Thermal power generating unit frequency modulation and energy storage collaborative frequency modulation system

Abstract

The invention discloses a thermal power generating unit frequency modulation and energy storage collaborative frequency modulation system, which comprises a data acquisition module, a control module, a frequency modulation module and an energy storage module, the data acquisition module comprises an acquisition unit and is used for monitoring power grid system data and environment data and transmitting the data to the control module; the control module comprises a data storage unit and a first prediction module; the first prediction module adopts a long-short term memory (LSTM) network model, combines historical data and real-time data, predicts a future power grid frequency fluctuation trend of a power grid system, and generates a frequency modulation strategy; the frequency modulation module receives the frequency modulation strategy transmitted by the first prediction module and adjusts the power grid frequency; the energy storage module comprises at least two energy storage devices and an energy storage management system, and the energy storage management system is connected with the frequency modulation module and used for monitoring the operation states of the energy storage devices and adjusting the power of the energy storage devices to respond to the frequency modulation requirement of the frequency modulation module. The frequency modulation response time of the thermal power generating unit can be shortened, and frequent frequency fluctuation can be quickly coped with.

Description

Technical Field

[0001] The present application relates to the technical field of power grid frequency regulation, and in particular to a frequency regulation system for thermal power units and energy storage coordinated frequency regulation. Background Art

[0002] As the main source of frequency regulation in traditional power systems, thermal power plants have certain frequency regulation capabilities. However, with the increase in grid frequency fluctuations, especially the fluctuations caused by the access of new energy sources, the frequency regulation capabilities of thermal power plants face huge challenges: the frequency regulation response time of traditional thermal power units is relatively long, usually reaching several minutes, and they cannot quickly respond to frequent frequency fluctuations.

[0003] In addition, frequent adjustments to generator set power result in reduced unit efficiency and increased fuel consumption. Summary of the invention

[0004] In response to the above problems, the present application provides a frequency regulation system for thermal power units and energy storage coordinated frequency regulation, which can reduce the frequency regulation response time of thermal power units and quickly respond to frequent frequency fluctuations.

[0005] To achieve the purpose of this application, this application provides the following technical solutions:

[0006] The present application provides a frequency regulation and energy storage coordinated frequency regulation system for a thermal power unit, comprising: a data acquisition module, a control module, a frequency regulation module and an energy storage module; the data acquisition module comprises a collection unit for monitoring the power grid system data and environmental data, and transmitting the data to the control module; the collection unit is used to connect to the power grid system, and the power grid system data comprises power grid frequency data and power grid power flow data; the control module comprises a data storage unit and a first prediction module; the data storage unit is used to store data transmitted by the collection device; the first prediction module adopts a long short-term memory network LSTM model, combines the historical data stored in the storage unit and the real-time data collected by the collection unit, predicts the future power grid frequency fluctuation trend of the power grid system, and generates a frequency regulation strategy; the frequency regulation module receives the frequency regulation strategy transmitted by the first prediction module, and adjusts the power grid frequency of the power grid system; the energy storage module comprises at least two energy storage devices and an energy storage management system, the energy storage management system is connected to the frequency regulation module, and is used to monitor the operating status of the energy storage device and adjust the power of the energy storage device to respond to the frequency regulation demand of the frequency regulation module; the operating status of the energy storage device comprises: charging and discharging power, remaining capacity, and health status.

[0007] In a possible implementation, the first prediction module includes a first data preprocessing module, and the first data preprocessing module is used to perform denoising, outlier correction and feature extraction on the historical data.

[0008] In one possible implementation, the first prediction module includes a first model training module, which is used to design the LSTM model, train the LSTM model based on historical data, and update and predict the LSTM model parameters according to real-time data to improve prediction accuracy.

[0009] In a possible implementation, the energy storage module includes a second prediction module and a charge and discharge optimization unit; the acquisition unit is also used to collect load data of the power grid system, and the load data includes: voltage, current, and power data. The second prediction module is based on the LSTM model, combined with real-time load data and historical data to predict future load fluctuation trends, and generate a power allocation strategy for the energy storage device; the charge and discharge optimization unit adjusts the charge and discharge process of the energy storage device according to the power allocation strategy of the energy storage device to ensure that the energy storage device meets the frequency modulation requirements.

