A wind power prediction system and method based on seasonal and climate change

Abstract

The present invention belongs to the technical field of wind power prediction, and specifically is a wind power prediction system and method based on seasonal and climate change. The monitoring module obtains the ambient temperature, wind speed, air density and grid frequency of the wind turbine in a period in real time, and the parameter processing module obtains the ambient temperature difference WDX, wind speed difference FSC, air density mean KSJ and frequency coefficient DWX respectively after analysis based on the above data; the power generation coefficient ZFD is analyzed through data, and a preset power generation coefficient threshold YZD is obtained from a cloud platform, and the power generation coefficient ZFD is compared with the preset power generation coefficient threshold YZD; when the power generation coefficient ZFD is greater than or equal to the preset power generation coefficient threshold YZD, an overload signal is generated, which not only combines the influence of seasonal and climate change factors on wind power, but also combines wind speed, air density and grid frequency factors to obtain wind power prediction results, thereby improving the accuracy of wind power prediction results.

Description

Technical Field

[0001] The present invention belongs to the technical field of wind power prediction, and in particular to a wind power prediction system and method based on seasonal and climate change. Background Art

[0002] Wind power forecasting refers to a wind farm output power forecasting model established using data such as wind farm historical power, historical wind speed, topography, numerical weather forecast, and wind turbine operating status. This forecast is of great significance for the dispatching, optimal operation, and energy management of the power system.

[0003] A patent with publication number CN106934094B discloses a wind power prediction method based on the 24 solar terms. This method starts from the perspective of establishing a prediction model and sample division. Taking into account that the meteorological data sample set within the same solar term can better represent the climate state at the current prediction time, the sample data is divided according to the time nodes of the 24 solar terms, and a prediction model is established with the wind farm data within the same solar term, thereby reducing the impact of seasonal and climate changes to a certain extent and improving the accuracy of wind power prediction. Excessive power generation may bring a series of problems.

[0004] The above-mentioned prior art only considers the impact of seasons and climate change factors on wind power when predicting wind power. However, there are many factors that affect wind power, which are also related to wind speed, air density, grid frequency and other factors, resulting in low accuracy of wind power prediction results. Moreover, the wind power prediction method cannot timely combine real-time data to update data, and cannot accurately judge whether there will be a problem of excessive power generation. Taking measures in advance to reduce power generation will lead to power generation losses.

[0005] To this end, the present invention provides a wind power prediction system and method based on seasonal and climate changes. Summary of the invention

[0006] In order to make up for the deficiencies of the prior art, at least one technical problem raised in the background technology is solved.

[0007] The technical solution adopted by the present invention to solve the technical problem is: the wind power prediction system based on seasonal and climate change described in the present invention comprises:

[0008] Time collection module: The time collection module is used to collect the current time, obtain the corresponding season of the current time month, set the collection cycle according to the current season, and divide the collection cycle into several time nodes, and mark the time nodes as i, where i = 1, 2, 3, ..., n, where n is a positive integer;

[0009] Parameter monitoring module: The monitoring module is used to obtain the ambient temperature, wind speed, air density and grid frequency of the wind turbine in real time during the cycle, and mark them as ambient temperature value WDi, wind speed value FSi, air density value KSi and grid frequency value DWi respectively;

[0010] Parameter processing module: The parameter processing module obtains the ambient temperature difference WDX, the wind speed difference FSC, the air density mean KSJ and the frequency coefficient DWX respectively through analysis according to the ambient temperature value WDi, the wind speed value FSi, the air density value KSi and the power grid frequency value DWi;

[0011] Data analysis module: The data analysis module obtains the ambient temperature difference WDX, wind speed difference FSC, air density mean KSJ and frequency coefficient DWX through the formula The power generation coefficient ZFD of the cycle is obtained, where a1, a2, a3, and a4 are all proportional coefficients, a1+a2+a3+a4=1, and a2>a4>a1>a3;

[0012] Obtain the preset power factor threshold value YZD from the cloud platform, and compare the power factor ZFD with the preset power factor threshold value YZD; when the power factor ZFD ≥ the preset power factor threshold value YZD, an overload signal is generated; when the power factor ZFD < the preset power factor YZD, an external coefficient value is obtained through the ambient temperature value WDi, the wind speed value FSi, the air density value KSi and the grid frequency value DWi, and the external coefficient value is input into the wind power prediction model to obtain the predicted power value;

