Hydropower station equipment state monitoring method and system

Abstract

The invention discloses a hydropower station equipment state monitoring method and system, and relates to the technical field of hydropower station equipment monitoring, and the method comprises the steps: collecting the hydropower station equipment state through multi-layer data, and dynamically adjusting the collection frequency according to the change of the equipment state; performing signal quality evaluation on the collected data to generate fusion data; inputting the fused data into a trend analysis module, and generating a trend model of the equipment state in combination with historical data and real-time data; according to an analysis result of the trend model, combining the plurality of abnormal states into an event mode, and comparing the event mode with a known fault mode to determine a risk level; and generating a corresponding early warning signal based on the risk level, and performing corresponding response processing on early warning events of different levels. According to the hydropower station equipment state monitoring method provided by the invention, through multi-level data acquisition and dynamic frequency adjustment, the accuracy of data acquisition and the resource utilization rate are effectively improved, so that the system can adapt to different equipment states.

Description

Technical Field

[0001] The present invention relates to the technical field of hydropower station equipment monitoring, and in particular to a method and system for monitoring the status of hydropower station equipment. Background Art

[0002] In the prior art, the status monitoring system of hydropower station equipment generally relies on a single-level data collection method and a fixed-frequency monitoring mechanism. Usually, these monitoring systems use a fixed collection frequency to continuously collect and process the status data of the equipment. However, this fixed-frequency collection method will cause a waste of resources when the equipment status is stable, and when the equipment status fluctuates or is abnormal, it is impossible to flexibly adjust the collection frequency to capture subtle changes in time, resulting in difficulty in timely warning of potential risks of the equipment at critical moments.

[0003] At the same time, the monitoring system in the existing technology lacks a means of dynamically evaluating signal quality, which makes it susceptible to noise and unstable signals during data collection, resulting in low accuracy of fused data. Especially in complex environments such as hydropower stations, equipment operation data is often interfered by external factors (such as temperature, humidity, etc.). The existing monitoring system fails to fully consider the impact of these environmental factors, making it difficult to ensure the reliability and effectiveness of the data.

[0004] In addition, traditional monitoring methods mostly use single-dimensional time series analysis and lack multi-level trend analysis, especially under the interaction of factors such as equipment aging, load changes and environmental fluctuations, it is difficult to fully reflect the changing trend of the equipment's operating status. The limitations of this trend analysis make it difficult for the system to accurately predict potential failures or abnormal conditions, affecting the safety and stability of the equipment.

[0005] In terms of risk assessment, existing technologies are usually based on abnormal judgment of a single parameter, fail to effectively combine multiple abnormal states, and lack event pattern recognition and risk level classification based on multi-parameter combination. Therefore, when a fault occurs, the existing monitoring system lacks a detailed assessment of the risk level and a corresponding graded warning mechanism, and cannot take flexible response measures for different degrees of risk, which limits the system's fault response capabilities. Summary of the invention

[0006] In view of the above-mentioned problems, the present invention is proposed.

[0007] Therefore, the technical problems solved by the present invention are: first, the existing monitoring systems mostly adopt a fixed-frequency data acquisition method, which is difficult to adapt to the real-time changes in the equipment status, resulting in resource waste in a stable state and inability to capture subtle change information in an abnormal state; second, the traditional system lacks dynamic evaluation and fusion processing means for the signal quality of the collected data, resulting in noise and instability in the data, affecting the accuracy of monitoring; in addition, the traditional monitoring method lacks multi-level analysis combining short-term and long-term trends in trend analysis, making it difficult to accurately predict the operating trend and potential failure risks of the equipment; finally, the traditional monitoring system lacks an event pattern recognition method based on a multi-parameter combination, resulting in a single risk assessment method, making it difficult to effectively identify complex failure events and provide graded response processing.

[0008] In order to solve the above technical problems, the present invention provides the following technical solutions: A method for monitoring the status of hydropower station equipment, comprising:

[0009] The equipment status of the hydropower station is collected through multi-layer data, and the collection frequency is dynamically adjusted according to the changes in the equipment status;

[0010] Perform signal quality assessment on the collected data and generate fused data;

[0011] The fused data is input into the trend analysis module, and the trend model of the equipment status is generated by combining historical data and real-time data;

[0012] Based on the analysis results of the trend model, multiple abnormal states are combined into event patterns, and the risk level is determined by comparing with known failure patterns;

[0013] Generate corresponding warning signals based on risk levels, and respond to warning events of different levels accordingly.

