Complementary collection station power distribution system of new energy distributed power station and control method of complementary collection station power distribution system

Abstract

The invention discloses a complementary collection station power distribution system of a new energy distributed power station and a control method of the complementary collection station power distribution system, and relates to the technical field of complementary collection station power distribution, the complementary collection station power distribution system comprises a monitoring data acquisition module, a new energy power generation monitoring module, a monitoring optimization module, a power distribution monitoring module and a power distribution control module, the data acquisition module is used for acquiring solar power generation monitoring data, wind power generation monitoring data and real-time power generation amount of the new energy distributed power station and preprocessing the acquired data, and the model fusion module is used for acquiring a new energy power generation monitoring model according to output results of the solar power generation monitoring model and the wind power generation monitoring model. According to the method, a multiple linear regression algorithm modeling technology, a genetic algorithm optimization model technology and a neural network algorithm model fusion technology are closely combined with a modern information technology, so that real-time and comprehensive monitoring on a dual-energy power generation process is achieved; and the intelligent degree in the power distribution process of the complementary collection station of the new energy distributed power station is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of complementary power collection and distribution, and in particular to a complementary power collection and distribution system for a new energy dispersed power station and a control method thereof. Background Art

[0002] With the global attention to environmental protection and sustainable development, the development and utilization of new energy has become the key, and new energy decentralized power stations have emerged. However, there are many disadvantages in single new energy power generation. Solar energy is affected by day and night and weather, and cannot generate electricity at night and on rainy days; wind energy is intermittent and unstable, and unstable wind speed leads to large fluctuations in power generation; hydropower is significantly affected by seasons and precipitation, and power generation drops sharply during the dry season. In order to solve these problems, the complementary station distribution system of new energy decentralized power stations and its control method have become a research hotspot. The complementary station distribution system organically integrates a variety of new energy power generation equipment and utilizes the complementary characteristics of different energy sources. Its control method uses advanced monitoring and communication technology to collect the operating data and environmental information of each power generation equipment in real time. According to the principles of power balance and voltage and frequency stability, the power generation equipment and energy storage system are intelligently controlled to achieve optimal configuration and efficient utilization of energy. A complementary station distribution system of new energy decentralized power stations and its control method have emerged, which has promoted the transformation of energy structure to clean and low-carbon, and has broad application prospects and important practical significance. Although the existing technology has made great progress in the direction of complementary collection and distribution of new energy decentralized power stations, there are still some problems that need to be optimized. The existing new energy power generation process is greatly affected by the climate, and the existing technology is difficult to monitor climate change in real time, resulting in the complementary collection and distribution system being unable to coordinate the power generation process of decentralized solar power stations and decentralized wind power stations, affecting the power generation efficiency, and further resulting in poor control of the complementary collection station over the distribution. Summary of the invention

[0003] The object of the present invention is to provide a complementary station-collection and distribution system for a new energy decentralized power station and a control method thereof, so as to solve the problems raised in the above-mentioned background technology.

[0004] In order to solve the above technical problems, the technical solution adopted by the present invention is: in the first aspect, a complementary station-collecting power distribution system of a new energy decentralized power station comprises a monitoring data acquisition module, a new energy power generation monitoring module, a monitoring optimization module, a power distribution monitoring module and a power distribution control module, wherein each module is communicatively connected; The monitoring data acquisition module is used to collect solar power generation monitoring data, wind power generation monitoring data and real-time power generation of new energy decentralized power stations, and pre-process the collected data to provide data support for the operation of subsequent modules; The new energy power generation monitoring module, based on the preprocessed data and combined with the multivariate linear regression algorithm, constructs a solar power generation monitoring model and a wind power generation monitoring model respectively, so as to monitor the power generation conditions of the solar power generation decentralized power station and the wind power generation decentralized power station; The monitoring optimization module optimizes the solar power generation monitoring model and the wind power generation monitoring model; The power distribution monitoring module constructs a new energy power generation monitoring model through the output results of the solar power generation monitoring model and the wind power generation monitoring model; The power distribution control module, combined with the output results of the new energy power generation monitoring model, takes corresponding measures to deploy decentralized solar power stations and decentralized wind power stations, and performs power distribution control according to load demand, which solves the problem that the existing technology is difficult to monitor climate change in real time and coordinate the power generation process of decentralized solar power stations and decentralized wind power stations, which in turn leads to poor control effect of complementary collection stations on power distribution.

