Active power oscillation fault monitoring and processing method for low-water-head hydraulic power plant

Abstract

The invention provides an active power oscillation fault monitoring and processing method for a low-water-head hydraulic power plant. The invention aims to solve the problems of active power fluctuation, aggravated equipment vibration, cavitation and cavitation and the like caused by lowering of a water purification head and deviation of a unit from an optimal operation area due to rising of a tail water level. According to the method, based on the water turbine power, the theoretical maximum output and the power set value under the current water head are compared in real time, and alarm is triggered once it is found that the set value exceeds the unit capacity; and meanwhile, multi-dimensional characteristics such as guide vane / paddle opening abnormity, frequent twitching, sudden oil pressure drop of a speed regulator, overtime operation of an oil pump and set vibration aggravation are comprehensively monitored, and a multi-parameter fusion fault recognition model is constructed. When an oscillation symptom is detected, the system automatically suggests or executes intervention measures such as reducing power setting, quitting active PID adjustment, switching the speed regulator to the local operation and the like, and machines and secondary professionals are linked to carry out on-site disposal.

Description

Technical Field

[0001] This invention relates to the field of active power oscillation monitoring and processing technology in low-head hydropower stations, and particularly to a method for monitoring and processing active power oscillation faults in low-head hydropower plants. Background Technology

[0002] Low-head hydropower stations, as an important component of hydropower production, are widely used in flood season peak-shaving power generation and seasonal energy allocation scenarios. Related technologies utilize the coordinated operation of head monitoring, governor control, and power regulation to construct a conventional operation system based on PID control. Specifically, this system covers the entire process from hydrological forecasting to unit output regulation, including key aspects such as dynamic water level modeling, guide vane opening control, and auxiliary system linkage. With the development of hydropower technology, existing monitoring methods often rely on single-parameter threshold judgments, lacking systematic modeling of the dynamic coupling relationship between head, flow, and power. This makes it difficult to promptly identify active power oscillation faults caused by insufficient net head under flood season tailwater level rise conditions.

[0003] However, existing hydropower station operation monitoring technologies directly rely on fixed power setpoints and single head thresholds for judgment, without establishing a dynamic monitoring model based on the real-time head-flow-power coupling relationship. This can lead to the unit maintaining its rated power setpoint under low head conditions. This abnormal operating state can trigger a chain reaction, such as frequent guide vane movement, delayed start-up of hydraulic systems, and power failure of auxiliary equipment. Specifically, the deviation between the actual active power generated by the unit and the setpoint exceeds a certain threshold. Phenomena such as short-term average vibration deviations from the long-term baseline exceeding a threshold are observed. Due to the lack of a multi-physical quantity collaborative analysis mechanism, existing technologies struggle to effectively intervene in the early stages of faults, thereby exacerbating turbine cavitation damage, threatening grid frequency stability, and ultimately resulting in both power generation efficiency losses and shortened equipment lifespan. Summary of the Invention

[0004] The present invention aims to at least partially solve one of the technical problems in the related art.

[0005] Therefore, the first objective of this invention is to propose a method for monitoring and handling active power oscillation faults in low-head hydropower plants.

[0006] The second objective of this invention is to provide a device for monitoring and handling active power oscillation faults in low-head hydropower plants.

[0007] To achieve the above objectives, a first aspect of the present invention provides a method for monitoring and handling active power oscillation faults in low-head hydropower plants, comprising: S1. Establish a real-time monitoring model based on mixed integer programming, introduce binary state variables and set an objective function, and determine whether the unit meets the rated power operation conditions by calculating the net water head and constraints. S2, based on a sliding time window, monitors the active power fluctuation of the unit in real time, calculates the range of actual generated values ​​within the window, and when the active power setpoint changes, the system delays for 3 minutes before counting the number of times the range of actual generated values ​​exceeds the preset fluctuation threshold and determines an alarm; when the setpoint does not change, when the range of actual generated values ​​exceeds the fluctuation threshold 10 times, it is determined to be a large fluctuation in active power and an alarm is triggered. S3 monitors the rate of change of oil pressure in the governor and the continuous running time of the oil pump. If the rate of drop in oil pressure exceeds the preset threshold and the running time of the oil pump exceeds the maximum allowable running time, an abnormal oil pressure alarm is triggered and the governor is switched to the local operation mode. S4. By comparing the current guide vane opening with the normal operating condition threshold range, if the guide vane opening exceeds the preset upper and lower limits or the number of times the guide vane opening direction changes within the sliding window exceeds the frequent tumbling threshold, a guide vane abnormality alarm is triggered and the active PID regulation is exited.

[0008] In one embodiment of the present invention, S1 includes: S11, Constraints In the middle, when Mandatory water purification head satisfy ,when The constraint is automatically made to hold by using a large positive number M. S12, Constraints In the middle, through real-time calculation The functional relationship, when Required power generation flow satisfy ,when The constraint is automatically set by using a large positive number M.

[0009] In one embodiment of the present invention, S2 includes: S21, the length of the sliding time window is 60 seconds, which includes the data of the most recent 60 sampling points; S22, the range is calculated by comparing the maximum and minimum values ​​of the real value sequence within the window point by point.

[0010] In one embodiment of the present invention, S3 includes: S31, The oil pressure drop rate is calculated by dividing the difference between the current oil pressure and the previous oil pressure by the sampling time interval; S32, the oil pump running time is determined by comparing the cumulative continuous running time with the preset maximum allowable running time.

[0011] In one embodiment of the present invention, S4 includes: S41, the statistics of the number of direction changes are obtained by calculating the first-order difference of the guide vane opening signal. And statistically analyze the relationship between adjacent sampling points. The number of sign changes; S42, the determination of the number of orientation changes within the sliding window adopts a configuration of a window length of 10 seconds and a threshold of 5 orientation changes.