[0013] Adjustment module: After receiving the overload signal, the adjustment module issues an alarm notification, and the staff adjusts the generator speed through the speed governor to adjust the power generation, and marks the real-time power generation before adjustment as TG;

[0014] Prediction and correction module: The prediction and correction module obtains the preset power generation YTG from the cloud platform, and calculates the power generation TG according to the formula Get the updated preset power generation GX, through the formula Get the correction coefficient ZU, combine the power generation coefficient ZFD and the preset power generation coefficient threshold YZD, and use the formula The updated preset power generation coefficient threshold ZTZ is obtained, and the updated preset power generation power GX and the preset power generation coefficient threshold ZTZ are sent to the cloud platform.

[0015] Furthermore, the ambient temperature difference WDX is specifically obtained in the following manner:

[0016] The ambient temperatures WDi corresponding to all time nodes in the acquisition cycle are sorted from large to small according to the temperature values, and the maximum ambient temperature at the top is marked as WDZ and the minimum ambient temperature at the bottom is marked as WDM. Get the ambient temperature difference WDX.

[0017] Furthermore, the wind speed difference FSC is specifically obtained in the following manner:

[0018] The maximum wind speed FSD and the minimum wind speed FSX corresponding to all time nodes within the collection period are obtained, and the difference between the maximum wind speed FSD and the minimum wind speed FSX is obtained to obtain the wind speed difference FSC.

[0019] Furthermore, the specific method of obtaining the mean air density KSJ is as follows:

[0020] The air density value KSi corresponding to all time nodes within the acquisition period is calculated by the formula Get the average air density KSJ.

[0021] Furthermore, the frequency coefficient DWX is specifically obtained as follows:

[0022] The grid frequency value DWi corresponding to all time nodes within the acquisition period is calculated by the formula The frequency coefficient DWX is obtained.

[0023] Furthermore, the monitoring module is used in real time to obtain the ambient temperature, wind speed, air density and grid frequency of the wind turbine during the cycle, specifically: the ambient temperature value WDi is measured by a temperature sensor, the wind speed value FSi is measured by a wind speed sensor, the air density value KSi is obtained by an air density sensor, and the grid frequency DWi is measured by a frequency sensor.

[0024] Furthermore, the wind power prediction model is obtained by:

[0025] The historical meteorological data of the current season and the corresponding wind power value are obtained periodically, wherein the historical meteorological data include the ambient temperature value LW, the wind speed value LF, the air density value LK and the grid frequency value LD of the wind turbine, and the external coefficient value WBX is obtained by calculation, and the obtained external coefficient value and the corresponding wind power value at the historical prediction moment are used as standard training data; the constructed artificial intelligence model is trained with the standard training data, and the trained artificial intelligence model is marked as a wind power prediction model; by inputting the external coefficient value WBX into the wind power prediction model, a wind power value can be output, and the wind power value is marked as a predicted power value, and the artificial intelligence model is a support vector machine model or a deep convolutional neural network model.

[0026] Furthermore, the external coefficient value WBX is obtained as follows:

[0027] Substitute the ambient temperature value LW, wind speed value LF, air density value LK and grid frequency value LD of the wind turbine into the formula WBX=LW×b1+LF×b2+LK×b3+LD×b4 to obtain the external coefficient value WBX, where b1, b2, b3, and b4 are all preset proportional coefficients, and b1+b2+b3+b4=1.

