A monitoring and control system for the operation status of a hydropower station

Abstract

The present invention discloses a monitoring and control system for the operation status of a hydropower station, which relates to the technical field of monitoring and control of the operation status of a hydropower station, including a power demand data acquisition and preprocessing module, a power demand feature analysis and prediction module, a power demand classification module, a water level monitoring module, a high water level threshold dynamic control module, and a generator set operation strategy adjustment module; the power demand data acquisition and preprocessing module obtains real-time power demand information from the monitoring platform of the smart grid, and after obtaining the power demand information of the power grid. The present invention dynamically adjusts the high water level threshold of the reservoir by acquiring power demand information in real time and using a long short-term memory network for prediction. The threshold is lowered during peak power periods to increase power generation, and the threshold is increased during trough periods to reduce power generation and avoid waste of resources. Through real-time monitoring and automatic adjustment, the efficiency of water energy utilization is improved to ensure efficient operation of the power station and stability of the power system.

Description

Technical Field

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

[0002] The hydropower station operation status monitoring and control system is an integrated management system used to monitor and control the operation status of the hydropower station in real time. The system continuously collects key parameters such as water level, flow, unit speed, power generation, etc. through sensor networks and data acquisition devices, and uses intelligent algorithms to analyze these data to detect and warn of potential operational anomalies. The system also includes automated control functions that can automatically adjust the operating status of the generator set based on real-time monitoring data, optimize the utilization efficiency of water resources, ensure the safe and efficient operation of the power station, and reduce maintenance costs and risks.

[0003] The hydropower station operation status monitoring and control system will automatically adjust the operating load of the generator set according to the change of water level during operation, as follows:

[0004] When the water level is too high:

[0005] The system will detect that the water level of the reservoir rises above the set high water level warning line, which usually means that there is enough water available for power generation. In order to prevent the reservoir from overflowing and make full use of the water energy brought by the high water level, the system will automatically increase the load of the generator set. This means increasing the speed of the unit or adding more generator sets to participate in the power generation process, thereby improving the power generation power and efficiency and speeding up the consumption of excess water.

[0006] When the water level is too low:

[0007] When the system detects that the water level in the reservoir has dropped below the set low water level warning line, it indicates that the available water resources are limited. In order to protect the power generation equipment and ensure the rational use of water resources, the system will automatically reduce the power generation load. This can reduce water consumption by reducing the speed of the generator set, reducing the number of units involved in power generation, or directly shutting down when necessary. In this way, the system can protect water resources and prevent the reservoir water level from further dropping to a level where power generation is not possible.

[0008] The prior art has the following deficiencies:

[0009] Existing technologies usually set a fixed high water level threshold based on experience, which is the maximum limit of the water level allowed in the reservoir of a hydropower station, and is used to determine the operating load and power generation of the generator set. When the water level reaches or exceeds this threshold, the power station will increase the power generation load to fully utilize the water energy and prevent overflow to ensure effective power generation. During peak power demand periods (such as hot summer days or cold winter nights), the demand on the power grid increases sharply. If the power station continues to operate according to the fixed high water level threshold, it may not be able to quickly increase power generation, resulting in insufficient power supply and even power outages, endangering the stability of the power system. During low power demand periods (such as late at night), power consumption is greatly reduced, but the fixed high water level threshold may cause the power station to continue to generate power at high load, resulting in excess power and energy waste. If the excess power cannot be absorbed by the power grid, the power station may have to shut down or reduce the load, affecting operating efficiency and equipment life.

[0010] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0011] The purpose of the present invention is to provide an operation status monitoring and control system for a hydropower station, which obtains power demand information in real time from a smart grid monitoring platform and uses a long short-term memory network to predict power demand, accurately identifying the peak period, normal operation period and valley period of power demand. Based on these prediction results, the high water level threshold of the reservoir is dynamically adjusted, and the threshold is lowered during the peak period of power demand to increase power generation and avoid power shortages; the threshold is raised during the valley period to reduce power generation load or shutdown, and avoid water resource waste and equipment wear. Through real-time monitoring and automated strategy adjustment, the system effectively improves water energy utilization efficiency and power generation benefits, ensures the efficient operation of the hydropower station and the stability of the power system, so as to solve the problems in the above-mentioned background technology.

[0012] In order to achieve the above-mentioned object, the present invention provides the following technical solutions: a hydropower station operation status monitoring and control system, including a power demand data acquisition and preprocessing module, a power demand feature analysis and prediction module, a power demand classification module, a water level monitoring module, a high water level threshold dynamic control module and a generator set operation strategy adjustment module;

[0013] The power demand data acquisition and preprocessing module obtains real-time power demand information from the monitoring platform of the smart grid, and performs preliminary processing on the acquired data after obtaining the power demand information of the power grid;

[0014] The power demand feature analysis and prediction module analyzes the pre-processed power demand information, extracts the key features that reflect the power demand of the power grid, analyzes the extracted key features, understands the structure and trend of power demand, and then inputs the extracted key features into the pre-trained long short-term memory network, and uses the long short-term memory network to make short-term predictions on the power demand of the power grid;

[0015] The power demand classification module divides the power demand into three types based on the prediction results of the long short-term memory network: peak power demand period, normal operation period and low power demand period;

[0016] The water level monitoring module monitors the water level of the hydropower station in real time, uses the water level sensor to collect the real-time data of the reservoir water level, and transmits it to the central control system through the data acquisition system to update the water level information in real time;

[0017] The high water level threshold dynamic control module dynamically controls the high water level threshold of the hydropower station according to the type of power demand. It lowers the high water level threshold during the peak period of power demand to increase the power generation load; it raises the high water level threshold during the low period of power demand to reduce the power generation load or shut down the power station.

