Monitoring processing method and device

Abstract

The invention discloses a monitoring processing method and device, and mainly aims to solve the problem of multi-physical field coupling during operation of a water-turbine generator set. Establishing a spatial decoupling rule, arranging sensors according to physical field spatial distribution, and performing spatial decoupling by using a coherence function; establishing a working condition decoupling rule, analyzing stable working condition data through clustering, and decoupling a dominant physical field under a fluctuation working condition; and establishing a fault mechanism characteristic frequency decoupling rule, and performing fault diagnosis based on characteristic frequency deviation. According to the method, efficient decoupling and accurate fault diagnosis of complex coupling monitoring data are realized, and the sensitivity and accuracy of fault identification are remarkably improved.

Description

Technical Field

[0001] This application relates to the field of hydropower technology, and in particular to a monitoring and processing method and apparatus. Background Technology

[0002] As the core equipment of a hydropower station, the operation of a hydro-turbine generator unit involves the complex coupling of multiple physical fields, including flow field, stress field, electromagnetic field, temperature field, and acoustic field. The turbine section uses water flow as its power source to drive the generator rotor to rotate, thereby converting mechanical energy into electrical energy through the principle of electromagnetic induction. In this process, the interactions and mutual excitation between these physical fields form a highly nonlinear dynamic system, posing significant challenges to the stable operation and health management of the unit.

[0003] Given the critical role of hydro-generator units in the power system, real-time monitoring and fault diagnosis of their operating status are of paramount importance. An effective health monitoring system can detect potential faults early, prevent unplanned outages, ensure the safe and stable operation of the power grid, reduce maintenance costs, and extend equipment lifespan.

[0004] However, existing health monitoring technologies largely rely on the simple accumulation and qualitative analysis of multidimensional parameters such as vibration, sway, and temperature, lacking a deep understanding and decoupling capability of the multi-physics coupling mechanism. This results in low accuracy of monitoring and diagnosis under complex operating conditions, making it difficult to meet the high standards required for unit health monitoring in modern hydropower stations. Summary of the Invention

[0005] The purpose of this application is to provide a monitoring and processing method and apparatus that can improve the above-mentioned problems.

[0006] The embodiments of this application are implemented as follows: On the one hand, this application provides a monitoring and processing method, which includes at least one of steps S1 to S3, wherein S1, S2, etc. are only step identifiers, and the execution order of the method is not necessarily in ascending order of numbers. For example, step S2 can be executed first and then step S1 can be executed, or S1, S2, and S3 can be executed simultaneously. This application does not impose any restrictions.

[0007] S1: Real-time acquisition of monitoring data from various sensors on the power equipment of the hydropower station; spatial decoupling rules are used to decouple the physical field monitoring data corresponding to the location of the monitoring point from each monitoring data.

[0008] S2, periodically acquires monitoring data from various sensors on the hydropower station's power equipment and key parameters reflecting the current operating status of the equipment within a preset time period, and uses operating condition decoupling rules to perform cluster analysis on the monitoring data under stable operating conditions, and / or decouples the monitoring data of the dominant physical field under fluctuating operating conditions.

[0009] S3 periodically acquires monitoring data from various sensors on the hydropower station's power equipment within a preset time period, and uses fault mechanism characteristic frequency decoupling rules to achieve fault diagnosis of the hydropower station's power equipment.

[0010] It is understandable that step S1 can accurately separate monitoring data from different physical fields in the spatial dimension, reduce cross-physical field signal aliasing, and improve the authenticity and accuracy of the data. Step S2 performs cluster analysis on stable operating condition data, decouples the dominant physical field monitoring data under fluctuating operating conditions, establishes a benchmark model of multi-physical field coupling, and provides a reference standard for subsequent fluctuating operating condition analysis; decoupling the dominant physical field monitoring data under fluctuating operating conditions, locking the dominant physical field, and accurately identifying abnormal monitoring data caused by changes in the dominant physical field, solving the diagnostic ambiguity problem caused by multi-field coupling in transient processes in traditional methods. Step S3 obtains sensor monitoring data and performs Fourier transform to obtain characteristic frequencies based on the mapping relationship between the characteristic frequencies of the monitoring data and each physical field. The fault cause is determined by the characteristic frequency shift, which can directly map the physical field, realize the directional stripping of coupled signals, output quantitative fault diagnosis conclusions, and improve the sensitivity and accuracy of fault identification.

[0011] In optional embodiments of this application, step S1 specifically includes steps S11 to S15, where S11, S12, etc. are merely step identifiers. The execution order of the method does not necessarily follow the order of numbers from smallest to largest. For example, step S12 can be executed first and then step S11 can be executed. This application does not impose any restrictions.

[0012] S11, the monitoring data of the target monitoring point on the target component is used as the target monitoring data, and the monitoring data of multiple reference monitoring points on another component is used as the reference monitoring data. The target monitoring data and the reference monitoring data are associated with the same physical field.

[0013] S12, the target monitoring data is compared with each of the reference monitoring data using a coherence function to obtain multiple coherence values.

[0014] S13, among the multiple coherence values, select the reference monitoring points whose coherence values ​​are greater than the first threshold as candidate monitoring points.

[0015] S14, select the reference monitoring point corresponding to the maximum coherence value from the candidate monitoring points and confirm it as the decoupling reference point.

[0016] S15, using the frequency amplitude of the reference monitoring data at the decoupling reference point and the positional relationship between the decoupling reference point and the target monitoring point, decouple the physical field monitoring data mapped to the target monitoring point in the target monitoring data.

[0017] It is understandable that spatial decoupling rules are mainly based on coherence function analysis and positional relationship correction to achieve multi-physics data separation. First, target monitoring point data and reference data from reference monitoring points associated with the same physics field are acquired. The reference point with the strongest coherence is selected as the decoupling benchmark through coherence function calculation. Second, the bandwidth amplitude of the decoupling reference point is used, combined with its spatial distance from the target monitoring point (through weighting coefficients). Based on the overall height parameters of the unit, an amplitude correction model is established. Finally, the cross-field coupling components in the target monitoring data are removed through a correction formula, achieving accurate extraction of single physical field data. This rule effectively solves the problem of signal aliasing in the spatial domain and improves the interpretability of the physical properties of multi-physics monitoring data.

[0018] In an optional embodiment of this application, step S12 specifically includes: Perform a Fourier transform on the target monitoring data to obtain the target frequency domain data; For each of the aforementioned reference monitoring data, a Fourier transform is performed on the reference monitoring data of a reference monitoring point to obtain target reference frequency domain data; Based on the target frequency domain data and the target reference frequency domain data, calculate the auto-power spectral density of the target frequency domain data, the auto-power spectral density of the target reference frequency domain data, and the cross-power spectral density of the target frequency domain data and the target reference frequency domain data; Based on the auto-power spectral density of the target frequency domain data, the auto-power spectral density of the reference frequency domain data, and the cross-power spectral density of the target frequency domain data and the reference frequency domain data, the coherence function of the target frequency domain data and the reference frequency domain data is obtained; The first three peak frequencies are selected and substituted into the coherence function point by point for calculation, and the peak frequencies with coherence values ​​greater than the second threshold are retained as candidate frequencies. The maximum coherence value is selected from the coherence values ​​corresponding to the candidate frequency points and used as the coherence value of the target monitoring data and the target reference monitoring data.

[0019] In an optional embodiment of this application, step S15 specifically includes: The weighting coefficients are determined based on the positional relationship between the decoupling reference point and the target monitoring point. ; The generator set coefficient of the hydropower station's power equipment is selected based on the unit power capacity of the hydropower station's power equipment. ; Calculate the unit coefficient The passband amplitude of the decoupling reference point 1 and the weighting coefficient The product of the differences is taken as the first product. ; Calculate the passband amplitude of the target monitoring data. with the first product The difference This serves as the physical field monitoring data corresponding to the target monitoring point within the target monitoring data.

[0020] In an optional embodiment of this application, the step of determining the weighting coefficient based on the positional relationship between the decoupling reference point and the target monitoring point is... ,include: Obtain the overall height of the power equipment of the hydropower station. ; Obtain the distance between the target monitoring point and the centerline of the other component. ; The distance and the overall height ratio As a weighting coefficient .

[0021] In optional embodiments of this application, step S2 specifically includes steps S21 to S23, where S21, S22, etc. are merely step identifiers. The execution order of the method does not necessarily follow the order of numbers from smallest to largest. For example, step S22 can be executed first and then step S21 can be executed. This application does not impose any restrictions.

[0022] S21, the monitoring data and the key parameters acquired at the same time are used as a single sample data.

[0023] S22, Based on the key parameters, determine whether the power equipment of the hydropower station is in a stable or fluctuating operating condition.

[0024] S23, perform at least one of the following steps S231 and S232: S231, Select stable operating condition sample data, and use the key parameters in each stable operating condition sample data as a clustering feature vector. Classify each stable operating condition sample data into different operating condition clusters through cluster analysis. The coupling relationship of each physical field within the same operating condition cluster is consistent. Conduct in-depth analysis of each physical field dimension of the same operating condition cluster to provide optimization suggestions for the hydropower station's power equipment. S232, filter out the sample data of fluctuating operating conditions, identify the dominant physical field corresponding to the fluctuating operating condition, and calculate the rate of change of the physical field monitoring data affected by the dominant physical field in the sample data of the fluctuating operating condition. This allows for the analysis of the impact of the dominant physical field.

[0025] It is understandable that the operating condition decoupling rule achieves dynamic separation of multiple physical fields through phased processing. First, sensor data and key parameters (such as guide vane opening and rotational speed) are periodically collected, and the operating condition type is determined by the rate of parameter change. For stable operating conditions, key parameters are extracted as clustering feature vectors, and the FCM algorithm is used to classify the data into operating condition clusters, ensuring consistent physical field coupling relationships within each cluster, providing a basis for equipment optimization. For fluctuating operating conditions (such as start-up and variable load), the dominant physical field (such as flow field and stress field) is identified based on the operating condition type, and the rate of change of vibration / acoustic data under its influence is calculated to quantify the intensity of the dominant field's effect. This rule, through operating condition classification and decoupling from the dominant field, solves the diagnostic ambiguity problem caused by multi-field coupling interference during transient processes.

[0026] In an optional embodiment of this application, the power equipment of the hydropower station includes a turbine generator set; the key parameters include one or more of the following: guide vane opening, generator rotor speed, excitation current, and generator output power.

[0027] In an optional embodiment of this application, step S22 includes at least one of steps S221 to S226.

