression. For example, “if the water turbine guide water system is not ready, then prohibit the execution of the unit start operation” will be converted into the formal rule of “IF (guide water system state ≠ ready) THEN (prohibit operation: unit start)”. In this way, all control logic relationships are standardized into rule expressions of a unified format, facilitating subsequent rule judgment and reasoning.

[0034] S2.3: Based on the standardized control rule set, a control logic rule library supporting fast query and reasoning is established through a rule library construction algorithm to obtain the control logic rule library containing an index mechanism.

[0035] The execution of the above rule library construction algorithm specifically includes: first compiling the standardized IF-THEN rule set output by S2.2 into an incremental updateable “decision graph”, then establishing a multi-level hybrid index on this graph, and further obtaining the control logic rule library containing an index mechanism, so that subsequent “query allowed actions given the current state” or “reverse preconditions given target actions” can be completed in sub-millisecond.

[0036] where the set of normalized IF-THEN rules is compiled into an incrementally updatable "decision graph" by first converting each rule into a CNF formula and then using the Bryant algorithm based on variable ordering to construct an OBDD: All rules are merged into the same multi-root OBDD by a "shared sub-graph" strategy, and the number of nodes after merging satisfies: which is essentially a compressed "if-else decision graph" with an average path length of .

[0037] Thirdly, the multi-level hybrid index is established, including: Two sets of indexes are established in parallel to support both "forward reasoning" (finding actions given a state) and "backward reasoning" (finding feasible states given an action): 1) Forward index: with the hash key pointing to the corresponding node in the OBDD; since the OBDD path is unique, a hash can locate it. 2) Reverse index: with the action atom as the key, all leaf node paths containing it are stored in reverse. As shown in Fig. 3, in the preferred embodiment of the present application, step S3 determines the operating parameter boundary values of each device and system based on the control logic rule base and the safe operation standard, and constructs a preset safety boundary system, specifically including: S3.1: Based on the control logic rule base and the safe operation standard of the digital power plant equipment such as the hydro-generator unit, the water pump and the transformer, the minimum value and the maximum value of the key operating parameters of each device are determined, and a device-level safety parameter boundary set is obtained; Based on the safe operation standard and the control logic rule of the device, the safety boundary value of each key parameter is determined. For the hydro-generator unit, the speed range, the load change rate limit, the bearing temperature upper limit and other parameter boundaries of its safe operation are calculated; for the water pump, the safe lift range, the minimum flow requirement, the maximum allowable vibration value and other boundary conditions are determined; for the transformer, the temperature rise limit under the rated capacity, the overvoltage protection setting value, the short-circuit current bearing capacity and other safety boundaries are determined. Through systematic boundary calculation of the key parameters of each device, a complete device-level safety parameter boundary set is formed.

[0038] Figure 4

[0039] ​​​​​S3.2: Based on the device-level safety parameter boundary set and system operation constraints, calculate the safety operation boundary of the system as a whole through a system-level constraint analysis algorithm to obtain system-level safety boundary constraint conditions; The execution of the system-level constraint analysis algorithm specifically includes: (1) Write all the device-level safety parameters, such as the speed of a certain unit cannot be higher than 1500 r / min, the water head of a certain pipe cannot drop by 20 m, into interval boxes of the same dimension to obtain a high-dimensional cuboid. This cuboid only guarantees that the single machine and single pipe do not exceed the limit, but does not consider the coupling between them.

[0040] (2) Simultaneously start four independent sub-engines, each of which outputs a "local safety corridor". Among them, the hydraulic engine uses the method of characteristic line to connect the pipe, surge tank, and tailwater tunnel into a transient model, and rolls to calculate which flow rate changes first under extreme pump switching or load rejection conditions. The electromechanical engine takes the fourth-order model of the unit to run the transient stability program to find the maximum output step that the unit can maintain synchronization under grid faults. The grid engine uses direct current flow to scan the thermal stability of bus voltage and line current together with the power angle stability to give the joint upper limit of power angle and power. The monitoring engine extrapolates quantities such as bearing temperature and winding temperature, which have large thermal inertia, to the limit according to the time constant.

[0041] (3) Boundary fusion, put the four local safety corridors into the same high-dimensional coordinate system, first find the intersection, and then use convex hull or minimum volume ellipsoid to package the intersection. Any operating point that falls inside this ellipsoid is equivalent to simultaneously satisfying all the limit constraints of hydraulic, electromechanical, grid, and monitoring.