[0018] The generator set operation strategy adjustment module automatically adjusts the generator set operation strategy based on the dynamically adjusted high water level threshold to avoid wasting water resources during the low electricity demand period and maximize power generation during the peak electricity demand period.

[0019] Preferably, real-time power demand information is obtained from the monitoring platform of the smart grid, and the specific steps are as follows:

[0020] Establishing the connection between the smart grid monitoring platform and the hydropower station control system;

[0021] Once the connection is established, the real-time power demand data is obtained from the smart grid monitoring platform through the hydropower station control system;

[0022] The acquired data is decoded by the hydropower station control system, and the received raw data is converted into a format that can be directly understood and used by the power station control system.

[0023] Preferably, the preprocessed power demand information is analyzed to extract the power load situation of the power grid and the maximum power demand of the power grid within the monitoring time window, and after analyzing the power load situation and the maximum power demand of the power grid within the monitoring time window, load quantization values ​​and peak demand quantization values ​​are generated respectively, and the load quantization values ​​and peak demand quantization values ​​are input into a pre-trained long short-term memory network to generate a power demand prediction index, and the power demand of the power grid within the monitoring time window is predicted in the short term through the power demand prediction index.

[0024] Preferably, the power demand forecast index generated after analyzing the power demand information of the power grid within the monitoring time window is compared and analyzed with a preset reference threshold range of the power demand forecast index to divide the power demand. The specific division steps are as follows:

[0025] If the power demand forecast index is within the power demand forecast index reference threshold range, the power demand in the monitoring time window is divided into a normal operation period;

[0026] If the power demand forecast index is greater than the maximum value of the power demand forecast index reference threshold range, the power demand in the monitoring time window is classified as a power demand peak period;

[0027] If the power demand forecast index is less than the minimum value of the power demand forecast index reference threshold range, the power demand in the monitoring time window is classified as a power demand low period.

[0028] Preferably, the specific steps of generating a load quantization value after analyzing the power load situation of the power grid within the monitoring time window are as follows:

[0029] In the monitoring time window, the instantaneous load value of the power grid is recorded at a fixed time interval, which is recorded as , where t is the time point, represents the instantaneous load value at time t, and , n is the total number of data points within the monitoring time window;

[0030] The load change rate is calculated by the instantaneous load value. The load change rate is defined as the rate of change of the instantaneous load value between two adjacent time points. The calculation expression is: , where Indicates the load change rate, Indicates a time interval;

[0031] Calculate the load fluctuation amplitude within the monitoring window to reflect the severity of load changes. The calculation expression is: , where Indicates the load fluctuation amplitude, which indicates the load change amplitude at time t. is the absolute change in load, which indicates the absolute change in load between two time points. It indicates the absolute value of the rate of load change, ensuring that the calculated amplitude is not affected by direction;

[0032] A fluctuation adjustment coefficient is introduced for the load fluctuation amplitude to emphasize the impact of large fluctuation periods on the total load. The calculation expression of the fluctuation adjustment coefficient is: , where It indicates the fluctuation adjustment coefficient, which amplifies or reduces the impact of load fluctuation on the total load. e is the base of natural logarithms, used for exponential calculations, k It represents the sensitivity adjustment parameter, which controls the reaction degree of the fluctuation adjustment coefficient to the load fluctuation amplitude;

[0033] The load quantification value within the monitoring time window is calculated by combining the load fluctuation amplitude and the fluctuation adjustment coefficient. The calculation expression is: , where It represents the quantified value of load, representing the overall load situation of the power grid within the monitoring time window. It is the weighted sum of the load value at each time point and the fluctuation adjustment coefficient, reflecting the comprehensive impact of load changes.

[0034] Preferably, the specific steps of generating a peak demand quantization value after analyzing the maximum power demand of the power grid within the monitoring time window are as follows:

[0035] Obtain real-time power demand data within the monitoring time window from the power grid monitoring system and convert the real-time power demand data into time series. To express, represents the power demand at time t;

[0036] Calculate the time-weighted average load within the monitoring window. The calculation expression is:

[0037] ,

[0038] In the formula, It represents the time-weighted average load, reflecting the comprehensive level of power demand in the entire monitoring window. Indicates time point The weight of reflects the relative importance of this time point in the overall load. u The index of the time point. N Indicates the total number of time points within the monitoring time window;

[0039] The peak load deviation coefficient is calculated to measure the deviation between the maximum power demand and the time-weighted average load. The peak load deviation coefficient reflects the relative intensity of load fluctuations in the power grid within the monitoring time window. The calculation expression is:

[0040] ,

[0041] Indicates the maximum power demand within the monitoring time window. is the peak load deviation coefficient, which indicates the degree of deviation of the peak load relative to the time-weighted average load. It represents the standard deviation of the load within the monitoring window, and measures the magnitude of load fluctuation;

[0042] In the monitoring time window, the peak demand quantification value is calculated, and the calculation expression is:

[0043] ,

[0044] In the formula, It represents the quantitative value of peak demand, reflecting the peak load characteristics of the power grid within the monitoring time window. It represents the logarithmic ratio of peak load to average load, and is used to measure the relative size of peak load to average load. Indicates the load change rate, reflecting the degree of load fluctuation within the monitoring time window. The peak load deviation coefficient , the logarithmic ratio of peak load to average load and load change rate The weight coefficient reflects the relative importance of the peak load deviation coefficient, the logarithmic ratio of peak load to average load, and the load change rate in the peak demand quantification value.

