New energy power generation prediction method and system based on extreme weather

Abstract

The invention discloses a new energy power generation prediction method and system based on extreme weather, and the method comprises the steps: collecting first comprehensive data in real time, and carrying out the first preprocessing; and generating a task label through first task calculation. And generating a scheduling strategy according to the task label and the pre-processed data, and adjusting the state of the first object. And performing first optimization processing on the first object state to obtain an optimized scheduling strategy. According to the invention, response can be made in time when extreme weather occurs; through fusing the weather event label and the equipment health state data, the generated scheduling strategy can dynamically adjust the operation state of the power generation equipment, and ensures that the equipment outputs according to the optimal power. And an uncertainty quantification method is adopted to quantify the uncertainty of power generation prediction, and the stability of the power grid is evaluated. Based on the uncertainty quantification result, the load distribution of the power generation equipment is further optimized, the overload operation of high-risk equipment is avoided, and the risks of equipment failure and power grid instability are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of renewable energy power generation prediction, and in particular to a renewable energy power generation prediction method and system based on extreme weather. Background Art

[0002] With the rapid development of new energy power generation technology, the proportion of renewable energy such as wind power and solar energy in the power system has gradually increased. However, since these energy generation processes are highly dependent on meteorological conditions, especially the occurrence of extreme weather events, it will cause drastic fluctuations in power generation capacity, posing huge challenges to grid dispatching and operation. At present, most new energy power generation prediction systems rely on meteorological forecast models and simple historical power generation data fitting, which cannot fully cope with power generation fluctuations under extreme weather conditions. In addition, the real-time collection and utilization of equipment health status data has not been effectively combined with the weather forecast system, resulting in the difficulty of early warning of the risks of equipment failure and power fluctuations.

[0003] The shortcomings of existing technologies are mainly reflected in the following two aspects: first, the accuracy of extreme weather forecasts is limited, especially when historical meteorological data and equipment health status data are not fully utilized, the reliability of power generation forecast models is low; second, the existing power generation equipment scheduling strategy lacks dynamic adjustment capabilities and cannot be optimized according to real-time changes in extreme weather and equipment health status, resulting in power generation equipment operating in a high-risk state, increasing the risk of grid instability. Therefore, how to combine meteorological data and equipment health status data to identify extreme weather events in real time and generate optimized scheduling strategies is an urgent problem to be solved in the current field of new energy power generation forecasting. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is that the accuracy of extreme weather forecasts is limited, especially when historical meteorological data and equipment health status data are not fully utilized, the reliability of the power generation forecast model is low. Secondly, the existing power generation equipment scheduling strategy lacks dynamic adjustment capabilities and cannot be optimized according to real-time changes in extreme weather and equipment health status, resulting in power generation equipment operating in a high-risk state, increasing the risk of grid instability. Therefore, how to combine meteorological data and equipment health status data to identify extreme weather events in real time and generate optimized scheduling strategies is a problem that urgently needs to be solved in the current field of new energy power generation forecasting.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for predicting new energy power generation based on extreme weather, comprising: real-time collection of first comprehensive data and first pre-processing.

[0007] Generate a task label through the first task calculation.

[0008] A scheduling strategy is generated according to the task label and the pre-processed data, and the state of the first object is adjusted.

[0009] A first optimization process is performed on the first object state to obtain an optimized scheduling strategy.

[0010] As a preferred solution of the method for predicting new energy power generation based on extreme weather described in the present invention, wherein: the first comprehensive data is the first object operation related data collected by the sensor.

[0011] As a preferred solution of the extreme weather-based renewable energy power generation prediction method described in the present invention, the first pre-processing includes pre-processing the collected first comprehensive data to improve data accuracy and standardization.

[0012] As a preferred solution of the extreme weather-based renewable energy power generation prediction method described in the present invention, the first task calculation includes extracting features from the data after the first pre-processing, calculating the first task through a classification algorithm, and obtaining a task label.

