New energy operation evaluation method and system in extreme meteorological environment

Abstract

The invention provides a new energy operation evaluation method and system in an extreme meteorological environment, and relates to the technical field of new energy power generation, and the method comprises the steps: determining extreme meteorological data in a target evaluation region; determining operation data of the new energy unit in the target evaluation area; carrying out feature extraction on the extreme meteorological data to obtain meteorological features, and analyzing the operation characteristics of the new energy unit according to the operation data and the meteorological features; training a pre-constructed neural network architecture by using the meteorological characteristics and the operation characteristics to obtain an operation evaluation model; performing future short-term operation prediction on the new energy unit based on the operation evaluation model to obtain prediction data; and performing index evaluation on the prediction data to optimize a power grid dispatching strategy according to an evaluation result. According to the invention, the technical problem that the stability and the operation efficiency of the power grid are affected due to the fact that the intermittency and the volatility of the new energy unit are not effectively controlled in the prior art can be solved, and the technical effect of improving the operation efficiency of the power grid is achieved.

Description

Technical Field

[0001] This application relates to the technical field of new energy power generation, and particularly to a new energy operation evaluation method and system under extreme meteorological environments. Background Art

[0002] With the rapid development of renewable energy, especially the wide application of new energy sources such as wind energy and solar energy, the operation of new energy units has become an important part of the power system. However, the power generation of new energy units has strong intermittency, volatility, and uncertainty, which brings great challenges to power grid dispatching and energy management.

[0003] Currently, there are obvious defects in the existing technologies for dealing with the intermittency and volatility of new energy units. Wind power generation and photovoltaic power generation are greatly affected by weather and environmental changes, and the power output has strong randomness. For example, when the wind speed changes or on cloudy days, the power output of wind power and photovoltaic generating units will fluctuate significantly, and even shut down. This unstable power output makes it difficult for power grid dispatching to achieve precise control. The traditional dispatching system cannot efficiently respond to this change, easily leading to unstable power supply and even causing fluctuations in the frequency and voltage of the power grid. Most of the existing power dispatching methods are based on the dispatching optimization of traditional units and lack in-depth consideration of the special operating characteristics of new energy units.

[0004] In summary, in the existing technologies, due to the ineffective control of the intermittency and volatility of new energy units, it is difficult for power grid dispatching to accurately respond to the random changes in new energy power generation, further affecting the stability and operation efficiency of the power grid, possibly leading to unstable power supply, and even causing fluctuations in the frequency and voltage of the power grid. Summary of the Invention

[0005] The purpose of this application is to provide a new energy operation evaluation method and system under extreme meteorological environments to solve the technical problems in the existing technologies, where due to the ineffective control of the intermittency and volatility of new energy units, it is difficult for power grid dispatching to accurately respond to the random changes in new energy power generation, further affecting the stability and operation efficiency of the power grid, possibly leading to unstable power supply, and even causing fluctuations in the frequency and voltage of the power grid.

[0006] In view of the above problems, this application provides a new energy operation evaluation method and system under extreme meteorological environments.

[0007] In a first aspect, the present application provides a new energy operation evaluation method under extreme meteorological conditions, which is implemented through a new energy operation evaluation system under extreme meteorological conditions, including: connecting to a smart meteorological system to determine extreme meteorological data within a target evaluation area, where the extreme meteorological data has historical time records; determining the operation data of new energy units within the target evaluation area by interacting with a smart operation and maintenance monitoring system; extracting features from the extreme meteorological data to obtain meteorological features, and analyzing the operation characteristics of the new energy units based on the operation data and the meteorological features, where the operation characteristics include intermittency, volatility, and uncertainty; training a pre-constructed neural network architecture with the meteorological features and the operation characteristics to obtain an operation evaluation model; performing short-term future operation prediction on the new energy units based on the operation evaluation model to obtain prediction data; and evaluating the prediction data to optimize the power grid dispatching strategy based on the evaluation results.

[0008] In a second aspect, the present application further provides a new energy operation evaluation system under extreme meteorological conditions for performing the new energy operation evaluation method under extreme meteorological conditions as described in the first aspect, including: an extreme meteorological data determination module for connecting to a smart meteorological system to determine extreme meteorological data within a target evaluation area, where the extreme meteorological data has historical time records; an operation data determination module for determining the operation data of new energy units within the target evaluation area by interacting with a smart operation and maintenance monitoring system; an operation characteristic analysis module for extracting features from the extreme meteorological data to obtain meteorological features, and analyzing the operation characteristics of the new energy units based on the operation data and the meteorological features, where the operation characteristics include intermittency, volatility, and uncertainty; an operation evaluation model acquisition module for training a pre-constructed neural network architecture with the meteorological features and the operation characteristics to obtain an operation evaluation model; a prediction data acquisition module for performing short-term future operation prediction on the new energy units based on the operation evaluation model to obtain prediction data; and a dispatching optimization module for evaluating the prediction data to optimize the power grid dispatching strategy based on the evaluation results.