[0045] Preferably, the high water level threshold of the hydropower station is dynamically regulated according to the type of power demand, and the specific process is as follows:

[0046] The deviation rate of the power demand forecast index relative to the center point of the power demand forecast index reference threshold range is calculated using the following expression: , where Represents the central value of the reference threshold range of the power demand forecast index, , and They represent the maximum and minimum values ​​of the reference threshold range of the power demand forecast index, It represents the current electricity demand forecast index, reflecting the estimated value of current electricity demand. It represents the deviation rate of the electricity demand forecast index, which is used to evaluate the relative change of current demand with respect to the reference level;

[0047] The high water level threshold adjustment factor is determined based on the comparison result between the power demand forecast index and the power demand forecast index reference threshold range.

[0048] Preferably, during peak power demand periods, the factor is adjusted according to the high water level threshold , calculate the new high water level threshold, which is used to lower the high water level threshold and increase the power generation load. The reduction of the high water level threshold will cause the reservoir to release more water for power generation to meet higher power demand. The calculation expression is: , where Indicates the initial high water level threshold, that is, the default water level limit when it is not adjusted. It represents the high water level threshold adjusted during the peak electricity demand period, which is used to cope with the peak electricity demand period;

[0049] During the low electricity demand period, the factor is also adjusted according to the high water level threshold. , calculate the new high water level threshold, the calculation expression is: , where It indicates the high water level threshold after adjustment during the low electricity demand period, which is used to cope with the low electricity demand period;

[0050] Calculate the control strategy update value, and adjust the control strategy of the hydropower station according to the control strategy update value to ensure that the water level regulation meets the current power demand. The calculation expression of the update value is:

[0051] ,

[0052] ,

[0053] In the formula, is the control strategy update value during the peak period of power demand, indicating the relative proportion of the change in the water level threshold. is the control strategy update value during the low power demand period, which also indicates the relative proportion of the water level threshold change;

[0054] The initial high water threshold operation continues to be maintained for power demand during normal operation.

[0055] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0056] The present invention obtains power demand information from the smart grid monitoring platform in real time, combines with advanced long-short term memory networks to predict power demand, accurately identifies the peak period, normal operation period and valley period of power demand, and dynamically adjusts the high water level threshold of the reservoir based on these prediction results, ensuring that during the peak period of power demand, by lowering the high water level threshold, the power generation is quickly increased to meet the power demand of the power grid and avoid the risks of insufficient supply and system instability; and during the valley period of power demand, by raising the high water level threshold, unnecessary power generation load or shutdown is reduced, water resource waste is avoided, and equipment life is extended. At the same time, through real-time monitoring and automatic operation strategy adjustment, it is possible to maximize the efficiency of water energy utilization, optimize power generation and economic benefits, improve the overall operation efficiency and response capability of the hydropower station, and ensure the safe and stable operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0058] Figure 1 The present invention is a module schematic diagram of a hydropower station operation status monitoring and control system. DETAILED DESCRIPTION

[0059] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0060] The present invention provides Figure 1 A hydropower station operation status monitoring and control system shown includes a power demand data acquisition and preprocessing module, a power demand feature analysis and prediction module, a power demand classification module, a water level monitoring module, a high water level threshold dynamic control module, and a generator set operation strategy adjustment module;

[0061] The power demand data acquisition and preprocessing module obtains real-time power demand information from the monitoring platform of the smart grid, and performs preliminary processing on the acquired data after obtaining the power demand information of the power grid;

[0062] To obtain real-time power demand information from the smart grid monitoring platform, the specific steps are as follows:

[0063] Establishing the connection between the smart grid monitoring platform and the hydropower station control system;

[0064] Through dedicated communication interfaces or protocols, such as the IEC 61850 standard, the smart grid monitoring platform can share power demand data in real time. Usually, this requires the deployment of a data acquisition and communication module at the power station, which can communicate directly with the monitoring platform of the grid dispatch center. This connection must not only have high bandwidth to ensure the real-time nature of the data, but also high reliability and security to prevent delays or data loss during data transmission.

[0065] Once the connection is established, the hydropower station control system will obtain real-time power demand data from the smart grid monitoring platform;

[0066] This data usually includes current grid load, historical load curves, power demand forecasts, demand fluctuations caused by emergencies, etc. The system can continuously receive this data through API calls or through the timed data transmission mechanism in the protocol. This process ensures that the power station always has the latest power demand situation, thus providing a reliable basis for subsequent decision-making.