[0010] In one possible implementation, the energy storage module is used to support primary frequency regulation and secondary frequency regulation; the primary frequency regulation quickly adjusts the output power of the energy storage device through millisecond-level response, and the secondary frequency regulation maintains the grid frequency stability of the power grid system by dynamically adjusting the output power of the energy storage device; the energy storage module is also used to provide emergency power supply when a thermal power unit fails.

[0011] In a possible implementation, the system adopts a hierarchical frequency modulation strategy according to the intensity of the frequency disturbance; the first prediction module predicts the intensity of the frequency disturbance; the hierarchical frequency modulation strategy includes: when the frequency fluctuation is in the range of -0.01HZ to 0.01Hz, frequency modulation is performed through the energy storage module; when the frequency fluctuation is in the range of -0.02Hz to 0.02Hz, the power output of the thermal power unit is adjusted through the frequency modulation module; when the frequency fluctuation is less than -0.02Hz or greater than 0.02Hz, frequency modulation is performed in combination with the frequency modulation module and the energy storage module.

[0012] In a possible implementation, the system includes a monitoring module; the monitoring module is used to monitor the operating status, frequency regulation effect and energy storage device status of the power grid system in real time.

[0013] In one possible implementation, the system includes a visualization module; the visualization module is used to receive and display the detection results of the monitoring module, including grid frequency fluctuations, energy storage device charging and discharging status, frequency regulation strategy execution status, and generate an operation analysis report.

[0014] In a possible implementation, the acquisition unit includes an acquisition device, which includes a synchronized phasor measurement device PMU and a sensor device; the synchronized phasor measurement device PMU is used to monitor the operating status of the power grid system; the sensor device is used to monitor environmental data, and the sensor device includes a temperature sensor, a humidity sensor, and a wind speed sensor.

[0015] Beneficial effect: The frequency regulation and energy storage coordinated frequency regulation system of a thermal power unit provided in this application improves the frequency regulation response speed and stability of the thermal power plant through the coordinated work of the intelligent frequency regulation system and the energy storage device, reduces the burden on traditional thermal power units, and improves the frequency regulation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings are used to provide a further understanding of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation to the present application.

[0017] Figure 1 A schematic diagram of the overall structure of the frequency modulation system provided in an embodiment of the present application;

[0018] Figure 2 A schematic diagram of the structure of a control module provided in an embodiment of the present application;

[0019] Figure 3 A schematic diagram of the structure of the energy storage module provided in an embodiment of the present application.

[0020] Reference numerals:

[0021] 1. Power grid system, 2. Data acquisition module, 3. Control module, 4. Frequency modulation module, 5. Energy storage module, 6. Monitoring module, 7. Visualization module. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0023] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features; in the description of this application, unless otherwise specified, "plurality" means two or more.

[0024] The inventors of the present invention have found that thermal power plants, as the main source of frequency regulation in traditional power systems, have certain frequency regulation capabilities. However, with the increase in grid frequency fluctuations, especially the fluctuations caused by the access of new energy sources, the frequency regulation capabilities of thermal power plants face huge challenges: specifically, the frequency regulation response time of traditional thermal power units is relatively long, usually reaching several minutes, and they cannot quickly respond to frequent frequency fluctuations. In addition, frequent adjustments to the power of generator sets lead to reduced unit efficiency and increased fuel consumption. Based on this, in the embodiment of the present invention provided by the inventor, the frequency regulation response speed and stability of thermal power plants are improved through the coordinated work of an intelligent frequency regulation system and an energy storage device, reducing the burden on traditional thermal power units and improving frequency regulation efficiency.

[0025] like Figure 1-Figure 3 As shown, a frequency regulation and energy storage coordinated frequency regulation system of a thermal power unit provided in an embodiment of the present application includes: a data acquisition module 2, a control module 3, a frequency regulation module 4 and an energy storage module 5; the data acquisition module 2 includes a collection unit for monitoring the data of the power grid system 1 and the environmental data, and transmitting them to the control module 3; the collection unit is used to connect with the power grid system 1, and the data of the power grid system 1 includes power grid frequency data and power grid power flow data; the control module 3 includes a data storage unit and a first prediction module; the data storage unit is used to store the data transmitted by the collection device; the first prediction module adopts a long short-term memory network LSTM model, combines the historical data stored in the storage unit and the real-time data collected by the collection unit, predicts the future power grid frequency fluctuation trend of the power grid system 1, and generates a frequency regulation strategy; the frequency regulation module 4 receives the frequency regulation strategy transmitted by the first prediction module, and adjusts the power grid frequency of the power grid system 1; the energy storage module 5 includes at least two energy storage devices and an energy storage management system, the energy storage management system is connected to the frequency regulation module 4, and is used to monitor the operating status of the energy storage device and adjust the power of the energy storage device to respond to the frequency regulation demand of the frequency regulation module 4; the operating status of the energy storage device includes: charging and discharging power, remaining capacity, and health status.