[0028] A method for predicting wind power based on seasonal and climate change, comprising the following steps:

[0029] S1: Collect the current time, obtain the corresponding season of the current month, set the collection cycle according to the current season, and divide the collection cycle into several time nodes, and mark the time nodes as i, where i = 1, 2, 3, ..., n, where n is a positive integer;

[0030] S2: Obtain the ambient temperature, wind speed, air density and grid frequency of the wind turbine during the period, and mark them as the ambient temperature value WDi, the wind speed value FSi, the air density value KSi and the grid frequency value DWi respectively;

[0031] S3: According to the ambient temperature value WDi, the wind speed value FSi, the air density value KSi and the power grid frequency value DWi, the ambient temperature difference value WDX, the wind speed difference value FSC, the air density mean value KSJ and the frequency coefficient DWX are obtained respectively;

[0032] S4: According to the ambient temperature difference WDX, wind speed difference FSC, air density average KSJ and frequency coefficient DWX, the formula is obtained and then The power generation coefficient ZFD of the cycle is obtained, where a1, a2, a3, and a4 are all proportional coefficients, a1+a2+a3+a4=1, and a2>a4>a1>a3;

[0033] Obtain the preset power factor threshold value YZD from the cloud platform, and compare the power factor ZFD with the preset power factor threshold value YZD; when the power factor ZFD ≥ the preset power factor threshold value YZD, an overload signal is generated; when the power factor ZFD < the preset power factor YZD, an external coefficient value is obtained through the ambient temperature value WDi, the wind speed value FSi, the air density value KSi and the grid frequency value DWi, and the external coefficient value is input into the wind power prediction model to obtain the predicted power value;

[0034] S5: After receiving the overload signal, an alarm notification is issued, and the staff adjusts the generator speed through the speed governor to adjust the power generation, and marks the real-time power generation before adjustment as TG;

[0035] S6: Obtain the preset power generation YTG from the cloud platform, and calculate the power generation according to the real-time power generation TG through the formula Get the updated preset power generation GX, through the formula Get the correction coefficient ZU, combine the power generation coefficient ZFD and the preset power generation coefficient threshold YZD, and use the formula The updated preset power generation coefficient threshold ZTZ is obtained, and the updated preset power generation power GX and the preset power generation coefficient threshold ZTZ are sent to the cloud platform.

[0036] Furthermore, the cloud platform storage is used to store a preset power generation coefficient threshold and a corresponding preset power generation power.

[0037] The beneficial effects of the present invention are as follows:

[0038] 1. The wind power prediction system and method based on seasonal and climate change described in the present invention obtains the ambient temperature, wind speed, air density and grid frequency of the wind turbine in real time during the period through the monitoring module, and the parameter processing module obtains the ambient temperature difference WDX, wind speed difference FSC, air density mean KSJ and frequency coefficient DWX respectively according to the analysis of the above data; the power generation coefficient ZFD is obtained through data analysis, and the preset power generation coefficient threshold YZD is obtained from the cloud platform, and the power generation coefficient ZFD is compared with the preset power generation coefficient threshold YZD; when the power generation coefficient ZFD≥the preset power generation coefficient threshold YZD, an overload signal is generated, which not only combines the influence of seasonal and climate change factors on wind power, but also combines wind speed, air density and grid frequency factors to obtain wind power prediction results, thereby improving the accuracy of wind power prediction results.

[0039] 2. A wind power prediction system and method based on seasonal and climate change described in the present invention obtains the current real-time power generation TG when the power generation coefficient ZFD ≥ the preset power generation coefficient threshold value YZD, obtains the preset power generation YTG through the cloud platform, sums the real-time power generation and the preset power generation and takes the average value to obtain the updated preset power generation GX, and can timely adjust the preset power generation value to make the preset power generation closer to the actual result; then the correction coefficient ZU is calculated by the updated preset power generation GX and the real-time power generation TG, and the updated preset power generation coefficient threshold value ZTZ is calculated by combining the correction coefficient ZU with the power generation coefficient ZFD and the preset power generation coefficient threshold value YZD, and the updated preset power generation GX and the preset power generation coefficient threshold value ZTZ are sent to the cloud platform for storage to replace the original preset power generation coefficient threshold value YZD, so that the subsequent power generation coefficient is judged more accurately, the accuracy of the power generation prediction result is improved, and the data can be updated in time in combination with the real-time data to accurately judge whether the power generation is too large, so as to avoid taking measures in advance to reduce the power generation and cause power generation loss. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The present invention will be further described below in conjunction with the accompanying drawings.

[0041] Figure 1 This is a system block diagram of a wind power prediction system based on seasonal and climate change according to Embodiment 1 of the present invention;

[0042] Figure 2 The present invention is a method flow chart of a method for predicting wind power based on seasonal and climate changes. DETAILED DESCRIPTION

[0043] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below in conjunction with specific implementation methods.