[0028] S221, in response to the zero time change rate of the key parameter, it is determined that the hydro-generator unit is in a stable operating condition.

[0029] S222, in response to the positive time change rate of the guide vane opening and the positive time change rate of the generator rotor speed, it is determined that the hydro-generator unit is in the start-up fluctuation condition.

[0030] S223, in response to the negative time change rate of the guide vane opening and the negative time change rate of the generator rotor speed, it is determined that the hydro-generator unit is in a shutdown fluctuation condition.

[0031] S224, in response to the fact that the time change rate of the generator rotor speed and the time change rate of the excitation current are zero, and the time change rate of the guide vane opening and the time change rate of the generator output power are non-zero, it is determined that the hydro-generator unit is in a variable load fluctuation condition.

[0032] S225, in response to the fact that the time change rate of the excitation current and the time change rate of the generator output power are zero, and the time change rate of the generator rotor speed and the time change rate of the guide vane opening are non-zero, it is determined that the hydro-generator unit is in a variable speed fluctuation condition.

[0033] S226, in response to the time change rate of the generator rotor speed, the time change rate of the guide vane opening and the generator output power being zero, and the time change rate of the excitation current being non-zero, it is determined that the hydro-generator unit is in a variable excitation fluctuation condition.

[0034] In an optional embodiment of this application, step S232 specifically includes at least one of the following steps S2321 to S2323.

[0035] S2321, Select sample data of variable load fluctuation under the variable load fluctuation condition, determine that the dominant physical field corresponding to the variable load fluctuation condition is the flow field, and calculate the rate of change of vibration monitoring data and the rate of change of acoustic text monitoring data in the sample data of the variable load fluctuation condition.

[0036] S2322, filter out the variable speed fluctuation condition sample data, determine that the dominant physical field corresponding to the variable speed fluctuation condition is the stress field, and calculate the rate of change of vibration monitoring data and the rate of change of acoustic text monitoring data in the variable speed fluctuation condition sample data.

[0037] S2323, filter out the sample data of the variable excitation fluctuation condition under the variable excitation fluctuation condition, clarify that the dominant physical field corresponding to the variable excitation fluctuation condition is the electromagnetic field, and calculate the rate of change of vibration monitoring data and the rate of change of acoustic text monitoring data in the sample data of the variable excitation fluctuation condition.

[0038] In optional embodiments of this application, step S3 specifically includes steps S31 to S32, where S31, S32, etc. are merely step identifiers. The execution order of the method does not necessarily follow the numerical order from smallest to largest. For example, step S32 can be executed first and then step S31 can be executed. This application does not impose any restrictions.

[0039] S31, Perform frequency domain transformation on the acquired monitoring data to obtain the characteristic frequency.

[0040] S32, in response to the mean and / or standard deviation of the characteristic frequency within the preset time period reaching a third threshold, it is determined that the characteristic frequency has shifted, and then the cause of the fault is determined based on the physical field and monitoring point corresponding to the characteristic frequency. The mapping relationship between each characteristic frequency and each physical field is clarified, and the fault diagnosis of the hydropower station's power equipment is achieved through the mean and / or standard deviation of the characteristic frequency within the preset time period.

[0041] It is understandable that the fault mechanism characteristic frequency decoupling rule achieves fault source tracing through frequency domain analysis and physical field mapping. First, sensor data is periodically acquired and subjected to Fourier transform to extract characteristic frequencies reflecting the equipment state. Second, a characteristic frequency library is constructed to clarify the mapping relationship between each frequency component (such as the hydroelectric generator frequency f1 and guide vane frequency f3) and physical fields such as flow field, stress field, and electromagnetic field. Finally, by statistically analyzing the mean and standard deviation of characteristic frequencies within a preset time period, when the offset exceeds a preset safety threshold, it is determined that an anomaly has occurred in the corresponding physical field. This rule utilizes the strong correlation between frequency characteristics and physical field faults to achieve directional decoupling of coupled signals and quantitative fault diagnosis, solving the problem of low accuracy in qualitative analysis using traditional methods.

[0042] On the other hand, this application discloses a monitoring and processing device, including at least one of a spatial decoupling module, an operating condition decoupling module, and a fault mechanism characteristic frequency decoupling module; The spatial decoupling module is configured to acquire monitoring data from various sensors on the hydropower station's power equipment in real time, and to decouple the physical field monitoring data corresponding to the monitoring point location from each monitoring data using spatial decoupling rules. The operating condition decoupling module is configured to periodically acquire monitoring data from various sensors on the hydropower station's power equipment and key parameters reflecting the current operating status of the equipment within a preset time period. It adopts operating condition decoupling rules to perform cluster analysis on the monitoring data under stable operating conditions and / or decouple the monitoring data of the dominant physical field under fluctuating operating conditions. The fault mechanism characteristic frequency decoupling module is configured to periodically acquire monitoring data from various sensors on the hydropower station's power equipment within a preset time period, and to perform fault diagnosis on the hydropower station's power equipment by adopting fault mechanism characteristic frequency decoupling rules.

[0043] In an optional embodiment of this application, the spatial decoupling rule execution module includes a first data acquisition submodule, an association submodule, a coherence value calculation submodule, a candidate monitoring point screening submodule, a decoupling reference point screening submodule, and a decoupling submodule; the first data acquisition submodule is configured to acquire monitoring data from various sensors on the hydropower station's power equipment in real time; the association submodule is configured to use the monitoring data from a target monitoring point on a target component as target monitoring data, and use the monitoring data from multiple reference monitoring points on another component as reference monitoring data, wherein the target monitoring data and the reference monitoring data are associated with the same physical field; the coherence value calculation submodule is configured to use the monitoring data from the target monitoring point on the ... The measured data are respectively compared with each of the reference monitoring data to calculate a coherence function, resulting in multiple coherence values; the candidate monitoring point screening submodule is configured to select the reference monitoring points whose coherence values ​​are greater than a first threshold from the multiple coherence values ​​as candidate monitoring points; the decoupling reference point screening submodule is configured to select the reference monitoring point corresponding to the largest coherence value from the candidate monitoring points and confirm it as the decoupling reference point; the decoupling submodule is configured to decouple the physical field monitoring data mapped to the target monitoring point in the target monitoring data by using the passband amplitude of the reference monitoring data at the decoupling reference point and the positional relationship between the decoupling reference point and the target monitoring point.

[0044] In an optional embodiment of this application, the coherence value calculation submodule includes a target frequency domain data calculation unit, a target reference frequency domain data calculation unit, a power spectral density calculation unit, a coherence function calculation unit, a candidate frequency point calculation unit, and a coherence value calculation unit. The target frequency domain data calculation unit is configured to perform a Fourier transform on the target monitoring data to obtain target frequency domain data. The target reference frequency domain data calculation unit is configured to perform a Fourier transform on the reference monitoring data of a reference monitoring point for each reference monitoring data point to obtain target reference frequency domain data. The power spectral density calculation unit is configured to calculate the auto-power spectral density of the target frequency domain data and the target reference frequency domain data based on the target frequency domain data and the target reference frequency domain data. The coherence function calculation unit is configured to obtain the coherence function of the target frequency domain data and the reference frequency domain data based on the auto-power spectral density of the target frequency domain data, the auto-power spectral density of the reference frequency domain data, and the cross-power spectral density of the target frequency domain data and the reference frequency domain data; the candidate frequency point calculation unit is configured to select the first three peak frequencies and substitute them point by point into the coherence function for calculation, and retain the peak frequencies with coherence values ​​greater than a second threshold as candidate frequency points; the coherence value calculation unit is configured to select the maximum coherence value from the coherence values ​​corresponding to the candidate frequency points as the coherence value of the target monitoring data and the target reference monitoring data.

[0045] In an optional embodiment of this application, the decoupling submodule includes a weight coefficient determination unit, a unit coefficient determination unit, a product calculation unit, and a decoupling calculation unit; the weight coefficient determination unit is configured to determine the weight coefficient based on the positional relationship between the decoupling reference point and the target monitoring point. The unit coefficient determination unit is configured to select the unit coefficient of the hydropower station's power equipment based on the unit power size of the hydropower station's power equipment. The product calculation unit is configured to calculate the unit coefficient. The passband amplitude of the decoupling reference point 1 and the weighting coefficient The product of the differences is taken as the first product. The decoupling calculation unit is configured to calculate the passband amplitude of the target monitoring data. with the first product The difference This serves as the physical field monitoring data corresponding to the target monitoring point within the target monitoring data.

[0046] In an optional embodiment of this application, the weighting coefficient determination unit includes an overall height acquisition unit, a spacing acquisition unit, and a coefficient calculation unit; the overall height acquisition unit is configured to acquire the overall height of the hydropower station's power equipment. The spacing acquisition unit is configured to acquire the distance between the target monitoring point and the centerline of the other component. The coefficient calculation unit is configured to calculate the distance. and the overall height ratio As a weighting coefficient .

[0047] In an optional embodiment of this application, the operating condition decoupling module includes at least one of a second data acquisition module, a sample confirmation submodule, an operating condition judgment module, and a stable operating condition decoupling module and a fluctuating operating condition decoupling module; the second data acquisition module is configured to periodically acquire monitoring data from various sensors on the hydropower station's power equipment and key parameters reflecting the current operating status of the equipment within a preset time period; the sample confirmation submodule is configured to use the monitoring data and key parameters acquired simultaneously as a single sample data; the operating condition judgment module is configured to judge the hydropower station's power equipment based on the key parameters. The system is either in a stable operating condition or a fluctuating operating condition. The stable operating condition decoupling module is configured to filter out stable operating condition sample data, use the key parameters in each stable operating condition sample data as a clustering feature vector, and classify each stable operating condition sample data into different operating condition clusters through cluster analysis. The coupling relationship of each physical field within the same operating condition cluster is consistent. The fluctuating operating condition decoupling module is configured to filter out fluctuating operating condition sample data, identify the dominant physical field corresponding to the fluctuating operating condition, and calculate the rate of change of the physical field monitoring data affected by the dominant physical field in the fluctuating operating condition sample data.

[0048] In an optional embodiment of this application, the power equipment of the hydropower station includes a turbine generator set; the key parameters include one or more of the following: guide vane opening, generator rotor speed, excitation current, and generator output power.