[0005] A further improvement of the technical solution of the present invention is that the process of the monitoring data acquisition module collecting solar power generation monitoring data, wind power generation monitoring data and real-time power generation of new energy decentralized power stations includes: Deploy collection equipment to collect solar power generation monitoring data, wind power generation monitoring data and real-time power generation of new energy decentralized power stations, wherein the collection equipment includes light intensity sensors, sunshine meters, ultrasonic anemometers, ultrasonic wind direction sensors and smart meters, etc.; The solar power generation monitoring data is the light intensity and light time of the decentralized solar power station environment; the wind power generation monitoring data is the wind speed and wind direction angle of the decentralized wind power station environment; the real-time power generation of the new energy decentralized power station includes the real-time power generation of the decentralized solar power station and the real-time power generation of the decentralized wind power station; Light intensity sensors and sunshine meters are used to collect light intensity and sunshine time in the environment of distributed solar power stations. Ultrasonic wind speed and ultrasonic wind direction sensors are used to collect wind speed and wind direction angle in the environment of distributed wind power stations. Smart meters are used to collect real-time power generation of distributed solar power stations and distributed wind power stations.

[0006] A further improvement of the technical solution of the present invention is that: the monitoring data acquisition module, the preprocessing process includes: Perform data cleaning and data normalization on the collected solar power generation monitoring data, wind power generation monitoring data, real-time power generation of distributed solar power stations, and real-time power generation of distributed wind power stations; The collected solar power generation monitoring data, wind power generation monitoring data, real-time power generation of distributed solar power stations and real-time power generation of distributed wind power stations are uploaded to the cloud. The cloud calibrates the time of the corresponding collection equipment according to the clock source, and transmits the calibrated time information to the corresponding collection equipment, so as to synchronize the collection time of the solar power generation monitoring data, wind power generation monitoring data, real-time power generation of distributed solar power stations and real-time power generation of distributed wind power stations.

[0007] A further improvement of the technical solution of the present invention is that the construction process of the new energy power generation monitoring module and the solar power generation monitoring model includes: The light intensity of the distributed solar power station environment, the light time of the distributed solar power station environment and the real-time power generation of the distributed solar power station are used as the first data set, which are divided into a training set and a test set in a ratio of 8:2, and the data in the first data set are used as input, and the solar power generation impact index is used as output; Through the training set data and the multivariate linear regression algorithm, the intercept term, the regression coefficient of the ambient light intensity of the distributed solar power station, the regression coefficient of the ambient light time of the distributed solar power station, and the regression coefficient of the real-time power generation of the distributed solar power station are adjusted respectively, and the linear relationship between the ambient light intensity of the distributed solar power station, the ambient light time of the distributed solar power station, the real-time power generation of the distributed solar power station and the solar power generation impact index is learned to train the solar power generation monitoring model; Input the test set data into the solar power generation monitoring model, evaluate the performance of the solar power generation monitoring model, adjust the parameters of the solar power generation monitoring model, optimize the solar power generation monitoring model, and deploy the optimized solar power generation monitoring model to the system.

[0008] A further improvement of the technical solution of the present invention is that the construction process of the wind power generation monitoring model of the new energy power generation monitoring module includes: The wind speed and wind direction of the distributed wind power station environment and the real-time power generation of the distributed wind power station are used as the second data set, which is divided into a training set and a test set in a ratio of 8:2, and the data in the second data set is used as input, and the wind power generation impact index is used as output; Through the training set data and the multivariate linear regression algorithm, the intercept term, the regression coefficient of the wind speed of the distributed wind power station environment, the regression coefficient of the wind direction angle of the distributed wind power station environment, and the regression coefficient of the real-time power generation of the distributed wind power station are adjusted respectively, and the linear relationship between the wind speed of the distributed wind power station environment, the wind direction angle of the distributed wind power station environment, the real-time power generation of the distributed wind power station and the wind power generation impact index is learned to train the wind power generation monitoring model; The test set data is input into the wind power generation monitoring model, the performance of the wind power generation monitoring model is evaluated, the parameters of the wind power generation monitoring model are adjusted, the wind power generation monitoring model is optimized, and the optimized wind power generation monitoring model is deployed into the system.

