Power system stability optimization method, equipment and device based on edge intelligence technology

Abstract

The invention discloses a power system stability optimization method, equipment and device based on an edge intelligence technology, and relates to the technical field of power systems, and the method comprises the following steps: collecting first feature data affecting the stability of a power system in real time, and constructing a power system stability prediction model; obtaining the output of a plurality of power system stability prediction models, drawing a power system-time broken line graph, and judging the power stability of the to-be-detected region; obtaining a scheduling area sorting table according to the judgment result and based on the direction of the power transmission line and the line loss; preferentially performing power dispatching on the key lines in the unstable area of the power system according to the dispatching area sorting table, and controlling to stop; according to the method, the key power transmission lines in the unstable region of the power system are subjected to priority cross-region power dispatching, so that the problem of reasonable resource allocation in stability optimization of the power system is solved.

Description

Technical Field

[0001] The present invention relates to the field of power system stability technology, and more specifically, to a power system stability optimization method, equipment, and device based on edge intelligence technology. Background Art

[0002] Power system stability is a key characteristic of its normal operation and reliable power supply. With the continuous adjustment of energy structures and changes in electricity demand, ensuring power system stability has become increasingly important. Power system stability not only affects the reliability of power supply but also involves the system's ability to resist disturbances, recover, and adapt to changing environments. Therefore, stability issues remain a core concern for technical research and practical application in the design, operation, dispatch, and maintenance of power systems.

[0003] Traditional power system stability optimization relies on remote central servers for data analysis and decision-making, which can result in slow response times and difficulty in responding to emergencies. Furthermore, excessive reliance on generators to regulate power system stability can lead to grid scheduling being overly focused on generator power regulation, while neglecting other more flexible resources.

[0004] In view of the above problems, the present invention proposes a solution. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide a method, equipment and device for optimizing power system stability based on edge intelligence technology, which solves the problem of reasonable resource allocation in power system stability optimization by prioritizing cross-regional power dispatching of key transmission lines in unstable areas of the power system.

[0006] To achieve the above object, the present invention provides the following technical solutions: The power system stability optimization method based on edge intelligence technology includes the following steps: real-time collection of first characteristic data affecting the stability of the power system and construction of a power system stability prediction model; obtaining the outputs of several power system stability prediction models and drawing a power system-time line graph to judge the power stability of the area to be tested; obtaining a dispatching area ranking table based on the judgment results and the direction and line loss of the transmission line; and giving priority to power dispatching of key lines in the unstable area of the power system according to the dispatching area ranking table, and controlling the stop.

[0007] In a preferred embodiment, the outputs of several power system stability prediction models are obtained and a power system-time line graph is drawn to judge the power stability of the area to be tested. The specific steps are as follows: obtain the outputs of several power system stability prediction models in the current area G time period, use the output of the power system stability prediction model as the Y-axis and the G time period as the X-axis, and draw a power system stability prediction line graph; obtain historical power system stability data, perform data cleaning and normalization operations, calculate the average value, apply the average value to the power system stability prediction line graph and set it as the power system stability threshold; if the entire line graph shows an upward trend and exceeds the power system stability threshold, mark the area as a power system unstable area; if the entire line graph shows a downward or stable trend and does not exceed the power system stability threshold, mark the area as a power system stable area.

[0008] In a preferred embodiment, the scheduling area ranking table is obtained based on the judgment result and the direction and line loss of the transmission line. The specific steps are as follows: obtaining the line distribution in the unstable area of the power system and the line distribution in the rich area; analyzing the line distribution in the unstable area of the power system based on the importance method to obtain the key transmission line; obtaining the voltage at both ends of the key transmission line; calculating the voltage drop for the voltage at both ends of the key transmission line; if the voltage drop is a positive number, the voltage at one end is higher than the voltage at the other end, and the current flows from the high-end voltage to the low-end voltage, and the power flow direction of the key transmission line is obtained; if it is a negative number, the opposite is true; analyzing the line distribution in the rich area based on the voltage drop to obtain the transmission lines with the same power flow direction as the key transmission lines, marking them, and obtaining the preliminary scheduling transmission lines in the rich area; obtaining the line loss of the preliminary scheduling transmission lines in all rich areas based on the power flow calculation method, and sorting the line losses in ascending order based on the bubble sort algorithm to obtain the scheduling area ranking table.

