State cognition demand-based on-line monitoring credibility evaluation method for pumped storage power station

Abstract

The application relates to a pumped storage power station online monitoring credibility evaluation method based on state cognitive demand, which can be used in the technical field of electric power. The method comprises the following steps: obtaining a current operation condition of a target device in a pumped storage power station, and obtaining online monitoring data of the target device collected by an online monitoring system of the pumped storage power station; according to the current operation condition, a target state space of the target device is determined, and an observation mapping relationship between the target state space and the online monitoring data is established; according to the current operation condition, state cognitive demand information of the target device is determined; based on the state cognitive demand information and the observation mapping relationship, state observability information of the target device is determined; and according to the state observability information, a credibility evaluation result of the online monitoring system is determined. The method can improve the accuracy of credibility evaluation of the online monitoring system.

Description

Technical Field

[0001] This application relates to the field of power technology, and in particular to a reliability assessment method for online monitoring of pumped storage power stations based on state awareness requirements. Background Technology

[0002] With the rapid development of new power systems, pumped storage power stations, with their advantages of peak shaving and valley filling and flexible response, have become an important hub for ensuring the safe and stable operation of the power grid. To ensure the safe and stable operation of the core equipment within the power station, online monitoring systems are typically installed to monitor the equipment status in real time.

[0003] Currently, the reliability assessment of online monitoring systems is mainly based on regular manual inspections or simple comparison of historical data. However, when faced with complex and ever-changing actual operating scenarios, this conventional approach is difficult to accurately reflect the true situation of the online monitoring system itself, resulting in low accuracy in the reliability assessment of the online monitoring system. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for online monitoring of pumped storage power stations based on state cognition requirements, which can improve the accuracy of online monitoring system reliability assessment and improve the reliability assessment of online monitoring systems.

[0005] Firstly, this application provides a method for assessing the reliability of online monitoring of pumped storage power stations based on state awareness requirements. The method includes:

[0006] The current operating status of the target equipment in the pumped storage power station is obtained, and the online monitoring data of the target equipment collected by the online monitoring system of the pumped storage power station is also obtained.

[0007] Based on the current operating conditions, determine the target state space of the target device, and establish an observation mapping relationship between the target state space and the online monitoring data;

[0008] Based on the current operating conditions, determine the status awareness requirements of the target device;

[0009] Based on the state awareness requirement information and the observation mapping relationship, the state observability information of the target device is determined;

[0010] Based on the state observability information, the reliability assessment result of the online monitoring system is determined.

[0011] In one embodiment, determining the reliability assessment result of the online monitoring system based on the state observability information includes:

[0012] Based on the state observability information, the state perception credibility of the target device is determined;

[0013] Obtain the system operation support capability information of the online monitoring system, and determine the comprehensive credibility of the online monitoring system based on the state cognition credibility and the system operation support capability information;

[0014] The credibility assessment result is determined based on the overall credibility.

[0015] In one embodiment, determining the state awareness credibility of the target device based on the state observability information includes:

[0016] Obtain the status recognition accuracy of the target device and the data integrity information of the online monitoring data;

[0017] The state observability information, the state recognition accuracy, and the data integrity information are fused together to obtain the state cognition credibility.

[0018] In one embodiment, determining the overall reliability of the online monitoring system based on the state awareness reliability and the system operation support capability information includes:

[0019] Obtain the data quality characteristics and time delay characteristics of the online monitoring data;

[0020] Based on the data quality characteristic information and the time delay characteristic information, the basic reliability of the target device is determined;

[0021] Obtain the physical logic verification result and historical trend verification result of the target device under the current operating condition. Based on the basic confidence level, the physical logic verification result and the historical trend verification result, perform correction processing on the state cognition confidence level to obtain the target state cognition confidence level.

[0022] The overall credibility is determined based on the credibility of the target state perception and the information on the system's operational support capabilities.

[0023] In one embodiment, determining the state observability information of the target device based on the state awareness requirement information and the observation mapping relationship includes:

[0024] Based on the state cognition requirement information and the observation mapping relationship, the monitoring coverage, state signal sensitivity and time resolution information of the target device under the current operating conditions are determined;

[0025] The state observability information is determined based on the monitoring coverage, the state signal sensitivity, and the time resolution information.

[0026] In one embodiment, determining the target state space of the target device based on the current operating conditions and establishing an observation mapping relationship between the target state space and the online monitoring data includes:

[0027] Based on the current operating conditions, determine the state evolution relationship of the target equipment;

[0028] Based on the aforementioned state evolution relationship, the target state space is constructed;

[0029] Obtain the measured noise information of the target device under the current operating conditions;

[0030] The observation mapping relationship is constructed based on the target state space and the measurement noise information.

[0031] Secondly, this application also provides a reliability assessment device for online monitoring of pumped storage power stations based on state awareness requirements. The device includes:

[0032] The operating condition acquisition module is used to acquire the current operating condition of the target equipment in the pumped storage power station and to acquire the online monitoring data of the target equipment collected by the online monitoring system of the pumped storage power station.

[0033] The space determination module is used to determine the target state space of the target device based on the current operating conditions, and to establish an observation mapping relationship between the target state space and the online monitoring data;

[0034] The requirement determination module is used to determine the status awareness requirement information of the target device based on the current operating conditions.

[0035] The information determination module is used to determine the state observability information of the target device based on the state cognition requirement information and the observation mapping relationship;

[0036] The result determination module is used to determine the reliability assessment result of the online monitoring system based on the state observability information.

[0037] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described above.

[0038] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0039] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in any of the above aspects.

[0040] The aforementioned method, apparatus, computer equipment, computer-readable storage medium, and computer program product for online monitoring reliability assessment of pumped storage power stations based on state cognition requirements acquire the current operating conditions and online monitoring data of the target equipment in the pumped storage power station. Based on the current operating conditions, it determines the target state space, observation mapping relationship, and state cognition requirement information. Then, based on the state cognition requirement information and the observation mapping relationship, it determines the state observability information. Finally, based on the state observability information, it determines the reliability assessment result of the online monitoring system. This approach facilitates the combination of static equipment monitoring data with dynamic operating conditions and real cognitive requirements, thereby accurately quantifying the online monitoring system's true perception capability of key states and avoiding misjudgments caused by sensor anomalies or changes in operating conditions. Ultimately, this effectively improves the accuracy of the reliability assessment of the online monitoring system. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a flowchart illustrating a method for assessing the reliability of online monitoring of a pumped storage power station based on state awareness requirements in one embodiment.

[0043] Figure 2 Here is a flowchart illustrating the overall process of online monitoring system reliability assessment in one embodiment;

[0044] Figure 3 This is a schematic diagram illustrating the relationship between the actual physical state, the digital state model, and the measurement state in one embodiment.

[0045] Figure 4 This is a schematic diagram of device-level reliability assessment in one embodiment;

[0046] Figure 5 This is a schematic diagram of unit-level reliability assessment in one embodiment;

[0047] Figure 6 This is a schematic diagram of the operating space mapping of a pumped storage power station in one embodiment;

[0048] Figure 7 This is a schematic diagram of dynamic effective topology construction based on switch state and operating conditions in one embodiment;

[0049] Figure 8 This is a structural block diagram of an online monitoring reliability assessment device for a pumped storage power station based on state cognition requirements in one embodiment.