[0009] A further improvement of the technical solution of the present invention is that: the process of optimizing the solar power generation monitoring model in the monitoring optimization module includes: Using genetic algorithm, the regression coefficient of solar power generation monitoring model is encoded by real numbers to obtain a real vector of length 4, which is expressed as [ρ0, ρ1, ρ2, ρ3]. The individuals are represented by real vectors, N individuals are randomly generated, and the N individuals are combined to obtain the initial population. The inverse of the mean square error is used as the fitness function, and the fitness threshold and the number of iterations are set. According to the fitness value of each individual in the population, the roulette wheel selection method is adopted. The larger the fitness value of each individual, the greater the probability of the individual being selected, and the individuals in the population are selected to participate in the reproduction of the next generation; the arithmetic crossover method is adopted to exchange genes for the selected individuals; the regression coefficient of the solar power generation monitoring model in the individuals after gene exchange is randomly selected, and the regression coefficient of the solar power generation monitoring model is randomly changed, and the iterative operations of selection, crossover and mutation are repeated until the fitness value of the individual in the population reaches the fitness threshold and the number of repeated iterative operations reaches the set number of iterations, the repeated iterative operation is terminated, and after the repeated iterative operation is terminated, the regression coefficient corresponding to the individual with the highest fitness value in the population is used to update the regression coefficient in the solar power generation monitoring model to optimize the solar power generation monitoring model.

[0010] A further improvement of the technical solution of the present invention is that: the process of optimizing the wind power generation monitoring model in the monitoring optimization module includes: Using genetic algorithm, the regression coefficient of wind power generation monitoring model is encoded by real numbers to obtain a real vector of length 4, which is expressed as [ρ0, ρ1, ρ2, ρ3]. The real vector represents the individual, M individuals are randomly generated, and the M individuals are combined to obtain the initial population. The inverse of the mean square error is used as the fitness function, and the fitness threshold and the number of iterations are set. According to the fitness value of each individual in the population, the roulette wheel selection method is adopted. The larger the fitness value of each individual, the greater the probability of the individual being selected, and the individuals in the population are selected to participate in the reproduction of the next generation; the arithmetic crossover method is adopted to perform gene exchange on the selected individuals; the regression coefficient of the wind power generation monitoring model in the individuals after gene exchange is randomly selected, and the regression coefficient of the wind power generation monitoring model is randomly changed, and the iterative operations of selection, crossover and mutation are repeated until the fitness value of the individual in the population reaches the fitness threshold and the number of repeated iteration operations reaches the set number of iterations, the repeated iteration operation is terminated, and after the repeated iteration operation is terminated, the regression coefficient corresponding to the individual with the highest fitness value in the population is used to update the regression coefficient in the wind power generation monitoring model to optimize the wind power generation monitoring model.

[0011] A further improvement of the technical solution of the present invention is that: the construction process of the power distribution monitoring module and the new energy power generation monitoring model includes: Input the solar power generation monitoring data and the real-time power generation of the distributed solar power station into the solar power generation monitoring model to obtain the solar power generation impact index; input the wind power generation monitoring data and the real-time power generation of the distributed wind power station into the wind power generation monitoring model to obtain the wind power generation impact index; The solar power generation monitoring data, the wind power generation monitoring data, the solar power generation impact index and the wind power generation impact index are used as the third data set, which are divided into a training set and a test set in a ratio of 7:3, the solar power generation monitoring data and the wind power generation monitoring data are used as input, and the solar power generation impact index and the wind power generation impact index are used as output; Using the training set data and the neural network algorithm, the nonlinear relationship between the solar power generation monitoring data and the solar power generation impact index and the nonlinear relationship between the wind power generation monitoring data and the wind power generation impact index are learned through repeated iterative training until the set number of iterative training times is reached, and the new energy power generation monitoring model is trained; The test set data is input into the renewable energy power generation monitoring model, and the MSE function is used to evaluate the errors between the solar power generation impact index and wind power generation impact index output by the renewable energy power generation monitoring model and the actual solar power generation impact index and wind power generation impact index. According to the evaluation results, the parameters of the renewable energy power generation monitoring model are adjusted to optimize the performance of the renewable energy power generation monitoring model, and the optimized renewable energy power generation monitoring model is deployed into the system.