[0009] The technical solutions provided in this application have at least the following technical effects or advantages: By connecting to the intelligent meteorological system, the extreme meteorological data within the target evaluation area is determined, where the extreme meteorological data has historical time records; By interacting with the intelligent operation and maintenance monitoring system, the operation data of the new energy units within the target evaluation area is determined; Feature extraction is performed on the extreme meteorological data to obtain meteorological features, and the operation characteristics of the new energy units are analyzed based on the operation data and the meteorological features, where the operation characteristics include intermittency, volatility, and uncertainty; The pre-constructed neural network architecture is trained with the meteorological features and the operation characteristics to obtain an operation evaluation model; Based on the operation evaluation model, short-term future operation prediction is performed on the new energy units to obtain prediction data; Index evaluation is performed on the prediction data, and the grid dispatching strategy is optimized based on the evaluation results. That is to say, by achieving the technical goals of accurate prediction of the operation characteristics of new energy units and dispatching optimization, the technical effects of improving the accuracy of grid dispatching, stabilizing power supply, and optimizing the operation efficiency of the grid are achieved.

[0010] The above description is only an overview of the technical solutions of this application. In order to be able to more clearly understand the technical means of this application, it can be implemented in accordance with the content of the specification. And in order to make the above and other purposes, features, and advantages of this application more obvious and understandable, the following specifically describes the embodiments of this application. It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of this application, nor is it used to limit the scope of this application. Other features of this application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in this application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings described below are only exemplary, and for those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0012] Figure 1 It is a flowchart of the new energy operation evaluation method under extreme meteorological conditions of this application;

[0013] Figure 2 It is a structural diagram of the new energy operation evaluation system under extreme meteorological conditions of this application.

[0014] Description of the reference numerals:

[0015] Extreme meteorological data determination module 11, operation data determination module 12, operation characteristic analysis module 13, operation evaluation model acquisition module 14, prediction data acquisition module 15, dispatching optimization module 16. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] By providing a new energy operation evaluation method and system under extreme meteorological conditions, the present application solves the technical problems existing in the prior art. Due to the intermittent and fluctuating nature of new energy units not being effectively controlled, it is difficult for power grid dispatching to accurately respond to the random changes in new energy power generation, further affecting the stability and operation efficiency of the power grid, potentially leading to unstable power supply, and even causing fluctuations in power grid frequency and voltage. The technical goal of accurately predicting the operating characteristics of new energy units and optimizing dispatching is achieved, and the technical effects of improving power grid dispatching accuracy, stabilizing power supply, and optimizing power grid operation efficiency are achieved.

[0017] Next, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited by the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application. Additionally, it should be noted that for the sake of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0018] Example 1. Please refer to the attached Figure 1 , the present application provides a new energy operation evaluation method under extreme meteorological conditions, which is applied to a new energy operation evaluation system under extreme meteorological conditions, and specifically includes:

[0019] Step 1: Connect to the intelligent meteorological system to determine the extreme meteorological data within the target evaluation area, where the extreme meteorological data has historical time records.

[0020] Specifically, connecting to the intelligent meteorological system means integrating meteorological data collection and processing technologies with an intelligent system so that it can provide real-time meteorological information within the region. This system includes an automated weather station, satellite data, and a meteorological prediction model, which provides accurate meteorological data such as wind speed, wind direction, precipitation, temperature, humidity, etc. The intelligent meteorological system can not only provide real-time data but also predict future weather based on historical data and prediction models to provide a basis for subsequent analysis.

[0021] Determining the extreme meteorological data within the target evaluation area means screening and collecting those extreme meteorological data that can have a significant impact on the operation of new energy units within a specific geographical area. Extreme meteorological data includes, but is not limited to, extreme weather phenomena such as storms, extreme temperatures, heavy precipitation, and hail. The target evaluation area can be a specific power plant area or a wide wind power or photovoltaic power generation area.

[0022] Among them, the extreme meteorological data has historical time records, which means that these data are not temporary data, but historical data with long-term records and accumulations. The historical data comes from various channels such as meteorological stations, satellite monitoring, and meteorological forecast models, covering multiple years or time periods. These historical data can help analyze past extreme meteorological events and their impacts on new energy units, providing data support for future prediction models and grid dispatching strategies. Through the accumulation and analysis of these data, the impacts of specific extreme meteorological conditions on the unit performance can be more accurately identified, providing a data basis for subsequent predictions and decisions.

[0023] Step 2: Determine the operation data of the new energy units in the target evaluation area by interacting with the intelligent operation and maintenance monitoring system.

[0024] Specifically, by interacting with the intelligent operation and maintenance monitoring system, the real-time monitoring and management of the new energy units in the target area can be realized. The intelligent operation and maintenance monitoring system is a system integrating sensors, data acquisition devices, remote communication technologies, and data analysis platforms, which can collect the operation status information of new energy units in real time, such as the power generation power, operation speed, unit fault conditions, temperature, humidity, etc. of wind turbines. Information interaction refers to the data sharing and real-time transmission between systems, which can ensure that the system can make decisions and dispatching arrangements based on the latest operation data.