[0050] Figure 9 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0052] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0053] In one exemplary embodiment, such as Figure 1 As shown, a reliability assessment method for online monitoring of pumped storage power stations based on state awareness requirements is provided. This embodiment illustrates the application of this method to a terminal; it is understood that this method can also be applied to a server, or to a system including a terminal and a server, and is implemented through interaction between the terminal and the server. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, tablets, etc.; the server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. In this embodiment, the method includes the following steps:

[0054] Step S101: Obtain the current operating status of the target equipment in the pumped storage power station, and obtain the online monitoring data of the target equipment collected by the online monitoring system of the pumped storage power station.

[0055] Step S102: Based on the current operating conditions, determine the target state space of the target equipment and establish the observation mapping relationship between the target state space and the online monitoring data;

[0056] Step S103: Determine the status awareness requirements of the target equipment based on the current operating conditions;

[0057] Step S104: Based on the state cognition requirement information and the observation mapping relationship, determine the state observability information of the target device;

[0058] Step S105: Determine the reliability assessment result of the online monitoring system based on the status observability information.

[0059] The target equipment can be key equipment in pumped storage power stations that require condition monitoring, such as generator circuit breakers, main transformers, hydro-generator sets, cables, or GIS (Gas Insulated Switchgear).

[0060] The current operating condition can be the physical operating mode of the target equipment at the current moment, such as power generation, pumping, no-load standby, start-stop switching, or abnormal disturbance.

[0061] Among them, the online monitoring system can be a set of hardware and software deployed in a pumped storage power station to continuously collect equipment operating parameters.

[0062] Among them, online monitoring data can be multi-source sensor data collected by sensors in the online monitoring system, such as temperature monitoring data, vibration monitoring data, partial discharge monitoring data, current monitoring data, oil chromatography monitoring data, or mechanical property monitoring data.

[0063] The target state space can be a set of real physical state variables used to describe the target equipment under the current operating conditions, such as the mechanical state of a generator circuit breaker, the insulation state of a main transformer, or the vibration state of a hydro-generator set.

[0064] The observation mapping relationship can be a mathematical correlation model between online monitoring data and various state variables in the target state space, used to characterize the different observation response characteristics corresponding to the same state variable under different operating conditions.

[0065] Among them, the state perception requirement information can be a set of key state variables that need to be perceived based on the operational safety requirements, fault early warning requirements, and equipment health management requirements of pumped storage power stations.

[0066] Among them, state observability information can be used to quantitatively assess the ability of an online monitoring system to perceive key state variables, such as monitoring coverage, state signal sensitivity, and time resolution.

[0067] The credibility assessment results can be quantitative indicators used to characterize whether the online monitoring system can truly, timely and accurately reflect the operating status of the target equipment, such as comprehensive credibility values, credibility levels or monitoring capability gap analysis results.

[0068] Optionally, the terminal identifies the current operating mode of the target equipment in the pumped storage power station in real time as the current operating condition, and simultaneously receives real-time operating data transmitted by various sensors in the online monitoring system as online monitoring data; based on the current operating condition, it calls the corresponding state evolution model to generate the target state space of the target equipment, and constructs a correlation model between the online monitoring data and each state variable in the target state space as an observation mapping relationship; based on the safety constraint target and fault warning target under the current operating condition, it extracts the set of key state variables that need to be focused on as state cognition requirement information; combining the observation mapping relationship and the state cognition requirement information, it calculates the monitoring coverage and signal sensitivity of the target equipment under the current sensor configuration as state observability information (if the monitoring coverage is less than the preset coverage threshold, it is determined that there is a monitoring blind spot); it integrates the state observability information and the data integrity index of the online monitoring system to calculate the comprehensive credibility value, and outputs the credibility level corresponding to the comprehensive credibility value as the credibility assessment result of the online monitoring system.

[0069] The aforementioned method for assessing the reliability of online monitoring of pumped storage power stations based on state cognition requirements involves acquiring the current operating conditions and online monitoring data of the target equipment in the pumped storage power station. Based on the current operating conditions, the target state space, observation mapping relationship, and state cognition requirement information are determined. Then, state observability information is determined based on the state cognition requirement information and the observation mapping relationship. Finally, the reliability assessment result of the online monitoring system is determined based on the state observability information. This method facilitates the combination of static equipment monitoring data with dynamic operating conditions and real cognitive requirements, thereby accurately quantifying the online monitoring system's true perception capability of key states and avoiding misjudgments caused by sensor anomalies or changes in operating conditions. Ultimately, this effectively improves the accuracy of the reliability assessment of the online monitoring system.

[0070] In an exemplary embodiment, determining the reliability assessment result of the online monitoring system based on state observability information includes: determining the state perception reliability of the target device based on the state observability information; obtaining the system operation support capability information of the online monitoring system; determining the comprehensive reliability of the online monitoring system based on the state perception reliability and the system operation support capability information; and determining the reliability assessment result based on the comprehensive reliability.

[0071] Among them, the credibility of state perception can be a quantitative evaluation index used to characterize the online monitoring system's ability to truly perceive and correctly identify the key operating states of the target equipment.

[0072] Among them, the system operation support capability information can be a set of parameters that reflect the stable operation capability of the online monitoring system's own software and hardware as well as its real-time status response capability. For example, it may include system reliability indicators and monitoring timeliness indicators.

[0073] Among them, the overall credibility can be the final value obtained by integrating the device-level state awareness level and the system-level operational support capability, which is used to evaluate the overall credibility status of the online monitoring system.

[0074] Optionally, the terminal determines the credibility of the target device's state perception based on the state observability information; obtains the system operation support capability information of the online monitoring system, performs a weighted fusion calculation based on the state perception credibility and the system operation support capability information, and determines the comprehensive credibility of the online monitoring system; and determines the corresponding credibility level as the credibility assessment result based on the preset numerical range of the comprehensive credibility.

[0075] The technical solution provided in this embodiment is beneficial for comprehensively measuring the actual sensing capability and operating status of the monitoring system under the current working conditions, thereby improving the comprehensiveness and accuracy of the reliability assessment results.

[0076] In an exemplary embodiment, determining the state perception credibility of a target device based on state observability information includes: obtaining the state recognition accuracy rate corresponding to the target device and the data integrity information of online monitoring data; and fusing the state observability information, state recognition accuracy rate, and data integrity information to obtain the state perception credibility.

[0077] Among them, the state recognition accuracy rate can be the accuracy indicator of the diagnostic algorithm in the online monitoring system when performing fault identification or trend analysis on the true state of the target equipment.

[0078] Among them, data integrity information can be a quality assessment parameter used to characterize whether there are packet loss, missing or dead values ​​in the online monitoring data during transmission and storage.