[0014] As a preferred solution of the method for monitoring the status of hydropower station equipment described in the present invention, wherein: the multi-layer data acquisition includes acquiring temperature, vibration frequency, current fluctuation and pressure change through the first acquisition layer, and acquiring ambient temperature, humidity and atmospheric pressure through the second acquisition layer;

[0015] When the change rate of the operating parameters of the first collection layer exceeds the threshold, the collection frequency is increased to the preset frequency; when the change rate returns to normal, it is restored to the initial frequency.

[0016] As a preferred solution of the method for monitoring the status of equipment in a hydropower station according to the present invention, the generating of the fusion data includes receiving raw data of temperature, vibration frequency, current fluctuation, pressure change, ambient temperature, humidity and atmospheric pressure from the first acquisition layer and the second acquisition layer; preprocessing the raw data, removing abnormal points that deviate from the data fluctuation range threshold, and generating a first data set;

[0017] Perform noise level analysis on the data in the first data set, measure the fluctuation range of the data, mark the data with a noise level less than the noise threshold as a high noise quality signal, and mark the data with a noise level greater than or equal to the noise threshold as a low noise quality signal, and perform stability analysis;

[0018] Analyze the data fluctuation consistency of each sensor and mark the signal stability of each sensor. The signal with fluctuation less than the fluctuation threshold is marked as a high-stability signal, and the signal with fluctuation greater than or equal to the fluctuation threshold is marked as a low-stability signal.

[0019] Based on the dual marking of noise quality and stability analysis, the signal priority screening is performed. Sensor data with both high noise quality signals and high stability signals are marked as priority signals; other combinations are marked as secondary signals.

[0020] Assign weights to each signal type and generate a weight parameter table;

[0021] Based on the weight parameter table, the data of each sensor is weighted according to the weight to generate a second data set.

[0022] As a preferred solution of the method for monitoring the status of hydropower station equipment of the present invention, wherein: the trend model for generating the status of the equipment includes obtaining equipment operating parameters from the second data set, comparing each equipment operating parameter with historical data, calculating the real-time deviation value and change rate of each operating parameter, and obtaining preliminary deviation information;

[0023] Compare the preliminary deviation information with the current trend of environmental parameters;

[0024] Based on preliminary deviation information and environmental correlation information, analyze the short-term changes in equipment operating parameters and generate a short-term trend model;

[0025] After the short-term trend model is generated, the long-term trend analysis of the equipment operating parameters is performed to generate a long-term trend model;

[0026] Integrate the short-term trend model and the long-term trend model to form an equipment status trend model.

[0027] As a preferred solution of the hydropower station equipment status monitoring method described in the present invention, the equipment status trend model includes data on short-term abnormal points, short-term environmental abnormal points, long-term deviation points and long-term environmental deviation points.

[0028] As a preferred solution of the method for monitoring the status of hydropower station equipment according to the present invention, wherein: the risk level determination includes aggregating the abnormal point data according to the equipment operation parameter category based on the short-term abnormal point, short-term environmental abnormal point, long-term deviation point and long-term environmental deviation point in the equipment status trend model, and constructing a parameter abnormality combination mode;

[0029] Analyze multiple abnormal states existing in the same time period, identify abnormal correlations between operating parameters of different devices, and generate abnormal event patterns with specific combination characteristics;

[0030] The generated abnormal event pattern is matched and compared with the historical failure patterns in the known failure pattern database of the equipment, and the risk level of the event pattern is determined according to the pattern similarity and severity.

[0031] As a preferred solution of the method for monitoring the status of hydropower station equipment described in the present invention, the risk levels are divided according to the matching results of the event patterns in the known fault patterns, and are marked as the first level, the second level and the third level.

[0032] Another purpose of the present invention is to provide a hydropower station equipment status monitoring system. The existing hydropower station equipment status monitoring methods have problems such as fixed data acquisition frequency, difficulty in dynamically evaluating signal quality, and single risk assessment method, as well as how to improve monitoring accuracy and optimize the prediction of equipment status trends. It can solve the problems in the existing hydropower station equipment monitoring system that the data acquisition frequency cannot adapt to real-time changes, the data fusion accuracy is insufficient, and the trend analysis is insufficient to capture short-term and long-term changes by constructing a multi-level data acquisition and dynamic risk assessment system.