[0012] A further improvement of the technical solution of the present invention is that the power distribution control module, in combination with the output result of the new energy power generation monitoring model, takes corresponding measures to allocate the distributed solar power station and the distributed wind power station, and performs power distribution control according to the load demand, including: The solar power generation monitoring data and the wind power generation monitoring data are respectively input into the new energy power generation monitoring model, and the new energy power generation monitoring model outputs the solar power generation impact index and the wind power generation impact index respectively; When the solar power generation impact index is between 0 and 0.3, the impact of solar power generation monitoring data on the real-time power generation of distributed solar power stations is low; when the solar power generation impact index is between 0.3 and 0.7, the impact of solar power generation monitoring data on the real-time power generation of distributed solar power stations is medium; when the solar power generation impact index is between 0.7 and 1, the impact of solar power generation monitoring data on the real-time power generation of distributed solar power stations is high; When the wind power generation impact index is between 0 and 0.3, the impact of wind power generation monitoring data on the real-time power generation of distributed wind power stations is low; when the wind power generation impact index is between 0.3 and 0.7, the impact of wind power generation monitoring data on the real-time power generation of distributed wind power stations is medium; when the wind power generation impact index is between 0.7 and 1, the impact of wind power generation monitoring data on the real-time power generation of distributed wind power stations is high; When the impact of solar power generation monitoring data on the real-time power generation of decentralized solar power stations is greater than the impact of wind power generation monitoring data on the real-time power generation of decentralized wind power stations, the decentralized solar power stations are shut down for power generation, and the decentralized wind power stations are deployed to operate at full load; When the impact of solar power generation monitoring data on the real-time power generation of decentralized solar power stations is lower than the impact of wind power generation monitoring data on the real-time power generation of decentralized wind power stations, the decentralized wind power stations are shut down for power generation, and the decentralized solar power stations are deployed to operate at full load; When the impact of solar power generation monitoring data on the real-time power generation of distributed solar power stations is the same as the impact of wind power generation monitoring data on the real-time power generation of distributed wind power stations and is low, the distributed solar power stations and distributed wind power stations are deployed to operate at full load to maximize power generation; When the impact of solar power generation monitoring data on the real-time power generation of decentralized solar power stations is the same as the impact of wind power generation monitoring data on the real-time power generation of decentralized wind power stations and is medium, load operation is allocated to decentralized solar power stations and decentralized wind power stations in a 1:1 ratio to dynamically balance the supply and demand of power distribution; When the impact of the solar power generation monitoring data on the real-time power generation of the distributed solar power station is the same as the impact of the wind power generation monitoring data on the real-time power generation of the distributed wind power station and is high, the power generation process of the distributed solar power station and the distributed wind power station is shut down and the backup power supply is started; The power distribution plan is dynamically adjusted according to the total power generation of the decentralized solar power stations and decentralized wind power stations and the load demand. When the total power generation of the decentralized solar power stations and decentralized wind power stations is greater than the load demand, the total power generation of the decentralized solar power stations and decentralized wind power stations is allocated according to the load demand, and the excess power generation is stored; when the total power generation of the decentralized solar power stations and decentralized wind power stations is less than the load demand, the load demand is reduced according to the load priority, and the stored power generation is allocated.

[0013] In a second aspect, a complementary station distribution control method for a new energy distributed power station is provided, which is used to implement the complementary station distribution system for a new energy distributed power station, and is composed of the following steps: Step 1: Use a light intensity sensor, a sunshine meter, an ultrasonic anemometer, an ultrasonic wind direction sensor and a smart meter to collect solar power generation monitoring data, wind power generation monitoring data and real-time power generation of new energy decentralized power stations, and pre-process the collected data; Step 2: Based on the preprocessed data, combined with the multivariate linear regression algorithm, a solar power generation monitoring model and a wind power generation monitoring model are constructed respectively; Step 3: Optimize the solar power generation monitoring model and the wind power generation monitoring model; Step 4: Obtain a new energy power generation monitoring model through the output results of the solar power generation monitoring model and the wind power generation monitoring model; Step 5: According to the output results of the new energy power generation monitoring model, corresponding measures are taken to deploy decentralized solar power stations and decentralized wind power stations, which solves the problem that the existing technology is difficult to monitor climate change in real time and coordinate the power generation process of decentralized solar power stations and decentralized wind power stations, which in turn leads to poor control effect of complementary collection stations on power distribution.