[0079] The fusion processing can be a data processing process that uses preset state weight coefficients to perform weighted summation or multiplication of multiple evaluation indicators of different dimensions.

[0080] Optionally, the terminal obtains the state recognition accuracy of the target device and the data integrity information of the online monitoring data; extracts the pre-configured weight coefficients corresponding to each evaluation indicator, and performs weighted fusion processing on the state observability information, state recognition accuracy and data integrity information to obtain the state cognition credibility.

[0081] The technical solution provided in this embodiment is conducive to multi-dimensional quantitative monitoring of the system's cognitive level of key states, thereby helping to avoid the one-sidedness caused by single indicator evaluation and further improving the accuracy of the credibility of state cognition.

[0082] In an exemplary embodiment, the overall credibility of the online monitoring system is determined based on the credibility of state perception and the system's operational support capability information. This includes: acquiring data quality characteristic information and time delay characteristic information of the online monitoring data; determining the basic credibility of the target device based on the data quality characteristic information and time delay characteristic information; acquiring the physical logic verification results and historical trend verification results of the target device under the current operating conditions; and correcting the credibility of state perception based on the basic credibility, physical logic verification results, and historical trend verification results to obtain the target state perception credibility; and determining the overall credibility based on the target state perception credibility and the system's operational support capability information.

[0083] Among them, data quality characteristic information can be parameters used to directly audit the characteristics of messages returned by sensors, such as sampling integrity, signal rationality, or quality bit status in industrial protocols.

[0084] Among them, the time delay feature information can be a parameter used to assess whether the time taken for data to be transmitted from the underground plant acquisition terminal to the ground receiving terminal is greater than a preset time delay threshold.

[0085] Among them, the basic reliability can be determined by both data quality characteristics and time delay characteristics, and is used as an initial trust index to characterize the purity and freshness of the underlying data.

[0086] Among them, the physical logic verification result can be a residual operator obtained after performing logical self-consistency audit by utilizing the coupling relationship between multiple physical quantities across space, such as the differential balance verification result of the high and low voltage side currents of a transformer.

[0087] Among them, the historical trend verification result can be the evolution consistency score obtained by comparing the overlap between the current action feature curve and the historical standard fingerprint curve.

[0088] Among them, the credibility of target state cognition can be a more accurate value of state cognition credibility obtained after auditing the underlying data and correcting it through multi-dimensional verification.

[0089] Optionally, the terminal acquires data quality characteristic information and time delay characteristic information of online monitoring data; determines the basic reliability of the target device by multiplying the data quality characteristic information and time delay characteristic information; acquires the physical logic verification results and historical trend verification results of the target device under the current operating conditions; and dynamically weights and corrects the state perception reliability based on the basic reliability, physical logic verification results, and historical trend verification results using preset operating condition adaptive weights to obtain the target state perception reliability; and determines the comprehensive reliability based on the target state perception reliability and system operation support capability information.

[0090] The technical solution provided in this embodiment is beneficial for accurately identifying abnormal situations such as sensor drift, communication congestion, or forged data, thereby further improving the reliability and anti-interference capability of comprehensive credibility calculation.

[0091] In an exemplary embodiment, the state observability information of the target device is determined based on the state awareness requirement information and the observation mapping relationship, including: determining the monitoring coverage, state signal sensitivity, and time resolution capability information of the target device under the current operating conditions based on the state awareness requirement information and the observation mapping relationship; and determining the state observability information based on the monitoring coverage, state signal sensitivity, and time resolution capability information.

[0092] Among them, monitoring coverage can be a numerical value used to characterize whether the types and number of currently deployed sensors can fully cover the key state variables required in the state cognition needs information.

[0093] Among them, state signal sensitivity can be a parameter used to measure the drastic response of online monitoring data to changes in the corresponding state variables of the target device.

[0094] Among them, time resolution information can be a parameter used to compare whether the actual sampling frequency of the sensor meets the required sampling frequency of the industry standard.

[0095] Optionally, the terminal calculates the monitoring coverage, status signal sensitivity, and time resolution information of the target device under the current operating conditions based on the status cognition requirement information and the observation mapping relationship; and makes a joint judgment based on the monitoring coverage, status signal sensitivity, and time resolution information to determine the status observability information. If the calculation result indicates that it is unobservable, it is determined that there is a problem of insufficient monitoring capability in the corresponding status.

[0096] The technical solution provided in this embodiment is conducive to refining the evaluation dimensions of observability, accurately identifying whether there are blind spots or insufficient perception capabilities in the current monitoring configuration, thereby providing an accurate data evaluation basis for subsequent credibility calculation.

[0097] In an exemplary embodiment, the target state space of the target device is determined based on the current operating conditions, and an observation mapping relationship is established between the target state space and the online monitoring data. This includes: determining the state evolution relationship of the target device based on the current operating conditions; constructing the target state space based on the state evolution relationship; obtaining the measurement noise information of the target device under the current operating conditions; and constructing the observation mapping relationship based on the target state space and the measurement noise information.

[0098] Among them, the state evolution relationship can be a model used to describe the dynamic changes of key state variables of the target equipment over time or with changes in control input under preset operating conditions.

[0099] Among them, the measured noise information can be random interference or environmental background noise parameters introduced by the online monitoring system during the data acquisition process.

[0100] Optionally, the terminal invokes the state evolution relationship of the target device that matches the current operating condition according to the current operating condition; constructs the target state space based on the state evolution relationship and the operation control input; obtains the measurement noise information of the target device under the current operating condition; and establishes a mathematical correlation between the real state variables and sensor observations based on the target state space and the measurement noise information, thereby constructing an observation mapping relationship.

[0101] The technical solution provided in this embodiment is conducive to truly restoring the physical evolution of equipment under different operating modes, and provides a benchmark for accurately assessing the reliability of monitoring data.

[0102] The following example illustrates the reliability assessment method for online monitoring of pumped storage power stations based on state awareness requirements provided in this application. Specifically, it can be a reliability assessment method for online monitoring systems of pumped storage power stations oriented towards state awareness requirements. (See reference below.) Figure 2 , Figure 3 , Figure 4 , Figure 5 , Figure 6 and Figure 7 .

[0103] This application relates to the field of online monitoring and condition assessment technology for pumped storage power stations. Specifically, it relates to a method for assessing the reliability of an online monitoring system at the equipment, unit, and power station system levels by constructing a condition-related digital condition model, mapping the actual physical condition to the measured condition, and comprehensively considering condition observability, monitoring system reliability, condition identification accuracy, and information timeliness, in order to meet the needs of power station condition cognition.

[0104] Pumped storage power stations include various types of equipment such as turbine generators, main transformers, generator output circuit breakers (GCBs), static frequency converters (SFCs), speed control systems, excitation systems, high-voltage cables, and GIS (Gas Insulation System). When the unit switches between operating conditions such as power generation, pumping, no-load standby, start-stop transitions, and abnormal disturbances, the energy transfer relationships, switching topologies, control sequences, and key states that need to be understood of each piece of equipment will all change.