[0042] S3.3: Based on the device-level safety parameter boundary set and the system-level safety boundary constraint conditions, construct a comprehensive safety boundary system covering electrical parameters, mechanical parameters, and hydraulic parameters to obtain the preset safety boundary system.

[0043] Map the boundary conditions of electrical parameters (voltage, current, power, etc.), mechanical parameters (speed, vibration, temperature, etc.), and hydraulic parameters (water head, flow, pressure, etc.) to a unified parameter space to construct a comprehensive safety boundary system. Through multi-dimensional space mapping and boundary surface fitting technology, a multi-dimensional safe operation region representation is formed, and any operating point inside this region is considered safe. This comprehensive boundary system can simultaneously consider the constraints of multiple physical quantities, providing comprehensive safety criteria for subsequent test flow correction.

[0044] The core of multidimensional spatial mapping technology lies in mapping different types of parameters (such as electrical, mechanical, and hydraulic parameters) from their respective independent parameter spaces to a unified multidimensional parameter space. This process requires standardization of each parameter to ensure that parameters with different dimensions and magnitudes can be effectively compared and analyzed in the same space. For example, electrical parameters such as voltage, current, and power; mechanical parameters such as rotational speed, vibration, and temperature; and hydraulic parameters such as head, flow rate, and pressure each have different physical meanings and numerical ranges. Through standardization, they can be converted into dimensionless relative values, thus enabling comprehensive consideration in a unified multidimensional space.

[0045] Boundary surface fitting technology is a method used to determine the boundary of the safe operating region after multidimensional spatial mapping. This technology analyzes the safe operating data of equipment and systems, combining the physical characteristics of the equipment and operating constraints, to fit a multidimensional surface that accurately describes the boundary of the safe operating region. This boundary surface clearly distinguishes the safe operating region from the unsafe operating region; any operating point located within the area enclosed by this boundary surface can be considered safe. The boundary surface fitting process needs to consider various factors, including the rated parameters of the equipment, safety margins, and the range of operating condition variations, to ensure that the boundary surface accurately reflects the safe operating limits of the equipment and system.

[0046] like Figure 5 As shown, in a preferred embodiment of the present invention, step S4 performs digital parsing processing on the existing test process document, converting the test steps in the test process document into structured test sequence data to obtain a standardized test process description, specifically including: S4.1: Based on the existing test process documents, extract the test step descriptions, operation objects and parameter setting information to obtain an unstructured test information set; The system intelligently analyzes existing test process documents to extract key information. First, it identifies the paragraph structure and heading hierarchy within the document. Then, it uses natural language processing (NLP) technology to extract the descriptive text for each test step. Through keyword matching and entity recognition, the system identifies the operational objects (e.g., "Unit 1," "main transformer"), operational actions (e.g., "start," "stop"), and parameter settings (e.g., "set speed to 1500 rpm") within the test steps. For complex test descriptions, syntactic analysis is used to understand the relationships between actions and objects, ensuring the accuracy of information extraction.

[0047] S4.2: Based on the unstructured test information set, identify the temporal relationships and dependencies between test steps to obtain a test step relationship graph; Based on the extracted unstructured information, the logical relationships between test steps are analyzed. Temporal keywords (such as "after" and "after completion") and conditional expressions (such as "when..." and "if...then") in the documents are identified to construct dependencies between steps. For implicit dependencies, reasoning is performed based on domain knowledge, such as automatically identifying the preparatory conditions before device startup. Through these analyses, a complete test step relationship graph is constructed, representing the logical structure and execution order of the test process.

[0048] S4.3: Based on the test step relationship diagram, the test process is converted into a standardized data format containing operation sequence number, device object, parameter value, preconditions, and postconditions to obtain the standardized test process description.

[0049] The test step relationship diagram is converted into a standardized data format. Each test step is formatted as a structured data item containing an operation sequence number, device object identifier, parameter value and unit, precondition expression, and poststate description. These data items are then standardized, unifying device naming, standardizing parameter units, and formalizing condition expressions, ultimately generating a fully standardized test process description. This provides a structured data foundation for subsequent conflict detection and correction optimization.