[0014] The beneficial effects of the present invention are: a complementary collection and distribution system for a new energy decentralized power station and a control method thereof. Compared with the complementary collection and distribution system for a traditional new energy decentralized power station and a control method thereof, the multivariate linear regression algorithm modeling technology, the genetic algorithm optimization model technology, and the neural network algorithm model fusion technology in the method of the present invention are closely combined with modern information technology. Through light intensity sensors, sunshine meters, ultrasonic anemometers, ultrasonic wind direction sensors and smart meters, solar power generation monitoring data, wind power generation monitoring data and real-time power generation of decentralized new energy power stations are accurately captured. Through the multivariate linear regression algorithm, a solar power generation monitoring model and a wind power generation monitoring model are respectively constructed. Combined with the genetic algorithm, the construction of the solar power generation monitoring model and the wind power generation monitoring model are optimized. The power generation monitoring model obtains quantitative evaluation indicators of the benefits of solar and wind power generation. Through the neural network algorithm, the solar power generation monitoring model and the wind power generation monitoring model are integrated to achieve real-time and comprehensive monitoring of the dual-energy power generation process. This solves the problem that the existing new energy power generation process is greatly affected by the climate, and the existing technology is difficult to monitor climate change in real time, resulting in the complementary collection station distribution system being unable to coordinate the power generation process of decentralized solar power stations and decentralized wind power stations, affecting the power generation efficiency, and then resulting in the complementary collection station The poor control effect of power distribution, ensures that the method in the present invention can be used for a new energy decentralized power station. The dynamic monitoring standard of the complementary collection station distribution system and its control method can be refined within a more precise range, so that the monitored data becomes a more accurate indicator under the same conditions. The development and application of this method significantly enhances the intelligence level of the complementary collection station distribution process of decentralized new energy power stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0016] Figure 1 It is a block diagram of a complementary station-collection and distribution system of a new energy decentralized power station of the present invention; Figure 2 The present invention is a flow chart of a complementary power distribution control method for a new energy decentralized power station. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] Embodiment 1, as Figure 1 As shown, the present invention provides a complementary station-distribution system for a new energy dispersed power station, including a monitoring data acquisition module, a new energy power generation monitoring module, a monitoring optimization module, a distribution monitoring module and a distribution control module, wherein each module is communicatively connected; The monitoring data acquisition module is used to collect solar power generation monitoring data, wind power generation monitoring data and real-time power generation of new energy decentralized power stations, and pre-process the collected data to provide data support for the operation of subsequent modules; The new energy power generation monitoring module, based on the pre-processed data and combined with the multivariate linear regression algorithm, constructs the solar power generation monitoring model and the wind power generation monitoring model respectively, so as to monitor the power generation of the solar power generation decentralized power station and the wind power generation decentralized power station; Monitoring optimization module, which optimizes the solar power generation monitoring model and the wind power generation monitoring model; The power distribution monitoring module obtains the new energy power generation monitoring model through the output results of the solar power generation monitoring model and the wind power generation monitoring model; The power distribution control module, combined with the output results of the new energy power generation monitoring model, takes corresponding measures to deploy decentralized solar power stations and decentralized wind power stations, and performs power distribution control according to load demand, solving the problem that existing technologies are difficult to monitor climate change in real time and coordinate the power generation process of decentralized solar power stations and decentralized wind power stations, which in turn leads to poor control effect of complementary collection stations on power distribution.

[0019] Preferably, the monitoring data collection module collects the solar power generation monitoring data, the wind power generation monitoring data and the real-time power generation of the new energy decentralized power station, including: Deploy collection equipment to collect solar power generation monitoring data, wind power generation monitoring data and real-time power generation of new energy decentralized power stations, wherein the collection equipment includes light intensity sensors, sunshine meters, ultrasonic anemometers, ultrasonic wind direction sensors and smart meters, etc.; The solar power generation monitoring data is the light intensity and light time of the decentralized solar power station environment; the wind power generation monitoring data is the wind speed and wind direction angle of the decentralized wind power station environment; the real-time power generation of the new energy decentralized power station includes the real-time power generation of the decentralized solar power station and the real-time power generation of the decentralized wind power station; Light intensity sensors and sunshine meters are used to collect light intensity and sunshine time in the environment of distributed solar power stations. Ultrasonic wind speed and ultrasonic wind direction sensors are used to collect wind speed and wind direction angle in the environment of distributed wind power stations. Smart meters are used to collect real-time power generation of distributed solar power stations and distributed wind power stations.

[0020] Preferably, the monitoring data acquisition module, the preprocessing process includes: Perform data cleaning and data normalization on the collected solar power generation monitoring data, wind power generation monitoring data, real-time power generation of distributed solar power stations, and real-time power generation of distributed wind power stations; The collected solar power generation monitoring data, wind power generation monitoring data, real-time power generation of distributed solar power stations and real-time power generation of distributed wind power stations are uploaded to the cloud. The cloud calibrates the time of the corresponding collection equipment according to the clock source, and transmits the calibrated time information to the corresponding collection equipment, so as to synchronize the collection time of the solar power generation monitoring data, wind power generation monitoring data, real-time power generation of distributed solar power stations and real-time power generation of distributed wind power stations.