[0033] In order to solve the above technical problems, the present invention provides the following technical solutions: a hydropower station equipment status monitoring system, comprising: an output acquisition module, a data fusion module, a trend analysis module, a risk assessment module and a response processing module; the output acquisition module is used to collect the status of hydropower station equipment through multi-layer data, and dynamically adjust the acquisition frequency according to the change of equipment status; the data fusion module is used to perform signal quality assessment on the collected data and generate fused data; the trend analysis module is used to input the fused data into the trend analysis module, and generate a trend model of the equipment status by combining historical data and real-time data; the risk assessment module is used to combine multiple abnormal states into event patterns according to the analysis results of the trend model, and determine the risk level by comparing with known fault patterns; the response processing module is used to generate corresponding warning signals based on the risk level, and perform corresponding response processing on warning events of different levels.

[0034] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned method for monitoring the status of hydropower station equipment when executing the computer program.

[0035] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the method for monitoring the status of hydropower station equipment as described above.

[0036] Beneficial effects of the present invention: The hydropower station equipment status monitoring method provided by the present invention effectively improves the accuracy of data collection and resource utilization through multi-level data collection and dynamic frequency adjustment, so that the system can adapt to different equipment states; through signal quality evaluation and weighted fusion, the accuracy and stability of data are significantly improved, and the data quality of real-time monitoring is guaranteed; through the combined analysis of short-term and long-term trends, the operating status change trend of the equipment is fully captured, and early warning of potential equipment failures is achieved; through the construction of multi-parameter event patterns and hierarchical risk assessment, flexible response processing to different risk levels is achieved, so that the system can adopt different response strategies according to the actual risk situation, thereby improving the safety of hydropower station equipment and overall operation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0038] Figure 1 An overall flow chart of a method for monitoring the status of equipment in a hydropower station provided by an embodiment of the present invention.

[0039] Figure 2 An overall structural diagram of a hydropower station equipment status monitoring system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0041] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0042] Example 1

[0043] Reference Figure 1 , as an embodiment of the present invention, provides a method for monitoring the status of hydropower station equipment, comprising:

[0044] The equipment status of the hydropower station is collected through multi-layer data, and the collection frequency is dynamically adjusted according to the changes in the equipment status;

[0045] Perform signal quality assessment on the collected data and generate fused data;

[0046] The fused data is input into the trend analysis module, and the trend model of the equipment status is generated by combining historical data and real-time data;

[0047] Based on the analysis results of the trend model, multiple abnormal states are combined into event patterns, and the risk level is determined by comparing with known failure patterns;

[0048] Generate corresponding warning signals based on risk levels, and respond to warning events of different levels accordingly.

[0049] The multi-layer data collection includes collecting temperature, vibration frequency, current fluctuation and pressure change through a first collection layer, and collecting ambient temperature, humidity and atmospheric pressure through a second collection layer;

[0050] When the change rate of the operating parameters of the first collection layer exceeds the threshold, the collection frequency is increased to the preset frequency; when the change rate returns to normal, it is restored to the initial frequency.

[0051] Generating the fused data includes receiving raw data of temperature, vibration frequency, current fluctuation, pressure change, ambient temperature, humidity and atmospheric pressure from the first acquisition layer and the second acquisition layer; preprocessing the raw data, removing abnormal points that deviate from the data fluctuation range threshold, and generating a first data set;

[0052] Perform noise level analysis on the data in the first data set, measure the fluctuation range of the data, mark the data with a noise level less than the noise threshold as a high noise quality signal, and mark the data with a noise level greater than or equal to the noise threshold as a low noise quality signal, and perform stability analysis;

[0053] Analyze the data fluctuation consistency of each sensor and mark the signal stability of each sensor. The signal with fluctuation less than the fluctuation threshold is marked as a high-stability signal, and the signal with fluctuation greater than or equal to the fluctuation threshold is marked as a low-stability signal.

[0054] Based on the dual marking of noise quality and stability analysis, the signal priority screening is performed. Sensor data with both high noise quality signals and high stability signals are marked as priority signals; other combinations are marked as secondary signals.

[0055] Assign weights to each signal type. Priority signals are given higher weights, and secondary signals are given lower weights, to ensure that priority signals have a larger share in the final result when data is fused.

[0056] Based on the results of these weight allocations, a weight parameter table is generated. This table lists the weight values ​​of each signal source and will serve as a reference for subsequent weighted fusion.

[0057] Based on the weight parameter table, the data of each sensor is weighted according to the weight to generate a second data set.

[0058] Specifically, for each parameter’s data (e.g., temperature, vibration frequency, current fluctuation, etc.), its value is first multiplied by its respective weight to ensure that the priority signal data occupies a larger proportion in the fusion result.