[0050] like Figure 6 As shown, in a preferred embodiment of the present invention, step S5, based on the standardized test process description, the control logic rule base, and the preset security boundary system, performs logic conflict identification, detects contradictions and redundant operations between test steps, and obtains a conflict detection result set, specifically including: S5.1: Based on the standardized test process description and the control logic rule base, the execution order of test steps is analyzed by reconnecting the sub-modulus maximization principle of the sequential greedy algorithm, identifying the combination of steps that violate the device start-stop logic, and obtaining a set of logical conflict candidates. The standardized testing process is modeled as a directed graph structure, where nodes represent test steps and edges represent execution order relationships. A submodule utility function is defined to evaluate the effectiveness of step combinations, considering logical consistency, resource utilization, and safety metrics among steps. A greedy strategy starts with the initial test sequence, and in each iteration selects the connection reorganization operation that maximizes the overall submodule utility, such as adjusting the execution order of certain steps or changing the dependencies between steps. In this way, combined with a control logic rule base, step combinations that violate device control logic are identified, such as performing operations in an unready state or operation sequences that violate interlock conditions, indicating potential logical conflicts.

[0051] S5.2: Based on the logical conflict candidate set and the preset security boundary system, verify whether the test parameters exceed the security range, filter out security conflicts, and obtain the confirmed logical conflict set; The identified set of logical conflict candidates is compared with the preset safety boundary system to verify whether the test parameters exceed the safety range. Multi-level safety verification is performed: first, check whether a single parameter exceeds the independent safety boundary; then check whether the parameter combination meets the coupling constraint condition; finally, check whether the overall system constraint is satisfied. For timing-related safety constraints, verify whether the parameter change rate is within the safety range. Through these safety boundary checks, the real safety conflicts are filtered out to form the confirmed logical conflict set.

[0052] S5.3: Based on the confirmed logical conflict set, identify test steps with function duplication or the same effect, and obtain the conflict detection result set containing conflict type, impact range, and severity; The functional redundancy and effect similarity of test steps are analyzed to identify redundant operations. Each step is abstracted as a triple of input state, operation action, and output state, and by comparing the function descriptions of different steps, functionally equivalent or highly similar step combinations are identified. The data flow is also analyzed to identify steps that produce the same or highly related test results. For each identified conflict and redundancy, its type (logical conflict, safety conflict, redundant conflict), impact range (local impact, system-level impact), and severity (high, medium, low risk) are evaluated to form a detailed conflict detection result set.

[0053] S5.4: Based on the conflict detection result set, evaluate the cascading impact of conflict steps on subsequent test processes, calculate the conflict propagation path and impact depth, and obtain the conflict impact evaluation report; Evaluating the cascading impact of conflict steps on subsequent test processes specifically involves constructing a step dependency graph, tracing the state changes and error propagation paths that conflicts may cause, and calculating the breadth (number of steps affected) and depth (severity of impact) of the impact.

[0054] The step dependency graph is a directed graph structure where nodes represent each step in the test process and directed edges represent the dependency relationship between steps. For example, if the execution of step B depends on the result of step A, there will be a directed edge from A to B in the dependency graph. This dependency relationship can be data dependency (i.e., the output of the previous step is needed as input for the subsequent step) or control dependency (i.e., the execution condition of the subsequent step is based on the state of the previous step). By constructing the step dependency graph and tracing the propagation path of the conflict, the breadth and depth of the impact can be calculated to comprehensively evaluate the overall impact of the conflict on the test process. For example, a certain error setting may affect multiple subsequent test steps, even causing the entire test process to fail. By considering these factors, a conflict impact evaluation report is formed.

[0055] S5.5: Based on the aforementioned conflict impact assessment report, the conflicts are classified and ranked according to their severity and scope of impact, and the key conflict items that need to be addressed first are identified, resulting in a priority list of conflicts.

[0056] Specifically, the severity of a conflict is assessed primarily from two aspects: the risk of equipment damage and the risk of a safety accident. If a conflict could lead to damage or malfunction of critical equipment, thereby affecting the normal operation of the power plant or even causing a safety accident, then the severity of the conflict is rated as high. For example, an incorrect operation might cause the turbine speed to exceed the safe range, resulting in equipment damage or even mechanical failure. This type of conflict is extremely severe because it directly threatens the safe operation of the power plant.

[0057] The scope of impact primarily considers the number of test steps affected by the conflict and the system components involved. If a conflict only affects a few test steps and does not involve critical system components, its scope of impact is relatively small. However, if a conflict may cause multiple test steps to fail or even affect the collaborative operation of multiple system components, its scope of impact is large. For example, an incorrect parameter setting may affect the execution of multiple subsequent test steps or even cause the entire test process to be interrupted; such a conflict has a significant scope of impact.