[0105] Existing online monitoring systems typically collect data on temperature, vibration, partial discharge, current, voltage, oil chromatography, mechanical travel, operating coil current, and protection events. However, the ability to collect data does not equate to a reliable reflection of the power plant's operating status. Individual measurement points can be affected by sensor drift, transient interference, communication delays, data gaps, time asynchrony, or insufficient measurement point configuration. For example, a short-duration current pulse on a current transformer does not alone prove that the corresponding switch has closed; only by combining operating commands, switch auxiliary contacts, coil current, upstream and downstream voltages, current conduction topology, and operating time window can it be determined whether the current signal is supported by other evidence.

[0106] Existing evaluation methods often focus on sensor health, communication availability, or individual data quality, rarely addressing the power plant's state awareness requirements to uniformly answer the following questions: What real physical states must be recognized under current operating conditions? Can existing measuring points observe these states? Is the data acquisition link stable? Are the measured states consistent with the physical, topological, and temporal context? Can state changes be reflected within the allowable timeframe? Therefore, a reliability assessment method for online monitoring systems oriented towards state awareness capabilities is needed.

[0107] This application aims to address the following issues in the online monitoring system of pumped storage power stations: lack of a unified state cognition model, inability to dynamically adjust the evaluation object according to the operating conditions, lack of contextual cross-verification of single-point monitoring data, difficulty in identifying blind spots in state monitoring, and difficulty in forming reliable conclusions at the unit and power station levels from equipment-level results.

[0108] The reliability of the online monitoring system referred to in this application refers to the system's ability to completely, stably, accurately, and promptly reflect the true physical state of the power plant and its changes under given operating conditions and status awareness requirements. Its evaluation result is jointly determined by four dimensions: status observability, monitoring system reliability, status identification accuracy, and information timeliness.

[0109] In this embodiment:

[0110] This method follows the sequence of "operating condition identification—digital state model construction—state cognition requirement generation—mapping between real state and measured state—four-dimensional reliability evaluation—three-level aggregation—defect localization". All scores are normalized to the interval [0,1], with larger values ​​indicating higher reliability.

[0111] S1: Identify power plant operating conditions:

[0112] Identify the current operating condition based on the monitoring system's operating mode, dispatch instructions, unit speed, power direction, GCB status, SFC status, and hydraulic machinery status. The set of operating conditions is defined as follows:

[0113] .

[0114] Power generation conditions;

[0115] Pumping operation;

[0116] No-load, shutdown, or standby operating conditions;

[0117] : Start-up, shutdown, synchronization, grid connection, disconnection, and other state transition conditions;

[0118] Disturbance conditions such as protection actions, sudden load changes, abnormal vibrations, and communication anomalies.

[0119] The operating condition identification should output the operating condition category, identification confidence level, and operating condition start time. When the operating condition identification confidence level is lower than a set threshold, multiple candidate operating conditions are retained, and a more stringent set of state recognition requirements is adopted in subsequent evaluations.

[0120] S2: Constructing a digital state model of the power plant based on operating conditions:

[0121] Real physical state Indicates the time of power plant and equipment The objectively existing operating state includes the equipment's open / closed state, rotational speed, vibration, temperature, insulation condition, conduction path, power transmission status, and control execution status. The true physical state is not a value directly given by a single measuring point, but rather an object that this method requires to be understood through an online monitoring system.

[0122] .

[0123] In response to the current working conditions Construct a digital state model:

[0124] .

[0125] : The set of real physical states that may be involved under the current operating conditions.

[0126] State evolution relationships and equipment mechanism relationships, including energy conversion, heat conduction, vibration response and control closed-loop relationships.

[0127] : A condition-related topology diagram formed by electrical connection points, equipment, and switch branches.

[0128] The timing constraints of state transitions, the allowable response time window, and various model parameters.

[0129] The state evolution relationship can be represented as:

[0130] .

[0131] Control inputs include scheduling instructions, control commands, and setpoints.

[0132] : head variation, ambient temperature, load disturbance, and unmodeled factors.

[0133] The role of a digital state model is to specify which real states need to be known under the current operating conditions, what physical and topological relationships should be satisfied between these states, and in what time sequence state changes should occur, rather than claiming that the absolute real state can be obtained directly from a single online monitoring quantity.

[0134] S3: Establish a mapping between the actual physical state and the measured state:

[0135] Measurement status It is the set of analog quantities, switch quantities, event quantities, and timestamps obtained by the online monitoring system through sensors, monitoring devices, control systems, protection devices, and communication links:

[0136] .

[0137] Establish a condition-related observation mapping based on the current operating conditions:

[0138] .

[0139] The condition-related mapping from the actual physical state to the measurement state includes the state-measuring point correspondence, direct / indirect observation relationship, range and sensitivity, topology switching relationship, and expected response time.

[0140] Measurement noise, transmission error, interference, and unmodeled error.

[0141] For a given real-world condition, multiple direct or indirect measurement evidences can be configured; for a single measurement, multiple conditions can be considered simultaneously. The mapping relationship varies with operating conditions and switch topology. For example, during SFC startup, the SFC output current, startup circuit switch status, speed, and frequency are key evidences; whether the high-voltage side CT of the main transformer is valid evidence depends on the current actual conduction path and cannot be pre-set as an unconditionally mandatory measurement point.

[0142] S4: Generate a set of state awareness requirements:

[0143] Based on current operating conditions, operational safety, fault warning, and equipment health management tasks, from... Extract the set of states that need to be understood at present. And specify the importance, allowable error, and maximum allowable delay for each state:

[0144] .

[0145] :state Importance weight, Furthermore, the weights are normalized within the same evaluation object.

[0146] State recognition allowable error or consistency tolerance.

[0147] The maximum permissible delay for the online monitoring system to effectively reflect state changes.

[0148] The minimum observability or minimum evidence coverage requirement for a state.

[0149] Table 1. Relationship between cognitive needs for working conditions and key states and typical relevant evidence.

[0150]

[0151] S5: Evaluate state observability:

[0152] State observability is used to evaluate whether the current monitoring configuration is structurally capable of sensing state awareness requirements. It does not directly use whether current data is missing, to avoid redundant calculations related to system reliability. (Regarding state...) and its associated measurement point set ,definition:

[0153] .

[0154] Operating conditions Lower measurement point State The mapping strength takes values ​​[0,1]; it is 0 when it is not in a valid topology or is not applicable.

[0155] : The structural detectability is determined by the range, sensitivity, sampling configuration and observation method, and takes a value of [0,1].

[0156] Evaluation object The overall observability is:

[0157] .

[0158] When critical state At that time, it is classified as a monitoring blind zone or a weakly observable state. If a critical state has no effective measuring points at all, then... Compensation cannot be made through high scores from other unrelated measurement points.

[0159] S6: Evaluate the reliability of the monitoring system:

[0160] Monitoring system reliability is used to evaluate whether sensors, monitoring devices, and communication links can continuously and stably provide valid data. This applies to monitoring channels. In the statistics window Internal definition:

[0161] .

[0162] Channel availability rate, which is the ratio of effective running time to statistical window time.