[0021] Preferably, the construction process of the new energy power generation monitoring module and the solar power generation monitoring model includes: The light intensity of the distributed solar power station environment, the light time of the distributed solar power station environment and the real-time power generation of the distributed solar power station are used as the first data set, which are divided into a training set and a test set in a ratio of 8:2, and the data in the first data set are used as input, and the solar power generation impact index is used as output; Through the training set data and the multivariate linear regression algorithm, the intercept term, the regression coefficient of the ambient light intensity of the distributed solar power station, the regression coefficient of the ambient light time of the distributed solar power station, and the regression coefficient of the real-time power generation of the distributed solar power station are adjusted respectively, and the linear relationship between the ambient light intensity of the distributed solar power station, the ambient light time of the distributed solar power station, the real-time power generation of the distributed solar power station and the solar power generation impact index is learned to train the solar power generation monitoring model; Input the test set data into the solar power generation monitoring model, evaluate the performance of the solar power generation monitoring model, adjust the parameters of the solar power generation monitoring model, optimize the solar power generation monitoring model, deploy the optimized solar power generation monitoring model into the system, and obtain the final solar power generation monitoring model.

[0022] Preferably, the construction process of the new energy power generation monitoring module and the wind power generation monitoring model includes: The wind speed and wind direction of the distributed wind power station environment and the real-time power generation of the distributed wind power station are used as the second data set, which is divided into a training set and a test set in a ratio of 8:2, and the data in the second data set is used as input, and the wind power generation impact index is used as output; Through the training set data and the multivariate linear regression algorithm, the intercept term, the regression coefficient of the wind speed of the distributed wind power station environment, the regression coefficient of the wind direction angle of the distributed wind power station environment, and the regression coefficient of the real-time power generation of the distributed wind power station are adjusted respectively, and the linear relationship between the wind speed of the distributed wind power station environment, the wind direction angle of the distributed wind power station environment, the real-time power generation of the distributed wind power station and the wind power generation impact index is learned to train the wind power generation monitoring model; The test set data is input into the wind power generation monitoring model, the performance of the wind power generation monitoring model is evaluated, the parameters of the wind power generation monitoring model are adjusted, the wind power generation monitoring model is optimized, and the optimized wind power generation monitoring model is deployed into the system to obtain the final wind power generation monitoring model.

[0023] Preferably, the monitoring optimization module optimizes the process of the solar power generation monitoring model including: Using genetic algorithm, the regression coefficient of solar power generation monitoring model is encoded by real numbers to obtain a real vector of length 4, which is expressed as [ρ0, ρ1, ρ2, ρ3]. The individuals are represented by real vectors, N individuals are randomly generated, and the N individuals are combined to obtain the initial population. The inverse of the mean square error is used as the fitness function, and the fitness threshold and the number of iterations are set. According to the fitness value of each individual in the population, the roulette wheel selection method is adopted. The larger the fitness value of each individual, the greater the probability of the individual being selected, and the individuals in the population are selected to participate in the reproduction of the next generation; the arithmetic crossover method is adopted to exchange genes for the selected individuals; the regression coefficient of the solar power generation monitoring model in the individuals after gene exchange is randomly selected, and the regression coefficient of the solar power generation monitoring model is randomly changed, and the iterative operations of selection, crossover and mutation are repeated until the fitness value of the individual in the population reaches the fitness threshold and the number of repeated iterative operations reaches the set number of iterations, the repeated iterative operation is terminated, and after the repeated iterative operation is terminated, the regression coefficient corresponding to the individual with the highest fitness value in the population is used to update the regression coefficient in the solar power generation monitoring model to optimize the solar power generation monitoring model.

[0024] Preferably, the monitoring optimization module optimizes the process of the wind power generation monitoring model including: Using genetic algorithm, the regression coefficient of wind power generation monitoring model is encoded by real numbers to obtain a real vector of length 4, which is expressed as [ρ0, ρ1, ρ2, ρ3]. The real vector represents the individual, M individuals are randomly generated, and the M individuals are combined to obtain the initial population. The inverse of the mean square error is used as the fitness function, and the fitness threshold and the number of iterations are set. According to the fitness value of each individual in the population, the roulette wheel selection method is adopted. The larger the fitness value of each individual, the greater the probability of the individual being selected, and the individuals in the population are selected to participate in the reproduction of the next generation; the arithmetic crossover method is adopted to perform gene exchange on the selected individuals; the regression coefficient of the wind power generation monitoring model in the individuals after gene exchange is randomly selected, and the regression coefficient of the wind power generation monitoring model is randomly changed, and the iterative operations of selection, crossover and mutation are repeated until the fitness value of the individual in the population reaches the fitness threshold and the number of repeated iteration operations reaches the set number of iterations, the repeated iteration operation is terminated, and after the repeated iteration operation is terminated, the regression coefficient corresponding to the individual with the highest fitness value in the population is used to update the regression coefficient in the wind power generation monitoring model to optimize the wind power generation monitoring model.