[0059] For example, when merging temperature data, a larger weight is assigned to the temperature data of the priority signal, and a smaller weight is assigned to the temperature data of the secondary signal. The weighted temperature data are then added together to generate the final temperature fusion value.

[0060] Generate a fused dataset:

[0061] Following the above steps, the weighting and addition process is repeated for each parameter (such as vibration frequency, current fluctuation, etc.), and all weighted parameters are fused into a complete "fused data set".

[0062] The fusion data set contains the fusion results of all parameters. It is the comprehensive data generated after the weight allocation of priority signals and secondary signals, which is used for subsequent trend analysis and status monitoring.

[0063] Generating the trend model of the equipment status includes obtaining equipment operating parameters from the second data set, comparing each equipment operating parameter with historical data, calculating the real-time deviation value and change rate of each operating parameter, and obtaining preliminary deviation information;

[0064] The preliminary deviation information is compared with the current change trend of the environmental parameters. During the comparison, if the deviation of the equipment operating parameters is consistent with the change trend of the environmental parameters, the operating parameters are marked as being affected by the environment, and the correlation between the environmental factors is recorded for reference in subsequent trend analysis.

[0065] Based on preliminary deviation information and environmental correlation information, analyze the short-term changes in equipment operating parameters and generate a short-term trend model;

[0066] Specifically, the steps for generating a short-term trend model include:

[0067] A low threshold is set as the abnormal detection standard for short-term changes. The threshold setting refers to the short-term fluctuation range of the device under normal conditions and is used to identify change points that exceed the normal fluctuation range in a short period of time.

[0068] Compare the current equipment operating parameters with the low threshold. If the real-time data exceeds the set low threshold, the point is marked as a short-term abnormal point.

[0069] If the deviation of a short-term outlier point is affected by environmental correlation, it is further marked as a short-term environmental outlier point in order to distinguish the deviation affected by the environment from independent equipment changes.

[0070] All short-term anomalies and short-term environmental anomalies are recorded in the short-term trend model to form a complete record of short-term data changes, which is used to identify short-term equipment status fluctuations.

[0071] After the short-term trend model is generated, the long-term trend analysis of the equipment operating parameters is performed to generate a long-term trend model;

[0072] Specifically include:

[0073] A high threshold is set for long-term anomaly detection. This threshold refers to the long-term fluctuation range of the equipment under normal operating conditions and is used to identify significant deviations of equipment operating parameters over a long period of time.

[0074] The current equipment operating parameters are compared. If the parameters deviate from the set high threshold, the point is recorded as a long-term deviation point.

[0075] If the change trend of the long-term deviation point is consistent with the periodic change trend of the environmental parameters, this point will be further marked as a long-term environmental deviation point to distinguish the long-term impact of environmental factors on the equipment.

[0076] Integrate the short-term trend model and the long-term trend model to form an equipment status trend model.

[0077] The equipment status trend model includes data of short-term abnormal points, short-term environmental abnormal points, long-term deviation points and long-term environmental deviation points.

[0078] Determining the risk level includes aggregating the abnormal point data according to the equipment operation parameter category based on the short-term abnormal point, short-term environmental abnormal point, long-term deviation point and long-term environmental deviation point in the equipment status trend model to construct a parameter abnormality combination pattern;

[0079] Analyze multiple abnormal states existing in the same time period, identify abnormal correlations between operating parameters of different devices, and generate abnormal event patterns with specific combination characteristics;

[0080] The generated abnormal event pattern is matched and compared with the historical failure patterns in the known failure pattern database of the equipment, and the risk level of the event pattern is determined according to the pattern similarity and severity.

[0081] The risk levels are divided according to the matching results of the event pattern in the known failure mode, and are marked as the first level, the second level and the third level.

[0082] Outlier data aggregation and parameter combination model construction

[0083] Extract data of short-term anomalies, short-term environmental anomalies, long-term deviations, and long-term environmental deviations from the equipment status trend model and classify them according to the equipment operating parameter categories, including temperature, vibration frequency, current fluctuations, pressure changes, etc. For each type of equipment operating parameter, according to the time of occurrence and fluctuation amplitude of the anomalies, aggregate the anomalies of the same category into a single parameter combination pattern in chronological order. Specifically, select a certain time window (such as 10 seconds or 1 minute), count the number, frequency of occurrence, and fluctuation intensity of each type of anomalies within the window, generate a single parameter combination pattern, and form an anomaly point combination structure of multiple types of parameters.