[0049] In an optional embodiment of this application, the operating condition judgment module includes a stable operating condition judgment module and a fluctuating operating condition judgment module; the stable operating condition judgment module is configured to determine that the hydro-generator unit is in a stable operating condition in response to the time change rate of the key parameter being zero; the fluctuating operating condition judgment module is configured to perform at least one of the following: in response to the time change rate of the guide vane opening being positive and the time change rate of the generator rotor speed being positive, determine that the hydro-generator unit is in a start-up fluctuating operating condition; in response to the time change rate of the guide vane opening being negative and the time change rate of the generator rotor speed being negative, determine that the hydro-generator unit is in a shutdown fluctuating operating condition; in response to the time change rate of the generator rotor speed... If the time change rate of the excitation current is zero, and the time change rates of the guide vane opening and the generator output power are non-zero, it is determined that the hydro-generator unit is under variable load fluctuation condition; if the time change rates of the excitation current and the generator output power are zero, and the time change rates of the generator rotor speed and the guide vane opening are non-zero, it is determined that the hydro-generator unit is under variable speed fluctuation condition; if the time change rates of the generator rotor speed, the guide vane opening, and the generator output power are zero, and the time change rate of the excitation current is non-zero, it is determined that the hydro-generator unit is under variable excitation fluctuation condition.

[0050] In optional embodiments of this application, the fluctuating operating condition decoupling module is configured to perform at least one of the following: filtering sample data of fluctuating operating conditions under variable load conditions, identifying the dominant physical field corresponding to the fluctuating operating condition as a flow field, and calculating the rate of change of vibration monitoring data and acoustic signature monitoring data in the sample data of fluctuating operating conditions; filtering sample data of fluctuating operating conditions under variable speed conditions, identifying the dominant physical field corresponding to the fluctuating operating condition as a stress field, and calculating the rate of change of vibration monitoring data and acoustic signature monitoring data in the sample data of fluctuating operating conditions; filtering sample data of fluctuating operating conditions under variable excitation conditions, identifying the dominant physical field corresponding to the fluctuating operating condition as an electromagnetic field, and calculating the rate of change of vibration monitoring data and acoustic signature monitoring data in the sample data of fluctuating operating conditions.

[0051] In an optional embodiment of this application, the fault mechanism characteristic frequency decoupling module includes a third data acquisition module, a characteristic frequency calculation submodule, and a fault judgment submodule. The third data acquisition module is configured to periodically acquire monitoring data from various sensors on the hydropower station's power equipment within a preset time period. The characteristic frequency calculation submodule is configured to perform frequency domain transformation on the acquired monitoring data to obtain the characteristic frequency. The fault judgment submodule is configured to determine that the characteristic frequency has shifted in response to the mean and / or standard deviation of the characteristic frequency within the preset time period reaching a third threshold, and then determine the cause of the fault based on the physical field and monitoring point corresponding to the characteristic frequency.

[0052] On the other hand, a computer device is provided, the device including a processor and a memory, the memory storing at least one piece of program code, the at least one piece of program code being loaded and executed by the processor to implement the above-described monitoring and processing method.

[0053] On the other hand, a computer-readable storage medium is provided, wherein at least one piece of program code is stored in the storage medium, the at least one piece of program code being loaded and executed by a processor to implement the above-described monitoring and processing method.

[0054] On the other hand, a computer program product or computer program is provided, which includes computer program code stored in a computer-readable storage medium. A processor of a computer device reads the computer program code from the computer-readable storage medium and executes the computer program code, causing the computer device to perform the monitoring and processing method described above.

[0055] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, optional embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0056] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 This is a structural schematic diagram of a hydro-generator set provided in this application.

[0058] Figure 2 This is a physical field-space domain mapping diagram of a hydro-generator unit provided in this application.

[0059] Figure 3This is a schematic diagram of the structure of a monitoring and processing device provided in this application.

[0060] Figure 4 This is a schematic diagram of the structure of a computer device provided in this application. Detailed Implementation

[0061] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0062] As the core equipment of a hydropower station, the turbine generator unit, such as Figure 1 As shown, the hydro-generator unit consists of a turbine, a generator, and connecting components. The turbine section includes a spiral casing 101, runner blades 102, guide vanes 103, a tailrace pipe 104, and a water-guided bearing 105. The spiral casing 101 guides water flow into the runner blades 102 to drive their rotation. The guide vanes 103 regulate the flow rate, the tailrace pipe 104 recovers energy, and the water-guided bearing 105 supports the main shaft. The generator section includes a stator 201, a rotor 202, an upper guide bearing 203, a lower guide bearing 204, and a thrust bearing 205. The stator 201 generates a magnetic field, the rotor 202 generates electricity, and the bearings collectively support the main shaft 301. The main shaft 301 connects the turbine and the generator via a flange 302, transmitting mechanical energy.

[0063] Current health monitoring of hydro-generator units faces the challenge of multi-physics coupling. The operation of a hydro-generator unit is a dynamic process involving deep coupling of flow, electromagnetic, stress, temperature, and acoustic fields. Nonlinear interactions exist between these different physical fields, and anomalies in a single field can propagate to other fields through coupling, causing a chain reaction. The measured data from unit condition monitoring contain coupling components from multiple sources, and changes in operating conditions and health status lead to the dynamic evolution of the multi-physics coupling state.

[0064] Traditional single-parameter monitoring cannot cover the complex relationships of multi-field coupling, making it difficult to locate the root cause of faults. Traditional monitoring technologies rely on the accumulation of multi-dimensional parameters and qualitative analysis, lacking a deep understanding of the coupling mechanism of multi-physics fields and the ability to decouple them. This results in poor real-time performance and low accuracy of monitoring and diagnosis under complex operating conditions, with a high rate of missed diagnoses and false alarms, making it difficult to meet the high standards of modern hydropower stations for unit health monitoring.

[0065] The purpose of this application is to provide a monitoring and processing method and apparatus that can improve the above-mentioned problems.

[0066] On the one hand, this application provides a monitoring and processing method, which includes at least one of steps S1 to S3, wherein S1, S2, etc. are only step identifiers, and the execution order of the method is not necessarily in ascending order of numbers. For example, step S2 can be executed first and then step S1 can be executed, or S1, S2, and S3 can be executed simultaneously. This application does not impose any restrictions.

[0067] S1 acquires real-time monitoring data from various sensors on the hydropower station's power equipment and uses spatial decoupling rules to decouple the physical field monitoring data corresponding to the location of each monitoring point.

[0068] Step S1, establishing spatial decoupling rules, aims to address the aliasing problem in monitoring data signals caused by multi-physical field coupling in hydro-generator units. Specifically, during the operation of hydro-generator units, multiple physical fields, such as flow field, stress field, electromagnetic field, temperature field, and acoustic field, intertwine, resulting in coupled components from multiple physical fields in the monitoring data collected by sensors. This coupling makes it difficult for data from a single sensor to accurately reflect the true state of a specific physical field, reducing the interpretability of the data's physical properties.

[0069] In an optional embodiment of this application, the sensors associated with each physical field are arranged on the corresponding monitoring points of the hydropower station's power equipment according to the spatial distribution of the physical fields of the hydropower station's power equipment.

[0070] The physical field spatial distribution of the aforementioned hydropower station power equipment may include, for example: Figure 2 The diagram shows the physical field-spatial domain mapping of the hydro-generator unit. From the diagram, it can be seen that in the generator section, the core physical fields of the upper guide bearing, lower guide bearing, and thrust bearing during operation include stress field, temperature field, and acoustic field; the core physical fields of the generator stator and rotor include electromagnetic field, stress field, temperature field, and acoustic field; the core physical field of the main shaft flange includes stress field and acoustic field; in the turbine section, the core physical fields of the turbine casing, runner blades, guide vanes, and draft tube include flow field, stress field, and acoustic field; and the core physical fields of the water guide bearing include stress field, temperature field, and acoustic field.

[0071] By binding sensor monitoring data to corresponding fields, the spatial allocation of sensors is achieved. This application realizes multi-physics field state monitoring of hydro-generator units through monitoring quantities such as hydraulic pulsation, vibration, current, temperature, and acoustic signature.

[0072] For the turbine components, including runner blades, guide vanes, volute, and draft tube, hydraulic pulsation sensors can effectively and directly sense the flow field state of these components. However, the number of hydraulic pulsation monitoring points is limited, and the flow field perception is only localized and cannot detect changes in the entire flow field. The turbine interior is subjected to strong coupling between the flow field and stress field, resulting in significant acoustic energy. Acoustic fingerprint monitoring can non-contactly monitor the turbine interior and obtain information on the internal flow field, stress field, and acoustic field.

[0073] For the generator section, the current sensor can directly sense the changes in the electromagnetic field, the vibration sensor can directly sense the stress field, and the temperature sensor can directly sense the temperature field information.

[0074] For key mechanical structural components, vibration, temperature, and acoustic monitoring are used for the upper guide bearing, lower guide bearing, and water guide bearing; vibration and acoustic monitoring are used for the main shaft flange; and vibration, acoustic monitoring, and temperature monitoring are used for the thrust bearing.

[0075] In optional embodiments of this application, step S1 specifically includes steps S11 to S15, where S11, S12, etc. are merely step identifiers. The execution order of the method does not necessarily follow the order of numbers from smallest to largest. For example, step S12 can be executed first and then step S11 can be executed. This application does not impose any restrictions.

[0076] S11, the monitoring data of the target monitoring point on the target component is used as the target monitoring data, and the monitoring data of multiple reference monitoring points on another component is used as the reference monitoring data. The target monitoring data and the reference monitoring data are associated with the same physical field.

[0077] The purpose of steps S11 to S15 is to decouple the target monitoring data of the target monitoring point on the target component by using the reference monitoring data of multiple reference monitoring points on another component.

[0078] The target component can be any physical component on the power equipment of a hydropower station. The other component can be another physical component on the power equipment of a hydropower station that is different from the target component. For example, when the power equipment of a hydropower station is a turbine generator set, if the target component is selected as the turbine casing in the turbine section, the other component can be another physical component such as the stator, rotor, upper guide bearing, lower guide bearing, and thrust bearing in the generator section.

[0079] The aforementioned target monitoring point is a location point on the target component where a sensor is installed, i.e., a monitoring point where data is to be decoupled. The multiple reference monitoring points are the location points corresponding to multiple sensors installed on another component.

[0080] The aforementioned target monitoring data refers to the monitoring data fed back by the sensors installed at the target monitoring points, i.e., the data to be decoupled; the aforementioned reference monitoring data refers to the monitoring data fed back by the sensors installed at the reference monitoring points, i.e., the reference data for the decoupling operation.