[0058] After comprehensively considering the severity and scope of the conflict, a comprehensive scoring mechanism can be used to quantitatively assess the conflict. Specifically, weights can be assigned to each dimension, and corresponding scores can be given based on the conflict's performance in each dimension. For example, severity may be given a higher weight because the risk of equipment damage and safety accidents is crucial to the operation of the power plant; the scope of impact will also be assigned corresponding weights based on its degree of influence on the testing process. A comprehensive score for each conflict is calculated by weighted summation.

[0059] Conflicts can be prioritized based on their overall scores to identify critical conflicts requiring immediate attention. Conflicts with the highest scores typically have the highest priority because they pose the greatest threat to the security, integrity, and efficiency of the testing process. Creating a prioritized conflict list provides clear guidance for generating subsequent corrective actions, ensuring that the most pressing and impactful conflicts are addressed within limited time and resources. This hierarchical prioritization method not only improves the efficiency of conflict resolution but also enhances the scientific and systematic nature of testing process optimization.

[0060] like Figure 7 As shown, in a preferred embodiment of the present invention, step S6, based on the conflict detection result set and test coverage requirements, generates multiple candidate correction schemes, specifically including: S6.1: Based on the conflict detection result set and the test coverage requirement, a multi-objective optimization problem is constructed through the fixed parameter approximation principle of Multiwinner rules, balancing test efficiency, safety, and coverage, to obtain an optimization objective function; The core of Multiwinner rules lies in decomposing complex multi-objective optimization problems into multiple sub-problems that can be approximately solved through fixed parameters, thereby achieving an effective balance between test efficiency, safety, and coverage.

[0061] To achieve a balance between these objectives, the fixed parameter approximation principle of Multiwinner rules simplifies the problem by introducing fixed parameters. These fixed parameters can be weight allocations between objectives or constraints on certain objectives. For example, a fixed weight vector can be set to assign different weights to test efficiency, safety, and coverage, thereby transforming the multi-objective optimization problem into a weighted single-objective optimization problem. In this way, a comprehensive optimal solution can be found by balancing different objectives.

[0062] Based on the fixed parameter approximation principle of Multiwinner rules, an optimization model is established to balance test efficiency, safety, and coverage. Three key objective functions are defined: the test efficiency objective function E(x) measures test execution time and resource consumption, the safety objective function S(x) evaluates the risk level of the test process, and the coverage objective function C(x) measures the degree of test coverage of system functions. The multi-objective optimization problem is converted into a series of single-objective optimization sub-problems, each corresponding to a different combination of weight vectors, representing different preference strategies. Through Pareto frontier analysis, the selected multiple schemes have good distribution and diversity in the objective space.

[0063] S6.2: Based on the optimization objective function, multiple sets of different test flow adjustment strategies are generated, each containing specific step deletion, modification, and rearrangement suggestions, obtaining a candidate correction scheme set; Based on the established optimization objective function, multiple test flow adjustment strategies are generated. Using a genetic algorithm framework, each correction scheme is encoded as a chromosome, with gene positions representing specific adjustment operations. Through mutation operations (such as step deletion, parameter modification, and order rearrangement) and crossover operations, diverse candidate schemes are generated. Each scheme contains detailed adjustment suggestions: redundant steps to be deleted, parameter values to be modified, step combinations to be reordered, and specific implementation methods and expected effects of each adjustment.

[0064] S6.3: Based on the set of candidate correction schemes, calculate the comprehensive score of each scheme, select the scheme with the best performance in safety, efficiency and coverage, and obtain the test flow correction suggestion; A hierarchical evaluation index system is established to comprehensively score each candidate scheme. The analytic hierarchy process is used to determine the index weight, and the standardized processing is used to convert the index values of different dimensions to a unified interval. For qualitative indicators, fuzzy evaluation method is used for quantitative processing. Through the weighted comprehensive evaluation model, the comprehensive score of each scheme is calculated, and the TOPSIS method is used to select the scheme closest to the ideal solution as the optimal test flow correction suggestion.

[0065] S6.4: Based on the set of candidate correction schemes, identify the set of non-dominated solutions, ensure good distribution and diversity of multiple candidate schemes in the target space, and obtain the Pareto optimal solution set; All non-dominated solutions are identified, that is, those solutions that cannot be improved simultaneously in any target. The fast non-dominated sorting algorithm is used to stratify the candidate schemes, and the optimal solution set on the Pareto front is identified. In order to ensure the diversity of the solution set, the crowded distance calculation method is used to evaluate the distribution of the solution space, and the solutions located in the sparse area are preferentially selected to cover a wider decision space. In this way, a set of candidate schemes with good performance under different combinations of target weights is generated.