[0009] In a preferred embodiment, the power dispatching is given priority to the key lines in the unstable area of the power system according to the dispatching area ranking table, and the stop is controlled. The specific steps are as follows: obtain the unstable area of the power system, and obtain all the stable areas of the power system based on the circular range with the unstable area of the power system as the center and the preset W as the radius; obtain the power demand and power supply in the stable area of the power system, and make a judgment. If the power supply is greater than the preset C times of the power demand, the area is marked as a rich area; analyze the unstable area of the power system and the rich area to obtain the key transmission lines in the unstable area of the power system and the dispatching area ranking table of the rich area; according to the dispatching area ranking table of the rich area, the key transmission lines in the unstable area of the power system are dispatched based on edge intelligence technology; obtain the power supply and power demand of the rich area in the dispatching, and if the power supply is less than the preset times the power demand, the dispatching of this area will be stopped, and the dispatching of the next area will be carried out according to the order of the dispatching area sorting table; a power system stability prediction line chart after the power system unstable area is dispatched is obtained, and if the predicted value of the power system stability in the line chart shows a downward trend and is lower than the power system stability threshold, the dispatching will be stopped.

[0010] In a preferred embodiment, the construction of the power system stability prediction model is specifically as follows: First characteristic data is collected in real time, where the first characteristic data includes disturbance characteristic data, generator response characteristic data, and power regulation characteristic data; the first characteristic data is input as an input item into a power system stability prediction model constructed based on machine learning, and a power system stability prediction value is used as an output item.

[0011] In a preferred embodiment, the disturbance characteristic data is specifically obtained by the following method: obtaining first data on the disturbance impact on the power system, the first data including changes in current, voltage, and power, historical disturbance data, and the natural frequency of the power system; obtaining a disturbance impact factor by performing a regression analysis on the historical disturbance data to determine the degree of impact of the historical disturbance on the power system state; obtaining the disturbance amplitude of the power system based on a statistical method according to the first data; and obtaining disturbance characteristic data based on a Fourier transform according to the first data of the disturbance impact combined with the disturbance amplitude.

[0012] In a preferred embodiment, the generator response characteristic data is specifically obtained by the following method: obtaining generator data, the generator data including the generator speed, rated angular velocity, and the generator damping coefficient and the generator dynamic response time; calculating the angular velocity of the generator based on the angular velocity conversion formula according to the generator speed; collecting the angular velocity of the generator in multiple time periods, and obtaining the change in angular velocity based on a statistical method; and calculating the generator response characteristic data based on the change in angular velocity and a preset generator dynamic response formula according to the combination of the generator data.

[0013] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. By acquiring power system data through edge intelligent devices and combining feature extraction and machine learning algorithms, it is possible to accurately predict power system stability, promptly identify potential risks, optimize system scheduling and operation, and improve the safety, stability, and operational efficiency of the power system.

[0014] 2. By utilizing edge intelligence technology for power dispatching and rationally planning the direction of transmission lines, the stability and reliability of the power system can be effectively improved, while optimizing power flow and reducing power losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of the structure of a power system stability optimization method based on edge intelligence technology provided in an embodiment of the present application.

[0016] Figure 2 Schematic diagram of the structure of the power system stability optimization device based on edge intelligence technology provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0018] Example 1, Figure 1 The schematic diagram of the structure of the power system stability optimization method based on edge intelligence technology provided in the embodiment of the present application includes the following steps: S1, collect first characteristic data in real time and build a power system stability prediction model.

[0019] In this example, power system data is obtained by installing edge intelligence devices in key areas of the power system. Edge intelligence refers to the process of migrating computing, storage, and intelligent processing capabilities from centralized data centers or the cloud to network edge devices. This technology enables edge devices to process data, run artificial intelligence models, make decisions, and execute tasks locally without having to transmit all data to remote servers for analysis. Edge intelligence devices collect various forms of data from the environment through hardware such as sensors and cameras, and use built-in computing resources and artificial intelligence models to process and analyze the data in real time. Power system data includes system voltage, current changes, frequency changes, and engine response data. Through feature extraction, data that can reflect the unstable impact of disturbances on the power system is found from the power system data, namely power regulation data.