[0081] The hydro-generator unit has a large building space span, and the stress field and sound field are coupled across fields in the spatial distribution. Therefore, a weighting coefficient is used to decouple the cross-field coupling problem of vibration signal (stress field) and acoustic signal (sound field).

[0082] For example, taking vibration monitoring in a hydro-generator unit as an example, suppose we need to decouple a certain target monitoring point on the turbine casing. Vibration monitoring data In order to accurately obtain the flow field-related vibration information of the target monitoring point, it is first necessary to simultaneously acquire multiple reference monitoring points on the generator section. , ... Vibration monitoring data , ... , as reference monitoring data.

[0083] S12, the target monitoring data is compared with each of the reference monitoring data using a coherence function to obtain multiple coherence values.

[0084] For example, the target monitoring point on the turbine casing Target monitoring data Vibration monitoring data from multiple reference monitoring points on the generator section were compared with those from other parts. , ... Perform coherence function calculation to obtain One coherent value.

[0085] In optional embodiments of this application, step S12 specifically includes steps S121 to S126.

[0086] S121, Perform a Fourier transform on the target monitoring data to obtain the target frequency domain data. .

[0087] S122, For each reference monitoring data point, perform a Fourier transform on the reference monitoring data to obtain the target reference frequency domain data. .

[0088] S123, Calculate the auto-power spectral density of the target frequency domain data, the auto-power spectral density of the target reference frequency domain data, and the cross-power spectral density of the target frequency domain data and the target reference frequency domain data based on the target frequency domain data and the target reference frequency domain data.

[0089] Specifically, the auto-power spectral density of the target frequency domain data can be calculated using the following formula: ;in, Represents the target frequency domain data. Represents modulo calculation, Represents the length of the data, i.e. ,in It is a positive integer; The self-power spectral density of the target reference frequency domain data is calculated using the following formula: ;in, Represents the target reference frequency domain data. The cross-power spectral density of the target frequency domain data and the target reference frequency domain data is calculated using the following formula: ;in, express The conjugate of complex numbers.

[0090] S124, based on the auto-power spectral density of the target frequency domain data, the auto-power spectral density of the reference frequency domain data, and the cross-power spectral density of the target frequency domain data and the reference frequency domain data, the coherence function of the target frequency domain data and the reference frequency domain data is obtained.

[0091] Specifically, the coherence function of the target frequency domain data and the reference frequency domain data can be calculated according to the following formula: .

[0092] S125, Select the first three peak frequencies and substitute them into the coherence function point by point for calculation, and retain the peak frequencies with coherence values ​​greater than the second threshold as candidate frequencies.

[0093] The second threshold can be set according to actual needs, and no specific restrictions are imposed here.

[0094] S126, Select the maximum coherence value from the coherence values ​​corresponding to the candidate frequency points, and use it as the coherence value of the target monitoring data and the target reference monitoring data.

[0095] For example, the target monitoring point on the turbine casing Target monitoring data Vibration monitoring data from multiple reference monitoring points on the generator section were sequentially compared. , ... Perform coherence function calculations. Prioritize calculating target monitoring data. Monitoring points on the generator section Vibration monitoring data The coherence value is obtained by performing steps S121 to S126 above; then, steps S121 to S126 above are repeated to calculate the target monitoring data. Reference monitoring points on the generator section , ... Vibration monitoring data , ... The coherence value.

[0096] S13. Among multiple coherence values, select monitoring points with coherence values ​​greater than the first threshold as candidate monitoring points.

[0097] The first threshold can be set according to actual needs, and no specific restrictions are imposed here. The first threshold can be the same as or different from the second threshold.

[0098] For example, the target monitoring data is calculated by repeating steps S121 to S126 above. Reference monitoring points on the generator section , , ... Vibration monitoring data , , ... The coherence value. Reference monitoring points with coherence values ​​greater than or equal to the first threshold are retained as candidate monitoring points. For example, assuming the first threshold is 0.5, the reference monitoring points... The corresponding coherence value is 0.3 (less than the first threshold), referring to the monitoring point. The corresponding coherence value is 0.6 (greater than the first threshold), referring to the monitoring point. If the maximum coherence value is 0.7 (greater than the first threshold), then the reference monitoring point is retained. and reference monitoring points As an alternative monitoring point.

[0099] S14. Select the reference monitoring point corresponding to the maximum coherence value from the candidate monitoring points and confirm it as the decoupling reference point.

[0100] If multiple reference monitoring points have coherence values ​​greater than the first threshold, then the reference monitoring point corresponding to the largest coherence value is further selected and retained as the decoupling reference point. For example, monitoring points The corresponding maximum coherence value is 0.6, monitoring point If the corresponding maximum coherence value is 0.7, then select a monitoring point. As a reference point for decoupling.

[0101] S15, using the frequency amplitude of the reference monitoring data at the decoupling reference point and the positional relationship between the decoupling reference point and the target monitoring point, decouple the physical field monitoring data mapped to the target monitoring point in the target monitoring data.

[0102] In an optional embodiment of this application, step S15 specifically includes the following steps S151 to S153.

[0103] S151, Determine the weighting coefficients based on the positional relationship between the decoupling reference point and the target monitoring point. .

[0104] Specifically, step S151 includes obtaining the overall height of the hydropower station's power equipment. ; Obtain the distance between the target monitoring point and the centerline of the other component. ; the distance and the overall height ratio As a weighting coefficient .

[0105] When the target monitoring point is on the turbine section, and the other component is the generator section, Represents the target monitoring point and the generator centerline (reference) Figure 1 The distance (shown by the blue dashed line); when the target monitoring point is on the generator section and the other component is the turbine section, Represents the target monitoring point and the centerline of the turbine (reference) Figure 1 The distance (shown by the red dashed line in the middle).

[0106] For example, decoupling reference points are used in the generator section. Reference monitoring data on the turbine casing is used to decouple the target monitoring points. When monitoring target data, assume the target monitoring point The distance from the centerline of the generator is 4 m, that is The overall height of the hydropower station's power equipment is 20 m, that is Then calculate the weighting coefficients: .

[0107] S152, Select the unit coefficient of the hydropower station's power equipment based on the unit power size of the hydropower station's power equipment. .

[0108] For example, in the range of unit power from 10MW to 1000MW, The value ranges from 0.05 to 0.20, and is selected based on the unit's power output. When the unit's power output is high, Choose a larger value; when the unit power is low, Choose a smaller value. By adjusting the weighting coefficients, the intensity of the cross-field spatial enhancement signal can be effectively reduced.

[0109] S153, Computer Group Coefficient , passband amplitude of decoupling reference point 1 and the weighting coefficient The product of the differences is taken as the first product. .

[0110] S154, Calculate the bandwidth of the target monitoring data. with the first product The difference This serves as the physical field monitoring data corresponding to the target monitoring point within the target monitoring data.

[0111] That is, the physical field monitoring data corresponding to the target monitoring point in the target monitoring data is decoupled according to the following formula: .

[0112] in, This represents the physical field monitoring data corresponding to the target monitoring point in the target monitoring data, i.e., the passband amplitude of the target monitoring data after decoupling.

[0113] For example, suppose there is a target monitoring point on the turbine casing. The passband amplitude is 5 mm / s, that is Decoupling reference point on generator section The passband amplitude is 3 mm / s, that is Target monitoring points The distance from the centerline of the generator is 4 m, that is The overall height of the hydropower station's power equipment is 20m, that is .

[0114] Calculate the weighting coefficients: .

[0115] If the power generation capacity of the hydropower station's generator units is 200 MW, then the generator unit coefficient is selected. .

[0116] Use formula Perform decoupled computation: .

[0117] It is understandable that the spatial decoupling rule primarily relies on coherence function analysis and positional relationship correction to separate multi-physics field data. First, it acquires data from the target monitoring point and reference data from reference monitoring points associated with the same physical field. The reference point with the strongest coherence is then selected as the decoupling benchmark through coherence function calculation. Second, using the bandwidth amplitude of the decoupling reference point, combined with its spatial distance from the target monitoring point (reflected by the weighting coefficient ζ) and the overall unit height parameter, an amplitude correction model is established. Finally, the cross-field coupling components in the target monitoring data are removed using a correction formula, achieving accurate extraction of single-physics field data. This rule effectively solves the problem of signal aliasing in the spatial domain and improves the interpretability of the physical properties of multi-physics field monitoring data.

[0118] S2, periodically acquires monitoring data from various sensors on the hydropower station's power equipment and key parameters reflecting the current operating status of the equipment within a preset time period, and uses operating condition decoupling rules to perform cluster analysis on the monitoring data under stable operating conditions, and / or decouples the monitoring data of the dominant physical field under fluctuating operating conditions.

[0119] Step S2 performs cluster analysis on the stable operating condition data, decouples the monitoring data of the dominant physical field under fluctuating operating conditions, establishes a benchmark model of multi-physical field coupling, and provides a reference standard for subsequent fluctuating operating condition analysis; decouples the monitoring data of the dominant physical field under fluctuating operating conditions, locks the dominant physical field, accurately identifies abnormal monitoring data caused by changes in the dominant physical field, and solves the diagnostic ambiguity problem caused by multi-field coupling in the transient process of traditional methods.

[0120] In optional embodiments of this application, step S2 specifically includes steps S21 to S23, where S21, S22, etc. are merely step identifiers. The execution order of the method does not necessarily follow the order of numbers from smallest to largest. For example, step S22 can be executed first and then step S21 can be executed. This application does not impose any restrictions.

[0121] S21 uses the monitoring data and key parameters acquired at the same time as a single sample data.

[0122] In an optional embodiment of this application, the power equipment of the hydropower station includes a turbine-generator unit. The turbine-generator unit operates under numerous and complex conditions, influenced by the coupling of hydraulic, mechanical, and electromagnetic forces. Hydraulic, mechanical, and electromagnetic forces are the force and energy sources of stress fields, electromagnetic fields, temperature fields, flow fields, and sound fields. Processes such as start-up, load shedding, shutdown, speed changes, excitation changes, and load changes directly affect the coupling strength and patterns of these multi-physics fields.

[0123] The sensor monitoring data includes one or more of the following: hydraulic pulsation, vibration, current, temperature, and acoustic signature. Key parameters reflecting the current operating status of the hydropower station's power equipment include at least one of the following: guide vane opening, generator rotor speed, excitation current, and generator output power. At least one of these key parameters is selected as a clustering feature vector to construct an operating condition sample set, with a sample size exceeding a certain number, such as more than 10,000 sets, covering all operating conditions.