[0013] As a preferred solution of the extreme weather-based renewable energy power generation prediction method described in the present invention, generating a scheduling strategy based on the task label and pre-processed data includes setting optimization objectives and constraints and solving them to generate a final scheduling strategy.

[0014] As a preferred solution of the extreme weather-based new energy power generation prediction method described in the present invention, the first comprehensive data includes meteorological data and health status data of the first object.

[0015] The first pre-processing includes but is not limited to data cleaning and data standardization.

[0016] The first task is to identify and judge weather events.

[0017] The first object includes, but is not limited to, power generation equipment.

[0018] The first optimization process includes, but is not limited to, uncertainty analysis.

[0019] As a preferred solution of the extreme weather-based new energy power generation prediction method described in the present invention, wherein: the first optimization processing of the state of the first object to obtain the optimized scheduling strategy includes using an uncertainty quantification method to perform uncertainty analysis on the adjusted operating state of the first object to generate an uncertainty quantification result. Based on the uncertainty quantification result, the scheduling strategy of the first object is optimized to adjust the load distribution of the first object.

[0020] A new energy power generation prediction system based on extreme weather, characterized by: including:

[0021] The data collection module collects the first comprehensive data in real time and performs the first pre-processing.

[0022] The calculation module generates a task tag by calculating the first task, generates a scheduling strategy according to the task tag and the pre-processed data, and adjusts the state of the first object.

[0023] The optimization module performs a first optimization process on the first object state to obtain an optimized scheduling strategy.

[0024] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0025] A computer-readable storage medium stores a computer program, which implements the steps of the method described above when executed by a processor.

[0026] The beneficial effects of the present invention are as follows: by collecting meteorological data and equipment health status data in real time, and combining classification algorithms to identify potential extreme weather events, corresponding weather event labels are generated, so that timely responses can be made when extreme weather occurs. By integrating weather event labels and equipment health status data, the generated scheduling strategy can dynamically adjust the operating status of power generation equipment to ensure that the equipment outputs at the optimal power. Using uncertainty quantification methods, the adjusted equipment operating status is deeply analyzed, the uncertainty of power generation forecasts is quantified, and the stability of the power grid is evaluated. Based on the uncertainty quantification results, the load distribution of power generation equipment is further optimized to avoid overload of high-risk equipment and reduce the risk of equipment failure and power grid instability. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0028] Figure 1 An overall flow chart of a method and system for predicting renewable energy power generation based on extreme weather provided in the first embodiment of the present invention.

[0029] Figure 2 A schematic diagram of a power grid stability risk assessment method and system for predicting renewable energy power generation based on extreme weather provided in the first embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0031] Example 1, reference Figure 1-2 , which is an embodiment of the present invention, provides a new energy power generation prediction method based on extreme weather, comprising:

[0032] S1: Collect first comprehensive data in real time and perform first pre-processing.

[0033] In the present invention, the first comprehensive data includes meteorological data and equipment health status data, and the meteorological data includes wind speed, wind direction, temperature, humidity and air pressure. The equipment health status data includes fan speed, power generation, bearing temperature, gearbox temperature and vibration signal. Specifically, wind speed is used to evaluate wind power generation capacity, especially the impact on the fan in extreme windy weather. Wind direction is used to determine the direction of wind energy and help the fan adjust the angle. Extreme low temperature in temperature may affect the operation of the equipment, especially the mechanical parts of the fan. High humidity in humidity may affect the electrical part of the equipment, especially in rainstorm weather. Air pressure is used to identify weather changes caused by changes in air pressure, especially storm precursors. Fan speed reflects the operating status of the fan, and too low or too high speed may affect the power generation capacity. Power generation represents the actual power generation capacity of the fan, which is directly related to wind speed and equipment status. Bearing temperature indicates that too high temperature may cause equipment wear, and too low temperature may cause equipment damage. Gearbox temperature indicates that overheating of the gearbox may cause fan failure, and temperature change is a key indicator of equipment health. Vibration signal indicates that abnormal vibration may indicate mechanical failure of the equipment, especially wear of fan blades or gears.