[0067] The acquired data may be transmitted in a specific format or encoding, and the hydropower station control system will decode the acquired data;

[0068] Through dedicated decoding algorithms and data processing software, the received raw data is converted into a format that can be directly understood and used by the power plant control system.

[0069] After obtaining the power demand information of the power grid, the acquired data is initially processed, which means basic data cleaning and formatting operations are performed on the raw power demand data just received, including removing noise from the data, processing outliers, filling missing data, and converting the data into a unified format for subsequent analysis and use. Preliminary processing ensures the accuracy and completeness of the data, providing a reliable data basis for further demand forecasting and power generation strategy adjustment.

[0070] The power demand feature analysis and prediction module analyzes the pre-processed power demand information, extracts the key features that reflect the power demand of the power grid, analyzes the extracted key features, understands the structure and trend of power demand, and then inputs the extracted key features into the pre-trained long short-term memory network, and uses the long short-term memory network to make short-term predictions on the power demand of the power grid;

[0071] The preprocessed power demand information is analyzed to extract the power load situation and the maximum power demand of the power grid within the monitoring time window. After analyzing the power load situation and the maximum power demand of the power grid within the monitoring time window, load quantization values ​​and peak demand quantization values ​​are generated respectively. The load quantization values ​​and peak demand quantization values ​​are input into a pre-trained long short-term memory network to generate a power demand prediction index. The power demand of the power grid within the monitoring time window is predicted in the short term through the power demand prediction index.

[0072] The pre-trained LSTM network is a deep learning model specifically designed to process and predict time series data. By introducing memory units and gate mechanisms, it can effectively capture and utilize long-term dependencies and complex temporal dynamics in data. The pre-trained LSTM network means that the model has been trained and optimized on a large amount of historical power demand data, enabling it to identify and learn patterns and trends in power demand changes. By inputting load quantization values ​​and peak demand quantization values ​​into the pre-trained LSTM network, the model can generate a power demand forecast index, accurately predict short-term power demand changes in the power grid within the monitoring window, and help power grid operators optimize power dispatch and resource management.

[0073] Changes in grid load often directly reflect changes in electricity demand. An increase in grid load means that the grid is consuming more electricity, which often indicates that electricity demand is rising. For example, during hot summer days, heavy use of air conditioning and cooling equipment can cause a significant increase in grid load, reflecting rising electricity demand. Similarly, factors such as increased industrial activity, accelerated economic development, or unexpected events (such as large public events) can also lead to an increase in grid load, further indicating an increase in electricity demand. Conversely, a decrease in grid load often indicates a decrease in electricity demand. For example, during the late night or holidays, household and industrial users use less electricity, resulting in a decrease in grid load. This load reduction may be caused by the shutdown of electrical equipment, a decrease in economic activity, or changes in climatic conditions (such as warm days in winter that reduce heating demand). In short, increases and decreases in grid load are a direct reflection of changes in electricity demand, with increases in load pointing to higher electricity demand, while decreases in load indicate a decrease in electricity demand. Understanding this load-demand relationship is important for the dispatch and optimization of power systems, helping to ensure the stability and efficient operation of the power grid.

[0074] The specific steps for generating load quantification values ​​after analyzing the power load situation of the power grid within the monitoring time window are as follows:

[0075] In the monitoring time window, the instantaneous load value (power consumption) of the power grid is recorded at a fixed time interval, which is recorded as , where t is the time point, represents the instantaneous load value at time t, and , n To monitor the total number of data points in the time window, the load change rate is calculated by the instantaneous load value. The load change rate is defined as the rate of change of the instantaneous load value between two adjacent time points. The calculation expression is: , where Indicates the load change rate, Indicates a time interval, usually a fixed duration;

[0076] Calculate the load fluctuation amplitude within the monitoring window to reflect the severity of load changes. The calculation expression is: , where Indicates the load fluctuation amplitude, which indicates the load change amplitude at time t. is the absolute change in load, which indicates the absolute change in load between two time points. It indicates the absolute value of the rate of load change, ensuring that the calculated amplitude is not affected by direction;

[0077] A fluctuation adjustment coefficient is introduced for the load fluctuation amplitude to emphasize the impact of large fluctuation periods on the total load. The calculation expression of the fluctuation adjustment coefficient is: , where It indicates the fluctuation adjustment coefficient, which amplifies or reduces the impact of load fluctuation on the total load. e is the base of the natural logarithm (approximately equal to 2.718), used for exponential calculations, k It represents the sensitivity adjustment parameter, which controls the reaction degree of the fluctuation adjustment coefficient to the load fluctuation amplitude. Usually, k The value range of is determined by empirical data;

[0078] The load quantification value within the monitoring time window is calculated by combining the load fluctuation amplitude and the fluctuation adjustment coefficient. The calculation expression is: , where It represents the quantified value of load, representing the overall load situation of the power grid within the monitoring time window. It is the weighted sum of the load value at each time point and the fluctuation adjustment coefficient, reflecting the comprehensive impact of load changes.

[0079] The load quantification value is used to quantify the power load of the power grid within the monitoring time window. The load quantification value generated by analyzing the load situation of the power grid reflects the level of power demand. Specifically, the larger the performance value of the load quantification value, the higher the power load of the power grid within the monitoring time window, indicating a strong demand for electricity; on the contrary, the smaller the performance value of the load quantification value, the lower the power load of the power grid within the monitoring time window, indicating a low demand for electricity. Therefore, the load quantification value provides an effective way to monitor and evaluate changes in power demand, which helps optimize power dispatching and energy management.