[0044] When predicting wind power, the existing technology only considers the impact of seasons and climate change factors on wind power. However, there are many factors that affect wind power, which are also related to wind speed, air density, grid frequency and other factors, resulting in low accuracy of wind power prediction results. In addition, the wind power prediction method cannot timely combine real-time data to update data, and cannot accurately judge whether there will be a problem of excessive power generation. Taking measures in advance to reduce power generation will lead to power generation losses.

[0045] The present invention obtains the ambient temperature, wind speed, air density and grid frequency of the wind turbine in real time during a period, and obtains the ambient temperature difference WDX, wind speed difference FSC, air density mean KSJ and frequency coefficient DWX through analysis; thereby obtaining the power generation coefficient ZFD, obtaining the preset power generation coefficient threshold YZD from the cloud platform, comparing the power generation coefficient ZFD with the preset power generation coefficient threshold YZD to determine whether there is a possibility of overload of the power generation equipment; if there is no possibility of overload, the external coefficient value can be obtained and input into the wind power prediction model to obtain the predicted power value.

[0046] Embodiment 1

[0047] like Figure 1 As shown, a wind power prediction system based on seasonal and climate change according to an embodiment of the present invention comprises:

[0048] Time acquisition module: The time acquisition module is used to collect the current moment, obtain the corresponding season of the current moment, set the acquisition cycle according to the current season, and divide the acquisition cycle into several time nodes, and mark the time nodes as i, where i = 1, 2, 3, ..., n, where n is a positive integer; specifically, due to seasonal and climatic changes, wind speed, air speed and ambient temperature will affect the power generation capacity of the wind energy system. For example, in some areas, wind power output reaches a peak in spring, decreases in summer, and rises again in autumn and winter. However, even if the wind force is the same in summer and winter, the power generation is also affected by the environment. Therefore, the acquisition cycle can be adjusted according to the current season. In summer, due to the short climate change cycle, the acquisition cycle can be set to 1 hour, while in winter, the climate does not change much in a short period of time, so the acquisition cycle time can be lengthened to reduce power consumption while ensuring the accuracy of wind power prediction.

[0049] Parameter monitoring module: The monitoring module is used to obtain the ambient temperature, wind speed, air density and grid frequency of the wind turbine in real time during the cycle, and mark them as ambient temperature value WDi, wind speed value FSi, air density value KSi and grid frequency value DWi respectively;

[0050] Parameter processing module: The parameter processing module obtains the ambient temperature difference WDX, the wind speed difference FSC, the air density mean KSJ and the frequency coefficient DWX respectively through analysis according to the ambient temperature value WDi, the wind speed value FSi, the air density value KSi and the power grid frequency value DWi;

[0051] Data analysis module: The data analysis module obtains the ambient temperature difference WDX, wind speed difference FSC, air density mean KSJ and frequency coefficient DWX through the formula The power generation coefficient ZFD of the cycle is obtained, where a1, a2, a3, and a4 are all proportional coefficients, a1+a2+a3+a4=1, and a2>a4>a1>a3;

[0052] Obtain the preset power factor threshold value YZD from the cloud platform, and compare the power factor ZFD with the preset power factor threshold value YZD; when the power factor ZFD ≥ the preset power factor threshold value YZD, an overload signal is generated; when the power factor ZFD < the preset power factor YZD, an external coefficient value is obtained through the ambient temperature value WDi, the wind speed value FSi, the air density value KSi and the grid frequency value DWi, and the external coefficient value is input into the wind power prediction model to obtain the predicted power value;

[0053] Adjustment module: After receiving the overload signal, the adjustment module issues an alarm notification, and the staff adjusts the generator speed through the speed governor to adjust the power generation, and marks the real-time power generation before the adjustment as TG; it can be understood that after the adjustment module receives the overload signal, it means that the real-time power generation of the generator is about to exceed the preset power generation. If it is not adjusted, the power generation will be too large in the future. Therefore, the staff needs to make adjustments. The power generation of the generator is related to the speed of the generator. The higher the speed, the greater the power generation. The speed of the generator is controlled by changing the mechanical transmission ratio through the speed governor. The speed can be adjusted according to the output power and load changes of the generator to achieve precise and stable control, and avoid the occurrence of excessive power generation in advance.