[0066] S6.5: Based on the Pareto optimal solution set, test the influence of key parameter changes on the ranking of schemes, verify the robustness and stability of the correction suggestion, and obtain a sensitivity analysis report.

[0067] Test the influence of key parameter changes on the ranking of schemes to verify the robustness of the correction suggestion. Small range perturbation is performed on important weight parameters and critical thresholds, and the change of scheme ranking is observed. Through the Monte Carlo simulation method, a large number of random parameter combinations are generated to evaluate the stability statistical characteristics of the scheme ranking. For high sensitivity parameters, more detailed analysis is performed to determine the safety boundary and recommended value of parameter changes. Through these sensitivity analyses, it is ensured that the selected scheme can maintain good performance when the parameters fluctuate, thereby enhancing the reliability and applicability of the correction suggestion.

[0068] In the preferred embodiment of the present application, the method further comprises the step of automatically correcting the test flow based on the test flow correction suggestion: S7.1: Based on the test flow correction suggestion and the standardized test flow description, perform the steps of deletion, parameter adjustment and sequence rearrangement operation to obtain the modified test sequence; According to the correction suggestions, specific test procedure modification operations are performed. For redundant steps, they are directly deleted from the test sequence, while the reference relationship of related steps is adjusted; for steps with unreasonable parameters, the parameter values are modified to recommended safe values; for steps with unreasonable sequence, their execution order is rearranged to ensure compliance with the device start-stop logic and safety specifications. These modification operations are performed in an atomic manner, and after each modification, the system checks the coherence and integrity of the modification.

[0069] S7.2: Based on the modified test sequence, check the logical consistency and safety compliance of the procedure, ensure that all modifications comply with the control logic rule library and the preset safety boundary system requirements, and obtain a verified test sequence; The modified test sequence is comprehensively checked to ensure its logical consistency and safety compliance. Verify whether the preconditions of each test step can be met, check whether the dependency relationship between steps is complete, verify whether the parameter settings are within the safety range, and confirm whether the operation sequence meets the requirements of the control logic rule library. For complex conditional branches and loop structures, path coverage analysis is used to ensure that all possible execution paths are safe and effective.

[0070] S7.3: Based on the verified test sequence, generate a final test procedure document and executable instructions that meet the requirements of the power plant execution system, and obtain a corrected complete test procedure scheme; The verified test sequence is converted into a final output format that meets the requirements of the power plant execution system. A standardized test procedure document is generated, which includes detailed step descriptions, parameter settings, operation guidelines, and safety precautions. At the same time, executable instruction sets that can be directly loaded into automated test systems can also be generated to support automated or semi-automated test execution. The output document conforms to the power industry standard format and naming conventions, ensuring that operators can accurately understand and execute.

[0071] S7.4: Based on the corrected complete test procedure scheme, simulate the test execution process, verify the feasibility and safety of the corrected procedure, and obtain simulation verification results; Using digital twin technology to simulate the test execution process, verify the feasibility and safety of the corrected procedure. The simulation environment includes device models, control system models, and operating environment models, which can truly reflect the dynamic response characteristics of the device and the overall behavior of the system. Execute the test sequence step by step, monitor the change trajectory of key parameters, and detect potential abnormal situations and risk points. Through detailed simulation analysis, it is confirmed that the corrected test procedure can be safely and effectively executed.

[0072] S7.5: Based on the simulation verification results, calculate the efficiency improvement rate, safety risk reduction rate, and coverage improvement degree of the corrected test procedure, and obtain a procedure optimization effect evaluation report.

[0073] In the performance evaluation stage, the test procedures before and after correction are compared and analyzed, and the optimization effect is quantitatively evaluated. The time efficiency improvement rate (percentage reduction of test execution time), the safety risk reduction rate (reduction degree of dangerous operation and risk point), and the test coverage improvement degree (extension degree of function coverage range) are calculated. Through these quantitative indicators, a detailed process optimization effect evaluation report is generated, which provides an intuitive optimization result display for power plant managers and provides basic data for subsequent continuous improvement.