[0026] In the embodiment of the present application, the acquisition device adopts, for example, a synchronized phasor measurement device PMU and a sensor device. The synchronized phasor measurement device PMU is used to monitor the operating state of the power grid system 1 in real time and provide high-precision and real-time measurement data. The main monitoring data includes: power grid frequency data and power grid power data. The sensor device is used to supplement the environmental data that the synchronized phasor measurement device PMU cannot measure. The environmental data will affect the operating state of the power grid system 1. For example, high temperature or high humidity may cause equipment overload. Therefore, it is necessary to collect environmental data with the help of the sensor device to provide support for the overall frequency modulation strategy of the system. Environmental data such as temperature, humidity and wind speed, etc. Correspondingly, the sensor device adopts, for example, a temperature sensor, a humidity sensor, and a wind speed sensor. After the data acquisition is completed, it is transmitted to the data storage unit. The first prediction module of the control module 3 will obtain the historical data of the data storage, perform model training, and obtain an LSTM model suitable for this system. In actual use, the collected real-time frequency data is input into the trained LSTM model to predict the power grid frequency changes of the power grid system 1 in the short term in the future in real time. The frequency modulation module 4 triggers the frequency modulation response of the thermal power unit in advance according to the prediction results to maintain the stability of the power grid frequency. In addition, the energy storage module 5 will be in a timely response state. If the energy storage device support is required during frequency modulation, the energy storage module 5 will provide rapid response support when the grid frequency fluctuates.

[0027] It should be noted that the Long Short-Term Memory (LSTM) network is a special recurrent neural network designed to solve the gradient vanishing and gradient exploding problems of traditional RNN when processing long sequence data. LSTM can effectively model long-term dependencies in time series by introducing memory cells (Cell State) and gating mechanisms (Gating Mechanism). LSTM cells mainly include forget gates (Forget Gate), input gates (InputGate), output gates (Output Gate) and memory cells (Cell State). The forget gate determines which parts of the information in the memory cell need to be "forgotten" or retained. In the prediction of power grid frequency fluctuations in this application, the forget gate helps the LSTM model discard historical information that is no longer useful. For example, discarding early irrelevant power grid frequency data ensures that the LSTM model focuses on information that has a greater impact on current and future frequency fluctuations. The input gate is responsible for introducing new information into the memory cell for updating and enhancing the long-term memory of the model. In the prediction of grid frequency fluctuations in the present application, the input gate enables the LSTM model to update its memory based on real-time data such as grid power flow, environmental data, etc., and to more accurately consider the current input when predicting grid frequency. The output gate controls the final output value, which is the prediction result of the model, that is, the content read from the memory unit. In the prediction of grid frequency fluctuations in the present application, the output gate generates the final prediction result based on the input at the current moment and the long-term memory information in the memory unit, such as the historical frequency fluctuation trend, for formulating the frequency modulation strategy.

[0028] The memory unit stores and updates the long-term information of the time series. The embodiment of the present application optimizes the model as follows:

[0029] Input feature optimization: First construct the input features: The input features of the model include historical grid frequency data, grid power data, and environmental data such as temperature, humidity, and wind speed. Specifically: Input the grid frequency data of the past N time steps, such as 5 minutes, 30 minutes, and 60 minutes. In order to smooth the data, calculate statistical features such as the weighted moving average and weighted variance in each time window. Calculate the frequency change rate, such as the difference between the current frequency and the frequency at the previous moment. After the input is completed, the input features are normalized to eliminate differences. The normalization method uses, for example, the minimum-maximum normalization method: map all feature values ​​to a fixed interval (usually [0,1]), which is suitable for data with a more consistent distribution.