[0012] To achieve the above objectives, a second aspect of the present invention provides a device for dynamically adjusting the active power setpoint of a low-head hydropower plant, comprising: The real-time monitoring model building module is used to build a real-time monitoring model based on mixed integer programming. It introduces binary state variables and sets an objective function. By calculating the net head and constraints, it determines whether the unit meets the rated power operating conditions. The active power fluctuation monitoring module is used to monitor the active power fluctuation of the unit in real time based on a sliding time window, calculate the range of the actual generated value within the window, and when the range exceeds the preset fluctuation threshold, it is determined to be a large fluctuation of active power and triggers an alarm. The governor oil pressure abnormality monitoring module is used to monitor the rate of change of governor oil pressure and the continuous running time of oil pump. If the rate of oil pressure drop exceeds the preset threshold and the oil pump running time exceeds the maximum allowable running time, an oil pressure abnormality alarm is triggered and the governor is switched to local operation mode. The guide vane anomaly monitoring module is used to compare the current guide vane opening with the normal operating condition threshold range. If the guide vane opening exceeds the preset upper and lower limits or the number of guide vane opening direction changes within the sliding window exceeds the frequent tumbling threshold, a guide vane anomaly alarm is triggered and the active PID regulation is exited.

[0013] The present invention discloses a method and apparatus for adjusting the dynamic setpoint of active power in a low-head hydropower plant, which can realize the early identification and dynamic handling of active power oscillation faults in low-head hydropower plants. Through head-flow-power coupling modeling and multi-parameter anomaly monitoring, it can effectively prevent the unit from deviating from the optimal efficiency zone and improve equipment safety and power generation efficiency.

[0014] To achieve the above objectives, a third aspect of this application provides a computer device comprising a processor and a memory; wherein the processor reads executable program code stored in the memory to run a program corresponding to the executable program code, for implementing a method for adjusting the dynamic setpoint of active power in a low-head hydropower plant as described in the first aspect embodiment.

[0015] To achieve the above objectives, the fourth aspect of this application proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a method for adjusting the dynamic setpoint of active power in a low-head hydropower plant as described in the first aspect embodiment.

[0016] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0017] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of a method for adjusting the dynamic setpoint of active power in a low-head hydropower plant according to an embodiment of the present invention; Figure 2 This is a structural diagram of a dynamic setpoint adjustment device for active power in a low-head hydropower plant according to an embodiment of the present invention. Figure 3 It is a computer device according to an embodiment of the present invention. Detailed Implementation

[0018] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0019] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0020] The following description, with reference to the accompanying drawings, describes a method and apparatus for adjusting the dynamic setpoint of active power in a low-head hydropower plant according to an embodiment of the present invention.

[0021] Example 1 Figure 1 This is a flowchart of a method for monitoring and handling active power oscillation faults in low-head hydropower plants according to an embodiment of the present invention.

[0022] like Figure 1 As shown, the method for monitoring and handling active power oscillation faults in low-head hydropower plants includes the following steps: S1. Establish a real-time monitoring model based on mixed integer programming, introduce binary state variables and set an objective function, and determine whether the unit meets the rated power operation conditions by calculating the net water head and constraints.

[0023] Specifically, this step establishes a real-time monitoring model based on mixed-integer programming to determine the status of hydropower plant units at any given time. Does it have the capability to operate at rated power? Operating conditions. This model introduces binary state variables. The unit's operating status is discretized into rated power operation or non-rated power operation, thereby achieving precise control and optimization of the operating status.

[0024] Furthermore, the objective function of the model is:

[0025] in, Indicates the rated power of the unit. Indicates in The actual active power output of the generator set at any given time. The time step is specified (e.g., 1 second or 1 minute). The objective function aims to maximize the scheduling cycle. The time during which the internal units operate at rated power, and when the rated power operating conditions cannot be met, the actual power generation should be increased as much as possible.

[0026] To achieve the above objectives, the model introduces a formula for dynamically calculating the net head:

[0027] in, The upstream water level is calculated based on the reservoir capacity curve. This is a nonlinear function relating the tailrace level to the discharge flow rate. This is the head loss constant. This formula is used to calculate the net head at the current moment in real time, serving as a basis for determining whether the unit meets the conditions for rated power operation.

[0028] Furthermore, the model ensures through the following two logical constraints At that time, the unit did indeed meet the conditions for operating at its rated power:

[0029]

[0030] in, The minimum clean water head required for the unit to deliver its rated power. The minimum power generation flow rate required to generate rated power under the current net head. Let it be a sufficiently large positive number to relax the constraints. When At that time, the above two constraints require that both the net water head and the power generation flow rate meet the rated power operating conditions; when When this occurs, the constraint automatically becomes invalid, allowing the unit to operate at its actual power.

[0031] Furthermore, this model is applicable to real-time operation monitoring systems for low-head hydropower stations, especially during floods when upstream inflow is high and tailrace water levels rise, leading to a decrease in net head. Through this model, operators can promptly identify whether the unit is unable to reach its rated power due to insufficient head, thereby adjusting operating strategies and preventing the unit from operating under inefficient or dangerous conditions.

[0032] Furthermore, by introducing a mixed-integer programming method, complex physical operating conditions are transformed into a solvable mathematical model, enabling dynamic judgment and optimized control of the unit's operating status. Its innovation lies in combining the physical constraints of head and flow rate with binary decision variables, thereby maximizing power generation efficiency while ensuring safety.

[0033] Furthermore, S1 includes: S11, Constraints In the middle, when Mandatory water purification head satisfy ,when The constraint is automatically set by using a large positive number M.

[0034] Furthermore, this constraint employs the Big M method to construct the logical relationship. Among these, This represents the unit's net water head at time t. It is the minimum clean head required for the unit to output its rated power, provided by the turbine manufacturer. It is a binary variable, with a value of 1 indicating that the unit is operating at rated power at time t, and a value of 0 indicating that the rated power operating condition has not been met. When When, the constraints are simplified to This means that the water purification head must meet the minimum requirements for rated power operation; when At that time, due to If it is a sufficiently large positive number (usually 10000 or higher), the constraint becomes The inequality holds automatically, thus not imposing any additional restrictions on the clean water head.