[0054] Prediction correction module: The prediction correction module obtains the preset power generation YTG from the cloud platform. Specifically, the initial preset power generation YTG can be set to the rated output power of the generator to ensure that the operating state of the generator is within the rated output power. Exceeding the rated output power may cause damage to the generator. According to the real-time power generation TG, the formula Get the updated preset power generation GX, through the formula Get the correction coefficient ZU, combine the power generation coefficient ZFD and the preset power generation coefficient threshold YZD, and use the formula The updated preset power generation coefficient threshold ZTZ is obtained, and the updated preset power generation power GX and the preset power generation coefficient threshold ZTZ are sent to the cloud platform. It can be understood that the initial value of the preset power generation coefficient threshold value YZD can be obtained by collecting the power generation coefficient when the generator is at rated output power according to the ambient temperature, wind speed, air density and grid frequency of the generator; when the power generation coefficient ZFD ≥ the preset power generation coefficient threshold value YZD, the current real-time power generation TG is obtained, the preset power generation YTG is obtained through the cloud platform, the real-time power generation and the preset power generation are summed and averaged to obtain the updated preset power generation GX, so that the preset power generation value can be adjusted in time, so that the preset power generation is closer to the actual result; the correction coefficient ZU is calculated by the updated preset power generation GX and the real-time power generation TG, and the updated preset power generation coefficient threshold value ZTZ is calculated by the correction coefficient ZU combined with the power generation coefficient ZFD and the preset power generation coefficient threshold value YZD, and the updated preset power generation GX and the preset power generation coefficient threshold value ZTZ are sent to the cloud platform for storage to replace the original preset power generation coefficient threshold value YZD, so that the subsequent power generation coefficient is judged more accurately, and the accuracy of the power generation prediction result is improved.

[0055] The specific method for obtaining the ambient temperature difference WDX is as follows:

[0056] The ambient temperatures WDi corresponding to all time nodes in the acquisition cycle are sorted from large to small according to the temperature values, and the maximum ambient temperature at the top is marked as WDZ and the minimum ambient temperature at the bottom is marked as WDM. The ambient temperature difference WDX is obtained. Specifically, wind power is related to ambient temperature. The maximum ambient temperature in the acquisition period is marked as WDZ and the minimum ambient temperature is marked as WDM, and the ambient temperature difference is calculated. The power factor can be calculated more accurately in combination with the ambient temperature factor.

[0057] The specific method for obtaining the wind speed difference FSC is as follows:

[0058] The maximum wind speed FSD and the minimum wind speed FSX corresponding to all time nodes in the collection period are obtained, and the difference between the maximum wind speed FSD and the minimum wind speed FSX is obtained to obtain the wind speed difference FSC. Specifically, wind speed directly affects wind power. By collecting the maximum wind speed FSD and the minimum wind speed FSX in the collection period, the wind speed difference FSC is calculated. Combined with the wind speed factor, the power generation coefficient can be calculated more accurately.

[0059] The specific method for obtaining the mean air density KSJ is:

[0060] The air density value KSi corresponding to all time nodes within the acquisition period is calculated by the formula The average air density KSJ is obtained. Specifically, the air density changes with temperature. The higher the temperature, the lower the air density, which leads to a decrease in wind power. By calculating the average air density in the cycle, the average air density KSJ is obtained. Combining the air density factor can more accurately calculate the power generation coefficient.

[0061] The specific method of obtaining the frequency coefficient DWX is as follows:

[0062] The grid frequency value DWi corresponding to all time nodes within the acquisition period is calculated by the formula The frequency coefficient DWX is obtained. Specifically, the grid frequency refers to the number of times the alternating current changes per second. Under steady-state conditions, the grid frequency is basically equal, but when the grid frequency fluctuates greatly, it will affect the operation of the entire power system. When the active power load of the grid changes and the input power of the generator prime mover cannot be adjusted immediately, the imbalance between the power load and the power generation load will cause the generator speed to change, thereby causing the grid frequency to fluctuate. For example, when the power generation output of the grid is constant, an increase in the grid load will lead to insufficient system active power, a decrease in the generator speed, and thus a decrease in the grid frequency; on the contrary, a decrease in the grid load will cause excess system active power, an increase in the generator speed, and an increase in the grid frequency. This frequency fluctuation has an impact on the operation of both electrical equipment and the power system. If the grid frequency deviates from the rated frequency of the equipment, it will affect the normal operation of the equipment and may even cause damage to the equipment. At the same time, the fluctuation of the grid frequency will also affect the stability of the power supply; if the grid frequency deviates from the rated frequency, it will affect the normal operation of the power generation equipment. By calculating the frequency coefficient DWX, combined with the grid frequency factor, the power generation coefficient can be calculated more accurately.

[0063] The monitoring module is used to obtain the ambient temperature, wind speed, air density and grid frequency of the wind turbine in real time during the cycle. Specifically, the ambient temperature value WDi is measured by a temperature sensor, the wind speed value FSi is measured by a wind speed sensor, the air density value KSi is obtained by an air density sensor, and the grid frequency DWi is measured by a frequency sensor.

[0064] The wind power prediction model is obtained by:

[0065] The historical meteorological data of the current season and the corresponding wind power value are obtained periodically, wherein the historical meteorological data include the ambient temperature value LW, the wind speed value LF, the air density value LK and the grid frequency value LD of the wind turbine, and the external coefficient value WBX is obtained by calculation, and the obtained external coefficient value and the corresponding wind power value at the historical prediction moment are used as standard training data; the constructed artificial intelligence model is trained with the standard training data, and the trained artificial intelligence model is marked as a wind power prediction model; by inputting the external coefficient value WBX into the wind power prediction model, a wind power value can be output, and the wind power value is marked as a predicted power value, and the artificial intelligence model is a support vector machine model or a deep convolutional neural network model.

[0066] The external coefficient value WBX is obtained as follows:

[0067] Substitute the ambient temperature value LW, wind speed value LF, air density value LK and grid frequency value LD of the wind turbine generator into the formula WBX=LW×b1+LF×b2+LK×b3+LD×b4 to obtain the external coefficient value WBX, where b1, b2, b3, b4 are all preset proportional coefficients, and b1+b2+b3+b4=1. When the power generation coefficient ZFD<preset power generation coefficient YZD, it indicates that the predicted wind power will not exceed the standard, so there is no need to adjust the generator speed in time, and the correct predicted wind power value can be obtained through the wind power prediction model.

[0068] Embodiment 2

[0069] like Figure 2 As shown in Comparative Example 1, another embodiment of the present invention is:

[0070] A method for predicting wind power based on seasonal and climate change, comprising the following steps:

[0071] S1: Collect the current time, obtain the corresponding season of the current month, set the collection cycle according to the current season, and divide the collection cycle into several time nodes, and mark the time nodes as i, where i = 1, 2, 3, ..., n, where n is a positive integer;

[0072] S2: Obtain the ambient temperature, wind speed, air density and grid frequency of the wind turbine during the period, and mark them as the ambient temperature value WDi, the wind speed value FSi, the air density value KSi and the grid frequency value DWi respectively;

[0073] S3: According to the ambient temperature value WDi, the wind speed value FSi, the air density value KSi and the power grid frequency value DWi, the ambient temperature difference value WDX, the wind speed difference value FSC, the air density mean value KSJ and the frequency coefficient DWX are obtained respectively;

[0074] S4: According to the ambient temperature difference WDX, wind speed difference FSC, air density average KSJ and frequency coefficient DWX, the formula is obtained and then The power generation coefficient ZFD of the cycle is obtained, where a1, a2, a3, and a4 are all proportional coefficients, a1+a2+a3+a4=1, and a2>a4>a1>a3;

[0075] Obtain the preset power factor threshold value YZD from the cloud platform, and compare the power factor ZFD with the preset power factor threshold value YZD; when the power factor ZFD ≥ the preset power factor threshold value YZD, an overload signal is generated; when the power factor ZFD < the preset power factor YZD, an external coefficient value is obtained through the ambient temperature value WDi, the wind speed value FSi, the air density value KSi and the grid frequency value DWi, and the external coefficient value is input into the wind power prediction model to obtain the predicted power value;