[0030] Network structure optimization: Layer optimization. The number of layers in the LSTM network determines the depth of time dependency that the model can capture. Too few layers may not be able to effectively capture complex time series patterns, while too many layers may lead to overfitting problems. According to the needs of the experiment, the LSTM network consists of 2 to 3 superimposed layers, which can capture long-term dependencies and avoid overfitting. Optimization of the number of hidden units: The number of hidden units controls the capacity of each layer of LSTM and determines the network's ability to process and remember information. It is very important to choose the right number of hidden units: 64 units are suitable for simpler tasks or when there are fewer data features. A smaller number of hidden units can reduce the computational burden and speed up the training process. 128 units: suitable for more complex tasks or when the amount of data is large. More units can improve the learning ability of the model and capture more complex patterns. In this application, the number of hidden units in each layer is set to 64 or 128, and the optimal configuration is determined by a grid search method. This optimization scheme can provide an efficient and accurate prediction model for the frequency regulation system of thermal power units, and provide reliable data support for power grid frequency regulation and energy storage coordinated frequency regulation.

[0031] Regularization: Dropout is a commonly used regularization technique to prevent neural networks from overfitting. It improves the generalization ability of the model by randomly "discarding" a portion of neurons during each training so that the network does not rely too much on certain specific neurons. In this application, a Dropout layer is added after each layer of LSTM, and the Dropout ratio is set to 0.2, that is, 20% of the neurons are randomly discarded during each training. This can effectively prevent the model from overfitting the training set during training, thereby improving its ability to predict new data. In applications such as power grid frequency prediction and frequency regulation strategy generation in this application, the model can adapt to different input data and avoid training only on specific historical data or local data patterns.

[0032] Through the above optimization, the accuracy and stability of the LSTM model in the prediction of power grid frequency fluctuations can be improved. After the model is built, it can enter the training phase. After the data is processed in the training phase, the model is trained with the data. After the training is completed and the performance is verified, it is put into use. When used, the historical data stored in the storage unit and the real-time data collected by the acquisition unit are combined to predict the future power grid frequency fluctuation trend of the power grid system 1 and generate a frequency modulation strategy.

[0033] By adopting the above technical solution, through the coordinated work of the intelligent frequency regulation system and the energy storage device, the frequency regulation response speed and stability of the thermal power plant are improved, the burden on traditional thermal power units is reduced, and the frequency regulation efficiency is improved.

[0034] As a feasible implementation, the first prediction module of the embodiment of the present application includes a first data preprocessing module and a first model training module, and the first data preprocessing module is used to denoise, correct outliers and extract features from historical data. The first data preprocessing module is used to denoise, correct outliers and extract features from historical data. The first model training module is used to design an LSTM model, train the LSTM model based on historical data, and update the prediction LSTM model parameters based on real-time data to improve prediction accuracy.

[0035] In the embodiment of the present application, in order to make the prediction unit predict the future power grid frequency fluctuation trend of the power grid system 1 more accurate and generate a more reasonable frequency modulation strategy, it is necessary to use a first data preprocessing module to preprocess the historical data, and the preprocessing includes:

[0036] For example, a low-pass filter is used to remove high-frequency noise from the collected historical data, and then the sliding average method is used to smooth the data. The missing data values ​​are filled using the neighboring data interpolation method or trend analysis method. This operation can ensure data quality. Then, key features are extracted from the data. According to the specific operation characteristics of the power grid and the environmental impact, features that have an important impact on the frequency regulation strategy are extracted, such as calculating the rate of change of the power grid frequency over time and analyzing the trend of voltage and power flow. The frequency fluctuation pattern in the historical data is analyzed, especially the characteristics of the load peak. For example, the changes in temperature, humidity, and wind speed over time.

[0037] The specific feature extraction method, taking environmental data as an example, the first step: statistically calculate the long-term average, variance and extreme value of environmental data: the average value can reflect the overall stability, the variance can measure the fluctuation range, and the extreme value can be used to identify the potential impact of extreme environment on the power grid, such as the highest temperature and the maximum wind speed. The second step: calculate the trend characteristics, and use the moving average method to extract the trend changes in a short period of time, such as the impact of the temperature rise trend on the power grid frequency demand. The third step: calculate the interactive characteristics, combine different environmental variables, and calculate the interactive impact, such as the temperature and humidity joint index, which reflects the impact of high temperature and high humidity on equipment heat dissipation. The fourth step: analyze the dynamic characteristics and use the time series analysis method to extract the dynamic characteristics. After the above steps are completed, the correlation analysis method, such as the Pearson correlation coefficient, is used to calculate the correlation between environmental characteristics and power grid frequency and power flow, and screen out environmental variables with significant impact. For example, if it is found that the correlation between temperature and power grid frequency fluctuation is high, the temperature change rate can be used as a feature for the next step. The extraction of power grid power data is the same as the above environmental extraction method. After all extractions are completed, the subsequent system operations are performed.