[0074] In the preferred embodiment of the present application, the sub-module maximization principle of the reconnection greedy algorithm includes: S9.1: Based on the standardized test procedure description, the test procedure is modeled as a directed graph structure, where the nodes represent test steps and the edges represent execution order relationships, obtaining a test procedure directed graph; The test procedure is converted into a mathematical directed graph structure G=(V,E), where the vertex set V represents the test steps and the edge set E represents the execution order relationship between the steps. For each test step, a corresponding graph node is created, containing step ID, operation type, device object, parameter setting and other attribute information. For the order relationship between steps, a directed edge is created to connect the related nodes, and the attributes of the edge include dependency type (strong dependency or weak dependency) and timing constraints (such as minimum interval time). In this way, a complete test procedure directed graph is constructed to provide data structure support for subsequent sub-module analysis.

[0075] S9.2: Based on the test procedure directed graph, the effectiveness of the test step combination is evaluated by the sub-module utility function, which considers the logical consistency, resource utilization and safety indicators, obtaining a sub-module utility score; An evaluation function U(S) is designed to measure the effectiveness of the sub-graph structure S, which considers multiple factors: a logical consistency indicator L(S) measures the rationality of the logical relationship between steps, a resource utilization indicator R(S) evaluates the efficiency of resource allocation, and a safety indicator F(S) measures the risk level of the operation sequence. The comprehensive utility function is defined as U(S)=w1L(S)+w2R(S)+w3F(S), where w1, w2, w3 are weight coefficients. Calculate the utility value of each potential sub-module in the test procedure, identify those high-utility and low-utility step combinations.

[0076] S9.3: Based on the sub-module utility score, the connection reorganization operation that can maximize the overall sub-module utility is selected by the greedy strategy selection algorithm in each iteration, obtaining a locally optimal sub-module structure.

[0077] From the initial test sequence, the iterative optimization method is used to reorganize the process structure. In each iteration, all possible connection reorganization operations (such as moving step positions, merging related steps, splitting complex steps, etc.) are evaluated, and the operation that can maximize the overall sub-module utility is selected. Specifically, each time the operation o = argmax (U (S U {o}) - U (S)), that is, the operation that maximizes the utility increment is selected. Through this greedy selection strategy, the test process structure is gradually improved, and finally a locally optimal sub-module organization form is obtained, which provides a basis for conflict identification and process optimization.

[0078] As Figure 8 shown, the application also provides an automatic correction system for a pumped storage power station test process, comprising: a device modeling module 10, a rule library establishment module 20, a safety boundary construction module 30, a process analysis module 40, a conflict identification module 50, and a correction scheme generation module 60.

[0079] The device modeling module 10 is used to construct a digital power station device model containing a hydro-generator unit, a pumping pump, and a transformer based on power station configuration parameters and operation data, to obtain a device relationship topology graph and a performance parameter set; The rule library establishment module 20 is used to analyze the control dependency relationship and operation constraint conditions between the digital power station devices of the hydro-generator unit, the pumping pump, and the transformer based on the device relationship topology graph and the performance parameter set, and to establish a control logic rule library; The safety boundary construction module 30 is used to determine the operating parameter boundary values of each device and system based on the control logic rule library and the safety operation standard, and to construct a preset safety boundary system; The process analysis module 40 is used to perform digital analysis processing on the existing test process document, to convert the test steps of the test process document into structured test sequence data, and to obtain a standardized test process description; The conflict identification module 50 is used to perform logical conflict identification for the standardized test process description, the control logic rule library, and the preset safety boundary system, to detect contradictions and redundant operations between test steps, and to obtain a conflict detection result set; The correction scheme generation module 60 is used to generate multiple sets of candidate correction schemes based on the conflict detection result set and the test coverage requirement.

[0080] The application realizes the conversion from static information to dynamic model by constructing a digital power plant equipment model, and provides a comprehensive and accurate data basis for subsequent analysis. A control logic rule library is established based on the equipment relationship topology graph and the performance parameter set, and the control dependency relationship and operation constraint conditions between the equipment are systematically extracted and formalized, so that the rules have interpretability and operability. A multi-dimensional preset safety boundary system is constructed, and the safety constraints of the equipment level and the system level are considered at the same time, so that comprehensive safety criteria are provided for test flow correction. Intelligent conflict detection is realized through logical conflict identification of the test flow. Based on the conflict detection result set and the test coverage requirement, the test efficiency, safety and coverage rate are balanced, and diversified test flow correction schemes are generated.