[0076] S5: After receiving the overload signal, an alarm notification is issued, and the staff adjusts the generator speed through the speed governor to adjust the power generation, and marks the real-time power generation before adjustment as TG;

[0077] S6: Obtain the preset power generation YTG from the cloud platform, and calculate the power generation according to the real-time power generation TG through the formula

[0078] Get the updated preset power generation GX, through the formula Get the correction coefficient ZU, combine the power generation coefficient ZFD and the preset power generation coefficient threshold YZD, and use the formula The updated preset power generation coefficient threshold ZTZ is obtained, and the updated preset power generation power GX and the preset power generation coefficient threshold ZTZ are sent to the cloud platform.

[0079] The cloud platform storage is used to store a preset power generation coefficient threshold and a corresponding preset power generation power.

[0080] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A wind power prediction system based on seasonal and climate change, characterized by: include Time collection module: The time collection module is used to collect the current time, obtain the corresponding season of the current time month, set the collection cycle according to the current season, and divide the collection cycle into several time nodes, and mark the time nodes as i, where i=1,2,3,...,n, where n is a positive integer; Parameter monitoring module: The monitoring module is used to obtain the ambient temperature, wind speed, air density and grid frequency of the wind turbine in real time during the cycle, and mark them as ambient temperature value WDi, wind speed value FSi, air density value KSi and grid frequency value DWi respectively; Parameter processing module: The parameter processing module obtains the ambient temperature difference WDX, the wind speed difference FSC, the air density mean KSJ and the frequency coefficient DWX respectively through analysis according to the ambient temperature value WDi, the wind speed value FSi, the air density value KSi and the power grid frequency value DWi; Data analysis module: The data analysis module obtains the ambient temperature difference WDX, wind speed difference FSC, air density mean KSJ and frequency coefficient DWX through the formula The power generation coefficient ZFD of this cycle is obtained, where a1, a2, a3, and a4 are all proportional coefficients, a1+a2+a3+a4=1, and a2>a4>a1>a3; Obtain the preset power factor threshold value YZD from the cloud platform, and compare the power factor ZFD with the preset power factor threshold value YZD; when the power factor ZFD ≥ the preset power factor threshold value YZD, an overload signal is generated; when the power factor ZFD < the preset power factor YZD, an external coefficient value is obtained through the ambient temperature value WDi, the wind speed value FSi, the air density value KSi and the grid frequency value DWi, and the external coefficient value is input into the wind power prediction model to obtain the predicted power value; Adjustment module: After receiving the overload signal, the adjustment module issues an alarm notification, and the staff adjusts the generator speed through the speed governor to adjust the power generation, and marks the real-time power generation before adjustment as TG; Prediction and correction module: The prediction and correction module obtains the preset power generation YTG from the cloud platform, and calculates the power generation TG according to the formula Get the updated preset power generation GX, through the formula Get the correction coefficient ZU, combine the power generation coefficient ZFD and the preset power generation coefficient threshold YZD, and use the formula The updated preset power generation coefficient threshold ZTZ is obtained, and the updated preset power generation power GX and the preset power generation coefficient threshold ZTZ are sent to the cloud platform.

2. A wind power prediction system based on seasonal and climate change according to claim 1, characterized in that: The specific method for obtaining the ambient temperature difference WDX is as follows: The ambient temperatures WDi corresponding to all time nodes in the acquisition cycle are sorted from large to small according to the temperature values, and the maximum ambient temperature at the top is marked as WDZ and the minimum ambient temperature at the bottom is marked as WDM. Get the ambient temperature difference WDX.

3. A wind power prediction system based on seasonal and climate change according to claim 1, characterized in that: The specific method for obtaining the wind speed difference FSC is as follows: The maximum wind speed FSD and the minimum wind speed FSX corresponding to all time nodes within the collection period are obtained, and the difference between the maximum wind speed FSD and the minimum wind speed FSX is obtained to obtain the wind speed difference FSC.

4. A wind power prediction system based on seasonal and climate change according to claim 1, characterized in that: The specific method for obtaining the mean air density KSJ is: The air density value KSi corresponding to all time nodes within the acquisition period is calculated by the formula Get the average air density KSJ.