[0034] The first pre-processing includes data cleaning and data standardization.

[0035] Specifically, data cleaning includes missing value processing and outlier removal. Missing value processing refers to using linear interpolation to fill in data that is missing in a short period of time. Outlier removal is to remove obviously abnormal data by setting upper and lower thresholds.

[0036] It should be noted that the first pre-processing includes but is not limited to data cleaning and data standardization processing, and may also be data completion processing and data denoising processing.

[0037] In an optional embodiment of the present invention, the first pre-processing is data completion processing. When part of the data is missing due to a sensor failure, network delay or other interference, the data integrity needs to be restored through a data completion method. Data completion is used to fill in the missing values ​​of meteorological data and equipment health status data. For example, wind speed data or equipment operating status data in certain time periods are missing due to sensor failure. Use time series analysis to locate the time period of missing data and identify which key data (such as wind speed and power generation) are missing. If the data is missing for a short time and the change trend is stable, use linear interpolation to complete the data. If the data is missing for a long time or the data changes have nonlinear characteristics, use a machine learning-based prediction method (such as a random forest or LSTM model) to complete the data. For short-term missing data, use the data of adjacent time points to calculate the linear interpolation value. For long-term missing data, train a machine learning model, input the complete data and related features of adjacent time periods (such as the correlation between wind speed and wind direction), and predict missing values.

[0038] In an optional embodiment of the present invention, the first pre-processing is data denoising. Sensor data may be affected by environmental interference (such as electromagnetic interference, mechanical vibration, etc.) and produce noise. It is necessary to extract the real part of the data through denoising. Data denoising is used to clean up the fluctuation noise of meteorological data (such as wind speed and temperature) and the random noise of equipment health status data (such as vibration signals). Detect abnormal fluctuations through statistical analysis. For example, a large change in wind speed data in a short period of time may be noise. Use frequency domain analysis (such as Fourier transform) to identify high-frequency noise components. Use the sliding average method to smooth time series data. The size of the sliding window is adjusted according to the data fluctuation characteristics (such as 5 minutes or 10 minutes). Use wavelet transform to decompose time series data, treat high-frequency components as noise and remove them, and retain low-frequency parts. Apply the sliding average method to data such as wind speed and temperature, and use the average value in the sliding window as the smoothing value at the current moment. Apply wavelet transform to vibration signal data, decompose it into different frequency components, and reconstruct the data after removing high-frequency noise. Compare the denoised data with the original data to verify whether the noise has been removed while retaining the real change trend. Compare the trends of the denoised data with the historical data to check whether the denoising process has introduced errors.

[0039] S2: Generate a task label through the first task calculation.

[0040] In the present invention, the first task is to identify and judge weather events.

[0041] Specifically, wind speed features, temperature change rate, humidity features, and air pressure change rate are extracted from the preprocessed meteorological data to form a meteorological feature vector. The random forest model is selected as the core classification model for extreme weather event identification. The random forest model is trained using historical meteorological data and historical extreme weather event labels. The meteorological feature vector is input into the trained random forest model for classification and prediction to identify potential extreme weather events. The type, severity, and duration of the identified potential extreme weather events are integrated to generate corresponding weather event labels.

[0042] Extreme weather events that need to be identified include high wind events, cold wave events, and heavy rain events.

[0043] Furthermore, a high wind event refers to an abnormal increase in wind speed, which may endanger wind power generation equipment. A cold wave event refers to a sudden drop in temperature, which may affect the performance or safety of equipment. A rainstorm event refers to an increase in rainfall, which may cause damage to the electrical part of the equipment.

[0044] S3: Generate a scheduling strategy based on the task label and the pre-processed data, and adjust the state of the first object.