[0020] The first characteristic data that affects the stability of the power system includes disturbance characteristic data and generator response characteristic data. Disturbance characteristic data refers to the degree of instability of the power system after being disturbed. Analyzing disturbance characteristic data for predicting the stability of the power system has the following advantages: Quantifying system stability: Disturbance signature data is a method for quantifying the degree of response of a power system to a disturbance. It provides a clear numerical indicator by analyzing the impact of the disturbance on the system's dynamic characteristics, such as frequency changes and power fluctuations. This helps quantify the stability of the power system and provides a basis for power system dispatchers or operators to assess system stability. Predicting system stability: Disturbance signature data can help predict the stability response of power systems to various disturbances. By assessing the response to different disturbance scenarios in advance, power system operators can better anticipate potential stability issues and take preventative measures or make scheduling adjustments in a timely manner.

[0021] Optimizing system design and scheduling: During the design and scheduling of power systems, analyzing disturbance signature data can provide a reference for optimizing decisions. For example, when planning grid topology, site selection, and control equipment configuration, analyzing disturbance signature data can be used to assess system robustness, leading to more rational decision-making.

[0022] Support for emergency response and dynamic protection: When a power system experiences a major disturbance, it can quickly become unstable. By calculating and tracking disturbance signature data in real time, potential risk areas can be identified, helping dispatchers or automated control systems take appropriate emergency measures. This improves the system's emergency response capabilities and prevents accidents such as large-scale power outages.

[0023] Adapting to complex and changing system dynamics: The power system is a highly complex and dynamic system, influenced by numerous factors, such as load fluctuations, generator startups and shutdowns, and changes in network topology. Disturbance signature data comprehensively considers the impact of various disturbance factors and can reflect the system's stability under varying disturbances. This approach adapts to the changing environmental conditions and dynamic characteristics of the power system, facilitating a more accurate assessment of system stability.

[0024] Providing a basis for advanced control strategies: In modern power systems, an increasing number of advanced control strategies (such as adaptive control and robust control) are being applied to improve system stability. Analysis of disturbance signature data can provide data support for the design and optimization of these control strategies.

[0025] The specific method for obtaining disturbance characteristic data is as follows: Acquiring first data on the impact of a power system disturbance, the first data including a change in current, a change in voltage, and a change in power, historical disturbance data, and a natural frequency of the power system; The impact of historical disturbances on the power system state is analyzed based on regression analysis of historical disturbance data to obtain the disturbance impact factor. Obtaining a disturbance amplitude of the power system based on a statistical method according to the first data; Disturbance characteristic data is obtained based on Fourier transform according to the first data of disturbance influence and the disturbance amplitude.

[0026] The specific calculation formula for the disturbance amplitude is as follows:

[0027] The specific calculation formula for disturbance characteristic data is as follows:

[0028] In the formula, is the disturbance characteristic data, is the disturbance amplitude, is the disturbance impact factor, is the natural frequency of the power system, is the stability margin, For the The voltage change caused by the secondary disturbance, For the The current change caused by the secondary disturbance, For the The power change caused by the secondary disturbance, For the The phase of the sub-disturbance harmonic, where i = 1, 2, 3, ..., R, R is an integer, and N is the number of disturbances.

[0029] It should be noted that the amplitude of the disturbance indicates the size of the disturbance. The natural frequency of the power system reflects the dynamic characteristics of the system.

[0030] Generator response characteristic data is used to quantify the instability generated by a generator during dynamic response to power system changes. This data can be used to assess power system stability. A large coefficient indicates that the generator may experience significant instability during dynamic response, posing a threat to the stable operation of the power system.