[0080] Changes in the maximum power demand directly reflect the rise or fall in power demand. The maximum power demand refers to the maximum amount of power that the power grid needs to supply during a specific period of time. It usually occurs during peak power demand periods, such as during hot summer days or cold winter nights, when air conditioners, heating equipment, and other power-consuming devices are heavily used. When the maximum power demand increases, it means that the power grid needs to provide more power to meet user needs, which usually reflects an overall increase in power demand. Possible reasons include weather changes, increased economic activity, or temporary peaks in power demand caused by certain special events. Conversely, when the maximum power demand decreases, it means that the load pressure on the power grid is reduced and the overall power demand is on a downward trend. This usually occurs during low power demand periods, such as late at night or during periods of relatively slow economic activity. A decrease in the maximum power demand may reflect a reduction in user demand for power, or it may be caused by energy-saving measures or the application of distributed energy generation. Changes in the maximum power demand are an important indicator for evaluating the dynamics of power demand and play a key role in the stable operation of the power grid and the formulation of power supply strategies.

[0081] The specific steps of generating the peak demand quantification value after analyzing the highest power demand of the power grid within the monitoring time window are as follows:

[0082] Obtain real-time power demand data within the monitoring time window from the power grid monitoring system and convert the real-time power demand data into time series. To express, represents the power demand (power) at time t;

[0083] This data usually includes the electrical load (power demand) per minute or per second.

[0084] Calculate the time-weighted average load within the monitoring window. The calculation expression is: , where It represents the time-weighted average load, reflecting the comprehensive level of power demand in the entire monitoring window. Indicates time point The weight reflects the relative importance of this time point in the overall load. The weight can be based on the actual demand change rate or strategy setting. u The index of the time point. N Indicates the total number of time points within the monitoring time window;

[0085] The peak load deviation coefficient is calculated to measure the deviation between the maximum power demand and the time-weighted average load. The peak load deviation coefficient reflects the relative intensity of load fluctuations in the power grid within the monitoring time window. The calculation expression of the peak load deviation coefficient is: , Indicates the highest power demand (peak load) within the monitoring time window. is the peak load deviation coefficient, which indicates the degree of deviation of the peak load relative to the time-weighted average load. It represents the standard deviation of the load within the monitoring window, and measures the magnitude of load fluctuation;

[0086] In the monitoring time window, the peak demand quantification value is calculated, and the calculation expression is:

[0087] ,

[0088] In the formula, It represents the quantitative value of peak demand, reflecting the peak load characteristics of the power grid within the monitoring time window. It represents the logarithmic ratio of peak load to average load, and is used to measure the relative size of peak load to average load. Indicates the load change rate, reflecting the degree of load fluctuation within the monitoring time window. The peak load deviation coefficient , the logarithmic ratio of peak load to average load and load change rate The weight coefficient reflects the relative importance of the peak load deviation coefficient, the logarithmic ratio of the peak load to the average load, and the load change rate in the peak demand quantification value. The weight coefficient can be adjusted according to specific needs and is not specifically limited here.

[0089] The peak demand quantification value is an indicator generated by analyzing the highest power demand of the power grid within the monitoring time window, which is used to quantify the intensity of power demand. When the peak demand quantification value is large, it indicates that the power demand of the power grid in the monitoring time window has reached its peak, which means that the user's power consumption is large and more power supply is needed to meet the demand. On the contrary, when the peak demand quantification value is small, it means that the power demand of the power grid in the monitoring time window is low, the power consumption is relatively small, and the load pressure of the power grid is small. Therefore, the high or low peak demand quantification value directly reflects the power demand of the power grid in a specific time period, which is helpful for power management and scheduling decisions.

[0090] The long short-term memory network is not specifically limited here, and can realize the load quantization value and peak demand quantification Conduct comprehensive analysis to generate power demand forecast index In order to realize the technical solution of the present invention, the present invention provides a specific implementation method;

[0091] Electricity Demand Forecast Index The resulting calculation formula is:

[0092] , where , Load quantification values and peak demand quantification The preset scaling factor of , Both are greater than 0.

[0093] It can be seen from the power demand forecasting index that the greater the performance value of the load quantization value generated after analyzing the power load situation of the power grid within the monitoring time window, the greater the performance value of the peak demand quantization value generated after analyzing the maximum power demand of the power grid within the monitoring time window. In other words, the greater the performance value of the power demand forecasting index generated after analyzing the power demand information of the power grid within the monitoring time window, the greater the power demand of the power grid within the monitoring time window, and vice versa.

[0094] The power demand classification module divides the power demand into three types based on the prediction results of the long short-term memory network: peak power demand period, normal operation period and low power demand period;

[0095] The power demand forecast index generated after analyzing the power demand information of the power grid within the monitoring time window is compared and analyzed with the pre-set reference threshold range of the power demand forecast index to divide the power demand. The specific division steps are as follows:

[0096] If the power demand forecast index is within the power demand forecast index reference threshold range, the power demand in the monitoring time window is divided into a normal operation period;

[0097] If the power demand forecast index is greater than the maximum value of the power demand forecast index reference threshold range, the power demand in the monitoring time window is classified as a power demand peak period;

[0098] If the power demand forecast index is less than the minimum value of the power demand forecast index reference threshold range, the power demand in the monitoring time window is classified as a power demand low period.