[0163] Data completeness rate, which is the ratio of the actual number of valid samples to the theoretical number of samples.

[0164] The data validity score is formed by protocol quality bits, range checks, self-test status, and communication verification.

[0165] , , : Non-negative weights, and .

[0166] state The reliability of associated channels is weighted according to the effective mapping strength; the equipment, unit, and power plant layers are then aggregated level by level according to the importance of the state. If all associated channels of a necessary state are unavailable, the reliability of that state is set to 0 and the reason for the interruption is output.

[0167] S7: Evaluate the accuracy of state recognition:

[0168] State recognition accuracy is used to evaluate whether the measured state matches the true state relationship specified by the digital state model. This applies to each physical, topological, temporal, or redundancy-related constraint. Calculate normalization consistency:

[0169] .

[0170] The residual between the measured relationship and the model's expected relationship.

[0171] Tolerances are determined by design parameters, equipment tolerances, health history data, or field tests.

[0172] clip(z,0,1): Restricts z to the range [0,1].

[0173] Consistency relationships can be categorized into the following four types:

[0174] Physical consistency Energy conservation, thermal response, frequency / phase synchronization, pressure-temperature relationship, and vibration condition envelope, etc.

[0175] Topological Consistency Based on the actual conduction path formed by the current switch combination, determine whether current, voltage, power, and position quantities should occur;

[0176] Timing consistency Determine whether the sequence of control commands, coil current, contact movement, position feedback, and current establishment is within the allowable time window;

[0177] Context consistency : Determine whether the target measurement is supported by redundant measurement points, upstream and downstream equipment, protection events, and control system information.

[0178] .

[0179] Among them each The weights are non-negative and sum to 1. For safety-critical states, an uncompensated severe conflict rule can be set: if there is no corresponding operating instruction, the effective topology is not conducting, and only isolated short-time current occurs, the accuracy of the target current measurement is directly limited to below the set upper limit; however, other evidence can still support the conclusion that the switch remains open.

[0180] S8: Evaluation of the timeliness of information:

[0181] Based on the original data collection timestamp Time of receipt by the evaluation platform The difference defines the end-to-end delay. Set a no-penalty threshold T0 and a failure threshold T1 for the status or measurement channel, where... :

[0182] .

[0183] T0 and T1 are determined by the equipment's operating speed and state recognition task. Smaller thresholds are used for start-stop transitions and protection-related states, while larger thresholds are used for slow variables such as steady-state temperature and oil chromatography. All embodiments use the same piecewise function for calculation, and different penalties are no longer given for the same delay.

[0184] S9: Form device layer reliability results.

[0185] For equipment Based on the set of applicable states under its current operating conditions, the following are obtained respectively: , , and The reliability of the equipment is determined by using a weighted geometric mean.

[0186] .

[0187] , , , The four-dimensional weights are all not less than 0 and .

[0188] Weighted geometric mean can prevent a serious weakness from being completely offset by other high scores. Normalized aggregation is performed only on states and indicators applicable to the current operating conditions; inapplicable indicators are not included in the denominator, and if key essential data is missing, the corresponding dimension is set to 0 and the reason for the missing data is output.

[0189] S10: Construct dynamic topology and establish unit-level credibility.

[0190] Using electrical connection points, main equipment, and monitored objects as nodes, and electrical connections, energy transfer, and control associations as edges, an effective topology is formed by activating or deactivating the corresponding edges based on switch positions and current operating conditions. The path set only includes paths that are actually likely to be activated or executed under the current operating conditions.

[0191] Consistency calculation for logical edge e in the valid topology To eliminate the inherent influence of path length and the number of logical edges on the score, a weighted geometric mean is used instead of direct multiplication:

[0192] .

[0193] : Importance weight of logical edges.

[0194] : To prevent small positive numbers from being undefined in terms of logarithms.

[0195] Unit Internal equipment basic reliability is ,and The reliability of the unit layer is defined as follows:

[0196] .

[0197] in . This assesses the overall observability of the unit's key states under current operating conditions. Exponential normalization ensures comparability between units with different numbers of equipment and different path lengths.

[0198] S11: Establishing the reliability of the power plant system layer.

[0199] A unified boundary equation is applied to the power plant energy chain. For any selected boundary, the following should be satisfied within the time window:

[0200] .

[0201] , : Boundary input and output power defined according to a unified power direction; power with signs is used for power generation and pumping operations.

[0202] Losses of main transformers, SFCs, cables, busbars, etc., located within the evaluation boundary, as well as plant service load.

[0203] : The rate of change of rotational inertial energy and other stored energy during startup, braking or disturbance; it can be approximated as 0 in steady state.

[0204] .

[0205] , The determination is based on measurement accuracy, rated capacity, and operating condition fluctuations. In standby conditions or when there is no effective energy transfer, power balancing is not enforced; instead, switch isolation, voltage status, and link availability constraints are employed.

[0206] The reliability of the power plant system layer is defined as follows:

[0207] .

[0208] in and Normalization is applied only to units currently in operation, standby, or undergoing state transition. Weights are reassigned when a unit exits the evaluation.

[0209] S12: Trustworthiness level, defect location and optimization output.

[0210] Table 2. Relationship between Levels, Rating Ranges, Meanings, and Recommendations

[0211]

[0212] In addition to outputting the total score and grade, the system should also output unobservable states, low reliability channels, collision types, timeout links, affected state awareness tasks, suspected abnormal devices, and optimization suggestions such as adding measurement points, time synchronization, calibration, link maintenance, or model parameter updates.

[0213] Accuracy model for identifying key equipment:

[0214] The following model is used to generate the physical and contextual consistency score in S7. All residuals are normalized to [0,1] using the clip function or exponential function, and all weights are non-negative and sum to 1.

[0215] 1. Hydroelectric generator motor.

[0216] The hydraulic input power under power generation conditions is The expected power is The expected input power amplitude under pumping conditions is . and Both refer to the efficiency under the corresponding operating conditions, and are no longer defined as the measured active power.

[0217] .

[0218] Vibration accuracy is calculated based on the deviation of the rotational frequency, blade frequency, and structural characteristic frequency within the corresponding operating condition envelope, and is jointly formed with the context of energy consistency, rotational speed, and guide vane opening. . Only traffic volume is represented; data quality is represented using a uniform symbol. .

[0219] 2. Main transformer.

[0220] The physical consistency of the main transformer includes the load-temperature rise relationship, three-phase current balance, cooling status, and oil chromatography trends. A temperature rise model is fitted using historical health data. and define:

[0221] .

[0222] .

[0223] in , and The three-phase current phasors contain phase angle information; if the system only provides current amplitude, then negative sequence current or three-phase amplitude imbalance rate is used instead, and phasors and formulas are not used directly.

[0224] If using the correlation coefficient, first normalize it to... The correlation coefficient, which may be negative, should not be used directly as the credibility score.

[0225] 3. GCB and GIS disconnect switch.

[0226] Switch status identification utilizes operating commands, coil current, mechanical travel, auxiliary contacts, upstream and downstream voltages, current, and gas pressure simultaneously. The command timing is set for each operation. Coil response time window, contact movement time window, and position feedback cutoff time .