[0035] Furthermore, The selection of must meet the following condition: its value should be much greater than With actual water purification head The possible differences are considered to ensure the correctness of the logical constraints. Meanwhile, The value is usually determined by the efficiency curve and design parameters of the turbine, and the unit is meter (m). The specific value needs to be set according to the unit model and the technical data provided by the manufacturer.

[0036] Furthermore, this constraint is widely used in the operation optimization and fault monitoring systems of low-head hydropower plants. When the upstream inflow is large and the tailrace level rises, causing a decrease in net head, the system uses this constraint to determine whether it can still maintain rated power operation. If the constraint is not met, it automatically switches to a suboptimal operating strategy to maximize actual output and avoid the unit operating under inefficient or dangerous conditions.

[0037] Furthermore, by introducing binary variables and the Big M method, this constraint achieves accurate modeling of the unit's operating state, effectively avoiding the mismatch between setpoints and actual output caused by insufficient head. Its clear logic and efficient computation provide a solid foundation for subsequent optimized scheduling and fault early warning, enhancing the safety and economy of hydropower plant operations during flood season.

[0038] S12, Constraints In the middle, through real-time calculation The functional relationship, when Required power generation flow satisfy ,when The constraint is automatically set by using a large positive number M.

[0039] Specifically, in this invention, the constraints involved in step two... It is one of the key logical constraints in mixed-integer programming models, and its core function lies in introducing binary variables. This allows for dynamic determination of whether the generating unit meets the conditions for rated power operation. The implementation of this constraint is based on the fundamental power formula of the hydro turbine. ,in For power generation flow, For the water purification head, For the efficiency of the water turbine, The density of water, This is the acceleration due to gravity. In actual operation, It's about a water purifier head. The function represents the rated power output of the unit under the current net head conditions. The minimum required flow rate for power generation. Since the lower the head, the greater the flow rate required to maintain rated power, therefore... Dynamic calculations are required based on real-time water head.

[0040] Furthermore, this constraint is achieved by introducing large positive numbers. To construct logical relationships. When When the unit is operating at its rated power, the constraint becomes... This means that the actual power generation flow rate must meet the minimum flow rate required for the rated power under the current water head. If Then the constraint will automatically relax to ,because For sufficiently large positive numbers, this inequality always holds, thus not imposing additional restrictions on power generation flow. This design allows the model to force units to operate at rated power when physical conditions are met, and automatically switch to non-rated operating mode otherwise, thereby achieving a balance between scheduling objectives and operational safety.

[0041] Furthermore, The value of needs to be large enough to ensure that constraint relaxation will not interfere with the model solution, but it cannot be too large to avoid affecting numerical stability. It can usually be taken as a fraction of the system's maximum flow rate. times. The calculation needs to be based on the turbine characteristic curve, combined with real-time head. Interpolation or function fitting is performed to ensure accuracy and real-time performance.

[0042] In practical applications, this constraint is mainly used for scheduling optimization during floods, especially when the rise in tailwater level leads to a decrease in net head. By judging in real time whether the power generation flow meets the rated power requirement, it determines whether to maintain rated power operation. If not, the power setpoint is adjusted in a timely manner to avoid the unit operating under low head and large opening, and to prevent equipment damage problems such as cavitation and vibration.

[0043] Furthermore, this constraint effectively enables logical control of the unit's operating status, ensuring maximum rated power operation time when physical conditions permit, while still maximizing power generation efficiency when conditions are limited. Its innovation lies in combining the physical characteristics of the turbine with scheduling logic, using mathematical modeling to achieve early warning of faults and automatic adjustment of operating strategies, thereby improving the operational safety and economy of low-head hydropower plants under complex hydrological conditions.

[0044] S2, based on a sliding time window, monitors the active power fluctuation of the unit in real time, calculates the range of actual generated values ​​within the window, and when the active power setpoint changes, the system delays for 3 minutes before counting the number of times the range of actual generated values ​​exceeds the preset fluctuation threshold and determines an alarm; when the setpoint does not change, when the range of actual generated values ​​exceeds the fluctuation threshold 10 times, it is determined to be a large fluctuation in active power and an alarm is triggered.

[0045] Specifically, this step aims to monitor the real-time fluctuations of the active power generated by low-head hydropower plant units through a sliding time window, thereby identifying abnormal operating conditions and triggering alarms in a timely manner to prevent equipment damage or grid disturbances caused by deviations from optimal operating conditions. This method, based on quantitative analysis of power fluctuation amplitude and combined with turbine operating characteristics and actual operating conditions, achieves early warning of active power oscillation faults.

[0046] Furthermore, the system employs a fixed-length sliding time window (e.g., 60 seconds) to collect actual active power generation data from the generating units at a sampling frequency of once per second. Within the window, the system calculates the range of power values ​​in real time, i.e., the maximum value within the window. and minimum value The difference is expressed as This range reflects the severity of power fluctuations over a short period of time and is an important indicator for determining whether the unit is in an unstable operating state.

[0047] Furthermore, the fluctuation threshold is set to the unit's rated power. of ,Right now When the range within the sliding window exceeds this threshold, the system determines that the active power is fluctuating significantly and triggers an alarm mechanism. This threshold is set based on actual operating experience and equipment safety boundaries to ensure no false alarms under normal operating conditions, while providing a timely response to abnormal fluctuations.

[0048] Furthermore, this step combines time series analysis with threshold judgment, using the range statistic to quantitatively assess power fluctuations. Its core lies in achieving continuous monitoring of the dynamic process through a sliding window mechanism, avoiding the omission of instantaneous fluctuations due to fixed-point judgments. Simultaneously, this method aligns with the basic operating formulas of water turbines. Correlation helps identify power output limitations caused by reduced head or abnormal guide vanes.