[0025] Extract the real-time operation data of the new energy units in a specific area from the intelligent operation and maintenance monitoring system, including the power generation power, equipment status, fault diagnosis, operation time, load fluctuation, etc. of the units. By extracting the operation data, it helps the manager accurately understand the current operation situation of the units, evaluate whether the units are in the best operation state, or whether there are abnormalities that require maintenance or adjustment.

[0026] Step 3: Extract features from the extreme meteorological data to obtain meteorological features, and analyze the operation characteristics of the new energy units with the operation data and the meteorological features, where the operation characteristics include intermittency, volatility, and uncertainty.

[0027] Specifically, extract the feature information closely related to the power generation performance of the new energy units from the extreme meteorological data. The extreme meteorological data includes wind speed, temperature, humidity, air pressure, etc., and feature extraction refers to extracting the key indicators that can affect the unit performance through data analysis methods such as statistics and machine learning. For example, by analyzing the changes in wind speed and temperature, important meteorological features affecting the performance of wind power generation and photovoltaic power generation units can be obtained. These meteorological features provide basic data for subsequent analysis.

[0028] Combine the data of new energy units during actual operation (such as power generation, equipment status, start-stop time, etc.) with the meteorological characteristics extracted from meteorological data to analyze the performance of the units under extreme meteorological conditions. The operation data reflects the actual operation of the units, while the meteorological characteristics show the impact of the external environment on the operation of the units. By combining these two types of data, the operation ability and characteristics of new energy units under complex meteorological conditions can be evaluated more accurately.

[0029] Among them, the operation characteristics include intermittency, volatility, and uncertainty. Intermittency means that new energy units (such as wind power, solar energy, etc.) have obvious intermittent power generation in different time periods, that is, the units may generate a large amount of electricity in some periods, while generating almost no electricity in other periods. Volatility refers to the obvious fluctuations in the power generation of new energy units with the change of the external environment. For example, the change in wind speed may cause drastic fluctuations in the output power of wind turbines. Uncertainty means that due to the instability of meteorological conditions, it is difficult to predict the power generation of new energy units, and there may be large errors, which poses challenges to power grid dispatching. By analyzing these characteristics, the performance of new energy units under different meteorological conditions can be better understood, and data support can be provided for optimizing power grid dispatching.

[0030] Step 4: Train a pre-constructed neural network architecture with the meteorological characteristics and the operation characteristics to obtain an operation evaluation model.

[0031] Specifically, use the meteorological characteristics and the operation characteristics to train a pre-constructed neural network architecture. The meteorological characteristics may include meteorological parameters such as wind speed and temperature, and the operation characteristics may involve information such as the operation status and load of the units. A neural network is a computational model that simulates the connection of human brain neurons and has powerful pattern recognition and prediction capabilities. The training process includes two stages: forward propagation and backward propagation. In the forward propagation stage, the input data is passed through the network layer by layer to generate a prediction result; in the backward propagation stage, according to the error between the prediction result and the actual result, the weights and biases in the network are adjusted to reduce the error. Through multiple iterative trainings, the neural network can gradually optimize its parameters and improve the prediction accuracy.

[0032] Step 5: Based on the operation evaluation model, conduct short-term future operation prediction for the new energy units to obtain prediction data.

[0033] Specifically, by using the constructed operation evaluation model and inputting the current meteorological data and the real-time operation data of the unit, the operation situation of the new energy unit in the future for a period of time is predicted. Short-term operation prediction usually refers to the prediction within the next few hours or days, making inferences based on the current environment and operation conditions. The results of the future operation state of the new energy unit obtained by calculating through the operation evaluation model. The prediction data includes the power generation power, load demand, operation efficiency, etc. of the unit.

[0034] Step Six: Conduct index evaluation on the said prediction data to optimize the grid dispatching strategy with the evaluation results.

[0035] Specifically, conduct systematic evaluation on the prediction results obtained from the operation evaluation model to determine the accuracy and reliability of the prediction data. Index evaluation usually includes the inspection of multiple parameters, such as statistical indicators like root mean square error, which can understand the error degree of the model prediction. For example, by calculating the difference between the predicted power and the actual power, an error index is obtained, thereby understanding the performance of the model in predicting the power generation power.

[0036] After obtaining the evaluation results, use these results to adjust the grid dispatching plan. The grid dispatching strategy involves issues such as how to reasonably arrange the start-stop sequence of generating units and how to balance the load and power generation demands. Through the evaluation of the prediction data, it can be identified which units have larger prediction errors, thereby adjusting the dispatching strategy to ensure the stability and efficient operation of the grid. For example, if the fluctuations of some units in the prediction are relatively large, the dispatching system can appropriately adjust the start-stop sequence of these units to avoid the situation of over-relying on a single unit.