[0045] In the present invention, the first object is the power generation equipment, and the extreme weather event label and the health status data of the equipment are obtained, and the initial power generation power and the maximum power generation power of each equipment are initialized, and the maximum power generation power is used as the optimization target. According to the extreme weather event and the health status data of the equipment, the constraints of extreme weather and equipment health status are formed. Based on the optimization target and the constraints, the linear programming simplex method is used to solve the optimal power generation power of each equipment. According to the optimal power generation power of each equipment solved, the final scheduling strategy of each equipment is generated.

[0046] Furthermore, the health status constraint of the equipment means that the bearing temperature and gearbox temperature of the fan cannot exceed the specified safety threshold, and the vibration signal cannot be too high. Extreme weather constraints mean that in windy weather, the fan speed cannot exceed the safety limit of the equipment to avoid mechanical damage. In cold weather, the load of the equipment should be appropriately reduced to prevent damage to the equipment caused by low temperature.

[0047] Establish a digital twin model of the power generation equipment, simulate the operating status of the equipment under extreme weather conditions based on the current meteorological data and equipment health status data, and generate simulation results.

[0048] Based on the simulation results of the digital twin model, the optimal operating state of the equipment under extreme weather conditions is identified, and combined with the generated final scheduling strategy, the control instruction set is obtained.

[0049] The instructions in the control instruction set are transmitted to the control modules of each power generation device through the control system of the power generation device.

[0050] According to the speed command in the scheduling strategy, the motor speed of the fan is controlled by the frequency converter.

[0051] Based on the dynamic load balancing mechanism, the load of each device is monitored in real time, and the load distribution of the power generation equipment is dynamically adjusted in combination with the power generation instruction in the scheduling strategy. The expression is:

[0052]

[0053] Where N represents the currently allocated load of the power generation equipment, M represents the maximum power that can be generated by the equipment, λ represents the adjustment coefficient, and H(t) represents the health status of the equipment at time t.

[0054] During the operation of the equipment, the equipment health status data is continuously monitored, and the equipment with potential failures is identified by combining the predictive maintenance algorithm. The expression is:

[0055] R f (t) = R0·e κ·H(t)

[0056] Among them, R f (t) represents the failure risk of the equipment at time, R0 represents the initial failure risk, and κ represents the failure risk coefficient.

[0057] Based on the equipment with potential failures identified, the event response mechanism in the scheduling strategy is automatically executed. According to the equipment health status and simulation prediction results, the backup equipment is automatically enabled and the equipment with potential failure risks is gradually disabled.

[0058] Specifically, when the load of the device is small (ie, the device load is light), the health state is large and the failure risk will increase rapidly. Conversely, when the load of the device is close to the maximum load, the health state is small and the failure risk will remain at a low level.

[0059] It should be noted that the first object includes but is not limited to power generation equipment, and may also be energy storage equipment and a power grid dispatching system.

[0060] In an optional embodiment of the present invention, the first object is an energy storage device, and the energy storage device (such as a battery energy storage system) plays a role in balancing supply and demand in renewable energy power generation. Under extreme weather conditions, the energy storage device needs to dynamically adjust the charging and discharging strategy according to the weather event label and the renewable energy power generation forecast results to smooth the power generation fluctuations.

[0061] According to the task labels and pre-processed data, a scheduling strategy is generated to adjust the state of the energy storage equipment as follows:

[0062] Identify weather types (e.g., high winds, cold snaps) from extreme weather event tags. Obtain health status data for energy storage devices, including current battery state of charge (SOC), battery temperature, voltage, and life status. Determine the maximum charge and discharge power of energy storage devices. Set the initial SOC to ensure that the SOC is within the operable range (e.g., 20%-80%). Aim to maximize the operating efficiency and life of energy storage devices. In extreme weather conditions, prioritize grid stability and try to avoid overcharging or over-discharging equipment.