[0025] Preferably, the construction process of the power distribution monitoring module and the new energy power generation monitoring model includes: Input the solar power generation monitoring data and the real-time power generation of the distributed solar power station into the solar power generation monitoring model to obtain the solar power generation impact index; input the wind power generation monitoring data and the real-time power generation of the distributed wind power station into the wind power generation monitoring model to obtain the wind power generation impact index; The solar power generation monitoring data, the wind power generation monitoring data, the solar power generation impact index and the wind power generation impact index are used as the third data set, which are divided into a training set and a test set in a ratio of 7:3, the solar power generation monitoring data and the wind power generation monitoring data are used as input, and the solar power generation impact index and the wind power generation impact index are used as output; Using the training set data and the neural network algorithm, the nonlinear relationship between the solar power generation monitoring data and the solar power generation impact index and the nonlinear relationship between the wind power generation monitoring data and the wind power generation impact index are learned through repeated iterative training until the set number of iterative training times is reached, and the new energy power generation monitoring model is trained; The test set data is input into the renewable energy power generation monitoring model, and the MSE function is used to evaluate the errors between the solar power generation impact index and wind power generation impact index output by the renewable energy power generation monitoring model and the actual solar power generation impact index and wind power generation impact index. According to the evaluation results, the parameters of the renewable energy power generation monitoring model are adjusted to optimize the performance of the renewable energy power generation monitoring model. The optimized renewable energy power generation monitoring model is deployed into the system to obtain the final renewable energy power generation monitoring model.

[0026] Preferably, the power distribution control module, in combination with the output result of the new energy power generation monitoring model, takes corresponding measures to allocate the distributed solar power station and the distributed wind power station, and performs power distribution control according to the load demand, including: The solar power generation monitoring data and the wind power generation monitoring data are respectively input into the new energy power generation monitoring model, and the new energy power generation monitoring model outputs the solar power generation impact index and the wind power generation impact index respectively; When the solar power generation impact index is between 0 and 0.3, the impact of solar power generation monitoring data on the real-time power generation of distributed solar power stations is low; when the solar power generation impact index is between 0.3 and 0.7, the impact of solar power generation monitoring data on the real-time power generation of distributed solar power stations is medium; when the solar power generation impact index is between 0.7 and 1, the impact of solar power generation monitoring data on the real-time power generation of distributed solar power stations is high; When the wind power generation impact index is between 0 and 0.3, the impact of wind power generation monitoring data on the real-time power generation of distributed wind power stations is low; when the wind power generation impact index is between 0.3 and 0.7, the impact of wind power generation monitoring data on the real-time power generation of distributed wind power stations is medium; when the wind power generation impact index is between 0.7 and 1, the impact of wind power generation monitoring data on the real-time power generation of distributed wind power stations is high; When the impact of solar power generation monitoring data on the real-time power generation of decentralized solar power stations is greater than the impact of wind power generation monitoring data on the real-time power generation of decentralized wind power stations, the decentralized solar power stations are shut down for power generation, and the decentralized wind power stations are deployed to operate at full load; When the impact of solar power generation monitoring data on the real-time power generation of decentralized solar power stations is lower than the impact of wind power generation monitoring data on the real-time power generation of decentralized wind power stations, the decentralized wind power stations are shut down for power generation, and the decentralized solar power stations are deployed to operate at full load; When the impact of solar power generation monitoring data on the real-time power generation of distributed solar power stations is the same as the impact of wind power generation monitoring data on the real-time power generation of distributed wind power stations and is low, the distributed solar power stations and distributed wind power stations are deployed to operate at full load to maximize power generation; When the impact of solar power generation monitoring data on the real-time power generation of decentralized solar power stations is the same as the impact of wind power generation monitoring data on the real-time power generation of decentralized wind power stations and is medium, load operation is allocated to decentralized solar power stations and decentralized wind power stations in a 1:1 ratio to dynamically balance the supply and demand of power distribution; When the impact of the solar power generation monitoring data on the real-time power generation of the distributed solar power station is the same as the impact of the wind power generation monitoring data on the real-time power generation of the distributed wind power station and is high, the power generation process of the distributed solar power station and the distributed wind power station is shut down and the backup power supply is started; The power distribution plan is dynamically adjusted according to the total power generation of the decentralized solar power stations and decentralized wind power stations and the load demand. When the total power generation of the decentralized solar power stations and decentralized wind power stations is greater than the load demand, the total power generation of the decentralized solar power stations and decentralized wind power stations is allocated according to the load demand, and the excess power generation is stored; when the total power generation of the decentralized solar power stations and decentralized wind power stations is less than the load demand, the load demand is reduced according to the load priority, and the stored power generation is allocated.