[0099] The water level monitoring module monitors the water level of the hydropower station in real time, uses the water level sensor to collect the real-time data of the reservoir water level, and transmits it to the central control system through the data acquisition system to update the water level information in real time;

[0100] The water level sensor is a device used to measure and monitor the height of liquid water levels. It is widely used in hydropower stations, reservoirs, rivers, and groundwater monitoring. It senses water level changes through physical principles (such as buoyancy, capacitance, conductivity, ultrasound, pressure, etc.) and converts the physical quantity of water level into electrical signals. These signals are then transmitted to the central control system through the data acquisition system to update the water level information of the reservoir in real time. The high precision and real-time performance of the water level sensor ensure the effective operation and safe management of the hydropower station, enabling the power station to dynamically adjust the power generation strategy according to the water level changes and optimize the use of water resources.

[0101] The high water level threshold dynamic control module dynamically controls the high water level threshold of the hydropower station according to the type of power demand. It lowers the high water level threshold during the peak period of power demand to increase the power generation load; it raises the high water level threshold during the low period of power demand to reduce the power generation load or shut down the power station.

[0102] The high water level threshold of the hydropower station is dynamically regulated according to the type of power demand. The specific process is as follows:

[0103] The deviation rate of the power demand forecast index relative to the center point of the power demand forecast index reference threshold range is calculated using the following expression: , where Represents the central value of the reference threshold range of the power demand forecast index, , and They represent the maximum and minimum values ​​of the reference threshold range of the power demand forecast index, It represents the current electricity demand forecast index, reflecting the estimated value of current electricity demand. It represents the deviation rate of the electricity demand forecast index, which is used to evaluate the relative change of current demand with respect to the reference level;

[0104] The high water level threshold adjustment factor is determined based on the comparison result between the power demand forecast index and the power demand forecast index reference threshold range.

[0105] Determining a high water level threshold adjustment factor based on a comparison result between the power demand forecast index and a reference threshold range of the power demand forecast index;

[0106] Determine the high water level threshold adjustment factor based on the comparison between the power demand forecast index and the power demand forecast index reference threshold range If the power demand forecast index is greater than the maximum value of the power demand forecast index reference threshold range (i.e. during the peak power demand period), will be greater than 1, indicating that the high water level threshold needs to be lowered to increase the power generation load. On the contrary, if the power demand forecast index is greater than the minimum value of the power demand forecast index reference threshold range (i.e., the power demand is low), will be less than 1, indicating that the high water level threshold needs to be increased to reduce the generation load.

[0107] During peak electricity demand periods, the factor is adjusted based on the high water level threshold , calculate the new high water level threshold, which is used to lower the high water level threshold and increase the power generation load. The reduction of the high water level threshold will cause the reservoir to release more water for power generation to meet higher power demand. The calculation expression is: , where Indicates the initial high water level threshold, that is, the default water level limit when it is not adjusted. It represents the high water level threshold adjusted during the low electricity demand period to cope with the peak electricity demand period;

[0108] During the low electricity demand period, the factor is also adjusted according to the high water level threshold. , calculate the new high water level threshold. At this time, the high water level threshold adjustment factor is usually less than 1, which causes the high water level threshold to rise, reduce the power generation load or shut down, retain more water resources, and avoid waste. The calculation expression is: , where It indicates the high water level threshold after adjustment during the low electricity demand period, which is used to cope with the low electricity demand period;

[0109] Calculate the control strategy update value, which represents the change ratio of the adjusted high water level threshold relative to the initial high water level threshold. According to the control strategy update value, adjust the control strategy of the hydropower station, such as increasing or decreasing the load of the generator set, to ensure that the water level control meets the current power demand. The calculation expression of the update value is:

[0110] , , where is the control strategy update value during the peak period of power demand, indicating the relative proportion of the change in the water level threshold. is the control strategy update value during the low power demand period, and also represents the relative proportion of the water level threshold change.

[0111] Continuing to operate at the initial high water level threshold for power demand during normal operation means that if there is no significant change in power demand, the hydropower station will perform regular operations according to the preset high water level threshold. Under this strategy, the power station does not need to make additional adjustments or controls to keep the water level of the reservoir within a stable range to ensure the normal operation of the generator set and the effective use of water resources. During normal operation, power demand is stable, neither reaching a peak nor entering a trough, so there is no need to increase or decrease the power generation load. Maintaining the initial high water level threshold can maintain the balance of the system, avoid unnecessary fluctuations caused by frequent adjustments, and ensure the continuity of power supply and the stability of the system.

[0112] The generator set operation strategy adjustment module automatically adjusts the generator set operation strategy based on the dynamically adjusted high water level threshold to avoid wasting water resources during the low power demand period and maximize power generation during the peak power demand period;

[0113] Adjustment of operation strategy during the low electricity demand period:

[0114] During the low electricity demand period, the power grid consumes less electricity, so it is necessary to reduce power generation to avoid wasting water resources. In this case, the automatic adjustment of the operation strategy of the generator set includes reducing the output load of the generator set, shutting down some generator sets, or suspending power generation operations. By raising the high water level threshold, the system allows the reservoir to maintain a higher water level, reducing the amount of water released, thereby retaining more water resources. Low-load operation or shutdown of generator sets can not only save water resources, but also reduce equipment wear, reduce operating costs, and extend the service life of equipment.