[0031] Analyzing generator response characteristic data has the following advantages for predicting power system stability: Promptly identify potential instability risks: A generator's dynamic response coefficient reflects its ability to respond to disturbances. By analyzing this coefficient, the generator's response characteristics can be identified promptly, allowing prediction of potential power system instability following a disturbance. This is crucial for early warning and stability management of power systems, particularly in large-scale power systems, enabling early detection of instability trends and the implementation of appropriate measures to prevent system instability.

[0032] Improved accuracy of system stability assessments: The generator's dynamic response factor is directly related to system stability, particularly in the short term. This factor quantifies the generator's ability to withstand system disturbances, enabling more accurate assessments of power system transient stability, frequency stability, and other factors. Compared to traditional stability analysis methods, the dynamic response factor provides a dynamic, quantitative metric that helps more accurately assess system stability under various operating conditions.

[0033] Optimizing generator unit operation and scheduling: By analyzing generator response characteristic data, power system dispatchers can more clearly understand the response capabilities of different generators under different operating conditions. This provides a basis for dispatching and selecting generator units, allowing them to rationally adjust unit load distribution based on the dynamic response characteristics of the generator units, ensuring system stability in the face of disturbances.

[0034] Supports system transient stability analysis: Generator dynamic response coefficients are a crucial factor in transient stability analysis. They directly impact the system's transient response, especially during large-scale disturbances. By studying these dynamic characteristics, system transient stability can be more accurately predicted, preventing system instability caused by slow or uncoordinated generator response.

[0035] Facilitates system frequency and voltage control: Generator response characteristics are closely related to system frequency and voltage control. During disturbances, the generator's response coefficient helps assess the system's frequency recovery capabilities and its effectiveness in suppressing voltage fluctuations. By analyzing these characteristics in advance, optimized frequency and voltage control strategies can be developed to avoid frequent frequency adjustments and voltage fluctuations.

[0036] The specific method for obtaining the generator response characteristic data is as follows: Acquire generator data, including generator speed, rated angular velocity, generator damping coefficient, and generator dynamic response time; The angular velocity of the generator is calculated by using the angular velocity conversion formula based on the rotational speed of the generator; Collect the angular velocity of the generator over multiple time periods and obtain the change in angular velocity based on statistical calculations; The generator response characteristic data is calculated based on the preset generator dynamic response calculation formula according to the change in angular velocity combined with the generator data.

[0037] The specific calculation formula for the change in angular velocity is as follows:

[0038] Generator response characteristic data, the specific calculation formula is as follows:

[0039] In the formula, is the generator response characteristic data, is the change in angular velocity, is the inertia of the generator, is the rated angular velocity, is the damping coefficient of the generator, is the dynamic response time of the generator, is the speed at time t2, is the speed at time t1, is the speed influencing factor at time t2, is the speed influencing factor at time t1.

[0040] It should be noted that the calculation formula for angular velocity is: , is the angular velocity, is the rotation speed, is the influencing factor of the rotation speed.

[0041] Power regulation characteristic data is used to quantify the impact of power regulation on power system stability. Analyzing power regulation characteristic data has the following advantages for predicting power system stability: Improving the accuracy of system stability prediction: Power regulation characteristic data measures the sensitivity of the power system to its power regulation capabilities under different operating conditions. Analyzing this coefficient allows for more accurate predictions of the system's response and stability under disturbances or changes, particularly during periods of significant fluctuation in power demand. It helps assess the impact of generator regulation capabilities and load changes on power system frequency, power balance, and stability. For dynamic stability analysis of power systems, power regulation characteristic data provides a quantitative method that accurately reflects changes in the system's regulation capabilities, enabling early identification of potential stability risks.

[0042] Effectively identify system weaknesses: By analyzing power regulation signature data, we can identify weaknesses in the power system, such as insufficient regulation capabilities of certain generators or unbalanced load regulation. This provides a basis for system optimization, enabling targeted measures such as adding energy storage equipment, strengthening regulation capabilities in specific areas, and adjusting generator dispatch priorities, thereby improving overall grid stability.