[0049] Furthermore, this step can be deployed in the hydropower plant's SCADA system or dedicated monitoring platform, and linked with modules such as unit operating status, governor control, and vibration monitoring. When abnormal power fluctuations are detected, the system can adjust setpoints, switch governor control modes, or initiate manual inspection procedures, thereby improving the safety and stability of unit operation.

[0050] Furthermore, by quantifying the power fluctuation amplitude, real-time perception and anomaly identification of the unit's operating status can be achieved, providing a reliable basis for subsequent fault handling and operation adjustment, effectively preventing equipment damage and grid disturbances caused by insufficient head or unstable governor, and improving the operating efficiency and safety of hydropower plants during flood season.

[0051] Furthermore, S2 includes: S21, the sliding time window is 60 seconds long and contains data from the most recent 60 sampling points.

[0052] Specifically, in step three, the present invention employs a sliding time window technique to monitor in real time the deviation between the actual active power generated by the unit and the set value, thereby identifying abnormal power fluctuations and taking timely control measures. This step is based on time series analysis and threshold judgment, exhibiting high real-time performance and sensitivity.

[0053] In some implementations, the sliding time window is set to 60 seconds, containing data from the most recent 60 sampling points (assuming a sampling frequency of 1Hz). This window slides along the time axis in fixed steps (e.g., 1 second), continuously updating the power dataset within the window. Through this window, the system can dynamically capture the changing trend of the unit's active power over a short period of time, avoiding misjudgments caused by instantaneous disturbances.

[0054] Optionally, the system sets a fluctuation threshold. Typically, this is the rated power of the unit. of When the calculated range When this occurs, the system determines that the active power is fluctuating significantly and triggers an alarm mechanism. At this time, the platform will automatically reduce the active power setpoint of the unit to alleviate the pressure on the speed regulation system and prevent the equipment from entering an unstable operating condition.

[0055] This step is primarily used in practical applications for monitoring the operation of low-head hydropower plants during floods. When the tailrace rises, causing a decrease in net head, the actual output of the generating units may not meet the set value, resulting in power fluctuations. Through sliding window analysis, the system can identify such anomalies early, providing operators with timely intervention information.

[0056] Furthermore, this step effectively improves the accuracy and response speed of power fluctuation identification, avoiding frequent governor operation and increased equipment vibration caused by unreasonable setpoints. Simultaneously, dynamically adjusting the setpoints helps maintain the unit operating within its optimal efficiency range, improving overall operational stability and power generation efficiency.

[0057] S22, the range is calculated by comparing the maximum and minimum values ​​of the real value sequence within the window point by point.

[0058] Specifically, the range The calculation is performed by comparing the maximum and minimum values ​​of the actual generated power sequence within a sliding window point by point. This step constructs a time-series-based power fluctuation monitoring model to identify abnormal fluctuations in the unit's active power near the setpoint, thereby providing a basis for subsequent fault diagnosis and control strategies.

[0059] Furthermore, the sliding window length is typically set to 60 seconds, with a sampling frequency of 1 second per point, meaning each window contains 60 consecutive active power generation data points. The window slides along the time axis in fixed steps (e.g., 1 second) to achieve continuous monitoring of real-time power fluctuations. Within each window, the system first extracts the maximum and minimum values ​​from the generation sequence, denoted as [values ​​to be filled in]. and Then calculate the difference, i.e., the range. This indicator directly reflects the magnitude of power output changes over a short period of time and is an important basis for determining whether the unit is in an abnormal operating state.

[0060] Furthermore, this is typically deployed within the SCADA system or dedicated power monitoring module of a hydropower plant, interfacing with the real-time acquired active power signal. Its operating environment must meet high sampling accuracy and low latency requirements to ensure the timeliness and accuracy of fluctuation detection. Under low head conditions, due to limited unit output, the actual power output is prone to periodic or random fluctuations. This step can effectively identify such anomalies, providing operators with timely intervention information.

[0061] Furthermore, this step achieves dynamic assessment of the unit's operating status by quantifying the degree of power fluctuation. Its innovation lies in combining a sliding window mechanism with a range index to construct a simple yet efficient fluctuation identification method, exhibiting good real-time performance and robustness. This method not only improves fault monitoring sensitivity but also provides data support for subsequent governor control strategy adjustments, thereby effectively ensuring the safe and stable operation of the unit under low head conditions.

[0062] S3 monitors the rate of change of oil pressure in the governor and the continuous running time of the oil pump. If the rate of drop in oil pressure exceeds the preset threshold and the running time of the oil pump exceeds the maximum allowable running time, an oil pressure abnormality alarm is triggered and the governor is switched to the local operation mode.

[0063] Specifically, in some implementations, step four achieves rapid identification and response to abnormal states in the governor's hydraulic system by real-time monitoring of the governor's oil pressure change rate and the continuous operating time of the oil pump. This step is technically based on quantitative analysis of the dynamic characteristics of the oil pressure and logical judgment of the operating time, exhibiting high real-time performance and reliability.

[0064] Furthermore, this procedure applies to governor status monitoring in low-head hydropower plants during flood season or high-load operation. When the system detects an abnormal drop in oil pressure and an oil pump timeout, it will trigger a "rapid drop in oil pressure, oil pump timeout" alarm and automatically switch the governor to local operation mode to prevent unit malfunction or equipment damage caused by hydraulic system failure.

[0065] Furthermore, this step effectively identifies potential faults in the governor's hydraulic system through a dual judgment mechanism that quantifies oil pressure changes and operating time, thereby improving the system's safety redundancy and response efficiency. Its innovation lies in combining physical parameters with operating logic to achieve early warning and automatic handling of abnormal governor conditions, thus ensuring stable operation of the unit under complex operating conditions.