[0084] Based on the parameter combination pattern, the abnormal point correlation between the operating parameters in the same time period is analyzed, and the abnormal event pattern is generated through the correlation conditions. The correlation conditions include time synchronization, correlation of abnormal point fluctuation intensity, and dependence on environmental parameters. If the number of abnormal points of multiple parameters in the same time window is relatively large, the fluctuation intensity is high, or they show synchronous changes with environmental parameters, these parameter combinations are marked as linkage anomalies. By analyzing the linkage anomalies in the parameter combination item by item, an abnormal event pattern with specific combination characteristics is generated. Each abnormal event pattern records the name of the relevant parameters, the number of abnormal points, the fluctuation amplitude, and the environmental correlation relationship to form a complete event description.

[0085] The generated abnormal event pattern is compared with the historical failure pattern in the equipment known failure pattern database, and a similarity score is generated through comparison analysis. In the specific comparison process, the similarity between the abnormal event pattern and the historical failure pattern is compared in the following aspects:

[0086] Fluctuation intensity: Analyze whether the fluctuation amplitude of each parameter in the event mode is consistent with the amplitude in the corresponding historical failure mode.

[0087] Number of abnormal points: Compare the abnormal event pattern with the distribution of the number of each type of abnormal points in the historical failure pattern to see if they are consistent.

[0088] Time synchronization: examine whether different parameters in the event pattern show synchronous change characteristics consistent with the historical failure pattern.

[0089] Environmental relevance: Analyze whether the environmental related parameters in the abnormal event pattern are consistent with the environmental related parameters of the historical failure pattern.

[0090] The above features are assigned comparison weights respectively, and the comprehensive similarity score is calculated according to the matching degree of different features. If the similarity score exceeds the preset threshold, it is considered that the abnormal event pattern matches a historical fault pattern, and the matching results are recorded and the risk level is further analyzed.

[0091] According to the similarity score of the matching comparison and the severity of the fault, the risk level of the event pattern is divided into three categories:

[0092] Level 1: The similarity score of the event pattern is lower than the set threshold, and contains a small number of short-term abnormal points, showing slight fluctuations, and does not reach the characteristics of serious failures. Such events are marked as low risk.

[0093] Level 2: The similarity score of the event pattern is close to or slightly higher than the set threshold, contains multiple short-term anomalies or short-term environmental anomalies, and some operating parameters show moderate deviations. Such event patterns are marked as medium risk.

[0094] Level 3: The similarity score of the event pattern is much higher than the set threshold, contains multiple long-term deviation points or long-term environmental deviation points, and shows strong correlation between multiple parameters. Such patterns are highly consistent with severe failure modes and are marked as high risk.

[0095] According to the results of risk level classification, each event pattern is marked as the first level, second level or third level. Each marked event pattern stores the corresponding parameter combination characteristics, similarity score and environmental association information to form a complete risk level database, so that the subsequent early warning response system can provide appropriate early warning and treatment measures according to the risk level.

[0096] When the event mode is at the first level, a low-priority event log is generated to record the type, frequency, time span and correlation data of the abnormal point with the environmental parameters. The adaptive acquisition frequency optimization process is started to dynamically adjust the acquisition frequency based on the characteristics of the first-level abnormal points. Specifically, the fluctuation rate and frequency characteristics of each abnormal point are analyzed, and a higher acquisition frequency is set for operating parameters with violent fluctuations (such as temperature and vibration frequency) to obtain more fine-grained data; the acquisition frequency is gradually reduced for parameters with stable changes (such as ambient temperature and humidity) to save system resources. If the first-level event is triggered multiple times within a set short time period (for example, 24 hours), the environmental association model reconstruction process is started to compare the impact of current and historical environmental parameters on the abnormal point, dynamically adjust the impact weight of environmental factors, and generate an updated environmental association model. The environmental association model is used to optimize the detection criteria of the first-level abnormal points to reduce the interference of environmental fluctuations and improve the accuracy and adaptability of monitoring.

[0097] When the event pattern is at the second level, a medium-priority risk event report is generated, and the report records in detail all the abnormal point data involved, the change characteristics of the short-term trend model and its mutual influence with the environmental parameters. The adjacent equipment status synchronization analysis process is started, and the core operating parameters of the second-level event pattern (such as current fluctuations and pressure changes) are interactively analyzed with the operating parameters of the adjacent equipment to generate a multi-device joint trend model. The joint trend model predicts the possible impact of adjacent equipment through correlation calculation and identifies potential linkage risks. If the joint trend model shows that the parameter trend of adjacent equipment has the risk of synchronous fluctuation, the pre-adjustment measures are triggered: actively intervene in the collection frequency and key operating parameters of adjacent equipment, increase its collection frequency to a higher monitoring mode, and adjust the equipment operation status to the buffer mode to avoid further expansion of chain anomalies. In addition, in the case of high-frequency fluctuations in environmental data, the system sets tracking monitoring within two subsequent cycles (for example, 1 hour per cycle) to accurately record the fluctuation frequency and amplitude of environmental influencing factors to improve sensitivity and adaptability to future environmental changes.