[0115] Adjustment of operation strategy during peak electricity demand period:

[0116] During peak electricity demand periods, the power demand of the power grid rises sharply, and the power generation needs to be maximized to meet the grid load. In this case, the operation strategy of automatically adjusting the generator set includes increasing the output load of the generator set, enabling the standby generator set, or putting all available units into power generation. By lowering the high water level threshold, the system increases the release of water and drives the generator set to operate under high load to maximize the utilization of water energy. This strategy ensures that during peak electricity demand periods, the hydropower station can provide sufficient electricity, maintain the stability of the power grid, and meet the electricity needs of users.

[0117] The present invention obtains power demand information from the smart grid monitoring platform in real time, combines with advanced long-short term memory networks to predict power demand, accurately identifies the peak period, normal operation period and valley period of power demand, and dynamically adjusts the high water level threshold of the reservoir based on these prediction results, ensuring that during the peak period of power demand, by lowering the high water level threshold, the power generation is quickly increased to meet the power demand of the power grid and avoid the risks of insufficient supply and system instability; and during the valley period of power demand, by raising the high water level threshold, unnecessary power generation load or shutdown is reduced, water resource waste is avoided, and equipment life is extended. At the same time, through real-time monitoring and automatic operation strategy adjustment, it is possible to maximize the efficiency of water energy utilization, optimize power generation and economic benefits, improve the overall operation efficiency and response capability of the hydropower station, and ensure the safe and stable operation of the power system.

[0118] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0119] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0120] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A hydropower station operation status monitoring and control system, characterized in that: It includes power demand data acquisition and preprocessing module, power demand feature analysis and prediction module, power demand classification module, water level monitoring module, high water level threshold dynamic control module and generator set operation strategy adjustment module; The power demand data acquisition and preprocessing module obtains real-time power demand information from the monitoring platform of the smart grid, and performs preliminary processing on the acquired data after obtaining the power demand information of the power grid; The power demand feature analysis and prediction module analyzes the pre-processed power demand information, extracts the key features that reflect the power demand of the power grid, analyzes the extracted key features, understands the structure and trend of power demand, and then inputs the extracted key features into the pre-trained long short-term memory network, and uses the long short-term memory network to make short-term predictions on the power demand of the power grid; The power demand classification module divides the power demand into three types based on the prediction results of the long short-term memory network: peak power demand period, normal operation period and low power demand period; The water level monitoring module monitors the water level of the hydropower station in real time, uses the water level sensor to collect the real-time data of the reservoir water level, and transmits it to the central control system through the data acquisition system to update the water level information in real time; The high water level threshold dynamic control module dynamically controls the high water level threshold of the hydropower station according to the type of power demand. It lowers the high water level threshold during the peak period of power demand to increase the power generation load; it raises the high water level threshold during the low period of power demand to reduce the power generation load or shut down the power station. The generator set operation strategy adjustment module automatically adjusts the generator set operation strategy based on the dynamically adjusted high water level threshold to avoid wasting water resources during the low power demand period and maximize power generation during the peak power demand period; The preprocessed power demand information is analyzed to extract the power load situation and the maximum power demand of the power grid within the monitoring time window. After analyzing the power load situation and the maximum power demand of the power grid within the monitoring time window, load quantization values ​​and peak demand quantization values ​​are generated respectively. The load quantization values ​​and peak demand quantization values ​​are input into a pre-trained long short-term memory network to generate a power demand prediction index. The power demand of the power grid within the monitoring time window is predicted in the short term through the power demand prediction index.

2. A hydropower station operation status monitoring and control system according to claim 1, characterized in that: To obtain real-time power demand information from the smart grid monitoring platform, the specific steps are as follows: Establishing the connection between the smart grid monitoring platform and the hydropower station control system; Once the connection is established, the real-time power demand data is obtained from the smart grid monitoring platform through the hydropower station control system; The acquired data is decoded by the hydropower station control system, and the received raw data is converted into a format that can be directly understood and used by the power station control system.

3. A hydropower station operation status monitoring and control system according to claim 1, characterized in that: The power demand forecast index generated after analyzing the power demand information of the power grid within the monitoring time window is compared and analyzed with the pre-set reference threshold range of the power demand forecast index to divide the power demand. The specific division steps are as follows: If the power demand forecast index is within the power demand forecast index reference threshold range, the power demand in the monitoring time window is divided into a normal operation period; If the power demand forecast index is greater than the maximum value of the power demand forecast index reference threshold range, the power demand in the monitoring time window is classified as a power demand peak period; If the power demand forecast index is less than the minimum value of the power demand forecast index reference threshold range, the power demand in the monitoring time window is classified as a power demand low period.