[0037] The new energy operation evaluation method under the extreme meteorological environment is applied to the new energy operation evaluation system under the extreme meteorological environment, which can achieve the technical goal of accurately predicting the operation characteristics of the new energy unit and optimizing the dispatching, and achieve the technical effects of improving the grid dispatching accuracy, stabilizing the power supply, and optimizing the grid operation efficiency.

[0038] Furthermore, this application also includes: establishing a data interface with the intelligent meteorological system, connecting with the intelligent meteorological system through the data interface to obtain extreme meteorological data, where the extreme meteorological data includes historical time data and real-time time data; inputting the real-time time data into the operation evaluation model for model prediction to obtain the prediction data.

[0039] Specifically, establishing a data interface with the intelligent meteorological system means connecting different systems so that the two systems can share data and information. A data interface refers to a standardized channel that enables the exchange of data between different technical platforms or systems. By establishing a data interface, meteorological data in the intelligent meteorological system can be extracted and transmitted in real time to other systems or platforms that need to use this data, ensuring interoperability between systems and enabling seamless flow of meteorological data to support subsequent analysis or decision-making.

[0040] Connecting to the intelligent meteorological system through a data interface to obtain extreme meteorological data means establishing a data interface to connect to the intelligent meteorological system in real time and obtain extreme meteorological data, including extreme weather events such as strong winds, heavy rains, extreme cold or extreme heat weather, etc. These phenomena may have a significant impact on the power generation performance of new energy units. Extreme meteorological data not only includes real-time observed meteorological conditions but also historical meteorological data. Historical data helps to understand the characteristics of past weather events and their impact on unit operation and provides a reference for future predictions.

[0041] Among them, extreme meteorological data includes historical time data and real-time time data, which means that the obtained data includes historical time data and real-time time data. Historical time data refers to meteorological records over a past period, such as wind speed, temperature, and precipitation in the past few years. Real-time time data refers to real-time observed data at the current moment or within a very short time range. These data reflect the current weather conditions and can help predict the impact of future weather on new energy units in real time.

[0042] Taking the meteorological data obtained in real time as input and inputting it into the established operation evaluation model to predict the power generation performance. The operation evaluation model is usually based on historical data and mathematical models. By analyzing real-time time data, it predicts the operation performance of new energy units under specific meteorological conditions, such as power generation power, unit load, etc. The prediction results provide support for power grid dispatching and unit maintenance and help to adjust the power grid operation strategy in real time.

[0043] After inputting the real-time time data into the operation evaluation model, the model calculates and analyzes based on the input data and finally obtains the prediction results. The prediction data includes the power generation power, equipment load, and possible fault conditions of the unit within a certain future time.

[0044] Furthermore, this application also includes: calculating the meteorological index value of the meteorological index through the meteorological characteristics, where the meteorological index at least includes the wind speed change rate and temperature fluctuation; identifying the start-stop mode and periodic characteristics with the meteorological index value and the operation data to obtain the intermittency; calculating the volatility index with the meteorological index value and the operation data to obtain the volatility; and calculating the prediction error with the meteorological index value and the operation data to obtain the uncertainty.

[0045] Specifically, representative and influential features are extracted from the collected meteorological data, and then the values of meteorological indexes are calculated, including wind speed, temperature, humidity, etc. The meteorological indexes are calculated based on these features, such as the wind speed change rate, temperature fluctuation, etc.

[0046] Among them, the meteorological index at least includes the wind speed change rate and temperature fluctuation, which means that in the meteorological data, at least these two important meteorological indexes, the wind speed change rate and temperature fluctuation, need to be calculated. The wind speed change rate can reflect the fluctuation of wind power, which is particularly important for wind power generation because the instability of wind speed directly affects the power generation capacity of wind turbines. Temperature fluctuation affects solar photovoltaic power generation because the power generation efficiency of photovoltaic panels is greatly affected by temperature, and temperature fluctuation will cause the power generation power of photovoltaic modules to change.

[0047] Using meteorological index values such as the wind speed change rate and temperature fluctuation and the actual operation data of the unit, analyze the start-stop mode and periodic characteristics of the new energy unit. The start-stop mode reflects the rules of the unit starting and stopping under different meteorological conditions, and the periodic characteristic refers to the regular fluctuation performance of the unit within a certain time range. By identifying these modes and characteristics, it can be revealed whether the operation of the unit shows intermittency. Intermittency means that the power generation output of the unit is not continuous but changes intermittently with the change of the external environment (such as wind speed, solar radiation, etc.).

[0048] Based on the meteorological index value and the unit operation data, calculate the volatility of the generated power. Volatility refers to the degree of fluctuation of the power generation output of the new energy unit within a certain time, which is caused by factors such as wind speed change and light change. By calculating the volatility, the instability degree of the power generation amount can be quantified, and then provide a reference for power grid dispatching. For example, if the wind speed of a wind turbine changes greatly within a certain period, it may cause a sharp fluctuation in the generated power, and at this time the volatility is relatively high.