[0124] S22, determine whether the hydropower station's power equipment is in a stable or fluctuating operating condition based on key parameters.

[0125] In an optional embodiment of this application, step S22 includes at least one of steps S221 to S226.

[0126] S221, in response to the zero time change rate of the key parameter, it is determined that the hydro-generator unit is in a stable operating condition.

[0127] In practical implementation, key parameters include guide vane opening, generator rotor speed, excitation current, and generator output power. When the time rate of change of all key parameters is zero, the hydro-generator unit is considered to be in a steady-state operating condition. and and and Under these circumstances, it is determined that the hydro-generator unit is in a stable operating condition, among which For guide vane opening, For generator rotor speed, For excitation current, This refers to the generator's output power. Represents time.

[0128] S222, in response to the positive time change rate of the guide vane opening and the positive time change rate of the generator rotor speed, it is determined that the hydro-generator unit is in a start-up fluctuation condition. That is, in and Under these circumstances, it can be determined that the hydro-generator unit is in a fluctuating start-up condition.

[0129] S223, in response to the negative time change rate of the guide vane opening and the negative time change rate of the generator rotor speed, it is determined that the hydro-generator unit is in a shutdown fluctuation condition. That is, in and Under these circumstances, it can be determined that the hydro-generator unit is in a shutdown fluctuation condition.

[0130] S224, in response to the time-varying rate of change of the generator rotor speed and the time-varying rate of change of the excitation current being zero, and the time-varying rate of change of the guide vane opening and the time-varying rate of change of the generator output power being non-zero, it is determined that the hydro-generator unit is under a variable load fluctuation condition. That is, in and and Under these circumstances, it can be determined that the hydro-generator unit is in a variable load fluctuation condition.

[0131] S225, in response to the time-varying rate of change of the excitation current and the time-varying rate of change of the generator output power being zero, and the time-varying rate of change of the generator rotor speed and the time-varying rate of change of the guide vane opening being non-zero, it is determined that the hydro-generator unit is in a variable speed fluctuation condition. That is, in and and Under these circumstances, it can be determined that the hydro-generator unit is in a variable speed fluctuation condition.

[0132] S226, in response to the time-varying rate of change of the generator rotor speed, the time-varying rate of change of the guide vane opening, and the generator output power being zero, and the time-varying rate of change of the excitation current being non-zero, it is determined that the hydro-generator unit is in a variable excitation fluctuation condition. That is, in and Under these circumstances, it is determined that the hydro-generator unit is in a variable excitation fluctuation condition.

[0133] S23, perform at least one of the following steps S231 and S232.

[0134] S231, Select stable operating condition sample data under stable operating conditions, use the key parameters in each stable operating condition sample data as cluster feature vectors, and classify each stable operating condition sample data into operating condition clusters through cluster analysis. The coupling relationship of each physical field within the same operating condition cluster is consistent.

[0135] In-depth analysis of various physical field dimensions within the same operating condition cluster provides optimization suggestions for hydropower station power equipment.

[0136] In practical applications, for sample data under stable operating conditions of hydro-generator units, guide vane opening, generator rotor speed, excitation current, and output power can be selected as clustering feature vectors to construct an operating condition sample set containing more than 10,000 data points. Using the Fuzzy C-Means Clustering Algorithm (FCM), with a cluster size of 8, cluster analysis is performed on the data to form different operating condition clusters.

[0137] For example, analysis revealed low efficiency of units within a certain operating condition cluster. Further investigation into the physical field coupling relationships within this cluster revealed strong coupling between the flow field and stress field, leading to energy loss. Based on this, optimization suggestions can be proposed: adjusting the guide vane opening to improve flow field distribution and reduce stress concentration; or optimizing the runner blade design to mitigate the impact of the flow field on the structure. Through these measures, the operating efficiency of the units under this condition can be improved, achieving optimization of the hydropower station's power equipment.

[0138] S232, filter out the sample data of fluctuating operating conditions, identify the dominant physical field corresponding to the fluctuating operating conditions, and calculate the rate of change of the physical field monitoring data affected by the dominant physical field in the sample data of fluctuating operating conditions.

[0139] Step S232 can analyze the influence of the dominant physical field.

[0140] In an optional embodiment of this application, step S232 specifically includes at least one of the following steps S2321 to S2323.

[0141] S2321, Select sample data of variable load fluctuation under the variable load fluctuation condition, determine that the dominant physical field corresponding to the variable load fluctuation condition is the flow field, and calculate the rate of change of vibration monitoring data and the rate of change of acoustic text monitoring data in the sample data of the variable load fluctuation condition.

[0142] The above sample data for variable load fluctuation conditions includes monitoring data from various sensors acquired simultaneously, as well as key parameters reflecting the current operating status of the equipment. The sensor monitoring data includes one or more monitoring data such as hydraulic pulsation, vibration, current, temperature, and acoustic signature. The key parameters reflecting the current operating status of the hydropower station's power equipment include at least one of the following: guide vane opening, generator rotor speed, excitation current, and generator output power.

[0143] The rate of change of the vibration monitoring data mentioned above refers to the change in vibration monitoring data within a unit load change.

[0144] The rate of change of the aforementioned voiceprint monitoring data refers to the change in the voiceprint monitoring data within a unit load change.

[0145] For example, the flow field is the product of the interaction between water flow and the flow components, and the dominant operating condition is a variable load process. By screening sample data of variable load fluctuation conditions, the flow field is identified as the dominant physical field.

[0146] During load fluctuations, the flow field undergoes abrupt changes. The monitored vibration and acoustic data reflect these abrupt changes. By eliminating interference from mechanical and electromagnetic forces and retaining only the stress and acoustic fields generated by the water flow (flow field), the changes in vibration and acoustic data of the turbine section under different unit load conditions can be determined from an engineering perspective as the result of the flow field acting alone. Load is The vibration data monitored in real time is Voiceprint data is , Load is The vibration data monitored in real time is Voiceprint data is The change value of vibration Voiceprint variation value .

[0147] S2322, filter out the variable speed fluctuation condition sample data, determine that the dominant physical field corresponding to the variable speed fluctuation condition is the stress field, and calculate the rate of change of vibration monitoring data and the rate of change of acoustic text monitoring data in the variable speed fluctuation condition sample data.

[0148] The above-mentioned variable speed fluctuation condition sample data includes monitoring data from various sensors acquired simultaneously, as well as key parameters reflecting the current operating status of the equipment; the sensor monitoring data includes one or more of the following: hydraulic pulsation, vibration, current, temperature, acoustic signature, etc.; the key parameters reflecting the current operating status of the hydropower station's power equipment include at least one of the following: guide vane opening, generator rotor speed, excitation current, and generator output power.

[0149] The rate of change of the vibration monitoring data mentioned above refers to the change in vibration monitoring data per unit change in rotational speed.

[0150] The rate of change of the aforementioned voiceprint monitoring data refers to the change in the voiceprint monitoring data per unit change in rotational speed.

[0151] For example, mechanical force originates from the coupling and friction of rotating components, and the dominant operating condition is variable speed fluctuation. By screening sample data of variable speed fluctuation operating conditions, the stress field is identified as the dominant physical field.

[0152] In variable speed fluctuation conditions, the monitored vibration and acoustic data map the gradual dynamic process of mechanical force changes; under different unit speed conditions, the changes in vibration and acoustic data monitored by the turbine and generator sections can be determined from an engineering perspective as the result of mechanical force acting alone. Rotation speed is The vibration data monitored in real time is Voiceprint data is , Rotation speed is The vibration data monitored in real time is Voiceprint data is The change value of vibration Voiceprint variation value .

[0153] S2323, filter out the sample data of the variable excitation fluctuation condition under the variable excitation fluctuation condition, clarify that the dominant physical field corresponding to the variable excitation fluctuation condition is the electromagnetic field, and calculate the rate of change of vibration monitoring data and the rate of change of acoustic text monitoring data in the sample data of the variable excitation fluctuation condition.

[0154] The above-mentioned variable excitation fluctuation condition sample data includes monitoring data from various sensors acquired simultaneously, as well as key parameters reflecting the current operating status of the equipment. The sensor monitoring data includes one or more of the following: hydraulic pulsation, vibration, current, temperature, acoustic signature, etc. The key parameters reflecting the current operating status of the hydropower station's power equipment include at least one of the following: guide vane opening, generator rotor speed, excitation current, and generator output power.

[0155] The rate of change of the vibration monitoring data mentioned above refers to the change in vibration monitoring data within a unit change in excitation current.

[0156] The rate of change of the aforementioned voiceprint monitoring data refers to the change in the voiceprint monitoring data within a unit change in excitation current.

[0157] For example, electromagnetic force is the product of the electromagnetic interaction between the stator and the rotor, the associated parameter is the excitation current, and the dominant operating condition is the variable excitation fluctuation condition.

[0158] By screening sample data of variable excitation fluctuation conditions, it was determined that the dominant physical field corresponding to the variable excitation fluctuation conditions is the electromagnetic field.

[0159] Under different excitation current conditions of the generator unit, the changes in the vibration and acoustic signature data of the monitored generator section can be determined from an engineering perspective as the result of the electromagnetic field acting alone. Excitation current is The vibration data monitored in real time is Voiceprint data is , Excitation current is The vibration data monitored in real time is Voiceprint data is The change value of vibration Voiceprint variation value .

[0160] It is understandable that the operating condition decoupling rule achieves dynamic separation of multiple physical fields through phased processing. First, sensor data and key parameters (such as guide vane opening and generator rotor speed) are periodically collected, and the operating condition type is determined by the rate of parameter change. For stable operating conditions, key parameters are extracted as clustering feature vectors, and the FCM algorithm is used to classify the data into operating condition clusters, ensuring consistent physical field coupling relationships within each cluster, providing a basis for equipment optimization. For fluctuating operating conditions (such as start-up and variable load), the dominant physical field (such as flow field and stress field) is identified based on the operating condition type, and the rate of change of vibration / acoustic data under its influence is calculated to quantify the intensity of the dominant field's effect. This rule, through operating condition classification and decoupling from the dominant field, solves the diagnostic ambiguity problem caused by multi-field coupling interference during transient processes.

[0161] S3 periodically acquires monitoring data from various sensors on the hydropower station's power equipment within a preset time period, and uses fault mechanism characteristic frequency decoupling rules to achieve fault diagnosis of the hydropower station's power equipment.