[0043] Dynamic Monitoring and Real-Time Adjustment: Power regulation characteristic data plays a vital role in real-time monitoring. By monitoring the changing trends of these coefficients, system dispatchers can better understand the dynamic characteristics of the power system and make real-time adjustments. This real-time prediction and adjustment capability can effectively avoid problems such as system overload and frequency fluctuations, improving the robustness and flexibility of system operations.

[0044] Assisting decision support and system planning: In the long-term planning of power systems, power regulation characteristic data, as a key predictive indicator, can provide decision makers with a scientific basis for system expansion, improvements, and investments. It helps assess the stability of the future power system and develop reasonable planning and response strategies based on actual needs and potential challenges.

[0045] By analyzing power regulation characteristic data, we can improve the prediction accuracy of power system stability from multiple dimensions, optimize system scheduling, identify potential risks, and enhance adaptability to renewable energy fluctuations, ultimately helping the power system achieve more efficient, safe, and sustainable operation. These advantages make power regulation characteristic data of great practical value in dynamic stability analysis, operational optimization, and fault prevention of power systems.

[0046] The specific method for obtaining power regulation characteristic data is as follows: Obtaining the amplitude of power regulation and the reference power of the power system and the frequency of power regulation; Obtaining the ratio of power regulation to reference power through the amplitude of power regulation and the reference power of the power system; Obtain the load changes of the power system, perform data analysis on the load changes and obtain the impact coefficient of the load changes on the system stability; Collect historical fault data of the power system to obtain the historical fault probability of the power system; The power regulation characteristic data is calculated based on a preset power regulation influence formula in combination with the historical fault probability, the influence coefficient of load change on system stability, and the ratio of power regulation to reference power.

[0047] The specific calculation formula for power regulation characteristic data is as follows:

[0048] In the formula, is the power regulation characteristic data, is the amplitude of power regulation, is the reference power of the power system, Frequency regulation for electricity, is the maximum regulation frequency allowed by the power system, is the current stability index of the power system, is the influence coefficient of load change on system stability, is the historical failure probability of the power system, 、 、 、 are weight coefficients respectively.

[0049] It should be noted that , these weights reflect the importance of each factor in a specific power system, is the ratio of power regulation to reference power, To adjust the frequency ratio.

[0050] Construct a power system stability prediction model, specifically: collecting first characteristic data in real time, the first characteristic data including disturbance characteristic data, generator response characteristic data, and power regulation characteristic data; Obtain historical disturbance characteristic data, generator response characteristic data, and power regulation characteristic data, preprocess the data, and divide it into training and test sets; The training set is used to build a model based on machine learning, and the test set is used to verify the model. Through automatic iteration, a power system stability prediction model is finally obtained; The newly acquired first feature data is used as an input item and input into a power system stability prediction model constructed based on machine learning, and the power system stability prediction value is used as an output item.

[0051] Using machine learning to construct power system stability prediction coefficients offers numerous advantages and can significantly improve power system operational efficiency and safety. Power system stability involves numerous factors, such as load fluctuations, generator regulation, network topology, and environmental conditions, and the relationships between these factors are highly complex. Traditional mathematical modeling methods often struggle to accurately capture these nonlinear and time-varying characteristics. However, machine learning algorithms can effectively process these complex patterns and provide more accurate stability predictions. Machine learning can analyze large amounts of sensor data and historical records in real time, helping the system predict potential stability issues in real time and thus prevent potential failures. As power systems dynamically change, traditional methods often require manual adjustments to models or parameters. However, machine learning algorithms, through adaptive training, can automatically adjust prediction models to adapt to new input data, maintaining high prediction accuracy. Machine learning models can automatically learn the relationships between stability and various factors from historical data, reducing manual intervention and improving prediction efficiency and accuracy.

[0052] The power system stability prediction model, the specific calculation formula is as follows:

[0053] In the formula, is the predicted value of power system stability, is the power regulation characteristic data, is the generator response characteristic data, is the disturbance characteristic data, is the weight of the power regulation characteristic data, is the weight of the generator response characteristic data, is the weight of the perturbation feature data.