[0049] Based on the meteorological index values and the actual operating data of the unit, calculate the error between the prediction result of the calculation model and the actual result, including the uncertainty of the prediction. Since the accuracy of the prediction model is affected by meteorological changes and the unit status, there may be a deviation between the prediction result and the actual situation. Uncertainty assessment can help identify areas with low prediction accuracy and provide a basis for improving the model.

[0050] Furthermore, this application also includes: evaluating the influence coefficient of the meteorological index value on the prediction error, determining the source of uncertainty according to the influence coefficient; using Monte Carlo simulation to generate uncertainty scenarios for the source of uncertainty; evaluating the infection coefficient of the uncertainty scenario on the prediction error, and performing incremental learning on the source of uncertainty with the infection coefficient until the infection coefficient is less than the infection coefficient threshold to generate meteorological index monitoring values; monitoring the neural network architecture based on the meteorological index monitoring values to obtain the operation evaluation model.

[0051] Specifically, by analyzing the degree of error generated by meteorological index values (such as wind speed change rate, temperature fluctuation, etc.) on the prediction result, the influence coefficient is calculated. The influence coefficient reflects how much the change of different meteorological indexes affects the inaccuracy of the prediction result. For example, if the wind speed change rate changes violently during a certain period, it may lead to a large prediction error, then the influence coefficient of the wind speed change rate will be higher. By evaluating these influence coefficients, it is possible to identify which meteorological indexes play a key role in the prediction process, thus providing a basis for further optimizing the prediction model.

[0052] Through the influence coefficient, the specific sources leading to the prediction error can be traced and found, including factors such as the instability of meteorological data itself, sensor errors, and incomplete model assumptions. For example, if it is found that the influence coefficient of the wind speed change rate is large, it indicates that the instability of the wind speed is the main factor leading to the prediction error.

[0053] Use the Monte Carlo method to simulate the possible results under different meteorological conditions and construct multiple scenarios for the source of uncertainty. The Monte Carlo simulation is a method for estimating the solution of a problem through random sampling and is suitable for dealing with complex systems containing uncertainties. By simulating a large number of possible uncertainty scenarios, it is used to evaluate the change range of the prediction error and its impact on the result under different meteorological conditions. For example, under different wind speeds and temperature fluctuations, simulate different power generation situations to more comprehensively analyze the impact of uncertainty on the prediction result.

[0054] By simulating different scenarios, calculate the impact degree of prediction errors on the entire system in each scenario. The infection coefficient represents the propagation and amplification effect of prediction errors in different scenarios, reflecting the degree of interference of uncertainty on prediction results. By evaluating the infection coefficient, we can further understand which factors have the greatest impact on prediction results in different meteorological scenarios. For example, in some extreme meteorological scenarios, the fluctuation of wind speed will lead to a significant increase in prediction errors, so the infection coefficient of these scenarios will be relatively high.

[0055] By continuously adjusting and optimizing the model and conducting incremental learning on the sources of uncertainty, the impact of prediction errors is gradually reduced. Incremental learning is a method to gradually improve the model. By continuously inputting new data and feedback, the prediction ability of the model is optimized until the infection coefficient is lower than the set threshold, which means that the prediction errors of the model are effectively controlled and the monitored values of meteorological indicators generated are more accurate. The model gradually reduces uncertainty through learning and improves the reliability of predictions.

[0056] Based on the monitored values of meteorological indicators as input, monitor and optimize the neural network architecture. Through learning historical data and monitoring data, the neural network architecture can generate a relatively accurate operation evaluation model to predict the power generation status, load, etc. of new energy units and assist in power grid dispatching decisions. By continuously optimizing the network architecture, its adaptability to uncertain factors is improved, thereby obtaining more accurate prediction results.

[0057] Furthermore, this application also includes: calculating the root mean square error with the root mean square as the evaluation index, evaluating the prediction accuracy based on the error calculation result to obtain an evaluation result; using the evaluation result as an optimization constraint to optimize the start-stop sequence and output level of the new energy units to obtain an optimized dispatching strategy.

[0058] Specifically, when evaluating the accuracy of prediction results, the root mean square error is used as a standard to measure the size of errors. The root mean square error is obtained by calculating the square root of the average of the squares of the differences between the predicted values and the actual values, which is a numerical value reflecting the prediction accuracy of the model. By calculation, the error degree of the model prediction results can be quantified. The smaller the error, the more accurate the prediction result. As an evaluation index, the root mean square error can clearly reflect the performance of the model in different scenarios and help further improve and optimize the model.

[0059] Evaluate the prediction accuracy using the calculation results. By comparing the root mean square error with the actual allowable error, it can be determined whether the prediction results meet the accuracy requirements. For example, in the power generation prediction of a wind farm, if the root mean square error between the actual wind speed and the predicted wind speed is large, it may mean that the model fails to capture the wind speed changes accurately, resulting in a large prediction error. Therefore, the error calculation results will directly affect the evaluation of the prediction accuracy and help to further identify and improve the weaknesses of the model. Through error calculation and prediction accuracy evaluation, specific evaluation results are obtained. The evaluation results can be a numerical value indicating the high or low prediction accuracy, or a range indicating the size of the prediction error.