[0227] exist Within the state transition time window, if the current build-up precedes the auxiliary contact flip, it can be identified as a transient timing difference and not considered a serious conflict; beyond that window... If the position still does not flip, the accuracy of position feedback and switch status recognition will be reduced; if there is no closing command, the effective topology is not conducting and there is only an isolated current pulse, the accuracy of the target current signal will be reduced.

[0228] .

[0229] Each sub-item is first normalized to [0,1], and there are no negative scores.

[0230] 4. Static Frequency Converter (SFC).

[0231] mechanical speed of the unit Converted to electrical frequency ,in For pole-logarithms. SFC accuracy simultaneously evaluates frequency, phase, and commutation logic:

[0232] .

[0233] .

[0234] .

[0235] in , It is obtained from the timing consistency between thyristor triggering, commutation pulse, output current zero crossing point and fault lockout state.

[0236] 5. Speed ​​control system and excitation system.

[0237] The speed control system and excitation system are modeled separately. The speed control system is evaluated based on guide vane opening command, measured opening, servo pressure, and unit power response.

[0238] .

[0239] in , To ensure consistency between the relay pressure and the action process, To ensure consistency between guide vane changes and unit power response, and .

[0240] The closed-loop relationship between excitation system evaluation terminal voltage, excitation current, reactive power, and setpoints is as follows:

[0241] .

[0242] in .

[0243] 6. High-voltage cables.

[0244] The evaluation of high-voltage cables includes load-temperature rise, continuity of the distributed temperature field, partial discharge propagation, and sheath circulation current balance. The consistency of partial discharge propagation is clearly defined as follows:

[0245] .

[0246] in and These represent the measured arrival time difference and amplitude attenuation of adjacent partial discharge sensors, respectively, with the capping amount being the expected value obtained according to the cable propagation model. The cable identification accuracy is the normalized weighted sum of the above sub-items, with a total weight of 1.

[0247] Missing measurements, outliers, and parameter calibration rules:

[0248] When there are no valid measurement points in the necessary state, the observability is set to 0 and the monitoring blind zone is output.

[0249] When non-critical optional metrics are missing, the system is removed from the current aggregation set and the remaining weights are renormalized, instead of directly penalizing the entire system with 0.

[0250] When key measurement points are missing, quality bits are invalid, or the link is interrupted, the corresponding reliability is set to 0.

[0251] If the denominator is 0 or the reference value is not applicable, the residual formula will not be executed, and a model inapplicability flag will be output.

[0252] Tolerances, time windows, operating condition envelopes, and weights can be determined by design parameters, equipment testing, health history data, and expert configuration, and versions should be recorded.

[0253] After the model parameters are updated, the old version and the effective time are retained to ensure that the evaluation results are traceable.

[0254] Working principle or usage method:

[0255] This application does not directly treat any single-point measurement value as the true state of the equipment. Instead, it first determines the true physical state that needs to be understood under the current operating conditions and its correlation through a digital state model, and then judges whether the online monitoring system can provide complete, stable, accurate and timely measurement evidence to support the understanding of these states.

[0256] Acquire scheduling instructions, unit operating modes, and key equipment status to identify the current operating conditions;

[0257] Call the digital state model and dynamic topology corresponding to the working condition to form a set of real states and state relationships;

[0258] A state awareness requirement matrix is ​​generated based on safety, early warning, and health management tasks;

[0259] Establish a mapping between the actual state and the measured state, and identify direct measurement points, indirect measurement points, and contextual evidence;

[0260] Calculate the observability of the state, the reliability of the monitoring system, the accuracy of state identification, and the timeliness of information respectively;

[0261] Establish equipment-level trustworthiness, and establish unit-level trustworthiness through effective topology and status coverage;

[0262] The reliability of the power plant system layer is formed by combining cross-unit energy chains and operating condition weights;

[0263] Outputs trust level, status blind spots, abnormal channels, logical conflicts, timeout links, and monitoring optimization suggestions.

[0264] Areas for improvement:

[0265] The reliability evaluation benchmark for the online monitoring system should be based on the state awareness requirements of pumped storage power stations rather than a fixed list of monitoring points.

[0266] Construct a working condition-related digital state model to clarify the mapping between the real physical state, state relationships, and measured state;

[0267] Unify state observability, monitoring system reliability, state identification accuracy, and information timeliness into the same evaluation framework;

[0268] Cross-validate single-point monitoring data using physical, topological, temporal, and multi-source contexts to avoid directly inferring device actions from isolated signals;

[0269] By using dynamic and effective topology, the measurement points and energy paths that should be evaluated are determined, thus avoiding taking CTs that are not on the actual conduction path as necessary evidence.

[0270] A normalized weighted geometric mean is used to evaluate the topology logic, eliminating the systematic bias of the number of logical edges and path length on the unit score;

[0271] It adopts a cross-space energy boundary equation that includes loss, plant load and energy storage variation terms, and is compatible with steady-state and dynamic operating conditions;

[0272] This results in a hierarchical structure of equipment, generating unit, and power plant system layers, and enables the reverse identification of specific status gaps.

[0273] Beneficial effects:

[0274] It can distinguish between "having data" and "being able to support state cognition," and identify monitoring systems that are fully configured but lack cognitive capabilities;

[0275] It can identify monitoring blind spots, link interruptions, latency exceeding limits, and physical and topology conflicts;

[0276] It can distinguish between permissible feedback timing differences and actual state conflicts under rapidly changing operating conditions such as start-stop transitions;

[0277] It can avoid misjudging isolated interference current as switching action, while retaining other evidence to determine the true switching state;

[0278] It can perform comparable evaluations of units with different numbers of equipment and topology scales;

[0279] It can provide a clear basis for adding measurement points, time calibration, sensor calibration, communication maintenance, and model optimization.

[0280] refer to Figure 2 , Figure 2 The overall flowchart for the online monitoring system reliability assessment includes inputting operating conditions and status awareness requirements, constructing a digital status model related to operating conditions, establishing a mapping between real status and measured status, status observability, monitoring system reliability, status identification accuracy, information timeliness, four-dimensional reliability fusion, aggregation of equipment layer, unit layer, and power plant layer, and outputting reliability level, monitoring blind spots, and defect location.

[0281] Figure 3 This diagram represents the relationship between the actual physical state, the digital state model, and the measured state, including the current operating condition S; the digital state model Ms, including state variables, state relationships, dynamic effective topology, and physical and temporal constraints; the actual physical state X(t), including equipment state, switching state, and energy state; the state-measurement mapping Hs + measurement noise v; and the measured state Y(t), including electrical quantities, mechanical quantities, temperature quantities, hydraulic quantities, and switching quantities.