[0098] When the event mode is at the third level, a high-priority emergency event report is generated, which contains the historical trend of abnormal parameters, associated environmental data, matching data between event mode and historical fault mode, and status data of adjacent devices. The process of triggering linkage shutdown and regional safety protection includes the following steps: First, the high-risk parameters of the current equipment are analyzed in time series, the adjacent equipment that may be affected is predicted, and the risk diffusion model in the area is generated. The risk diffusion model analyzes the potential impact of the high-risk operation status of the current equipment on the overall system and generates a priority list of affected equipment. For high-priority equipment in the risk diffusion model, the backup operation mode is triggered, and the key operating parameters of the relevant equipment are transferred to real-time backup to ensure data protection during the high-risk event. The operation mode of high-priority equipment is gradually reduced to the safe mode, and the operation data of the equipment is periodically backed up to maintain the integrity and continuity of the data. During the expansion of the high-risk event, a remote feedback loop for fault data is established, and the fault event data stream is transmitted to the remote expert system in real time through dedicated line communication, so that external experts can intervene in real time, update the equipment status diagnosis, and formulate temporary optimization measures. If the risk level of other equipment in the area also gradually increases, the system will automatically send a comprehensive early warning report to the regional management center and recommend the initiation of emergency response procedures.

[0099] Example 2

[0100] Reference Figure 2 , as an embodiment of the present invention, provides a hydropower station equipment status monitoring system, comprising:

[0101] Output acquisition module 100, data fusion module 200, trend analysis module 300, risk assessment module 400 and response processing module 500;

[0102] The output acquisition module 100 is used to acquire the status of the hydropower station equipment through multi-layer data, and dynamically adjust the acquisition frequency according to the change of the equipment status;

[0103] The data fusion module 200 is used to perform signal quality evaluation on the collected data and generate fused data;

[0104] The trend analysis module 300 is used to input the fused data into the trend analysis module, and generate a trend model of the device status by combining the historical data and the real-time data;

[0105] The risk assessment module 400 is used to combine multiple abnormal states into event modes according to the analysis results of the trend model, and determine the risk level by comparing with known failure modes;

[0106] The response processing module 500 is used to generate corresponding warning signals based on risk levels, and perform corresponding response processing on warning events of different levels.

[0107] Example 3

[0108] An embodiment of the present invention is different from the first two embodiments in that:

[0109] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0110] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0111] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0112] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0113] Example 4

[0114] As an embodiment of the present invention, a method for monitoring the status of equipment in a hydropower station is provided. In order to verify the beneficial effects of the present invention, a scientific demonstration is carried out through simulation experiments.

[0115] In the traditional hydropower station equipment status monitoring method, the monitoring system adopts a fixed frequency data acquisition mechanism. The experiment is set to a fixed acquisition frequency, and data is collected every 5 minutes. The data includes temperature, vibration frequency, current fluctuation, pressure, ambient temperature, humidity and atmospheric pressure. The monitoring system directly collects all data and performs simple mean denoising processing, without distinguishing data quality and its fluctuation frequency. Each data under the fixed acquisition frequency is regarded as valid data. This method does not consider the interference of the external environment on the equipment, and the data processing is only based on the fixed threshold judgment. During the experiment, when the equipment vibrates or the pressure changes suddenly, the traditional method may not be able to adapt to this fluctuation in the acquisition frequency. The trend analysis of the traditional method relies on time series analysis, which only calculates the real-time change amplitude of each parameter, and cannot combine the characteristic trends of short-term and long-term changes, and it is easy to ignore the initial fault. If a fault occurs, it is compared with a single threshold of each parameter. Once the threshold is exceeded, an alarm is generated, but the combined abnormality of multiple parameters cannot be identified. In addition, the response mechanism under different risk levels is relatively simple. When the warning level is lower than the set value, no alarm is triggered, and when it is higher than the set value, an alarm is directly triggered and a warning is triggered. This process is suitable for small-scale anomalies or short-term emergencies in testing, but it may produce false positives or false negatives in complex conditions.