[0038] The first model training module includes stages such as model design, training data division, model training, and evaluation and verification of model performance. The model design stage refers to building a neural network containing one or more LSTM layers, the input data includes the historical frequency data of the power grid system 1, and the output is a model of the future frequency prediction value of the power grid system 1. After the model design is completed, the data division stage can be entered, and the data processed by the first data preprocessing module is divided into 80% training set and 20% test set. The training set is used in the model training stage, and the test set is used in the evaluation and verification of model performance. In the model training stage, the designed model is trained using the training set data, and the parameters are adjusted during the training process to reduce the prediction error. After the training is completed, in order to ensure the accuracy of the prediction results, the training results are evaluated and verified, specifically using the test set to evaluate the model performance, and using indicators such as mean square error and mean absolute error to measure the prediction accuracy. After continuous model training and performance evaluation, the application stage can be entered. During use, the collected real-time frequency data is input into the trained LSTM model to predict the power grid frequency changes of the power grid system 1 in the short term in the future in real time. The frequency regulation module 4 triggers the frequency regulation response of the thermal power unit or the energy storage system in advance according to the prediction results to maintain the stability of the grid frequency.

[0039] By adopting the above technical solutions, LSTM can effectively handle the long-term dependency problem in time series and is suitable for complex pattern recognition of power grid frequency fluctuations. Through deep learning, the LSTM model shows high accuracy in frequency prediction and supports the stable operation of the power grid.

[0040] In a possible implementation, the energy storage module 5 includes a second prediction module and a charge and discharge optimization unit; the collection unit is also used to collect load data of the power grid system 1, the load data includes: voltage, current, and power data, the second prediction module is based on the LSTM model, combines real-time load data and historical data to predict future load fluctuation trends, and generates a power allocation strategy for the energy storage device; the charge and discharge optimization unit adjusts the charge and discharge process of the energy storage device according to the power allocation strategy of the energy storage device to ensure that the energy storage device meets the frequency regulation requirements.

[0041] In the embodiment of the present application, the main task of the charge and discharge optimization unit is to optimize the charge and discharge strategy of the energy storage device according to the frequency fluctuation trend of the power grid, so as to achieve the goal of improving the frequency regulation efficiency and extending the service life of the energy storage device. Improving the frequency regulation efficiency means to reasonably allocate the power output of the energy storage system according to the predicted power grid frequency fluctuation, so that the system can quickly respond to frequency disturbances in a short time and ensure the stability of the power grid. Extending the service life of the energy storage device means to reasonably arrange the charge and discharge operations and avoid overly frequent charge and discharge operations, which can not only reduce the aging speed of the energy storage device, but also improve the long-term stability of the equipment. For example, when the power grid frequency is stable, the system can reduce the charge and discharge frequency of the energy storage device to avoid equipment loss due to excessive charging and discharging.

[0042] It should be noted that the second prediction module and the first prediction module have the same technical means, except that the input data and prediction targets are different. The second prediction module predicts the load fluctuation trend, thereby generating a power allocation strategy for the energy storage device. The second prediction includes a second data preprocessing module and a second model training module. The second data preprocessing module processes data with the same technical means as the first preprocessing module, except that the data processed is different. The data processed by the second data preprocessing module is load data and historical load data. The second model training module also has the same technical means as the first model training module, except that the processed data is different from the output data, which will not be described in detail here.

[0043] In a possible implementation, the energy storage module 5 is used to support primary frequency regulation and secondary frequency regulation; the primary frequency regulation quickly adjusts the output power of the energy storage device through millisecond-level response, and the secondary frequency regulation maintains the grid frequency stability of the power grid system 1 by dynamically adjusting the output power of the energy storage device; the energy storage module 5 is also used to provide emergency power supply when the thermal power unit fails.