[0282] Figure 4 This is a schematic diagram of device-level reliability assessment, including: device tree, basic device information, GCB output switch, reliability assessment results, reliability trend, contribution of influencing factors, data quality overview, model parameters, assessment instructions, and historical assessment records. Figure 5 This is a schematic diagram of unit-level reliability assessment, including: collaborative monitoring reliability assessment system, operating conditions, overall reliability, reliability trend, system topology, equipment reliability details, reliability assessment indicators, reliability level distribution, and alarm information. Figure 6 This is a schematic diagram of the operating space mapping of a pumped storage power station, including: component library, station topology diagram, model logic editing area, model parameters, basic information, rated parameters, structural parameters, and other parameters. Figure 7 This is a schematic diagram of dynamic and effective topology construction based on switch states and operating conditions, including: complete physical topology G phy Dynamically effective topology G S act 500kV busbar, high-voltage cable, high-voltage side disconnect switch, main transformer, generator motor, GCB, starting circuit disconnect switch, SFC, operating condition S, actual switch position K, screened according to operating condition rules and switch status, retain actual conducting branches, eliminate disconnected and irrelevant branches, determine associated measuring points, starting circuit disconnect switch, current operating condition does not participate in evaluation, valid measuring points: current, voltage, position, frequency, valid conducting path, invalid or exit path.

[0283] For example:

[0284] Example 1: Reliability assessment of a general online monitoring system.

[0285] This embodiment evaluates a GCB-related state cognition task of a pumped storage unit to illustrate the method for distinguishing between isolated current disturbances and actual switching actions.

[0286] 1. Requirements for understanding operating conditions and status.

[0287] The system has identified that it is currently in standby mode. The monitoring system does not have a GCB closing command; the GCB auxiliary contact indicates that the circuit is open; and the voltage difference between the upstream and downstream terminals remains within the expected opening range. The status recognition requirement is to determine the actual opening / closing status of the GCB and evaluate whether the CT current signal can serve as valid evidence of this status.

[0288] 2. Measurement phenomena.

[0289] A CT channel experiences a pulse with an amplitude of 8% of the rated current within 10 ms, which then returns to zero. No coil current, mechanical travel, auxiliary contact changes, upstream / downstream voltage changes, or protection events occur within the same time window. Dynamic topology indicates that the GCB branch is not conducting.

[0290] 3. CT channel reliability calculation.

[0291] This example uses four-dimensional weights. CT channel structure observability Availability and completeness are formed The delay was normal. Because there are no operational instructions, the branches are not conductive, and there is no other contextual support, the accuracy of identification is low. .

[0292] .

[0293] Therefore, the CT pulse was determined to be a low-confidence measurement, suspected to be transient interference or channel anomaly. This conclusion only reduces the confidence of evidence regarding the CT channel and its related status, and does not interpret the CT pulse as indicating that the GCB has activated.

[0294] 4. Conclusions regarding GCB state cognition.

[0295] The GCB tripping status is supported by evidence including the absence of a closing command, auxiliary contact tripping, upstream and downstream voltage difference, and the absence of a conducting topology. Therefore, the accuracy of identifying the actual GCB tripping status remains high. The system outputs "GCB remains tripped; CT shows an isolated low-reliability current pulse," instead of outputting contradictory switching states.

[0296] 5. Example of a state transition time window.

[0297] If the system has already A closing command is issued, coil current appears within 15 ms, continuous loop current is established within 65 ms, auxiliary contacts flip within 95 ms, and position feedback cutoff time is set. If the above signal is within the allowed timing range, it is not considered a serious conflict. If there is no position flip after 150 ms, the accuracy of position feedback and GCB status recognition will be reduced; if there is still no feedback after the failure time window, the output position feedback will be abnormal.

[0298] Example 2: Reliability evaluation of SFC startup of Unit 5 based on the main wiring diagram of Meixu B Plant.

[0299] This embodiment uses the example of Unit 5 starting from a static state and entering the pumping preparation process via SFC. The equipment numbers are based on the main wiring diagram of Meixu B Plant: generator motor is Equipment 1, GCB is Equipment 7, main transformer is Equipment 21, high-voltage side disconnect switch and CT bay is Equipment 22, 500 kV high-voltage cable is Equipment 24, and SFC is Equipment 36. Equipment 19 is a current transformer and is no longer used as the main transformer number.

[0300] 1. Operating condition identification and dynamic topology.

[0301] The system identifies the generator based on the start command, SFC activation status, unit speed, and power direction. Constructed based on the actual locations of the starting circuit disconnect switch, GCB, and related disconnectors. And search for an effective power supply path from SFC 36 device through the 18 kV start-up circuit to generator 5 #1 device.

[0302] Only current, voltage, and position measurement points located on the current valid path are considered necessary evidence. Whether the current transformer (CT) in bay 22 on the high-voltage side of main transformer equipment 21 should detect current is determined by the current switch combination and conduction path; if the starting current does not pass through this bay, then the absence of current in this CT is normal and should not be used as evidence of SFC startup failure.

[0303] 2. State cognition needs and mapping.

[0304] Table 3. Relationship between State, Weight, Main Measurement State, and Maximum Allowable Delay

[0305]

[0306] 3. Four-dimensional evaluation.

[0307] The measurement point configuration has direct or indirect observation capabilities for the first four states; the fifth state is not applicable in this startup topology and is excluded from the current requirement set. (Calculation based on mapping intensity) The statistical window includes channel availability, integrity, and quality bit formation. .

[0308] SFC frequency and press The converted rotor frequency deviation is 20% of the allowable value, and the phase deviation is 25% of the allowable value; the positions of all switches in the starting circuit are consistent with the path current; all actions are completed within the allowable time window, resulting in... We set the four-dimensional weights to 0.20, 0.25, 0.40, and 0.15.

[0309] .

[0310] Therefore, the reliability of the online monitoring during the startup process of Unit 5 is rated as Level A. If the CT in bay 22 has no current but the current valid path does not include the main transformer branch, no points will be deducted; if the topology shows that the bay should be within the path but the CT continues to have no current, a topology conflict will occur and the identification accuracy will be reduced.

[0311] 4. Example of unit and whole-site aggregation.

[0312] Assuming the foundation reliability of Unit 5 equipment is B5=0.92, and topology consistency is... The observability of the critical state is O5 = 0.93. :

[0313] .

[0314] There is also Unit 6 The cross-space energy chain scores for Units #5 and #6 are 0.96 and 0.94, respectively, with operating condition weights of 0.60 and 0.40. :

[0315] .

[0316] The power plant system was rated A. The system retains scores for each dimension and specific defects, and does not hide low-reliability test points due to a high overall score.

[0317] 5. Dynamic energy chain verification.

[0318] During the startup acceleration phase, time window integration or boundary equations incorporating energy storage variations are used to calculate the residuals between SFC input, SFC output, unit absorbed power, conversion losses, and rotor kinetic energy increase. After the unit enters steady-state pumping, the steady-state power and loss relationship is then used. This avoids the erroneous evaluation of the dynamic startup process using static power equality relationships.

[0319] Example 3: Applicable rules for five types of working conditions.

[0320] To ensure that this method covers all declared operating conditions, different demand sets, topologies, and time thresholds are used for each operating condition, but a unified four-dimensional evaluation framework is employed.