[0066] Furthermore, S3 includes: S31, the oil pressure drop rate is calculated by dividing the difference between the current oil pressure and the previous oil pressure by the sampling time interval.

[0067] Specifically, the rate of decrease in governor oil pressure is calculated using the current oil pressure. Compared with the previous hydraulic pressure The difference divided by the sampling time interval This is achieved through a real-time monitoring system that continuously samples and dynamically analyzes the governor's oil pressure. Its core principle lies in determining whether there are any abnormalities in the governor's hydraulic system, such as oil pipeline leaks or oil pump malfunctions, by quantifying the rate of change in oil pressure.

[0068] Furthermore, a high-precision pressure sensor is deployed in the speed governor control system to collect oil pressure data at a fixed sampling period. The difference between adjacent sampling points is calculated and divided by the time interval using an embedded logic module or a higher-level monitoring system to obtain the oil pressure drop rate. When this rate exceeds a preset threshold, the system will trigger a "rapid oil pressure drop" alarm and simultaneously determine whether the oil pump has been running continuously for longer than the maximum allowable operating time (e.g., 30 seconds) to further confirm whether there is any abnormality in the oil pump's timeout operation.

[0069] In practical applications, this step is mainly used for health monitoring of the governor hydraulic system in low-head hydropower plants. During flood season or when the tailrace water level rises and causes a drop in head, frequent governor operations may lead to abnormalities in the hydraulic system. Abnormal changes in the rate of oil pressure drop can serve as an early sign of failure, helping to detect oil pipeline leaks or oil pump malfunctions in a timely manner, preventing governor failure due to insufficient oil pressure, which could then lead to unit vibration, power fluctuations, or even shutdown accidents.

[0070] Furthermore, by quantifying the rate of change of oil pressure, the accuracy and response speed of identifying abnormalities in the governor's hydraulic system are improved, providing operators with clear criteria for fault diagnosis. This enables real-time monitoring and early warning of the governor's operating status, ensuring the stable operation of the unit under complex hydraulic conditions.

[0071] S32, the oil pump running time is determined by comparing the cumulative continuous running time with the preset maximum allowable running time.

[0072] Specifically, the "judgment of oil pump running time is achieved by comparing the cumulative continuous running time with the preset maximum allowable running time" is based on the real-time monitoring and logical judgment of the oil pump's operating status. It aims to detect abnormal drops in oil pressure and oil pump overruns in a timely manner, thereby avoiding governor control instability due to insufficient oil pressure, which could lead to oscillations in the unit's active power or equipment damage.

[0073] Furthermore, the system can be configured with a periodic sampling mechanism to collect oil pump operating status signals at fixed time intervals (e.g., every second) and perform cumulative calculations in the background. If the oil pump operating time exceeds the limit within any consecutive time period, an alarm is triggered, prompting operators to check for problems such as oil pressure drop or oil pipeline leaks in the hydraulic system. Simultaneously, the system can incorporate pressure drop rate indicators... (Current oil pressure - previous oil pressure) / sampling time interval, comprehensively judge the degree of abnormality of the oil pressure system.

[0074] Furthermore, this applies to monitoring scenarios of governor systems in low-head hydropower plants during flood season or high-load operation. When the unit's net head decreases, causing frequent adjustments to the guide vanes, the governor's hydraulic system frequently starts to maintain stable oil pressure. If the continuous operation time of the oil pump exceeds the set threshold, it indicates that the system may have insufficient oil pressure or oil pump overload risk, requiring timely intervention.

[0075] This step enables the system to effectively identify abnormal operating states of the hydraulic system, providing timely alarm information to operators. This allows for preventative measures to be taken before serious equipment failures occur, such as disabling active power PID regulation or switching to local operation, ensuring the stability of the speed control system and the safe operation of the unit. This method offers advantages such as timely response, clear judgment logic, and integration with existing monitoring systems, making it a crucial component for early warning of active power oscillation faults in low-head hydropower plants.

[0076] S4. By comparing the current guide vane opening with the normal operating condition threshold range, if the guide vane opening exceeds the preset upper and lower limits or the number of times the guide vane opening direction changes within the sliding window exceeds the frequent tumbling threshold, a guide vane abnormality alarm is triggered and the active PID regulation is exited.

[0077] Specifically, in some implementations, the "frequent guide vane twitching" detection mechanism in step five is based on statistical analysis of the number of guide vane opening direction changes within a sliding window, aiming to identify abnormal dynamic behavior of the speed control system under low head conditions. The technical implementation principle of this step mainly relies on the time-series analysis of the guide vane opening signal. By setting reasonable "dead zones" and "sliding time windows," measurement noise and small-amplitude fluctuations under normal operating conditions are filtered out, thereby accurately capturing frequent reciprocating motions caused by governor parameter mismatch or system instability.

[0078] Furthermore, a sliding time window, such as 10 seconds, is set to analyze the changing trend of the guide vane opening in real time. Within this window, the guide vane opening signal is sampled, typically at a sampling frequency of 1Hz or higher, to ensure data continuity and accuracy. Next, the change in opening between two adjacent sampling points is calculated. And count the number of times its sign (positive / negative) changes. Whenever When the sign changes from positive to negative or vice versa, it is counted as a "direction change," representing the start of one reciprocating motion. Furthermore, a "frequent twitching threshold" is set, for example, 5 times / 10 seconds. When the number of direction changes within the window exceeds this threshold, it is determined as frequent guide vane twitching, triggering an abnormal alarm.

[0079] Furthermore, the dead zone range is typically set as follows: This is to avoid misjudgments caused by sensor inaccuracy or normal adjustment. The sliding window length and direction change threshold can be adjusted according to the dynamic characteristics of the specific unit to adapt to the detection sensitivity requirements under different operating environments.