[0060] According to the obtained evaluation results, use them as the constraints for optimization to adjust the start-stop sequence and output level of the new energy units. The start-stop sequence and output level of the new energy units are directly related to the load regulation ability of the power grid and the stability of the system. By optimizing the start-stop sequence, the maximum utilization efficiency of the units can be ensured, and unnecessary losses can be avoided under extreme meteorological conditions. Optimizing the output level means reasonably adjusting the output of each unit according to the predicted power grid load demand and meteorological conditions to ensure the stable operation of the power grid, and finally obtaining a reasonable scheduling strategy that can schedule the units according to the prediction results under extreme meteorological conditions or during daily operation. Optimizing the scheduling strategy not only helps to improve the operation efficiency of the new energy units but also reduces the impact on the stability of the power grid.

[0061] Furthermore, this application also includes: classifying the units according to the start-stop time, response speed, and fuel cost of the new energy units to obtain classified units; obtaining the start-stop sequence of the classified units through integer programming of the start-stop sequence; setting start-stop sequence constraint conditions and optimizing the start-stop sequence to obtain the optimal start-stop sequence.

[0062] Specifically, classify the units according to the different characteristics of each new energy unit, such as start-stop time, response speed, and fuel cost. The start-stop time refers to the time required for the unit to start up and be fully operational, the response speed is the reaction speed of the unit when adjusting the load, and the fuel cost refers to the cost of fuel consumed during the power generation process of the unit. These characteristics affect the operation efficiency of the unit and its adaptability in power grid scheduling. Therefore, classifying the units according to these characteristics helps to achieve reasonable scheduling. The classified units can be operated and scheduled differently according to their characteristics, thereby improving the operation efficiency of the entire power grid.

[0063] Through integer programming, optimize the start-stop sequences of different types of units. Integer programming is a mathematical method for solving optimization problems, usually used for problems that require determining the best sequence, location, or configuration, etc. The start-stop sequence refers to the order arrangement of unit startup and shutdown, and integer programming finds the start-stop sequence that makes the entire system operate most efficiently through model calculation. In this way, the start-stop sequences of units can be reasonably arranged to minimize energy consumption, maximize efficiency, meet the grid load demand, and reduce unnecessary energy waste.

[0064] During the process of integer programming, set constraint conditions such as start-stop time, load demand, unit capacity, etc. to ensure that the start-stop sequences of units meet the requirements of stable grid operation. Then, adjust the start-stop sequences through optimization algorithms to minimize the operating cost of the system and maximize the power generation efficiency, and finally obtain the optimal start-stop sequence. The optimal start-stop sequence can not only improve the economic benefits of units, but also effectively reduce the grid load fluctuation and improve the stability and reliability of the grid.

[0065] Furthermore, this application also includes: obtaining the output level of the new energy unit according to the unit performance of the new energy unit; performing flexible scheduling of the new energy unit with the output level to obtain an optimized output level.

[0066] Specifically, according to the specific performance characteristics of each new energy unit, such as power generation capacity, efficiency, response time, etc., determine its power generation output level at a specific moment. The unit performance is usually affected by factors such as the technical parameters of the unit, operating status, and environmental conditions, and the output level refers to the power that the unit can output under given conditions. For example, the unit performance of a wind turbine may be directly related to the wind speed. The greater the wind speed, the stronger the power generation capacity of the unit. Therefore, determining the output level according to meteorological conditions such as wind speed helps to optimize the operation and scheduling of the unit.

[0067] Flexibly adjust the operating mode and output power of new energy units according to the actual output level of each unit to meet the load demand of the power grid and ensure system stability. Flexible scheduling refers to dynamically adjusting the output of units in a power system according to real-time load demand, meteorological conditions, and unit operating status. For example, if solar radiation is strong during a certain period, the output level of photovoltaic power generation units may increase, and the scheduling system will adjust the output of the units according to this change to avoid over-generation or under-generation. Through flexible scheduling, the volatility of new energy can be maximally utilized to improve the efficiency of the power grid. During the process of flexible scheduling, the best power generation output plan is found by adjusting the output level of new energy units. The purpose of optimization is to maximize power generation efficiency, reduce energy waste, and ensure the stable operation of the system on the premise of meeting the load demand of the power grid. For example, in the case of changing wind speeds, some units may face problems of excessive or insufficient output. By optimizing the output level, the power generation of these units can be balanced, ensuring the stability of the power grid and avoiding unnecessary energy losses.