[0044] In the embodiment of the present application, the primary frequency modulation emphasizes rapid response at the millisecond level, while the secondary frequency modulation focuses on maintaining the stability of the grid frequency. When the grid frequency fluctuates greatly, the energy storage module 5 responds quickly, and provides rapid power regulation with a millisecond response time through the energy storage device, participating in the immediate recovery of the grid frequency. When the grid frequency fluctuation is small or the fluctuation is restored in a short period of time, the energy storage module 5 enters the secondary frequency modulation mode, and maintains the stability of the grid frequency by dynamically adjusting the output power of the energy storage device. During the primary frequency modulation process, the energy storage module 5 uses an efficient rapid charging and discharging control strategy to ensure that the required power is provided in a very short time. During the secondary frequency modulation process, the energy storage module 5 reduces frequency fluctuations by optimizing the power output balance adjustment, and ensures that the load fluctuations of the thermal power units are minimized.

[0045] By adopting the above technical solution, the specific response scenarios of the energy storage module 5 in the primary frequency modulation and the secondary frequency modulation can demonstrate how the system flexibly adjusts the output power of the energy storage device to adapt to grid frequency changes of different amplitudes.

[0046] In a possible implementation, the system adopts a hierarchical frequency modulation strategy according to the intensity of the frequency disturbance; the first prediction module predicts the intensity of the frequency disturbance; the hierarchical frequency modulation strategy includes: when the frequency fluctuation is in the range of -0.01HZ to 0.01Hz, frequency modulation is performed through the energy storage module 5; when the frequency fluctuation is in the range of -0.02Hz to 0.02Hz, the power output of the thermal power unit is adjusted through the frequency modulation module 4; when the frequency fluctuation is less than -0.02Hz or greater than 0.02Hz, frequency modulation is performed in combination with the frequency modulation module 4 and the energy storage module 5.

[0047] In a possible implementation, the system includes a monitoring module 6 ; the monitoring module 6 is used to monitor the operating status, frequency regulation effect and energy storage device status of the power grid system 1 in real time.

[0048] In a possible implementation, the system includes a visualization module 7; the visualization module 7 is used to receive and display the detection results of the monitoring module 6, including grid frequency fluctuations, energy storage device charging and discharging status, frequency regulation strategy execution status, and generate an operation analysis report.

[0049] Working principle: The system of this application obtains grid frequency, power flow, and environmental data in real time through PMU and sensor devices, and stores the data in the storage unit of the control module 3. Afterwards, the first prediction module uses the LSTM model to analyze historical data, predict future grid frequency fluctuation trends, and generate a frequency modulation strategy. The frequency modulation module 4 dynamically adjusts the output power of the thermal power unit according to the frequency modulation strategy, or triggers the energy storage module 5 to respond, giving priority to meeting the needs of slight fluctuations. The energy storage module 5 provides a fast response to solve the primary frequency modulation demand; when continuous frequency modulation is required, it works with the thermal power unit through the secondary frequency modulation strategy to reduce the burden on the unit. The control module 3 automatically selects the layered frequency modulation strategy according to the intensity of frequency fluctuations, and the energy storage module 5 dynamically adjusts the power distribution. The system monitors the operating status of the thermal power unit and the energy storage module 5 in real time, and generates a frequency modulation effect analysis report through the visualization module 7 for the operator's reference.

[0050] The beneficial effects of the embodiments of the present application are as follows:

[0051] 1. The energy storage module quickly responds to primary frequency modulation requirements and significantly improves frequency stability.

[0052] 2. The LSTM model combines historical and real-time data to predict frequency fluctuation trends and improve the accuracy of the frequency modulation strategy.

[0053] 3. The energy storage module coordinates the frequency regulation with the thermal power units to improve the overall system efficiency and reduce the burden of frequent frequency regulation of the units.

[0054] 4. The layered frequency modulation strategy dynamically adjusts the response subject to frequency disturbances of different intensities to achieve efficient operation.

[0055] In the several embodiments provided in the present application, it should be understood that the disclosed systems, modules and methods can be implemented in other ways. For example, the module embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, modules or units, which can be electrical, mechanical or other forms.

[0056] The above embodiments are only used to illustrate the technical solution of the present application, but not to limit it. The present application is not limited to the exact structure described above and illustrated in the accompanying drawings, and it cannot be determined that the specific implementation of the present application is limited to these descriptions. For ordinary technicians in the technical field to which the present application belongs, various changes and modifications made without departing from the concept of the present application should be deemed to belong to the protection scope of the present application.