[0063] Equipment constraints: The battery SOC cannot exceed 80% or fall below 20%. The battery temperature cannot exceed the safe range (such as 0°C to 40°C).

[0064] Weather constraints: In cold weather, limit charging power to prevent low temperatures from damaging the battery. In windy weather, prioritize energy storage devices to quickly respond to fluctuations.

[0065] Use dynamic programming methods to calculate the charging and discharging power distribution of energy storage devices. Output the charging and discharging plan for each time period to ensure grid stability and extend battery life.

[0066] In an optional embodiment of the present invention, the first object is a power grid dispatching system, which needs to dynamically adjust the operating status of the power grid (such as load distribution, flow regulation and backup dispatching) according to the new energy power generation forecast results and extreme weather event labels to ensure the safe and stable operation of the power grid.

[0067] According to the task labels and pre-processed data, a dispatch strategy is generated to adjust the state of the power grid dispatch system as follows:

[0068] Extract event type and duration (e.g., heavy rain lasts for 3 hours) from extreme weather event labels. Obtain the current operating status of the power grid, including line load, voltage level, power flow distribution and other data. Determine the maximum load capacity of key lines. Set the initial power flow distribution to ensure that the line capacity is not exceeded. Minimize the grid dispatching costs caused by extreme weather with the safety and economy of the grid as the goal. During extreme weather, give priority to ensuring stable power supply in important load areas.

[0069] Grid constraints: Line load cannot exceed the safety threshold, and bus voltage must be maintained within a specified range (such as ±5%).

[0070] Weather constraints: Heavy rain may cause some lines to fail, and the load distribution of these lines needs to be reduced. During cold waves, the power supply to the heating load area needs to be dispatched first. Use power flow optimization algorithms (such as power flow calculation based on the Newton-Raphson method) to optimize load distribution. Adjust the grid operation status according to the optimization results and generate dynamic dispatch strategies, including the activation of backup power sources and the redistribution of line loads.

[0071] S4: Perform a first optimization process on the first object state to obtain an optimized scheduling strategy.

[0072] In the present invention, the first optimization process is uncertainty analysis, which uses uncertainty quantification methods to perform uncertainty analysis on the adjusted operating state of the power generation equipment to generate uncertainty quantification results.

[0073] The health status data of the current equipment is obtained from the adjusted operating status of the power generation equipment.

[0074] According to the historical meteorological data and power generation data, the prior distribution of the impact of meteorological data on power generation is set.

[0075] Using real-time meteorological data and current equipment health status data, the posterior distribution is calculated using the Bayesian update formula, which is expressed as:

[0076]

[0077] Where θ represents the impact of wind speed on power generation, D represents real-time meteorological data and current equipment health status data, P(θ|D) represents the probability distribution of the impact of wind speed on power generation given real-time meteorological data and current equipment health status data, P(D|θ) represents the possibility of observing real-time meteorological data and current equipment health status data given the impact of wind speed on power generation, P(θ) represents the cognition of the impact of wind speed on power generation before observing real-time meteorological data and current equipment health data, and P(D) represents the overall probability of observing real-time meteorological data and current equipment health data.

[0078] Monte Carlo simulation method is used to sample from the calculated posterior distribution. Based on the sampling results, the upper and lower limits of power generation forecast are calculated. Based on the calculation results, the stability of the power grid is evaluated to obtain the grid stability risk assessment results.

[0079] Specifically, low risk: low uncertainty, the upper and lower limits of power generation forecasts are within a reasonable range, and the power grid supply and demand are well balanced. Medium risk: high uncertainty, large differences in power grid supply and demand, but still within controllable range. High risk: excessive uncertainty, serious differences in power grid supply and demand, and the need to immediately activate the emergency plan.

[0080] Based on the uncertainty quantification results, the dispatch strategy of power generation equipment is optimized and the load distribution of power generation equipment is adjusted.