[0068] In summary, the new energy operation evaluation method provided by this application in an extreme meteorological environment has the following technical effects: By connecting to the intelligent meteorological system, the extreme meteorological data within the target evaluation area is determined, where the extreme meteorological data has historical time records; By interacting with the intelligent operation and maintenance monitoring system, the operation data of new energy units within the target evaluation area is determined; Feature extraction is performed on the extreme meteorological data to obtain meteorological features, and the operation characteristics of the new energy units are analyzed based on the operation data and the meteorological features, where the operation characteristics include intermittency, volatility, and uncertainty; The pre-constructed neural network architecture is trained with the meteorological features and the operation characteristics to obtain an operation evaluation model; Based on the operation evaluation model, short-term future operation prediction of the new energy units is performed to obtain prediction data; Index evaluation is performed on the prediction data, and the grid scheduling strategy is optimized based on the evaluation results. That is to say, by achieving the technical goals of accurate prediction of the operation characteristics of new energy units and scheduling optimization, the technical effects of improving the accuracy of grid scheduling, stabilizing power supply, and optimizing the operation efficiency of the power grid are achieved.

[0069] Embodiment 2, based on the new energy operation evaluation method in an extreme meteorological environment in the foregoing embodiment and with the same inventive concept, this application also provides a new energy operation evaluation system in an extreme meteorological environment. Please refer to the appendix Figure 2, including: an extreme meteorological data determination module 11, which is used to connect to the intelligent meteorological system to determine the extreme meteorological data within the target evaluation area, where the extreme meteorological data has historical time records; an operation data determination module 12, which is used to determine the operation data of the new energy units within the target evaluation area by interacting with the intelligent operation and maintenance monitoring system; an operation characteristic analysis module 13, which is used to extract features from the extreme meteorological data to obtain meteorological features, and analyze the operation characteristics of the new energy units based on the operation data and the meteorological features, where the operation characteristics include intermittency, volatility, and uncertainty; an operation evaluation model acquisition module 14, which is used to train a pre-constructed neural network architecture with the meteorological features and the operation characteristics to obtain an operation evaluation model; a prediction data acquisition module 15, which is used to perform short-term future operation prediction on the new energy units based on the operation evaluation model to obtain prediction data; a dispatching optimization module 16, which is used to evaluate the indicators of the prediction data and optimize the power grid dispatching strategy based on the evaluation results.

[0070] Furthermore, the new energy operation evaluation system under extreme meteorological conditions is also used to: establish a data interface with the intelligent meteorological system, connect to the intelligent meteorological system through the data interface, and obtain extreme meteorological data, where the extreme meteorological data includes historical time data and real-time data; input the real-time data into the operation evaluation model for model prediction to obtain the prediction data.

[0071] Furthermore, the new energy operation evaluation system under extreme meteorological conditions is also used to: calculate the meteorological index values of meteorological indicators through the meteorological features, where the meteorological indicators at least include wind speed change rate and temperature fluctuation; identify the start-stop mode and periodic characteristics with the meteorological index values and the operation data to obtain the intermittency; calculate the volatility index with the meteorological index values and the operation data to obtain the volatility; calculate the prediction error with the meteorological index values and the operation data to obtain the uncertainty.

[0072] Further, the new energy operation evaluation system under the extreme meteorological environment is also used for: evaluating the influence coefficient of the meteorological index value on the prediction error, determining the uncertainty source according to the influence coefficient; generating an uncertainty scenario for the uncertainty source by using Monte Carlo simulation; evaluating the infection coefficient of the uncertainty scenario on the prediction error, and performing incremental learning on the uncertainty source with the infection coefficient until the infection coefficient is less than the infection coefficient threshold to generate a meteorological index monitoring value; and monitoring the neural network architecture based on the meteorological index monitoring value to obtain the operation evaluation model.

[0073] Further, the new energy operation evaluation system under the extreme meteorological environment is also used for: calculating the root mean square error with the root mean square as the evaluation index, evaluating the prediction accuracy with the error calculation result to obtain an evaluation result; and optimizing the start-stop sequence and output level of the new energy unit with the evaluation result as the optimization constraint to obtain an optimized scheduling strategy.

[0074] Further, the new energy operation evaluation system under the extreme meteorological environment is also used for: classifying the new energy units according to the start-stop time, response speed, and fuel cost of the new energy units to obtain classified units; obtaining the start-stop sequence of the classified units through integer programming of the start-stop sequence; setting start-stop sequence constraint conditions, and optimizing the start-stop sequence to obtain the optimal start-stop sequence.

[0075] Further, the new energy operation evaluation system under the extreme meteorological environment is also used for: obtaining the output level of the new energy unit according to the unit performance of the new energy unit; and performing flexible scheduling of the new energy unit with the output level to obtain an optimized output level.

[0076] The various embodiments in this specification are described in a progressive manner, and the key points of each embodiment are the differences from other embodiments. The new energy operation evaluation method and specific examples under the extreme meteorological environment in the foregoing Embodiment 1 are equally applicable to the new energy operation evaluation system under the extreme meteorological environment in this embodiment. Through the foregoing detailed description of the new energy operation evaluation method under the extreme meteorological environment, those skilled in the art can clearly know the new energy operation evaluation system under the extreme meteorological environment in this embodiment. Therefore, for the sake of brevity of the specification, it will not be described in detail here.