[0116] The method of the present invention is carried out in the same experimental environment, but uses multi-layer data acquisition, dynamic frequency adjustment, signal quality assessment and fusion. During the experiment, the equipment operation status is divided into normal state, slight fluctuation state and severe fluctuation state. The method of the present invention sets a dynamic adjustment mechanism, which increases the acquisition frequency to once every 2 minutes when there is slight fluctuation, and reduces the acquisition frequency to 10 minutes when the equipment state is stable, so as to avoid unnecessary data redundancy. In the signal quality assessment stage, the collected temperature, vibration frequency, current fluctuation and pressure data are subjected to noise filtering and stability analysis to generate fusion data. Unlike the traditional method, the method of the present invention performs short-term and long-term trend modeling based on environmental conditions, and uses the results of the trend model to identify possible fault modes. In addition, the present invention adopts event pattern recognition with a combination of multiple parameters to avoid false alarms due to fluctuations in a single parameter. Finally, hierarchical early warning processing is performed according to the risk level of the event mode, and emergency processing and external support are triggered first at a high risk level to ensure that the fault is responded to quickly, and only internal monitoring is triggered at a low risk level. The experimental results are shown in Table 1.

[0117] Table 1 Comparison of experimental results

[0118] project Traditional methods Method of the present invention Acquisition frequency adjustment 5 minutes (fixed) 2-10 minutes (dynamic) Data volume (8 hours) 960 items 640 items Percentage of valid data 78％ 94％ Noise Level 22％ 6％ Fault prediction accuracy 68％ 92％ False Positive Rate 27％ 7％ Response time 12 minutes 6 minutes

[0119] The present invention reduces the amount of collected data when the equipment is stable by dynamically adjusting the collection frequency, reduces the pressure on system storage, and quickly increases the collection frequency when the equipment status fluctuates, ensuring that more detailed real-time data is obtained under critical conditions. Compared with the fixed collection frequency of traditional methods, the present invention quickly collects changing data under fluctuating conditions, allowing the system to capture early signals in the early stage of abnormalities, providing key data support for trend analysis and risk assessment. This flexible adjustment brings about a dual improvement in resources and analysis efficiency.

[0120] The present invention significantly reduces the impact of low-quality data by screening signal quality before data processing, ensuring that subsequent data analysis is based on high-precision data. This improvement is particularly significant in environments with large noise and volatility, making the data more credible. In contrast, traditional methods often involve noisy data in analysis due to the lack of dynamic signal quality assessment, causing the system to deviate when identifying device status. The priority signal screening and fusion process of the present invention effectively reduces the impact of data noise, making trend analysis results more accurate and representative, and significantly improving the stability of data in subsequent processing.

[0121] Traditional methods mainly rely on the threshold judgment of a single parameter, ignoring the combined changes between different parameters of the equipment. The present invention, through the multi-dimensional integration of trend analysis, can identify the slight abnormal linkage of multiple equipment parameters and capture potential risks in the early stages of a failure. This multi-parameter combined analysis can more keenly detect abnormal changes in equipment status, especially when small fluctuations occur. The present invention can provide early warnings, greatly reducing the possibility of false alarms or missed alarms. Compared with the single threshold judgment of traditional methods, the multi-dimensional trend analysis of the present invention is more accurate, reduces unnecessary warnings, and improves the prediction reliability of the system.

[0122] Traditional methods use a single response mechanism when dealing with failures, resulting in the failure to mobilize resources in a timely manner in high-risk situations. The present invention uses a hierarchical response to ensure that high-risk events can quickly trigger an emergency response, while appropriate measures are taken in low-risk situations to avoid wasting resources. This hierarchical response mechanism significantly shortens the response time in high-risk events, allowing the system to respond quickly when faced with emergencies, truly achieving high efficiency and accuracy in fault handling. System resources are reasonably allocated under each level of risk, demonstrating the adaptability and flexibility of the monitoring system in different environmental changes.

[0123] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for monitoring the status of hydropower station equipment, characterized in that: include: Collect the status of hydropower station equipment through multi-layer data, and dynamically adjust the collection frequency according to the changes in equipment status; Perform signal quality assessment on the collected data and generate fused data; The fused data is input into the trend analysis module, and the trend model of the equipment status is generated by combining historical data and real-time data; Based on the analysis results of the trend model, multiple abnormal states are combined into event patterns, and the risk level is determined by comparing with known failure patterns; Generate corresponding warning signals based on risk levels, and respond to warning events of different levels accordingly.