[0080] Furthermore, it is mainly used for real-time monitoring of the unit's operating status under low head conditions, especially when the tailwater level rises and the net head falls, to prevent the governor from frequently operating the guide vanes due to unreasonable PID parameters or system interference, which could lead to increased unit vibration and frequent starts of the hydraulic system. By promptly triggering alarms and suggesting the withdrawal of active PID regulation, it can effectively prevent the equipment from entering an unstable operating range, thereby improving the safety and stability of the unit's operation.

[0081] Furthermore, this step, by quantifying the dynamic behavior of the guide vanes, enables early identification and intervention of anomalies in the speed control system, which helps reduce equipment wear, extend service life, and avoid system frequency fluctuations caused by power oscillations in power grid dispatch, thereby improving the economy and reliability of hydropower plant operation.

[0082] Furthermore, S4 includes: S41, the statistics of the number of direction changes are obtained by calculating the first-order difference of the guide vane opening signal. And statistically analyze the relationship between adjacent sampling points. The number of sign changes.

[0083] Furthermore, in order to identify whether the guide vane is frequently twitching, this method statistically... The sign change refers to the number of sign changes between two adjacent difference values, i.e., from positive to negative or from negative to positive. This change usually reflects the characteristic of the guide vane opening reciprocating within a short period of time, and is a typical manifestation of unstable governor control or improper parameter settings. In practical implementation, a sliding time window (such as 10 seconds) can be used to perform local analysis on the difference sequence, and each sign change within the window is counted as a "direction change".

[0084] Optionally, to eliminate misjudgments caused by measurement noise or minor disturbances, a "dead zone" threshold can be set before statistical analysis, for example... The change in guide vane opening. When When the absolute value is less than the dead zone, the direction change statistics are not included, thereby improving the robustness of the algorithm.

[0085] Furthermore, this method is widely used in the operation monitoring systems of low-head hydropower plants, especially during floods or when the tailrace rise causes a drop in head. By monitoring the changing trend of the guide vane opening in real time, operators can promptly detect abnormal governor behavior, such as PID parameter mismatch, feedback signal distortion, or hydraulic system failure, and take corresponding measures, such as disengaging active PID regulation or switching to local control, to prevent equipment damage and grid disturbances.

[0086] Furthermore, by quantifying the dynamic behavior of the guide vane opening, an operable indicator is provided for the stability assessment of the governor. Its innovation lies in combining first-order differential with sign change statistics to form an efficient identification mechanism for frequent governor fluctuations, thereby improving the operational safety and control accuracy of hydropower plants under complex operating conditions.

[0087] S42, the determination of the number of orientation changes within the sliding window adopts a configuration of a window length of 10 seconds and a threshold of 5 orientation changes.

[0088] Specifically, in some implementations, the determination of the number of directional changes within the sliding window is configured with a window length of 10 seconds and a threshold of 5 directional changes. This technique is based on the dynamic monitoring and analysis of the governor guide vane opening signal, aiming to identify whether the speed control system experiences frequent fluctuations, thereby determining its operational stability. This step involves setting a fixed-length sliding time window (e.g., 10 seconds) to continuously sample and process the real-time changes in the guide vane opening. Typically, the sampling frequency is 1Hz or higher to ensure the timing accuracy of the signal.

[0089] Furthermore, within the sliding window, a first-order difference calculation is performed on the guide vane opening signal, that is, the change between two adjacent sampling points. .when A change in the sign (positive / negative) is counted as a directional change, indicating that the guide vane opening has changed direction once during the reciprocating motion. In practical implementation, to avoid misjudgments due to measurement noise or minor fluctuations, a deadband is usually set, for example... Only when Only when the direction change exceeds the dead zone is it counted as a number of times.

[0090] Furthermore, the sliding window length is 10 seconds, meaning that each calculation only considers data points from the most recent 10 seconds to reflect the short-term dynamic behavior of the governor. The threshold for the number of directional changes is set to 5, indicating that if the guide vane opening signal undergoes more than 5 directional changes within 10 seconds, it is judged as "frequent jerking" and an alarm is triggered. This threshold can be adjusted according to the specific operating characteristics of the unit, but it must meet the power system equipment operation stability assessment standards such as IEC61400-25.

[0091] In practical applications, this step is mainly used in the active power oscillation fault monitoring system of low-head hydropower plants. Especially when the governor control loop is abnormal or the PID parameters are mismatched, it can promptly identify abnormal reciprocating motion of the guide vanes, preventing power fluctuations and equipment damage caused by control instability. Through this method, operators can obtain early warning signals and take emergency measures such as discontinuing active power PID regulation or switching to local operation, ensuring the safety and stability of the unit's operation.

[0092] Furthermore, this step effectively improves the identification accuracy and response speed of abnormal operating conditions of the speed control system, avoids mechanical wear and electrical disturbances caused by frequent twitching, and provides a key basis for subsequent fault location and handling, which has significant engineering practical value.

[0093] The active power oscillation fault monitoring and handling method for low-head hydropower plants according to the present invention can effectively monitor the early signs of active power oscillation faults in low-head hydropower plants, adjust the unit operating status in a timely manner, avoid the equipment from entering adverse operating conditions such as vibration and cavitation, and improve the safety and power generation efficiency of the unit operation.

[0094] Example 2 In one embodiment of the present invention, a method for monitoring and locating faults in thermal resistors in hydropower plants further includes the following steps: S101. When the inflow of water to the hydropower station is large, the reservoir capacity is freed up in advance to reduce the rise of the tailwater level during the flood peak, prevent the net head of the unit from being significantly reduced, maximize the time that the unit operates at rated power during the flood, and maximize the total power generation when the rated power cannot be met.