[0162] Step S3 involves acquiring sensor monitoring data and performing a Fourier transform to obtain the characteristic frequency based on the mapping relationship between the characteristic frequency of the monitoring data and each physical field. The fault cause is determined by the characteristic frequency offset. The physical field can be directly mapped to achieve directional stripping of the coupled signal, outputting a quantitative fault diagnosis conclusion, and improving the sensitivity and accuracy of fault identification.

[0163] In optional embodiments of this application, step S3 specifically includes steps S31 to S32, where S31, S32, etc. are merely step identifiers. The execution order of the method does not necessarily follow the numerical order from smallest to largest. For example, step S32 can be executed first and then step S31 can be executed. This application does not impose any restrictions.

[0164] S31, Perform frequency domain transformation on the acquired monitoring data to obtain the characteristic frequency.

[0165] It is understandable that the sensor's monitoring data includes data on hydraulic pulsation, vibration, current, temperature, acoustic signature, etc. Monitoring signals from different physical fields have unique frequency characteristics; by separating signals of different frequencies through frequency domain analysis, a precise mapping between frequency and physical field can be directly achieved.

[0166] Table 1 Feature Frequency Library Data

[0167] ①The frequency of the hydroelectric generator (f1=Nr / 60) and the mapped fields include: flow field, stress field, and sound field; Among them, the frequency of the hydropower unit is the fundamental frequency of the unit's rotation, which corresponds to the fundamental frequency flow state of the water flow as the impeller rotates in the flow field, and determines the overall stability of the flow field; the stress field is the source of centrifugal stress in rotating components such as the generator rotor, main shaft, and impeller, and dominates the rotational vibration characteristics; the sound field generates low-frequency mechanical noise, which is the fundamental frequency component of the overall noise of the unit.

[0168] ② The generator's main pole frequency (f2=50Hz) and the mapped fields include electromagnetic field, stress field, and sound field; Among them, the generator's main pole frequency is the fundamental frequency of the generator's electromagnetic field, which determines the electromagnetic coupling strength and electromagnetic vibration characteristics, indirectly causing the generator's stator and rotor structure to vibrate (stress field) and generate steady-state electromagnetic noise (sound field).

[0169] ③ The frequency of the turbine guide vanes (f3=Nr×Z1 / 60) and the mapped field include the flow field, stress field, and sound field; Among them, the turbine guide vane frequency is the characteristic frequency of the guide vane rotating and cutting the water flow, and the corresponding water flow pulsation in the guide vane region of the flow field affects the uniformity of the flow; the stress frequency corresponding to the guide vane vibration in the stress field affects the fatigue life of the guide vane; the sound field is generated by the flow-induced vibration of the guide vane, producing discrete frequency sound.

[0170] ④ The frequency of the turbine runner blades (f4=Nr×Z2 / 60) and the mapped field include the flow field, stress field, and sound field; Among them, the turbine runner blade frequency is the fundamental frequency of the water flow driven by the rotation of the runner blade. In the flow field, it determines the pulsation period of the flow field around the blade and affects flow states such as cavitation and vortex. In the stress field, it is the dominant frequency of the bending and torsional stress of the blade, which is related to the fatigue damage of the blade. In the sound field, it corresponds to the characteristic noise of the flow-induced vibration of the blade.

[0171] ⑤ The tailrace vortex frequency f5 (f5=f1×Z1×Z2 / a) and the mapped field includes the flow field, stress field, and acoustic field; Among them, the vortex frequency of the tailrace tube is the vortex shedding of the tailrace tube in the flow field, which causes the tailrace tube wall to vibrate (stress field) and generate alternating stress; in the acoustic field, it manifests as vortex-induced noise, and the frequency changes with the coupling relationship between the guide vane / blade number.

[0172] ⑥ Tailwater pipe pressure pulse frequency (f6=f1 / (3~5)), the mapped field includes flow field, stress field and sound field; Among them, the tailrace pressure pulse frequency is the frequency of the pressure fluctuation of the tailrace water flow. In the flow field, it reflects the dynamic characteristics of backflow and vortex in the tailrace. The pressure pulse is transmitted to the pipe wall and causes structural vibration (stress field), which may lead to resonance damage. In the sound field, it manifests as low-frequency pressure pulse noise, which is the acoustic characterization of the unstable flow state of the tailrace.

[0173] ⑦ The vortex frequency of the rotor (f7=(0.18~0.2)×W2 / b), the mapped field includes the flow field, stress field, and sound field; Among them, the rotor vortex frequency is the frequency at which the blade vortex detaches, which is the dominant energy loss in the flow field of the blade; the vortex impact on the blade causes alternating loads (stress field), which accelerates blade fatigue; and the vortex-induced noise generated in the sound field is an auxiliary indicator for cavitation early warning.

[0174] ⑧ The generator thrust bearing frequency (f8=f1×Z3) and the mapped field include the stress field and the sound field; Among them, the frequency of the generator thrust bearing is the characteristic frequency of the thrust bearing under force vibration, and the stress field corresponds to the alternating stress generated by the contact friction between the thrust bearing and the main shaft mirror plate and the load fluctuation; the sound field is the low-to-medium frequency mechanical noise caused by the vibration of the thrust bearing, without flow field or electromagnetic field correlation.

[0175] ⑨ Rotary pressure frequency of the impeller seal (f9=f1×C), the mapped field includes flow field, stress field, and sound field; Among them, the rotational pressure frequency of the impeller seal and the water flow pressure fluctuation frequency in the impeller seal gap reflect the dynamic characteristics of the leakage flow at the seal in the flow field and affect the sealing performance; pressure fluctuations cause the sealing ring to vibrate (stress field), accelerating seal wear; and discrete frequency noise is generated in the sound field.

[0176] ⑩ Generator air gap frequency (f 10 =f1×K); the mapped field includes electromagnetic field, stress field, and acoustic field; Among them, the air gap frequency of the generator is the pulsation frequency of the electromagnetic force in the air gap of the generator. It is caused by the harmonics of the air gap magnetic flux density in the electromagnetic field and determines the intensity of electromagnetic vibration. The electromagnetic force is transmitted to the stator and rotor, causing structural vibration (stress field). Electromagnetic noise is generated in the sound field.

[0177] ⑪ Stator extreme frequency (f) 11 =2f2=f1×P), the mapped field includes electromagnetic field, stress field, and sound field; Among them, the stator pole frequency vibration frequency is the stator magnetic pole electromagnetic frequency, which is the frequency harmonic component of the stator electromagnetic force and is related to the number of magnetic poles P; it induces the stator core and frame to vibrate (stress field); and generates medium-frequency electromagnetic noise in the sound field, with the frequency coupled to the power frequency.

[0178] ⑫ Generator electromagnetic vibration frequency (f) 12 =2f2), the mapped field includes electromagnetic field, stress field, and sound field; Among them, the electromagnetic vibration frequency of the generator is the characteristic frequency caused by the imbalance of electromagnetic force. It is generated by the coupling of the stator and rotor magnetic fields in the electromagnetic field and dominates the electromagnetic vibration of the generator. It directly causes alternating stress (stress field) in the stator winding and frame, affecting the structural durability. Steady-state electromagnetic noise is generated in the sound field and is related to the harmonic of the main pole frequency.

[0179] ⑬ Spindle torsional vibration frequency (f) 13 =KP0+1), the mapped field includes the stress field and the sound field; Among them, the spindle torsional vibration frequency is the torsional vibration frequency of the spindle, which is caused by the coupling of load fluctuation of the transmission system and electromagnetic force in the stress field; torsional vibration causes the overall vibration of the unit (stress field) and generates low-frequency mechanical noise (sound field).

[0180] ⑭ Runner blade exit frequency (f 14 =f4×(1-V n2 / u)), the mapped field includes the flow field, stress field, and sound field; Among them, the runner blade outlet frequency is the characteristic frequency of the water flow velocity at the blade outlet, which reflects the uniformity of the outlet flow field in the flow field; flow velocity pulsation causes blade vibration (stress field), increasing the risk of fatigue; flow-induced noise is generated in the acoustic field, which is a sensitive parameter for cavitation and flow separation.

[0181] ⑮ Tailwater pipe intermediate frequency pressure pulsation frequency (f 16 =(1.8~3.6)×f1), the mapped field includes the flow field, stress field, and sound field; Among them, the medium-frequency pressure pulsation frequency of the tailrace pipe is the medium-frequency water flow pulsation frequency of the tailrace pipe. It reflects the dynamic characteristics of medium-intensity vortex bands and backflow in the flow field, which is between low-frequency and high-frequency pulsation; it induces medium-frequency vibration (stress field) on the tailrace pipe wall, accelerating structural fatigue; and generates medium-frequency noise in the sound field.

[0182] ⑯ High-frequency pressure pulsation frequency of tailrace pipe (f 17 =(3.6~6.0)×f1); the mapped field includes the flow field, stress field, and sound field; Among them, the high-frequency pressure pulsation frequency of the tailrace pipe is the high-frequency water flow fluctuation frequency of the tailrace pipe, which corresponds to the rapid pulsation of strong turbulence and small-scale vortices in the flow field, and is prone to causing flow field disturbance; the high-frequency pressure pulsation causes high-frequency vibration (stress field) of the tailrace pipe structure; and manifests as high-frequency noise in the sound field.

[0183] S32, in response to the mean and / or standard deviation of the characteristic frequency reaching a third threshold within the preset time period, it is determined that the characteristic frequency has shifted, and then the cause of the fault is determined based on the physical field and monitoring point corresponding to the characteristic frequency.

[0184] The third threshold can be set according to the actual situation, and is not limited here.

[0185] The mapping relationship between each characteristic frequency and each physical field is clarified, and the fault diagnosis of hydropower station power equipment is realized by using the mean and / or standard deviation of the characteristic frequencies within a preset time period.

[0186] It is understandable that the fault mechanism characteristic frequency decoupling rule achieves fault source tracing through frequency domain analysis and physical field mapping. First, sensor data is periodically collected and Fourier transformed to extract characteristic frequencies reflecting the equipment state. Second, a characteristic frequency library is constructed to clarify the mapping relationship between each frequency component (such as the hydroelectric generator frequency f1 and guide vane frequency f3) and physical fields such as flow field, stress field, and electromagnetic field. Finally, by statistically analyzing the mean and standard deviation of characteristic frequencies within a preset time period, when the offset exceeds a preset safety threshold, an anomaly is determined in the corresponding physical field. This rule utilizes the strong correlation between frequency characteristics and physical field faults to achieve directional decoupling of coupled signals and quantitative fault diagnosis, solving the problem of low accuracy in qualitative analysis using traditional methods.