[0081] Based on the uncertainty quantification results, the uncertainty range of power generation forecasts is analyzed and equipment with large power generation fluctuations is identified.

[0082] Based on the equipment health status data, identify the equipment with health risks.

[0083] According to the health status, power generation capacity and uncertainty quantification results of the identified equipment, the optimization goals are set, the scheduling strategy of the power generation equipment is optimized, and the load distribution of the power generation equipment is adjusted.

[0084] It should be noted that the first optimization process includes but is not limited to uncertainty analysis, and may also be robust optimization and real-time feedback optimization.

[0085] In an optional embodiment of the present invention, the first optimization process is robust optimization, which is mainly used to deal with uncertainties and extreme changes that may occur in the system and ensure the stability and executability of the scheduling strategy under different circumstances.

[0086] Extract key influencing factors such as wind speed, temperature, and humidity from real-time meteorological data, and set the upper and lower limits of these parameters. Set an operating range for equipment health status data (such as power generation, speed, and temperature) to ensure that the equipment does not exceed the safety threshold. Establish a robust optimization model containing objective functions and constraints. The objective function is to maximize the power generation of renewable energy and the stability of the power grid. Take the uncertainty of meteorological data and equipment health status (such as the possible fluctuation range of wind speed) as part of the model constraints. Use a robust optimization method based on scenario generation to generate a set of optimized scheduling strategies suitable for different uncertainty scenarios. Select the strategy that performs best in all scenarios to ensure that power generation equipment can still operate safely in extreme weather. Assign a safe power range to each power generation equipment. Under the robust strategy, monitor the operating status of the equipment in real time to ensure that the equipment load distribution is always within the safe range. Output the scheduling strategy and pass it to the equipment control module to adjust the equipment operating status in real time.

[0087] In an optional embodiment of the present invention, the first optimization process is real-time feedback optimization, which dynamically adjusts the scheduling strategy according to the latest changes in the equipment operating status and meteorological conditions to ensure the real-time optimality of the equipment operation. Wind speed, temperature, humidity, power generation and equipment health status data are collected in real time through sensors. Data is transmitted using a fast communication network to ensure that the feedback delay is less than 10 seconds. Combined with meteorological data and equipment health status data, the current operating status is evaluated to determine whether there is a risk of power fluctuation or equipment overload. The power generation power and equipment status at the next moment are estimated using a real-time prediction model. The scheduling strategy is updated based on the latest data, and the operating status of risky equipment is adjusted first, such as reducing the operating power or allocating backup equipment to take over part of the load. If extreme weather conditions persist and there are many risky equipment, the backup energy storage system is started to compensate for the power generation fluctuation. Data is collected again and the optimization strategy is updated at fixed time intervals (such as 5 minutes). During the entire extreme weather event, the equipment operating status is continuously optimized to ensure that the system remains stable in a real-time changing environment. The adjusted operating power of the power generation equipment, the activation status of the backup equipment and the load distribution strategy are output. According to the optimization results, the operating status of the equipment is dynamically adjusted to avoid overload or equipment failure.

[0088] The computer device may be a server. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data cluster data of the power monitoring system. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a new energy power generation prediction method based on extreme weather is implemented.

[0089] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0090] Example 2 is an embodiment of the present invention, which provides a method and system for predicting new energy power generation based on extreme weather. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.

[0091] The test environment is a wind farm equipped with 10 wind turbines. The experimental period is 30 days, and meteorological data and equipment health data are collected every hour. Meteorological data include wind speed, wind direction, temperature, humidity and air pressure, and equipment health status data include wind turbine speed, power generation, bearing temperature, gearbox temperature and vibration signal. Two groups were designed for the experiment: one group used the traditional power generation prediction and scheduling method (existing technology group), and the other group used the method of the present invention (the present invention group). The power generation power, equipment health status and grid stability of each wind turbine were recorded in detail throughout the experimental period. The experimental record data are shown in Table 1.