[0095] S201, utilizing the basic principle of water turbines (P=η*ρ*g*Q*H), compares the deviation between the theoretical maximum output and the actual power setting value under the current net head. Once it detects that the power setting value exceeds the maximum output capacity of the unit corresponding to the current head due to a decrease in net head, the platform should immediately trigger an alarm to guide operators to adjust operations in a timely manner and prevent the equipment from entering unfavorable operating conditions. S301 establishes an effective model to identify deviations between the actual active power output of the generator unit and the set value, as well as large fluctuations in the active power output of the generator unit. An alarm is triggered when the deviation between the "actual output value" and the "set value" is too large, and the set value of the active power output of the generator unit is reduced at the same time.

[0096] S401: If the governor oil pressure drops rapidly and the oil pump operation timeout occurs, an alarm "Rapid Oil Pressure Drop, Oil Pump Operation Timeout" will be triggered, reminding operators to check for oil leaks in the oil lines. If necessary, active PID regulation can be discontinued, and the governor can be switched to local operation. Pressure drop rate = (Current oil pressure - Previous oil pressure) / Sampling time interval; Oil pump operation timeout = Continuous oil pump operation time > Maximum allowable operation time.

[0097] S501, check if the guide vane opening is normal (compare with normal operating conditions), and whether there is frequent twitching of the governor guide vanes and blades. If any abnormality is found, issue an alarm immediately and remind the operators to check. If necessary, the active PID control can be discontinued and the governor can be switched to local operation.

[0098] Is the guide vane opening normal? If the current guide vane opening is below the normal lower limit, a guide vane opening too small alarm is triggered. If the current guide vane opening is above the normal upper limit, a guide vane opening too large alarm is triggered. Is there frequent jerking of the governor guide vanes and blades? This algorithm detects whether the guide vanes are reciprocating at a high frequency and small amplitude, which is a typical sign of governor instability or inappropriate parameters.

[0099] Simple Algorithm: Based on the number of crossings within a sliding window, a small dead zone is set: for example, ±0.5%. This dead zone is used to filter out normal measurement noise and small, harmless fluctuations. A sliding time window is set: for example, 10 seconds. A frequent twitching threshold is set: for example, 5 times (within the 10-second window). Core Logic - Counting Crossings: Within the sliding window, record the number of times the guide vane opening signal crosses its average value (or a reference value) upwards. A simpler method is to calculate the first-order difference of the guide vane opening within the window (i.e., the change ΔGv between two adjacent sampling points). Counting Direction Changes: When the sign (positive / negative) of ΔGv changes from positive to negative or from negative to positive, it is counted as one direction change. This represents the start of one reciprocating motion.

[0100] Judgment: If the number of direction changes exceeds the frequent twitching threshold within the sliding window, it is determined that the guide vane is frequently twitching.

[0101] S601, alarm indicating increased unit vibration, reminding operators to check the operation of secondary equipment such as the governor, excitation system, and auxiliary equipment system.

[0102] Short-term average (V_avg_short): Reflects the recent level of volatility (e.g., the last 1-10 minutes). This value is highly sensitive to changes.

[0103] Long-term average (V_avg_long): Reflects the baseline of normal vibration over a longer period (such as the last few hours or a day). This value is very stable.

[0104] Calculate the deviation: Deviation = V_avg_short - V_avg_long Judgment: If the deviation is greater than the deviation threshold, an alarm is triggered.

[0105] The temperature fault monitoring and location method proposed in this invention can effectively improve the reliability and safety of the unit temperature monitoring system, providing an effective tool for real-time monitoring and anomaly detection of unit temperature, and helping to ensure the safe and stable operation of the unit.

[0106] Example 3 To achieve the above embodiments, such as Figure 2 As shown, this embodiment also provides a dynamic setpoint adjustment device 10 for active power of a low-head hydropower plant, comprising: The real-time monitoring model establishment module 100 is used to establish a real-time monitoring model based on mixed integer programming. It introduces binary state variables and sets an objective function. By calculating the net water head and constraints, it determines whether the unit meets the rated power operation conditions. The active power fluctuation monitoring module 200 is used to monitor the active power fluctuation of the unit in real time based on a sliding time window, calculate the range of actual generated values ​​within the window, and when the active power setpoint changes, the system delays for 3 minutes before counting the number of times the range of actual generated values ​​exceeds the preset fluctuation threshold and determines to alarm; when the setpoint does not change, when the range of actual generated values ​​exceeds the fluctuation threshold 10 times, it is determined to be a large fluctuation in active power and an alarm is triggered. The governor oil pressure abnormality monitoring module 300 is used to monitor the rate of change of governor oil pressure and the continuous running time of oil pump. If the rate of oil pressure drop exceeds the preset threshold and the oil pump running time exceeds the maximum allowable running time, an oil pressure abnormality alarm is triggered and the governor is switched to the local operation mode. The guide vane anomaly monitoring module 400 is used to compare the current guide vane opening with the normal operating condition threshold range. If the guide vane opening exceeds the preset upper and lower limits or the number of guide vane opening direction changes within the sliding window exceeds the frequent tumbling threshold, a guide vane anomaly alarm is triggered and the active PID regulation is exited.

[0107] Furthermore, the real-time monitoring model building module 100 is also used for: when Mandatory water purification head satisfy ,when When the constraint is made by a large positive number M Automatically established; when Required power generation flow satisfy ,when When the constraint is made by a large positive number M It was created automatically.

[0108] Furthermore, the active power fluctuation monitoring module 200 is also used for: The sliding time window is 60 seconds long and contains data from the most recent 60 sampling points; The range is calculated by comparing the maximum and minimum values ​​of the actual value sequence within the window point by point.

[0109] Furthermore, the governor oil pressure abnormality monitoring module 300 is also used for: The oil pressure drop rate is calculated by dividing the difference between the current oil pressure and the previous oil pressure by the sampling time interval. The oil pump operating time is determined by comparing the cumulative continuous operating time with the preset maximum allowable operating time.