2. The method for monitoring the status of a hydropower station equipment according to claim 1, characterized in that: The multi-layer data collection includes collecting temperature, vibration frequency, current fluctuation and pressure change through a first collection layer, and collecting ambient temperature, humidity and atmospheric pressure through a second collection layer; When the change rate of the operating parameters of the first collection layer exceeds the threshold, the collection frequency is increased to the preset frequency; when the change rate returns to normal, it is restored to the initial frequency.

3. The method for monitoring the status of a hydropower station equipment according to claim 2, characterized in that: Generating fused data includes receiving raw data of temperature, vibration frequency, current fluctuation, pressure change, ambient temperature, humidity and atmospheric pressure from the first acquisition layer and the second acquisition layer; Preprocessing the original data, removing abnormal points that deviate from the data fluctuation range threshold, and generating a first data set; Perform noise level analysis on the data in the first data set, measure the fluctuation range of the data, mark the data with a noise level less than the noise threshold as a high noise quality signal, and mark the data with a noise level greater than or equal to the noise threshold as a low noise quality signal, and perform stability analysis; Analyze the data fluctuation consistency of each sensor and mark the signal stability of each sensor. The signal with fluctuation less than the fluctuation threshold is marked as a high-stability signal, and the signal with fluctuation greater than or equal to the fluctuation threshold is marked as a low-stability signal. Based on the dual marking of noise quality and stability analysis, the signal priority screening is performed. Sensor data with both high noise quality signals and high stability signals are marked as priority signals; other combinations are marked as secondary signals. Assign weights to each signal type and generate a weight parameter table; Based on the weight parameter table, the data of each sensor is weighted according to the weight to generate a second data set.

4. The method for monitoring the status of a hydropower station equipment according to claim 3, characterized in that: Generating the trend model of the equipment status includes obtaining equipment operating parameters from the second data set, comparing each equipment operating parameter with historical data, calculating the real-time deviation value and change rate of each operating parameter, and obtaining preliminary deviation information; Compare the preliminary deviation information with the current trend of environmental parameters; Based on preliminary deviation information and environmental correlation information, analyze the short-term changes in equipment operating parameters and generate a short-term trend model; After the short-term trend model is generated, the long-term trend analysis of the equipment operating parameters is performed to generate a long-term trend model; Integrate the short-term trend model and the long-term trend model to form an equipment status trend model.

5. The method for monitoring the status of a hydropower station equipment according to claim 4, characterized in that: The equipment status trend model includes data of short-term abnormal points, short-term environmental abnormal points, long-term deviation points and long-term environmental deviation points.

6. The method for monitoring the status of a hydropower station equipment according to claim 5, characterized in that: Determining the risk level includes aggregating the abnormal point data according to the equipment operation parameter category based on the short-term abnormal point, short-term environmental abnormal point, long-term deviation point and long-term environmental deviation point in the equipment status trend model to construct a parameter abnormality combination pattern; Analyze multiple abnormal states existing in the same time period, identify abnormal correlations between operating parameters of different devices, and generate abnormal event patterns with specific combination characteristics; The generated abnormal event pattern is matched and compared with the historical failure patterns in the known failure pattern database of the equipment, and the risk level of the event pattern is determined according to the pattern similarity and severity.

7. The method for monitoring the status of a hydropower station equipment according to claim 6, characterized in that: The risk levels are divided according to the matching results of the event pattern in the known failure mode, and are marked as the first level, the second level and the third level.

8. A system using the method for monitoring the status of hydropower station equipment according to any one of claims 1 to 7, characterized in that: include: An output acquisition module (100), a data fusion module (200), a trend analysis module (300), a risk assessment module (400) and a response processing module (500); The output acquisition module (100) is used to acquire the status of hydropower station equipment through multi-layer data, and dynamically adjust the acquisition frequency according to changes in the equipment status; The data fusion module (200) is used to perform signal quality assessment on the collected data and generate fused data; The trend analysis module (300) is used to input the fused data into the trend analysis module, and generate a trend model of the equipment status by combining historical data and real-time data; The risk assessment module (400) is used to combine multiple abnormal states into an event mode according to the analysis results of the trend model, and determine the risk level by comparing with known failure modes; The response processing module (500) is used to generate a corresponding warning signal based on the risk level, and perform corresponding response processing on warning events of different levels.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for monitoring the status of hydropower station equipment according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for monitoring the status of hydropower station equipment according to any one of claims 1 to 7 are implemented.

Images