[0092] Table 1 Experimental data

[0093]

[0094]

[0095] It can be seen from the above experimental data that when the wind speed is high (such as device 1 and device 3), the power generation of the prior art group fluctuates greatly, reaching 12% and 15% respectively, while the present invention group adjusts the wind turbine speed and power generation in real time, and the fluctuation is effectively controlled within 4% and 5%. This shows that the scheduling strategy generation process of the present invention can better cope with extreme weather, reduce the power fluctuation of power generation equipment, and ensure the stability of the power grid.

[0096] In addition, the equipment health status data of the present invention group is also better than that of the prior art group. Under the same meteorological conditions, the bearing temperature, gearbox temperature and vibration signal of the present invention group are lower than those of the prior art group. For example, in the test of equipment 1, the bearing temperature of the prior art group was 75°C, while that of the present invention group was 72°C, and the gearbox temperature also dropped from 65°C to 62°C. This shows that the scheduling strategy of the present invention not only effectively reduces the fluctuation of power generation, but also can extend the service life of the equipment and reduce the loss caused by equipment overheating and vibration.

[0097] Embodiment 3 is an embodiment of the present invention, including a new energy power generation prediction system based on extreme weather, specifically:

[0098] The data collection module collects the first comprehensive data in real time and performs the first pre-processing.

[0099] The calculation module generates a task tag by calculating the first task, generates a scheduling strategy according to the task tag and the pre-processed data, and adjusts the state of the first object.

[0100] The optimization module performs a first optimization process on the first object state to obtain an optimized scheduling strategy.

[0101] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for predicting new energy power generation based on extreme weather, characterized in that: include: Collecting first comprehensive data in real time and performing first pre-processing; Generate a task label through the first task calculation; Generate a scheduling strategy based on the task label and the pre-processed data, and adjust the state of the first object; A first optimization process is performed on the first object state to obtain an optimized scheduling strategy.

2. The method for predicting new energy power generation based on extreme weather according to claim 1, characterized in that: The first comprehensive data is first object operation related data collected by the sensor.

3. The method for predicting new energy power generation based on extreme weather as claimed in claim 2, characterized in that: The first pre-processing includes pre-processing the collected first comprehensive data to improve data accuracy and standardization.

4. The method for predicting new energy power generation based on extreme weather as claimed in claim 3, characterized in that: The first task calculation includes extracting features from the data after the first pre-processing, calculating the first task through a classification algorithm, and obtaining a task label.

5. The method for predicting new energy power generation based on extreme weather as claimed in claim 4, characterized in that: Generating a scheduling strategy based on the task labels and pre-processed data includes setting and solving optimization objectives and constraints to generate a final scheduling strategy.

6. The method for predicting new energy power generation based on extreme weather according to claim 5, characterized in that: The first comprehensive data includes meteorological data and first object health status data; The first pre-processing includes but is not limited to data cleaning and data standardization; The first task is to identify and judge weather events; The first object includes but is not limited to power generation equipment; The first optimization process includes, but is not limited to, uncertainty analysis.

7. The method for predicting new energy power generation based on extreme weather according to claim 6, characterized in that: The performing a first optimization process on the state of the first object to obtain an optimized scheduling strategy includes performing uncertainty analysis on the adjusted operating state of the first object using an uncertainty quantification method to generate an uncertainty quantification result; Based on the uncertainty quantification result, the scheduling strategy of the first object is optimized, and the load distribution of the first object is adjusted.

8. A new energy power generation prediction system based on extreme weather using the method according to any one of claims 1 to 7, characterized in that: A data collection module collects first comprehensive data in real time and performs first pre-processing; A calculation module generates a task label by calculating the first task; Generate a scheduling strategy based on the task label and the pre-processed data, and adjust the state of the first object; The optimization module performs a first optimization process on the first object state to obtain an optimized scheduling strategy.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images