5. The wind power prediction system based on seasonal and climate change according to claim 1, characterized in that: The specific method of obtaining the frequency coefficient DWX is as follows: The grid frequency value DWi corresponding to all time nodes within the acquisition period is calculated by the formula The frequency coefficient DWX is obtained.

6. A wind power prediction system based on seasonal and climate change according to claim 1, characterized in that: The monitoring module is used to obtain the ambient temperature, wind speed, air density and grid frequency of the wind turbine in real time during the cycle.

7. The wind power prediction system based on seasonal and climate change according to claim 1, characterized in that: The wind power prediction model is obtained by: The historical meteorological data of the current season and the corresponding wind power value are obtained periodically, wherein the historical meteorological data includes the ambient temperature value LW, the wind speed value LF, the air density value LK and the grid frequency value LD of the wind turbine, and the historical external coefficient value WBX is obtained by calculation, and the obtained historical external coefficient value and the corresponding wind power value at the historical prediction time are used as standard training data; The constructed artificial intelligence model is trained with standard training data, and the trained artificial intelligence model is marked as a wind power prediction model; by inputting external coefficient values ​​into the wind power prediction model, a wind power value can be output, and the wind power value is marked as a predicted power value. The artificial intelligence model is a support vector machine model or a deep convolutional neural network model.

8. A wind power prediction system based on seasonal and climate change according to claim 7, characterized in that: The external coefficient value WBX is obtained as follows: Substitute the ambient temperature value LW, wind speed value LF, air density value LK and grid frequency value LD of the wind turbine into the formula The external coefficient value WBX is obtained, where b1, b2, b3, and b4 are all preset proportional coefficients, and b1+b2+b3+b4=1.

9. A method for predicting wind power based on seasonal and climate change, characterized in that: The specific steps include: S1: Collect the current time, obtain the corresponding season of the current month, set the collection cycle according to the current season, and divide the collection cycle into several time nodes, and mark the time nodes as i, where i=1,2,3,...,n, where n is a positive integer; S2: Obtain the ambient temperature, wind speed, air density and grid frequency of the wind turbine during the period, and mark them as the ambient temperature value WDi, the wind speed value FSi, the air density value KSi and the grid frequency value DWi respectively; S3: According to the ambient temperature value WDi, the wind speed value FSi, the air density value KSi and the power grid frequency value DWi, the ambient temperature difference value WDX, the wind speed difference value FSC, the air density mean value KSJ and the frequency coefficient DWX are obtained respectively; S4: According to the ambient temperature difference WDX, wind speed difference FSC, air density average KSJ and frequency coefficient DWX, the formula is obtained and then The power generation coefficient ZFD of this cycle is obtained, where a1, a2, a3, and a4 are all proportional coefficients, a1+a2+a3+a4=1, and a2>a4>a1>a3; Obtain the preset power factor threshold value YZD from the cloud platform, and compare the power factor ZFD with the preset power factor threshold value YZD; when the power factor ZFD ≥ the preset power factor threshold value YZD, an overload signal is generated; when the power factor ZFD < the preset power factor YZD, an external coefficient value is obtained through the ambient temperature value WDi, the wind speed value FSi, the air density value KSi and the grid frequency value DWi, and the external coefficient value is input into the wind power prediction model to obtain the predicted power value; S5: After receiving the overload signal, an alarm notification is issued, and the staff adjusts the generator speed through the speed governor to adjust the power generation, and the real-time power generation before adjustment is marked as TG; S6: Obtain the preset power generation YTG from the cloud platform, and calculate the power generation according to the real-time power generation TG through the formula Get the updated preset power generation GX, through the formula Get the correction coefficient ZU, combine the power generation coefficient ZFD and the preset power generation coefficient threshold YZD, and use the formula The updated preset power generation coefficient threshold ZTZ is obtained, and the updated preset power generation power GX and the preset power generation coefficient threshold ZTZ are sent to the cloud platform.

10. A method for predicting wind power based on seasonal and climate changes according to claim 9, characterized in that: The cloud platform storage is used to store a preset power generation coefficient threshold and a corresponding preset power generation power.

Images