[0321] Table 4. Relationship between operating conditions, key evaluation relationships, and inapplicable or special treatments.

[0322]

[0323] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0324] Based on the same inventive concept, this application also provides a state-awareness-based online monitoring reliability assessment device for pumped storage power stations, used to implement the aforementioned state-awareness-based online monitoring reliability assessment method. The solution provided by this device is similar to the implementation described in the above method. Therefore, the specific limitations of one or more state-awareness-based online monitoring reliability assessment device embodiments provided below can be found in the limitations of the state-awareness-based online monitoring reliability assessment method described above, and will not be repeated here.

[0325] In one exemplary embodiment, such as Figure 8 As shown, a reliability assessment device for online monitoring of pumped storage power stations based on state awareness requirements is provided. This reliability assessment device 800 may include:

[0326] The operating condition acquisition module 801 is used to acquire the current operating condition of the target equipment in the pumped storage power station and to acquire the online monitoring data of the target equipment collected by the online monitoring system of the pumped storage power station.

[0327] The space determination module 802 is used to determine the target state space of the target equipment based on the current operating conditions, and to establish an observation mapping relationship between the target state space and the online monitoring data;

[0328] The requirement determination module 803 is used to determine the status awareness requirement information of the target equipment based on the current operating conditions.

[0329] The information determination module 804 is used to determine the state observability information of the target device based on the state cognition requirement information and the observation mapping relationship;

[0330] The result determination module 805 is used to determine the reliability assessment result of the online monitoring system based on the status observability information.

[0331] In an exemplary embodiment, the result determination module 805 is further configured to determine the state perception credibility of the target device based on the state observability information; obtain the system operation support capability information of the online monitoring system; determine the comprehensive credibility of the online monitoring system based on the state perception credibility and the system operation support capability information; and determine the credibility assessment result based on the comprehensive credibility.

[0332] In an exemplary embodiment, the result determination module 805 is further configured to obtain the state recognition accuracy rate corresponding to the target device and the data integrity information of the online monitoring data; and to perform fusion processing on the state observability information, state recognition accuracy rate and data integrity information to obtain the state cognition credibility.

[0333] In an exemplary embodiment, the result determination module 805 is further configured to acquire data quality characteristic information and time delay characteristic information of online monitoring data; determine the basic reliability of the target device based on the data quality characteristic information and time delay characteristic information; acquire the physical logic verification results and historical trend verification results of the target device under the current operating conditions; and, based on the basic reliability, physical logic verification results, and historical trend verification results, correct the state perception reliability to obtain the target state perception reliability; and determine the comprehensive reliability based on the target state perception reliability and the system operation support capability information.

[0334] In an exemplary embodiment, the information determination module 804 is further configured to determine the monitoring coverage, state signal sensitivity, and time resolution capability information of the target device under the current operating conditions based on the state cognition requirement information and the observation mapping relationship; and to determine the state observability information based on the monitoring coverage, state signal sensitivity, and time resolution capability information.

[0335] In an exemplary embodiment, the space determination module 802 is further configured to determine the state evolution relationship of the target device based on the current operating conditions; construct the target state space based on the state evolution relationship; obtain the measurement noise information of the target device under the current operating conditions; and construct the observation mapping relationship based on the target state space and the measurement noise information.

[0336] The modules in the aforementioned online monitoring reliability assessment device for pumped storage power stations based on state awareness requirements can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0337] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 9 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a reliability assessment method for online monitoring of pumped-storage power stations based on state-awareness requirements. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0338] Those skilled in the art will understand that Figure 9 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0339] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0340] In one exemplary embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above-described method embodiments.

[0341] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0342] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0343] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0344] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A reliability assessment method for online monitoring of pumped storage power stations based on state cognition requirements, characterized in that, The method includes: The current operating status of the target equipment in the pumped storage power station is obtained, and the online monitoring data of the target equipment collected by the online monitoring system of the pumped storage power station is also obtained. Based on the current operating conditions, determine the target state space of the target device, and establish an observation mapping relationship between the target state space and the online monitoring data; Based on the current operating conditions, determine the status awareness requirements of the target device; Based on the state awareness requirement information and the observation mapping relationship, the state observability information of the target device is determined; Based on the state observability information, the reliability assessment result of the online monitoring system is determined.

2. The method according to claim 1, characterized in that, The step of determining the reliability assessment result of the online monitoring system based on the state observability information includes: Based on the state observability information, the state perception credibility of the target device is determined; Obtain the system operation support capability information of the online monitoring system, and determine the comprehensive credibility of the online monitoring system based on the state cognition credibility and the system operation support capability information; The credibility assessment result is determined based on the overall credibility.

3. The method according to claim 2, characterized in that, The step of determining the state awareness credibility of the target device based on the state observability information includes: Obtain the status recognition accuracy of the target device and the data integrity information of the online monitoring data; The state observability information, the state recognition accuracy, and the data integrity information are fused together to obtain the state cognition credibility.

4. The method according to claim 2, characterized in that, The determination of the overall reliability of the online monitoring system based on the state awareness reliability and the system operation support capability information includes: Obtain the data quality characteristics and time delay characteristics of the online monitoring data; Based on the data quality characteristic information and the time delay characteristic information, the basic reliability of the target device is determined; Obtain the physical logic verification result and historical trend verification result of the target device under the current operating condition. Based on the basic confidence level, the physical logic verification result and the historical trend verification result, perform correction processing on the state cognition confidence level to obtain the target state cognition confidence level. The overall credibility is determined based on the credibility of the target state perception and the information on the system's operational support capabilities.

5. The method according to claim 1, characterized in that, The determination of the state observability information of the target device based on the state awareness requirement information and the observation mapping relationship includes: Based on the state cognition requirement information and the observation mapping relationship, the monitoring coverage, state signal sensitivity and time resolution information of the target device under the current operating conditions are determined; The state observability information is determined based on the monitoring coverage, the state signal sensitivity, and the time resolution information.

6. The method according to claim 1, characterized in that, The step of determining the target state space of the target device based on the current operating conditions and establishing an observation mapping relationship between the target state space and the online monitoring data includes: Based on the current operating conditions, determine the state evolution relationship of the target equipment; Based on the aforementioned state evolution relationship, the target state space is constructed; Obtain the measured noise information of the target device under the current operating conditions; The observation mapping relationship is constructed based on the target state space and the measurement noise information.

7. A reliability assessment device for online monitoring of pumped storage power stations based on state cognition requirements, characterized in that, The device includes: The operating condition acquisition module is used to acquire the current operating condition of the target equipment in the pumped storage power station and to acquire the online monitoring data of the target equipment collected by the online monitoring system of the pumped storage power station. The space determination module is used to determine the target state space of the target device based on the current operating conditions, and to establish an observation mapping relationship between the target state space and the online monitoring data; The requirement determination module is used to determine the status awareness requirement information of the target device based on the current operating conditions. The information determination module is used to determine the state observability information of the target device based on the state cognition requirement information and the observation mapping relationship; The result determination module is used to determine the reliability assessment result of the online monitoring system based on the state observability information.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images