Claims

1. A frequency regulation system for thermal power units and energy storage coordinated frequency regulation, characterized in that: include: Data acquisition module, control module, frequency modulation module and energy storage module; The data acquisition module includes an acquisition unit, which is used to connect to the power grid system, monitor the power grid system data and environmental data, and transmit them to the control module; the power grid system data includes power grid frequency data and power grid power flow data; The control module includes a data storage unit and a first prediction module; The data storage unit is used to store the data transmitted by the acquisition device; The first prediction module adopts a long short-term memory network LSTM model, combines the historical data stored in the storage unit and the real-time data collected by the collection unit, predicts the future power grid frequency fluctuation trend of the power grid system, and generates a frequency modulation strategy; The frequency modulation module receives the frequency modulation strategy transmitted by the first prediction module, and adjusts the grid frequency of the grid system; The energy storage module includes at least two energy storage devices and an energy storage management system, wherein the energy storage management system is connected to the frequency modulation module and is used to monitor the operating status of the energy storage device and adjust the power of the energy storage device to respond to the frequency modulation demand of the frequency modulation module; The operating status of the energy storage device includes: charging and discharging power, remaining capacity, and health status.

2. A thermal power unit frequency regulation and energy storage coordinated frequency regulation system according to claim 1, characterized in that: The first prediction module includes a first data preprocessing module, and the first data preprocessing module is used to perform denoising, outlier correction and feature extraction on the historical data.

3. A thermal power unit frequency regulation and energy storage coordinated frequency regulation system according to claim 2, characterized in that: The first prediction module includes a first model training module, which is used to design the LSTM model, train the LSTM model based on historical data, and update and predict the LSTM model parameters according to real-time data to improve prediction accuracy.

4. A thermal power unit frequency regulation and energy storage coordinated frequency regulation system according to claim 1, characterized in that: The energy storage module includes a second prediction module and a charge and discharge optimization unit; The acquisition unit is also used to collect power grid system load data, the load data includes: voltage, current, power data, The second prediction module predicts future load fluctuation trends based on the LSTM model in combination with real-time load data and historical data, and generates a power allocation strategy for the energy storage device; the charging and discharging optimization unit adjusts the charging and discharging process of the energy storage device according to the power allocation strategy of the energy storage device to ensure that the energy storage device meets the frequency modulation requirements.

5. A thermal power unit frequency regulation and energy storage coordinated frequency regulation system according to claim 1, characterized in that: The energy storage module is used to support primary frequency modulation and secondary frequency modulation; The primary frequency modulation quickly adjusts the output power of the energy storage device through millisecond-level response, and the secondary frequency modulation maintains the grid frequency stability of the grid system by dynamically adjusting the output power of the energy storage device; The energy storage module is also used to provide emergency power supply when the thermal power unit fails.

6. A thermal power unit frequency regulation and energy storage coordinated frequency regulation system according to claim 1, characterized in that: The system adopts a hierarchical frequency modulation strategy according to the intensity of the frequency disturbance; The first prediction module predicts the intensity of the frequency disturbance; The layered frequency modulation strategy includes: when the frequency fluctuation is within the range of -0.01HZ to 0.01Hz, frequency modulation is performed through the energy storage module; when the frequency fluctuation is within the range of -0.02Hz to 0.02Hz, the power output of the thermal power unit is adjusted through the frequency modulation module; when the frequency fluctuation is less than -0.02Hz or greater than 0.02Hz, frequency modulation is performed in combination with the frequency modulation module and the energy storage module.

7. A thermal power unit frequency regulation and energy storage coordinated frequency regulation system according to claim 1, characterized in that: The system includes a monitoring module; the monitoring module is used to monitor the operating status, frequency modulation effect and energy storage device status of the power grid system in real time.

8. A thermal power unit frequency regulation and energy storage coordinated frequency regulation system according to claim 7, characterized in that: The system includes a visualization module; the visualization module is used to receive and display the detection results of the monitoring module, including grid frequency fluctuations, energy storage device charging and discharging status, frequency modulation strategy execution status, and generate an operation analysis report.

9. A thermal power unit frequency regulation and energy storage coordinated frequency regulation system according to claim 1, characterized in that: The acquisition unit includes an acquisition device, which includes a synchronized phasor measurement device PMU and a sensor device; the synchronized phasor measurement device PMU is used to monitor the operating status of the power grid system; the sensor device is used to monitor environmental data, and the sensor device includes a temperature sensor, a humidity sensor, and a wind speed sensor.

Images