[0081] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein can be applied to other aspects without departing from the scope of the disclosure. Thus, the present disclosure is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0082] The above description has been given for the purpose of illustration and description. Furthermore, this description does not intend to limit the embodiments of the present disclosure to the forms disclosed herein. Although a number of example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. An automatic calibration method for the testing process of a pumped storage power station, characterized in that, Includes the following steps: Based on the power plant configuration parameters and operating data, a digital power plant equipment model including hydro-generator units, pumps, and transformers is constructed to obtain the equipment relationship topology diagram and performance parameter set; Based on the equipment relationship topology diagram and performance parameter set, the control dependencies and operational constraints among the digital power station equipment such as hydro-generator units, pumps, and transformers are analyzed, and a control logic rule base is established. Based on the control logic rule base and safe operation standards, the boundary values ​​of the operating parameters of each device and system are determined, and a preset safety boundary system is constructed. The existing test process documents are digitally parsed and processed to convert the test steps in the test process documents into structured test sequence data, resulting in a standardized test process description. Logical conflict identification is performed on the standardized test process description, the control logic rule base and the preset security boundary system to detect contradictions and redundant operations between test steps and obtain a conflict detection result set. Based on the conflict detection result set and test coverage requirements, multiple candidate correction schemes are generated.

2. The method according to claim 1, characterized in that, Based on the actual configuration parameters and operating data of the power station, a digital power station equipment model including hydro-generator units, pumps, and transformers is constructed, resulting in an equipment relationship topology diagram and a set of performance parameters, including: Based on the power plant configuration documents and equipment list, the technical parameters of the hydro-generator units, pumps, transformers, switchgear, and monitoring systems are obtained, resulting in a database of basic equipment information. Based on the aforementioned equipment basic information database, the physical and electrical connection relationships between equipment are identified through topology analysis algorithms, and an adjacency matrix and connectivity graph between equipment are constructed to obtain the equipment relationship topology graph. Based on the device relationship topology and operating data, a dynamic characteristic model of each device is established through a performance modeling algorithm to obtain the performance parameter set, wherein the dynamic characteristic model of the device includes power characteristics, efficiency curves, and response time.

3. The method according to claim 1, characterized in that, Based on the aforementioned equipment relationship topology and performance parameter set, the control dependencies and operational constraints among the digital power station equipment, including hydro-generator units, pumps, and transformers, are analyzed, and a control logic rule base is established, including: Based on the device relationship topology and performance parameter set, the device start-up and shutdown sequence, interlocking relationship and protection conditions are analyzed by logic extraction algorithm to obtain the device control logic relationship matrix; Based on the device control logic relationship matrix, the control logic is converted into a rule expression in IF-THEN format to obtain a standardized set of control rules. Based on the standardized set of control rules, a control logic rule base supporting fast querying and reasoning is established through a rule base construction algorithm, resulting in the control logic rule base containing an indexing mechanism.

4. The method according to claim 1, characterized in that, Based on the control logic rule base and safe operation standards, the boundary values ​​of the operating parameters of each device and system are determined, and a preset safety boundary system is constructed, including: Based on the control logic rule base and the digital power station equipment safety operation standards for hydro-generator sets, pumps, and transformers, the minimum and maximum values ​​of the key operating parameters of each device are determined, resulting in a set of equipment-level safety parameter boundaries. Based on the set of device-level safety parameter boundaries and system operation constraints, the overall safety operation boundary of the system is calculated using a system-level constraint analysis algorithm to obtain the system-level safety boundary constraint conditions. Based on the set of equipment-level safety parameters and the system-level safety boundary constraints, a comprehensive safety boundary system covering electrical parameters, mechanical parameters, and hydraulic parameters is constructed to obtain the preset safety boundary system.

5. The method according to claim 1, characterized in that, The existing test process documents are digitally parsed to convert the test steps into structured test sequence data, resulting in a standardized test process description, including: Based on existing test process documents, extract test step descriptions, operation objects, and parameter setting information to obtain a set of unstructured test information; Based on the unstructured test information set, the temporal relationships and dependencies between test steps are identified to obtain a test step relationship graph; Based on the test step relationship diagram, the test process is converted into a standardized data format containing operation sequence number, device object, parameter value, preconditions, and postconditions to obtain the standardized test process description.