[0054] It should be noted that the predicted value of power system stability is affected by power regulation characteristic data, generator response characteristic data and disturbance characteristic data. When the power regulation characteristic data, generator response characteristic data and disturbance characteristic data become larger and larger, the predicted value of power system stability will become larger and larger, and the power system will become more and more unstable. Moreover, depending on the weight coefficient, the degree of influence of each coefficient is different.

[0055] S2: Draw a power system-time line graph based on the prediction model to determine the power stability of the area to be tested. The specific steps are as follows: Obtain the outputs of several power system stability prediction models within the current region's G time period, use the outputs of the power system stability prediction models as the Y-axis and the G time period as the X-axis to draw a power system stability prediction line chart; Obtain historical power system stability data, perform data cleaning and normalization operations, calculate the average value, apply the average value to the power system stability prediction line chart and set it as the power system stability threshold; If the entire line graph shows an upward trend and exceeds the power system stability threshold, the area is marked as a power system unstable area; If the entire line graph shows a downward or stable trend and does not exceed the power system stability threshold, the area is marked as a power system stable area.

[0056] S3, according to the judgment result and based on the direction of the transmission line and the line loss, a scheduling area ranking table is obtained. The specific steps are as follows: Obtain the line distribution in unstable areas and abundant areas of the power system; The distribution of power lines in unstable areas of the power system is analyzed based on the importance method to obtain key transmission lines; Obtain the voltage at both ends of key transmission lines; Calculate the voltage drop across the critical transmission lines; If the voltage drop is positive, the voltage at one end is higher than the voltage at the other end, and the current flows from the voltage at the high end to the voltage at the low end, obtaining the power flow direction of the critical transmission line. If it is negative, the opposite is true; The line distribution in the affluent area is analyzed based on voltage drop to obtain the transmission lines that are consistent with the power flow direction of the key transmission lines, and the preliminary dispatching transmission lines in the affluent area are obtained; Based on the power flow calculation method, the line losses of the preliminary dispatched transmission lines in all wealthy areas are obtained, and the line losses are sorted in ascending order based on the bubble sort algorithm to obtain the dispatch area sorting table.

[0057] It's important to note that with the widespread adoption of renewable energy, power systems need to be able to effectively manage and dispatch energy resources distributed across various nodes. Edge intelligence technology can move computing and decision-making capabilities down to edge devices, enabling each node to independently manage and dispatch energy. This distributed energy management approach reduces the complexity and latency of centralized dispatch, improving the stability and reliability of the power system.

[0058] The direction of transmission lines has a significant impact on power system stability. It not only determines the path of power flow but also influences multiple aspects, including load distribution, voltage stability, power loss, and network reliability. When conducting cross-regional power dispatch, the direction of transmission lines must be rationally planned to ensure efficient and stable power transmission to the target area.

[0059] First, considering the direction of transmission lines helps optimize the flow of electricity. By properly planning the direction of transmission lines, it is possible to achieve a balanced distribution of power loads, avoid load concentration in certain areas, and thus reduce the risk of overloading and overloading of transmission lines. This helps improve the overall stability of the power system.

[0060] Secondly, the direction of transmission lines also affects voltage stability and power losses. Improper transmission line direction can lead to voltage instability and increased power losses, further impacting the stability of the power system. Therefore, when conducting power dispatch, it is necessary to comprehensively consider the impact of transmission line direction on voltage stability and power losses to ensure the effectiveness and cost-effectiveness of power dispatch.

[0061] Finally, considering the direction of transmission lines can also help optimize network structure and improve network reliability. By properly planning the direction of transmission lines, the power system's network structure can be optimized, improving system redundancy and reliability. This helps reduce the impact of network failures on power supply and improves the overall stability and reliability of the power system.

[0062] In summary, when dispatching power from more affluent areas to unstable regions to improve stability, the direction of transmission lines in each region must be fully considered. By rationally planning the direction of transmission lines, we can ensure the effectiveness and economy of power dispatch and improve the overall stability and reliability of the power system.