[0110] Furthermore, the guide vane anomaly monitoring module 400 is also used for: The statistics of the number of directional changes are obtained by calculating the first-order difference of the guide vane opening signal. And statistically analyze the relationship between adjacent sampling points. The number of sign changes; The determination of the number of orientation changes within the sliding window is based on a configuration where the window length is 10 seconds and the threshold for the number of orientation changes is 5.

[0111] The active power oscillation fault monitoring and handling device for low-head hydropower plants according to the present invention can effectively monitor the early signs of active power oscillation faults in low-head hydropower plants, adjust the unit operating status in a timely manner, avoid the equipment from entering adverse operating conditions such as vibration and cavitation, and improve the safety of unit operation and power generation efficiency.

[0112] Example 4 The present invention also provides an electronic device such as Figure 3 As shown, it includes a processor and a memory. The memory stores executable instructions. When the processor executes the instructions, it implements the above-mentioned method for dynamically adjusting the active power setpoint of a low-head hydropower plant.

[0113] Example 5 The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned method for adjusting the dynamic setpoint of active power in a low-head hydropower plant.

[0114] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0115] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for monitoring and handling active power oscillation faults in low-head hydropower plants, characterized in that, include: S1. Establish a real-time monitoring model based on mixed integer programming, introduce binary state variables and set an objective function, and determine whether the unit meets the rated power operation conditions by calculating the net water head and constraints. S2, based on a sliding time window, monitors the active power fluctuation of the unit in real time, calculates the range of actual generated values ​​within the window, and when the active power setpoint changes, the system delays for 3 minutes before counting the number of times the range of actual generated values ​​exceeds the preset fluctuation threshold and determines an alarm; when the setpoint does not change, when the range of actual generated values ​​exceeds the fluctuation threshold 10 times, it is determined to be a large fluctuation in active power and an alarm is triggered. S3 monitors the rate of change of oil pressure in the governor and the continuous running time of the oil pump. If the rate of drop in oil pressure exceeds the preset threshold and the running time of the oil pump exceeds the maximum allowable running time, an abnormal oil pressure alarm is triggered and the governor is switched to the local operation mode. S4. By comparing the current guide vane opening with the normal operating condition threshold range, if the guide vane opening exceeds the preset upper and lower limits or the number of times the guide vane opening direction changes within the sliding window exceeds the frequent tumbling threshold, a guide vane abnormality alarm is triggered and the active PID regulation is exited.

2. The method as described in claim 1, characterized in that, S1 includes: S11, Constraints In the middle, when Mandatory water purification head satisfy ,when The constraint is automatically made to hold by using a large positive number M. S12, Constraints In the middle, through real-time calculation The functional relationship, when Required power generation flow satisfy ,when The constraint is automatically set by using a large positive number M.

3. The method as described in claim 1, characterized in that, S2 includes: S21, the length of the sliding time window is 60 seconds, which includes the data of the most recent 60 sampling points; S22, the range is calculated by comparing the maximum and minimum values ​​of the real value sequence within the window point by point.

4. The method as described in claim 1, characterized in that, The S3 includes: S31, The oil pressure drop rate is calculated by dividing the difference between the current oil pressure and the previous oil pressure by the sampling time interval; S32, the oil pump running time is determined by comparing the cumulative continuous running time with the preset maximum allowable running time.

5. The method as described in claim 1, characterized in that, The S4 includes: S41, the statistics of the number of direction changes are obtained by calculating the first-order difference of the guide vane opening signal. And statistically analyze the relationship between adjacent sampling points. The number of sign changes; S42, the determination of the number of orientation changes within the sliding window adopts a configuration of a window length of 10 seconds and a threshold of 5 orientation changes.

6. A device for monitoring and handling active power oscillation faults in low-head hydropower plants, characterized in that, include: The real-time monitoring model building module is used to build a real-time monitoring model based on mixed integer programming. It introduces binary state variables and sets an objective function. By calculating the net head and constraints, it determines whether the unit meets the rated power operating conditions. The active power fluctuation monitoring module is used to monitor the active power fluctuation of the unit in real time based on a sliding time window, calculate the range of actual generated values ​​within the window, and when the active power setpoint changes, the system delays for 3 minutes before counting the number of times the range of actual generated values ​​exceeds the preset fluctuation threshold and determines to alarm; when the setpoint does not change, when the range of actual generated values ​​exceeds the fluctuation threshold 10 times, it is determined to be a large fluctuation in active power and an alarm is triggered. The governor oil pressure abnormality monitoring module is used to monitor the rate of change of governor oil pressure and the continuous running time of oil pump. If the rate of oil pressure drop exceeds the preset threshold and the oil pump running time exceeds the maximum allowable running time, an oil pressure abnormality alarm is triggered and the governor is switched to local operation mode. The guide vane anomaly monitoring module is used to compare the current guide vane opening with the normal operating condition threshold range. If the guide vane opening exceeds the preset upper and lower limits or the number of guide vane opening direction changes within the sliding window exceeds the frequent tumbling threshold, a guide vane anomaly alarm is triggered and the active PID regulation is exited.

7. The apparatus as claimed in claim 6, characterized in that, The real-time monitoring model building module is also used for: when Mandatory water purification head satisfy ,when When the constraint is made by a large positive number M Automatically established; when Required power generation flow satisfy ,when When the constraint is made by a large positive number M It was created automatically.

8. The apparatus as claimed in claim 6, characterized in that, The active power fluctuation monitoring module is also used for: The sliding time window is 60 seconds long and contains data from the most recent 60 sampling points; The range is calculated by comparing the maximum and minimum values ​​of the actual value sequence within the window point by point.

9. A computer device, characterized in that, Including processor and memory; The processor reads executable program code stored in the memory to run a program corresponding to the executable program code, so as to implement the active power oscillation fault monitoring and handling method of a low-head hydropower plant as described in any one of claims 1-5.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements a method for monitoring and handling active power oscillation faults in low-head hydropower plants as described in any one of claims 1-5.

Images