4. A hydropower station operation status monitoring and control system according to claim 1, characterized in that: The specific steps for generating load quantification values ​​after analyzing the power load situation of the power grid within the monitoring time window are as follows: In the monitoring time window, the instantaneous load value of the power grid is recorded at a fixed time interval, which is recorded as , where t is the time point, represents the instantaneous load value at time t, and , n is the total number of data points within the monitoring time window; The load change rate is calculated by the instantaneous load value. The load change rate is defined as the rate of change of the instantaneous load value between two adjacent time points. The calculation expression is: , where Indicates the load change rate, Indicates a time interval; Calculate the load fluctuation amplitude within the monitoring window to reflect the severity of load changes. The calculation expression is: , where Indicates the load fluctuation amplitude, which indicates the load change amplitude at time t. is the absolute change in load, which indicates the absolute change in load between two time points. It indicates the absolute value of the rate of load change, ensuring that the calculated amplitude is not affected by direction; A fluctuation adjustment coefficient is introduced for the load fluctuation amplitude to emphasize the impact of large fluctuation periods on the total load. The calculation expression of the fluctuation adjustment coefficient is: , where It indicates the fluctuation adjustment coefficient, which amplifies or reduces the impact of load fluctuation on the total load. e is the base of natural logarithms, used for exponential calculations, k It represents the sensitivity adjustment parameter, which controls the reaction degree of the fluctuation adjustment coefficient to the load fluctuation amplitude; The load quantification value within the monitoring time window is calculated by combining the load fluctuation amplitude and the fluctuation adjustment coefficient. The calculation expression is: , where It represents the quantified value of load, representing the overall load situation of the power grid within the monitoring time window. It is the weighted sum of the load value at each time point and the fluctuation adjustment coefficient, reflecting the comprehensive impact of load changes.

5. A hydropower station operation status monitoring and control system according to claim 1, characterized in that: The specific steps of generating the peak demand quantification value after analyzing the highest power demand of the power grid within the monitoring time window are as follows: Obtain real-time power demand data within the monitoring time window from the power grid monitoring system and convert the real-time power demand data into time series. To express, represents the power demand at time t; Calculate the time-weighted average load within the monitoring window. The calculation expression is: ， In the formula, It represents the time-weighted average load, reflecting the comprehensive level of power demand in the entire monitoring window. Indicates time point The weight of reflects the relative importance of this time point in the overall load. u The index of the time point. N Indicates the total number of time points within the monitoring window; The peak load deviation coefficient is calculated to measure the deviation between the maximum power demand and the time-weighted average load. The peak load deviation coefficient reflects the relative intensity of load fluctuations in the power grid within the monitoring time window. The calculation expression is: ， Indicates the maximum power demand within the monitoring time window. is the peak load deviation coefficient, which indicates the degree of deviation of the peak load relative to the time-weighted average load. It represents the standard deviation of the load within the monitoring window, and measures the magnitude of load fluctuation; In the monitoring time window, the peak demand quantification value is calculated, and the calculation expression is: ， In the formula, It represents the quantitative value of peak demand, reflecting the peak load characteristics of the power grid within the monitoring time window. It represents the logarithmic ratio of peak load to average load, and is used to measure the relative size of peak load to average load. Indicates the load change rate, reflecting the degree of load fluctuation within the monitoring time window. The peak load deviation coefficient , the logarithmic ratio of peak load to average load and load change rate The weight coefficient reflects the relative importance of the peak load deviation coefficient, the logarithmic ratio of peak load to average load, and the load change rate in the peak demand quantification value.

6. A hydropower station operation status monitoring and control system according to claim 3, characterized in that: The high water level threshold of the hydropower station is dynamically regulated according to the type of power demand. The specific process is as follows: The deviation rate of the power demand forecast index relative to the center point of the power demand forecast index reference threshold range is calculated using the following expression: , where Represents the central value of the reference threshold range of the power demand forecast index, , and They represent the maximum and minimum values ​​of the reference threshold range of the power demand forecast index, It represents the current electricity demand forecast index, reflecting the estimated value of current electricity demand. It represents the deviation rate of the electricity demand forecast index, which is used to evaluate the relative change of current demand with respect to the reference level; The high water level threshold adjustment factor is determined based on the comparison result between the power demand forecast index and the power demand forecast index reference threshold range.

7. A hydropower station operation status monitoring and control system according to claim 6, characterized in that: During peak electricity demand periods, the factor is adjusted based on the high water level threshold , calculate the new high water level threshold, which is used to lower the high water level threshold and increase the power generation load. The reduction of the high water level threshold will cause the reservoir to release more water for power generation to meet higher power demand. The calculation expression is: , where Indicates the initial high water level threshold, that is, the default water level limit when it is not adjusted. It represents the high water level threshold adjusted during the peak electricity demand period, which is used to cope with the peak electricity demand period; During the low electricity demand period, the factor is also adjusted according to the high water level threshold. , calculate the new high water level threshold, the calculation expression is: , where It indicates the high water level threshold after adjustment during the low electricity demand period, which is used to cope with the low electricity demand period; Calculate the control strategy update value, and adjust the control strategy of the hydropower station according to the control strategy update value to ensure that the water level regulation meets the current power demand. The calculation expression of the update value is: ， ， In the formula, is the control strategy update value during the peak period of power demand, indicating the relative proportion of the change in the water level threshold. is the control strategy update value during the low power demand period, which also indicates the relative proportion of the water level threshold change; The initial high water threshold operation continues to be maintained for power demand during normal operation.

Images