[0063] S4: Prioritize power dispatching for key lines in unstable areas of the power system according to the dispatching area ranking table and control their shutdown. Specifically: Obtain the unstable area of the power system, and obtain all the stable areas of the power system based on the circular range with a preset radius of W and the unstable area as the center. The circular range with a radius of W needs to be determined according to the actual stable area of the power system; Obtain the power demand and power supply within the stable area of the power system and make a judgment. If the power supply is greater than a preset C times the power demand, mark the area as a rich area. The constant C in the C times the power demand needs to be determined based on the actual power demand and power supply within the stable area of the power system. According to the unstable area and abundant area of the power system, the key transmission lines in the unstable area and the dispatch area ranking table of the abundant area are obtained; Based on the dispatch area ranking table of wealthy areas, power dispatch is carried out on key transmission lines in unstable power system areas based on edge intelligence technology, giving priority to improving the stability of key transmission lines in unstable power system areas. Obtain the power supply and power demand of the rich area in the scheduling. If the power supply is less than the preset times the power demand, the dispatch of this area will be stopped and the next area will be dispatched according to the order of the dispatch area sorting table. Times is greater than 1; Obtain a power system stability prediction line chart after dispatching in the unstable power system area. If the predicted value of the power system stability in the line chart shows a downward trend and is lower than the power system stability threshold, stop dispatching.

[0064] Example 2, Figure 2 This is a schematic diagram of the structure of the power system stability optimization device based on edge intelligence technology provided in an embodiment of the present application, including a data acquisition module, a test area determination module, a scheduling area division module, and a priority scheduling module. There are connections between the modules: A data acquisition module is used to collect first characteristic data affecting the stability of the power system in real time and build a power system stability prediction model; The test area determination module is used to obtain the output of several power system stability prediction models and draw a power system-time line chart to determine the power stability of the test area; A dispatching area division module is used to obtain a dispatching area ranking table based on the judgment result and the direction and line loss of the transmission line; The priority dispatching module is used to prioritize power dispatching for key lines in unstable areas of the power system according to the dispatching area sorting table and control the shutdown.

[0065] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0066] The above embodiments may be implemented in whole or in part through software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0067] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0068] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0069] The above are only specific embodiments of the present application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0070] Finally: The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A power system stability optimization method based on edge intelligence technology, characterized in that: The steps include: Collecting first characteristic data in real time and building a power system stability prediction model; Draw a power system-time line graph based on the prediction model to determine the power stability of the area to be tested; Obtaining a dispatch area ranking table based on the judgment result and the direction and line loss of the transmission line; According to the dispatching area ranking table, priority is given to power dispatching of key lines in unstable areas of the power system, and control is exercised to stop them.

2. The power system stability optimization method based on edge intelligence technology according to claim 1 is characterized in that: The specific steps of drawing a power system-time line graph based on the prediction model to determine the power stability of the area to be tested are as follows: Obtain the outputs of several power system stability prediction models within the current region's G time period, use the outputs of the power system stability prediction models as the Y-axis and the G time period as the X-axis to draw a power system stability prediction line chart; Obtain historical power system stability data, process the data, calculate the average value, apply the average value to the power system stability prediction line chart and set it as the power system stability threshold; The line graph is compared with the power system stability threshold to obtain a comparison result.

3. The power system stability optimization method based on edge intelligence technology according to claim 2 is characterized in that: The line graph is compared with the power system stability threshold to obtain the comparison result, which is specifically: If the entire line graph shows an upward trend and exceeds the power system stability threshold, the area is marked as a power system unstable area; If the entire line graph shows a downward or stable trend and does not exceed the power system stability threshold, the area is marked as a power system stable area.

4. The power system stability optimization method based on edge intelligence technology according to claim 1 is characterized in that: The scheduling area ranking table is obtained according to the judgment result and based on the direction and line loss of the transmission line. The specific steps are as follows: Obtain the line distribution in unstable areas of the power system and analyze the key transmission lines based on the importance method; Obtain the voltage at both ends of the key transmission line, calculate the voltage drop at both ends, and obtain the power flow direction of the key transmission line; Obtain the line distribution in the rich area of the power system and obtain the preliminary dispatch transmission lines in the rich area based on voltage drop analysis; The line losses of the preliminary dispatched transmission lines in the affluent area are obtained based on the power flow calculation method, and the line losses are sorted based on the bubble sort algorithm to obtain a dispatch area sorting table.