6. The method according to claim 1, characterized in that, Based on the standardized test process description, the control logic rule base, and the preset security boundary system, logical conflict identification is performed to detect contradictions and redundant operations between test steps, resulting in a conflict detection result set, including: Based on the standardized test process description and the control logic rule base, the execution order of test steps is analyzed by reconnecting the submodulus maximization principle of the sequential greedy algorithm, identifying the combination of steps that violate the device start-stop logic, and obtaining a set of logical conflict candidates. Based on the logical conflict candidate set and the preset security boundary system, verify whether the test parameters exceed the security range, filter out security conflicts, and obtain the confirmed logical conflict set. Based on the confirmed set of logical conflicts, test steps that are functionally repetitive or have the same effect are identified, and a set of conflict detection results containing conflict type, scope of impact, and severity is obtained. Based on the conflict detection result set, the chain effect of the conflict step on the subsequent testing process is evaluated, the conflict propagation path and the depth of impact are calculated, and a conflict impact assessment report is obtained. Based on the conflict impact assessment report, the conflicts are classified and ranked according to their severity and scope of impact to identify the key conflict items that need to be addressed first, resulting in a priority list of conflicts.

7. The method according to claim 1, characterized in that, Based on the aforementioned conflict detection result set and test coverage requirements, multiple candidate correction schemes are generated, including: Based on the conflict detection result set and test coverage requirements, a multi-objective optimization problem is constructed using the fixed parameter approximation principle of the Multiwinner rule to balance test efficiency, security, and coverage, thereby obtaining the optimization objective function. Based on the optimization objective function, multiple different test process adjustment strategies are generated. Each strategy includes specific suggestions for deletion, modification, and rearrangement, resulting in a set of candidate correction schemes. Based on the set of candidate correction schemes, the comprehensive score of each scheme is calculated, and the scheme that performs best in terms of security, efficiency and coverage is selected to obtain the test process correction suggestions. Based on the set of candidate correction schemes, a set of non-dominated solutions is identified to ensure that multiple candidate schemes have good distribution and diversity in the target space, thereby obtaining a Pareto optimal solution set. Based on the Pareto optimal solution set, the impact of changes in key parameters on the scheme ranking is tested, the robustness and stability of the correction suggestions are verified, and a sensitivity analysis report is obtained.

8. The method according to claim 1, characterized in that, It also includes a step of automatically correcting the process based on the test process correction recommendations: Based on the test process correction suggestions and the standardized test process description, step deletion, parameter adjustment and sequence rearrangement operations are performed to obtain the modified test sequence. Based on the modified test sequence, check the logical consistency and security compliance of the process to ensure that all modifications comply with the control logic rule base and the preset security boundary system requirements, and obtain a verified test sequence. Based on the verified test sequence, a final test process document and executable instructions that meet the requirements of the power plant execution system are generated, resulting in a corrected and complete test process scheme. Based on the corrected complete test process scheme, the test execution process is simulated to verify the feasibility and security of the corrected process and obtain simulation verification results. Based on the simulation verification results, the efficiency improvement rate, security risk reduction rate, and coverage improvement rate of the corrected test process are calculated to obtain a process optimization effect evaluation report.

9. The method according to claim 6, characterized in that, The submodulus maximization principle of the reconnection sequential greedy algorithm includes: Based on the standardized test process description, the test process is modeled as a directed graph structure, where nodes represent test steps and edges represent execution order relationships, resulting in a directed graph of the test process. Based on the directed graph of the test process, the effectiveness of the test step combination is evaluated by the submodule utility function calculation algorithm. This function considers the logical consistency, resource utilization and security indicators between steps to obtain the submodule utility score. Based on the submodule utility score, a greedy strategy selection algorithm is used to select the connection recombination operation that maximizes the overall submodule utility in each iteration, thereby obtaining the locally optimal submodule structure.

10. An automatic calibration system for the testing process of a pumped storage power station, characterized in that, include: The equipment modeling module is used to construct a digital power plant equipment model, including hydro-generator units, pumps, and transformers, based on power plant configuration parameters and operating data, and to obtain equipment relationship topology diagrams and performance parameter sets. The rule base establishment module is used to analyze the control dependencies and operational constraints between the digital power station equipment such as hydro-generator sets, pumps, and transformers based on the equipment relationship topology diagram and performance parameter set, and to establish a control logic rule base. The safety boundary construction module is used to determine the boundary values ​​of the operating parameters of each device and system based on the control logic rule base and the safety operation standards, and to construct a preset safety boundary system. The process parsing module is used to perform digital parsing of existing test process documents, converting the test steps in the test process documents into structured test sequence data, and obtaining a standardized test process description. The conflict identification module is used to identify logical conflicts based on the standardized test process description, the control logic rule base and the preset security boundary system, detect contradictions and redundant operations between test steps, and obtain a conflict detection result set. The correction scheme generation module is used to generate multiple candidate correction schemes based on the conflict detection result set and test coverage requirements.

Images