[0077] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0078] Obviously, those skilled in the art can make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of this application and its equivalent technologies, this application also intends to cover these changes and modifications.

Claims

1. A new energy operation evaluation method under extreme weather conditions, characterized in that: include: Connecting to a smart meteorological system to determine extreme meteorological data in a target assessment area, wherein the extreme meteorological data has a historical time record; Determine the operation data of the new energy units in the target assessment area by exchanging information with the intelligent operation and maintenance monitoring system; Extracting features from the extreme meteorological data to obtain meteorological features, and analyzing the operating characteristics of the new energy generator set using the operating data and the meteorological features, wherein the operating characteristics include intermittency, volatility, and uncertainty; Training a pre-built neural network architecture with the meteorological characteristics and the operational characteristics to obtain an operational assessment model; Based on the operation evaluation model, a future short-term operation forecast is performed on the new energy unit to obtain forecast data; An indicator evaluation is performed on the predicted data to optimize the power grid dispatching strategy based on the evaluation results.

2. The new energy operation evaluation method under extreme weather conditions according to claim 1, characterized in that: include: Establishing a data interface with the smart meteorological system, connecting with the smart meteorological system through the data interface, and acquiring extreme meteorological data, wherein the extreme meteorological data includes historical time data and real-time time data; The real-time data is input into the running evaluation model to perform model prediction to obtain the predicted data.

3. The new energy operation evaluation method under extreme weather conditions according to claim 1, characterized in that: include: Calculating a meteorological index value of a meteorological index through the meteorological characteristics, wherein the meteorological index at least includes a wind speed change rate and a temperature fluctuation; Using the meteorological index value and the operating data to identify the start-stop mode and periodic characteristics, and obtain the intermittent; Calculating a volatility index using the meteorological index value and the operating data to obtain the volatility; The uncertainty is obtained by performing prediction error calculation using the meteorological index value and the operating data.

4. The new energy operation evaluation method under extreme weather conditions as claimed in claim 3, characterized in that: include: Evaluate the influence coefficient of the meteorological index value on the prediction error, and determine the source of uncertainty according to the influence coefficient; Using Monte Carlo simulation to generate uncertainty scenarios for the sources of uncertainty; Evaluate the infection coefficient of the forecast error through the uncertainty scenario, perform incremental learning on the uncertainty source with the infection coefficient until the infection coefficient is less than an infection coefficient threshold, and generate a meteorological indicator monitoring value; The neural network architecture is monitored based on the meteorological indicator monitoring value to obtain the operation evaluation model.

5. The new energy operation evaluation method under extreme weather conditions according to claim 1, characterized in that: include: The root mean square error is calculated using the root mean square as the evaluation index, and the prediction accuracy is evaluated based on the error calculation result to obtain the evaluation result; Taking the evaluation results as optimization constraints, the start and stop sequence and output level of the new energy units are optimized to obtain an optimized scheduling strategy.

6. The new energy operation evaluation method under extreme weather conditions as claimed in claim 5, characterized in that: include: Classify the units according to the start and stop time, response speed, and fuel cost of the new energy units to obtain classified units; The start and stop sequence of the classified units is obtained through integer programming of the start and stop sequence; The start-stop sequence constraints are set, and the start-stop sequence is optimized to obtain the optimal start-stop sequence.

7. The new energy operation evaluation method under extreme weather conditions as claimed in claim 5, characterized in that: include: According to the unit performance of the new energy unit, obtaining the output level of the new energy unit; The new energy generating unit is flexibly dispatched at the output level to obtain an optimized output level.

8. The new energy operation evaluation system under extreme weather environment is characterized by: The steps for implementing the new energy operation evaluation method under extreme meteorological environment described in any one of claims 1 to 7 include: An extreme weather data determination module, the extreme weather data determination module is used to connect to the smart weather system to determine extreme weather data in the target assessment area, wherein the extreme weather data has a historical time record; An operation data determination module, the operation data determination module is used to determine the operation data of the new energy units in the target assessment area by exchanging information with the intelligent operation and maintenance monitoring system; An operation characteristic analysis module, the operation characteristic analysis module is used to extract features from the extreme meteorological data to obtain meteorological features, and analyze the operation characteristics of the new energy unit with the operation data and the meteorological features, wherein the operation characteristics include intermittency, volatility and uncertainty; An operation evaluation model acquisition module, the operation evaluation model acquisition module is used to train a pre-built neural network architecture with the meteorological characteristics and the operation characteristics to obtain an operation evaluation model; A prediction data acquisition module, the prediction data acquisition module is used to perform future short-term operation prediction of the new energy unit based on the operation evaluation model to obtain prediction data; A dispatch optimization module is used to perform an index evaluation on the predicted data to optimize the power grid dispatch strategy based on the evaluation results.

Images