5. The power system stability optimization method based on edge intelligence technology according to claim 1 is characterized in that: The specific steps of giving priority to power dispatching and controlling the stopping of key lines in unstable areas of the power system according to the dispatching area ranking table are as follows: Obtaining an unstable area of the power system, and obtaining the stable area of the power system according to a circular range with the unstable area of the power system as the center and a preset radius W; Obtain the power demand and power supply in the stable area of the power system and make a judgment. If the power supply is greater than the preset C times of the power demand, mark the area as a rich area; According to the unstable area and abundant area of the power system, the key transmission lines in the unstable area and the dispatch area ranking table of the abundant area are obtained; Based on the dispatch area ranking table of wealthy areas, power dispatch is carried out on key transmission lines in unstable power system areas based on edge intelligence technology; Obtain the power supply and power demand of the rich area in the scheduling. If the power supply is less than the preset If the power demand is times greater than the specified value, the dispatch of the area will be stopped and the next area will be dispatched according to the order of the dispatch area ranking table; Obtain a power system stability prediction line chart after dispatching in the unstable power system area. If the predicted value of the power system stability in the line chart shows a downward trend and is lower than the power system stability threshold, stop dispatching.

6. The power system stability optimization method based on edge intelligence technology according to claim 1 is characterized in that: The construction of the power system stability prediction model is specifically as follows: collecting first characteristic data in real time, the first characteristic data including disturbance characteristic data, generator response characteristic data, and power regulation characteristic data; The first characteristic data is used as an input item and input into a power system stability prediction model constructed based on machine learning, and the power system stability prediction value is used as an output item.

7. The power system stability optimization method based on edge intelligence technology according to claim 6 is characterized in that: The specific method for obtaining the disturbance characteristic data is as follows: Acquiring first data on the impact of a power system disturbance, the first data including a change in current, a change in voltage, and a change in power, historical disturbance data, and a natural frequency of the power system; The impact of historical disturbances on the power system state is analyzed based on regression analysis of historical disturbance data to obtain the disturbance impact factor. Obtaining a disturbance amplitude of the power system based on a statistical method according to the first data; Disturbance characteristic data is obtained based on Fourier transform according to the first data of disturbance influence and the disturbance amplitude.

8. The power system stability optimization method based on edge intelligence technology according to claim 6 is characterized in that: The generator response characteristic data is specifically obtained as follows: Acquire generator data, including generator speed, rated angular velocity, generator damping coefficient, and generator dynamic response time; The angular velocity of the generator is calculated based on the angular velocity conversion formula according to the rotation speed of the generator; Collect the angular velocity of the generator over multiple time periods and obtain the change in angular velocity based on statistical methods; The generator response characteristic data is calculated based on the change in angular velocity and the generator data based on a preset generator dynamic response formula.

9. The power system stability optimization method based on edge intelligence technology according to claim 8 is characterized in that: The specific calculation formula for the change in angular velocity is as follows: The specific calculation formula of the generator response characteristic data is as follows: In the formula, is the generator response characteristic data, is the change in angular velocity, is the inertia of the generator, is the rated angular velocity, is the damping coefficient of the generator, is the dynamic response time of the generator, is the speed at time t2, is the speed at time t1, is the speed influencing factor at time t2, is the speed influencing factor at time t1.

10. A device using the power system stability optimization method based on edge intelligence technology according to any one of claims 1 to 9, characterized in that: It includes data acquisition module, test area determination module, scheduling area division module and priority scheduling module. There are connections between modules: A data acquisition module is used to collect first characteristic data affecting the stability of the power system in real time and build a power system stability prediction model; The test area determination module is used to obtain the output of several power system stability prediction models and draw a power system-time line chart to determine the power stability of the test area; A dispatching area division module is used to obtain a dispatching area ranking table based on the judgment result and the direction and line loss of the transmission line; The priority dispatching module is used to prioritize power dispatching for key lines in unstable areas of the power system according to the dispatching area sorting table and control the shutdown.

Images