[0187] On the other hand, such as Figure 3 As shown, this application discloses a monitoring and processing device 100, which includes at least one of a spatial decoupling module 101, an operating condition decoupling module 102, and a fault mechanism characteristic frequency decoupling module 103.

[0188] The spatial decoupling module is configured to acquire monitoring data from various sensors on the power equipment of the hydropower station in real time, and to decouple the physical field monitoring data corresponding to the location of the monitoring point from each monitoring data using spatial decoupling rules. The operating condition decoupling module is configured to periodically acquire monitoring data from various sensors on the hydropower station's power equipment and key parameters reflecting the current operating status of the equipment within a preset time period. It adopts operating condition decoupling rules to perform cluster analysis on the monitoring data under stable operating conditions and / or decouple the monitoring data of the dominant physical field under fluctuating operating conditions. The fault mechanism characteristic frequency decoupling module is configured to periodically acquire monitoring data from various sensors on the hydropower station's power equipment within a preset time period, and to realize fault diagnosis of the hydropower station's power equipment by adopting fault mechanism characteristic frequency decoupling rules.

[0189] In an optional embodiment of this application, the spatial decoupling rule execution module includes a first data acquisition submodule, an association submodule, a coherence value calculation submodule, a candidate monitoring point screening submodule, a decoupling reference point screening submodule, and a decoupling submodule; the first data acquisition submodule is configured to acquire monitoring data from various sensors on the hydropower station's power equipment in real time; the association submodule is configured to use the monitoring data from the target monitoring point on the target component as the target monitoring data, and use the monitoring data from multiple reference monitoring points on another component as the reference monitoring data, wherein the target monitoring data and the reference monitoring data are associated with the same physical field; the coherence value calculation submodule... The system is configured to perform coherence function calculations on the target monitoring data and each reference monitoring data to obtain multiple coherence values; the candidate monitoring point screening submodule is configured to select reference monitoring points with coherence values ​​greater than a first threshold from the multiple coherence values ​​as candidate monitoring points; the decoupling reference point screening submodule is configured to select the reference monitoring point corresponding to the maximum coherence value from the candidate monitoring points and confirm it as the decoupling reference point; the decoupling submodule is configured to decouple the physical field monitoring data mapped to the target monitoring point in the target monitoring data by using the passband amplitude of the reference monitoring data at the decoupling reference point and the positional relationship between the decoupling reference point and the target monitoring point.

[0190] In an optional embodiment of this application, the coherence value calculation submodule includes a target frequency domain data calculation unit, a target reference frequency domain data calculation unit, a power spectral density calculation unit, a coherence function calculation unit, a candidate frequency point calculation unit, and a coherence value calculation unit. The target frequency domain data calculation unit is configured to perform a Fourier transform on the target monitoring data to obtain target frequency domain data. The target reference frequency domain data calculation unit is configured to perform a Fourier transform on the reference monitoring data of a reference monitoring point to obtain target reference frequency domain data. The power spectral density calculation unit is configured to calculate the auto-power spectral density of the target frequency domain data and the auto-power spectral density of the target reference frequency domain data based on the target frequency domain data and the target reference frequency domain data. The system comprises: a self-power spectral density, a cross-power spectral density of the target frequency domain data and the target reference frequency domain data; a coherence function calculation unit, configured to obtain the coherence function of the target frequency domain data and the reference frequency domain data based on the self-power spectral density of the target frequency domain data, the self-power spectral density of the reference frequency domain data, and the cross-power spectral density of the target frequency domain data and the reference frequency domain data; a candidate frequency point calculation unit, configured to select the first three peak frequencies and substitute them into the coherence function point by point for calculation, retaining the peak frequencies with coherence values ​​greater than the second threshold as candidate frequency points; and a coherence value calculation unit, configured to select the maximum coherence value from the coherence values ​​corresponding to the candidate frequency points as the coherence value of the target monitoring data and the target reference monitoring data.

[0191] In an optional embodiment of this application, the decoupling submodule includes a weight coefficient determination unit, a unit coefficient determination unit, a product calculation unit, and a decoupling calculation unit; the weight coefficient determination unit is configured to determine the weight coefficients based on the positional relationship between the decoupling reference point and the target monitoring point. The unit coefficient determination unit is configured to select the unit coefficients of the hydropower station's power equipment based on the unit power output of the hydropower station's power equipment. The product calculation unit is configured as a computer group coefficient. , passband amplitude of decoupling reference point 1 and weighting coefficient The product of the differences is taken as the first product. The decoupled computing unit is configured to calculate the bandwidth amplitude of the target monitoring data. with the first product The difference This serves as the physical field monitoring data corresponding to the target monitoring point within the target monitoring data.

[0192] In an optional embodiment of this application, the weighting coefficient determination unit includes an overall height acquisition unit, a spacing acquisition unit, and a coefficient calculation unit; the overall height acquisition unit is configured to acquire the overall height of the hydropower station's power equipment. The spacing acquisition unit is configured to acquire the distance between the target monitoring point and the centerline of another component. The coefficient calculation unit is configured to calculate the distance. and overall height ratio As a weighting coefficient .

[0193] In an optional embodiment of this application, the operating condition decoupling module includes at least one of a second data acquisition module, a sample confirmation submodule, an operating condition judgment module, and a stable operating condition decoupling module and a fluctuating operating condition decoupling module; the second data acquisition module is configured to periodically acquire monitoring data from various sensors on the hydropower station's power equipment and key parameters reflecting the current operating status of the equipment within a preset time period; the sample confirmation submodule is configured to use the monitoring data and key parameters acquired simultaneously as a single sample data; the operating condition judgment module is configured to judge the hydropower station's power equipment based on the key parameters. The system is configured to operate under either stable or fluctuating conditions. The stable operating condition decoupling module is configured to filter out stable operating condition sample data, using the key parameters in each sample as clustering feature vectors. Through cluster analysis, each stable operating condition sample is classified into different operating condition clusters, ensuring consistency in the coupling relationships between physical fields within the same cluster. The fluctuating operating condition decoupling module is configured to filter out fluctuating operating condition sample data, identify the dominant physical field corresponding to the fluctuating operating condition, and calculate the rate of change of the physical field monitoring data affected by the dominant physical field in the fluctuating operating condition sample data.

[0194] In optional embodiments of this application, the power equipment of the hydropower station includes a turbine generator set; key parameters include one or more of the following: guide vane opening, generator rotor speed, excitation current, and generator output power.

[0195] In an optional embodiment of this application, the operating condition judgment module includes a stable operating condition judgment module and a fluctuating operating condition judgment module; the stable operating condition judgment module is configured to determine that the hydro-generator unit is in a stable operating condition in response to a zero time change rate of a key parameter; the fluctuating operating condition judgment module is configured to perform at least one of the following: in response to a positive time change rate of the guide vane opening and a positive time change rate of the generator rotor speed, determine that the hydro-generator unit is in a start-up fluctuating operating condition; in response to a negative time change rate of the guide vane opening and a negative time change rate of the generator rotor speed, determine that the hydro-generator unit is in a shutdown fluctuating operating condition; in response to a negative time change rate of the generator rotor speed... If the time-varying rate of change of the excitation current and the time-varying rate of change of the guide vane opening and the time-varying rate of change of the generator output power are zero, it is determined that the hydro-generator unit is under variable load fluctuation condition; if the time-varying rate of change of the excitation current and the time-varying rate of change of the generator output power are zero, and the time-varying rate of change of the generator rotor speed and the time-varying rate of change of the guide vane opening are non-zero, it is determined that the hydro-generator unit is under variable speed fluctuation condition; if the time-varying rate of change of the generator rotor speed, the time-varying rate of change of the guide vane opening and the generator output power are zero, and the time-varying rate of change of the excitation current is non-zero, it is determined that the hydro-generator unit is under variable excitation fluctuation condition.

[0196] In optional embodiments of this application, the fluctuating operating condition decoupling module is configured to perform at least one of the following: filtering sample data of fluctuating operating conditions under variable load conditions, identifying the dominant physical field corresponding to the fluctuating operating conditions as a flow field, and calculating the rate of change of vibration monitoring data and acoustic fingerprint monitoring data in the sample data of fluctuating operating conditions; filtering sample data of fluctuating operating conditions under variable speed conditions, identifying the dominant physical field corresponding to the fluctuating operating conditions as a stress field, and calculating the rate of change of vibration monitoring data and acoustic fingerprint monitoring data in the sample data of fluctuating operating conditions; filtering sample data of fluctuating operating conditions under variable excitation conditions, identifying the dominant physical field corresponding to the fluctuating operating conditions as an electromagnetic field, and calculating the rate of change of vibration monitoring data and acoustic fingerprint monitoring data in the sample data of fluctuating operating conditions.

[0197] In an optional embodiment of this application, the fault mechanism characteristic frequency decoupling module includes a third data acquisition module, a characteristic frequency calculation submodule, and a fault judgment submodule. The third data acquisition module is configured to periodically acquire monitoring data from various sensors on the hydropower station's power equipment within a preset time period. The characteristic frequency calculation submodule is configured to perform frequency domain transformation on the acquired monitoring data to obtain the characteristic frequency. The fault judgment submodule is configured to determine that the characteristic frequency has shifted when the mean and / or standard deviation of the characteristic frequency reaches a third threshold within a preset time period, and then determine the cause of the fault based on the physical field and monitoring point corresponding to the characteristic frequency.

[0198] The monitoring and processing device and the monitoring and processing method provided in the above embodiments belong to the same concept. For details of their specific implementation process, please refer to the method embodiments, which will not be repeated here.

[0199] Figure 4 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application.

[0200] The computer device 200 may be a server. The computer device 200 can vary considerably depending on its configuration or performance, and includes one or more Central Processing Units (CPUs) 201 and one or more memories 202. The memories 202 store at least one line of program code, which is loaded and executed by the processor 201 to implement the aforementioned operating state identification method. Of course, the computer device 200 also has wired or wireless network interfaces, a keyboard, and input / output interfaces for input and output. The computer device 200 also includes other components for implementing device functions, which will not be elaborated upon here.

[0201] In an exemplary embodiment, a computer-readable storage medium is also provided, such as a memory including program code that can be executed by a processor in a computer device to perform the above-described operating state identification method. For example, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical data storage device, etc.

[0202] In an exemplary embodiment, a computer program product or computer program is also provided, which includes computer program code stored in a computer-readable storage medium. The processor of a computer device reads the computer program code from the computer-readable storage medium and executes the computer program code, causing the computer device to perform the above-described operating state identification method.

[0203] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0204] The above description is merely an optional embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A monitoring and processing method, characterized in that, Includes at least one of the following: Real-time acquisition of monitoring data from various sensors on the power equipment of the hydropower station; and the use of spatial decoupling rules to decouple the physical field monitoring data corresponding to the location of the monitoring point from each monitoring data. The monitoring data of various sensors on the power equipment of the hydropower station and the key parameters reflecting the current operating status of the equipment are periodically acquired within a preset time period. The monitoring data of each stable operating condition are clustered and / or the monitoring data of the dominant physical field under fluctuating operating conditions are decoupled using the operating condition decoupling rule. The monitoring data of various sensors on the power equipment of the hydropower station are acquired periodically within a preset time period, and the fault diagnosis of the power equipment of the hydropower station is realized by adopting the fault mechanism characteristic frequency decoupling rule.

2. The monitoring and processing method according to claim 1, characterized in that, The real-time acquisition of monitoring data from various sensors on the hydropower station's power equipment employs spatial decoupling rules to decouple the physical field monitoring data corresponding to the monitoring point locations from each set of monitoring data, including: The monitoring data of the target monitoring point on the target component is used as the target monitoring data, and the monitoring data of multiple reference monitoring points on another component is used as the reference monitoring data. The target monitoring data and the reference monitoring data are associated with the same physical field. The target monitoring data is compared with each of the reference monitoring data using a coherence function to obtain multiple coherence values; Among the multiple coherence values, the reference monitoring points whose coherence values ​​are greater than a first threshold are selected as candidate monitoring points; The reference monitoring point corresponding to the maximum coherence value is selected from the candidate monitoring points and confirmed as the decoupling reference point; By utilizing the frequency amplitude of the reference monitoring data at the decoupling reference point and the positional relationship between the decoupling reference point and the target monitoring point, the physical field monitoring data mapped to the target monitoring point in the target monitoring data is decoupled.

3. The monitoring and processing method according to claim 2, characterized in that, The step of performing coherence function calculations on the target monitoring data and each of the reference monitoring data to obtain multiple coherence values ​​includes: Perform a Fourier transform on the target monitoring data to obtain the target frequency domain data; For each of the aforementioned reference monitoring data, a Fourier transform is performed on the reference monitoring data of a reference monitoring point to obtain target reference frequency domain data; Based on the target frequency domain data and the target reference frequency domain data, calculate the auto-power spectral density of the target frequency domain data, the auto-power spectral density of the target reference frequency domain data, and the cross-power spectral density of the target frequency domain data and the target reference frequency domain data; Based on the auto-power spectral density of the target frequency domain data, the auto-power spectral density of the reference frequency domain data, and the cross-power spectral density of the target frequency domain data and the reference frequency domain data, the coherence function of the target frequency domain data and the reference frequency domain data is obtained; The first three peak frequencies are selected and substituted into the coherence function point by point for calculation, and the peak frequencies with coherence values ​​greater than the second threshold are retained as candidate frequencies. The maximum coherence value is selected from the coherence values ​​corresponding to the candidate frequency points and used as the coherence value of the target monitoring data and the target reference monitoring data.

4. The monitoring and processing method according to claim 3, characterized in that, The step of decoupling the physical field monitoring data mapped to the target monitoring point from the reference monitoring data by utilizing the passband amplitude of the reference monitoring data at the decoupling reference point and the positional relationship between the decoupling reference point and the target monitoring point includes: The weighting coefficients are determined based on the positional relationship between the decoupling reference point and the target monitoring point. ; The generator set coefficient of the hydropower station's power equipment is selected based on the unit power capacity of the hydropower station's power equipment. ; Calculate the unit coefficient The passband amplitude of the decoupling reference point 1 and the weighting coefficient The product of the differences is taken as the first product. ; Calculate the passband amplitude of the target monitoring data. with the first product The difference This serves as the physical field monitoring data corresponding to the target monitoring point within the target monitoring data.

5. The monitoring and processing method according to claim 4, characterized in that, The weighting coefficient is determined based on the positional relationship between the decoupling reference point and the target monitoring point. ,include: Obtain the overall height of the power equipment of the hydropower station. ; Obtain the distance between the target monitoring point and the centerline of the other component. ; The distance and the overall height ratio As a weighting coefficient .

6. The monitoring and processing method according to claim 1, characterized in that, The process of periodically acquiring monitoring data from various sensors on the hydropower station's power equipment within a preset time period, as well as key parameters reflecting the current operating status of the equipment, employs operating condition decoupling rules to perform cluster analysis on the monitoring data under stable operating conditions, and / or decouples the dominant physical field monitoring data under fluctuating operating conditions, including: The monitoring data and key parameters acquired at the same time are used as a single sample data. Based on the key parameters, it can be determined whether the power equipment of the hydropower station is in a stable or fluctuating operating condition. Stable operating condition sample data are selected under stable operating conditions. The key parameters in each stable operating condition sample data are used as clustering feature vectors. The stable operating condition sample data are classified into different operating condition clusters through clustering analysis. The coupling relationship of each physical field within the same operating condition cluster is consistent. And / or, fluctuating operating condition sample data are selected under fluctuating operating conditions. The dominant physical field corresponding to the fluctuating operating condition is identified, and the rate of change of the physical field monitoring data affected by the dominant physical field in the fluctuating operating condition sample data is calculated.

7. The monitoring and processing method according to claim 6, characterized in that, The power equipment of the hydropower station includes a turbine generator set; The key parameters include one or more of the following: guide vane opening, generator rotor speed, excitation current, and generator output power.

8. The monitoring and processing method according to claim 7, characterized in that, The step of determining whether the hydropower station's power equipment is in a stable operating condition based on the key parameters includes: If the time change rate of the key parameter is zero, it is determined that the hydro-generator unit is in a stable operating condition. The determination that the hydropower station's power equipment is in a fluctuating operating condition based on the key parameters includes at least one of the following: If the time change rate of the guide vane opening is positive and the time change rate of the generator rotor speed is positive, it is determined that the hydro-generator unit is in a start-up fluctuation condition. If the time change rate of the guide vane opening is negative and the time change rate of the generator rotor speed is negative, it is determined that the hydro-generator unit is in a shutdown fluctuation condition. In response to the fact that the time change rate of the generator rotor speed and the time change rate of the excitation current are both zero, and the time change rate of the guide vane opening and the time change rate of the generator output power are both non-zero, it is determined that the hydro-generator unit is in a variable load fluctuation condition. If the time change rate of the excitation current and the time change rate of the generator output power are both zero, and the time change rate of the generator rotor speed and the time change rate of the guide vane opening are both non-zero, it is determined that the hydro-generator unit is in a variable speed fluctuation condition. In response to the fact that the time change rate of the generator rotor speed, the time change rate of the guide vane opening, and the generator output power are all zero, and the time change rate of the excitation current is non-zero, it is determined that the hydro-generator unit is in a variable excitation fluctuation condition.

9. The monitoring and processing method according to claim 8, characterized in that, The process of selecting sample data of fluctuating operating conditions, identifying the dominant physical field corresponding to the fluctuating operating condition, and calculating the rate of change of the physical field monitoring data affected by the dominant physical field in the sample data of fluctuating operating conditions includes at least one of the following: Select sample data of variable load fluctuation under the variable load fluctuation condition, identify the dominant physical field corresponding to the variable load fluctuation condition as the flow field, and calculate the rate of change of vibration monitoring data and the rate of change of acoustic text monitoring data in the sample data of the variable load fluctuation condition. Select sample data of variable speed fluctuation under the variable speed fluctuation condition, identify the dominant physical field corresponding to the variable speed fluctuation condition as the stress field, and calculate the rate of change of vibration monitoring data and the rate of change of acoustic text monitoring data in the sample data of the variable speed fluctuation condition. Select sample data of variable excitation fluctuation conditions, identify the dominant physical field corresponding to the variable excitation fluctuation conditions as the electromagnetic field, and calculate the rate of change of vibration monitoring data and acoustic fingerprint monitoring data in the sample data of variable excitation fluctuation conditions.

10. The monitoring and processing method according to claim 1, characterized in that, The process of periodically acquiring monitoring data from various sensors on the hydropower station's power equipment within a preset time period, and employing fault mechanism characteristic frequency decoupling rules to achieve fault diagnosis of the hydropower station's power equipment includes: The acquired monitoring data is subjected to frequency domain transformation to obtain the characteristic frequencies; In response to the mean and / or standard deviation of the characteristic frequency reaching a third threshold within the preset time period, it is determined that the characteristic frequency has shifted, and then the cause of the fault is determined based on the physical field and monitoring point corresponding to the characteristic frequency.

11. A monitoring and processing device, characterized in that, It includes at least one of the following: spatial decoupling module, operating condition decoupling module, and fault mechanism characteristic frequency decoupling module; The spatial decoupling module is configured to acquire monitoring data from various sensors on the hydropower station's power equipment in real time, and to decouple the physical field monitoring data corresponding to the monitoring point location from each monitoring data using spatial decoupling rules. The operating condition decoupling module is configured to periodically acquire monitoring data from various sensors on the hydropower station's power equipment and key parameters reflecting the current operating status of the equipment within a preset time period. It adopts operating condition decoupling rules to perform cluster analysis on the monitoring data under stable operating conditions and / or decouple the monitoring data of the dominant physical field under fluctuating operating conditions. The fault mechanism characteristic frequency decoupling module is configured to periodically acquire monitoring data from various sensors on the hydropower station's power equipment within a preset time period, and to perform fault diagnosis on the hydropower station's power equipment by adopting fault mechanism characteristic frequency decoupling rules.

12. A computer device, characterized in that, The device includes a processor and a memory, the memory storing at least one line of program code, which is loaded and executed by the processor to implement the monitoring and processing method as described in any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that, The storage medium stores at least one piece of program code, which is loaded and executed by a processor to implement the monitoring and processing method as described in any one of claims 1 to 10.

14. A computer program product or computer program, characterized in that, The computer program product or computer program includes computer program code stored in a computer-readable storage medium, a processor of a computer device reads the computer program code from the computer-readable storage medium, and the processor executes the computer program code, causing the computer device to perform the monitoring processing method as described in any one of claims 1 to 10.

Images