[0027] Embodiment 2, as Figure 2 As shown, based on Example 1, the present invention provides a technical solution: a complementary station distribution control method for a new energy distributed power station, which is used to implement the complementary station distribution system for the above-mentioned new energy distributed power station, and is composed of the following steps: Step 1: Use a light intensity sensor, a sunshine meter, an ultrasonic anemometer, an ultrasonic wind direction sensor and a smart meter to collect solar power generation monitoring data, wind power generation monitoring data and real-time power generation of new energy decentralized power stations, and pre-process the collected data; Step 2: Based on the preprocessed data, combined with the multivariate linear regression algorithm, a solar power generation monitoring model and a wind power generation monitoring model are constructed respectively; Step 3: Optimize the solar power generation monitoring model and the wind power generation monitoring model; Step 4: Obtain a new energy power generation monitoring model through the output results of the solar power generation monitoring model and the wind power generation monitoring model; Step 5: According to the output results of the new energy power generation monitoring model, corresponding measures are taken to deploy decentralized solar power stations and decentralized wind power stations, which solves the problem that the existing technology is difficult to monitor climate change in real time and coordinate the power generation process of decentralized solar power stations and decentralized wind power stations, which in turn leads to poor control effect of complementary collection stations on power distribution.

[0028] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A complementary station distribution system for a new energy decentralized power station, comprising a monitoring data acquisition module, a new energy power generation monitoring module, a monitoring optimization module, a power distribution monitoring module and a power distribution control module, wherein: Each module is connected in communication, characterized by: The monitoring data collection module is used to collect solar power generation monitoring data, wind power generation monitoring data and real-time power generation of new energy decentralized power stations, and pre-process the collected data; The new energy power generation monitoring module constructs a solar power generation monitoring model and a wind power generation monitoring model based on the preprocessed data and combined with a multivariate linear regression algorithm; The monitoring optimization module optimizes the solar power generation monitoring model and the wind power generation monitoring model; The model fusion module constructs a new energy power generation monitoring model based on the output results of the solar power generation monitoring model and the wind power generation monitoring model; The power distribution monitoring module, in combination with the output result of the new energy power generation monitoring model, takes corresponding measures to allocate the distributed solar power stations and the distributed wind power stations, and performs power distribution control according to the load demand.

2. According to claim 1, a complementary station distribution system for a new energy decentralized power station is characterized in that: The monitoring data collection module collects the solar power generation monitoring data, the wind power generation monitoring data and the real-time power generation of the new energy decentralized power station, including: Deploy collection equipment to collect solar power generation monitoring data, wind power generation monitoring data and real-time power generation of new energy decentralized power stations, wherein the collection equipment includes light intensity sensors, sunshine meters, ultrasonic anemometers, ultrasonic wind direction sensors and smart meters; The solar power generation monitoring data includes the light intensity and illumination time of the decentralized solar power station environment; the wind power generation monitoring data includes the wind speed and wind direction angle of the decentralized wind power station environment; the real-time power generation of the decentralized new energy power station includes the real-time power generation of the decentralized solar power station and the real-time power generation of the decentralized wind power station.

3. According to claim 2, a complementary station distribution system for a new energy decentralized power station is characterized in that: The monitoring data acquisition module, the pre-processing process includes: Perform data cleaning and data normalization on the collected solar power generation monitoring data, wind power generation monitoring data, real-time power generation of distributed solar power stations, and real-time power generation of distributed wind power stations; The collected solar power generation monitoring data, wind power generation monitoring data, real-time power generation of distributed solar power stations and real-time power generation of distributed wind power stations are uploaded to the cloud. The cloud calibrates the time of the corresponding collection equipment according to the clock source, and transmits the calibrated time information to the corresponding collection equipment, so as to synchronize the collection time of the solar power generation monitoring data, wind power generation monitoring data, real-time power generation of distributed solar power stations and real-time power generation of distributed wind power stations.

4. According to claim 3, a complementary station distribution system for a new energy decentralized power station is characterized in that: The construction process of the new energy power generation monitoring module and the solar power generation monitoring model includes: The light intensity of the distributed solar power station environment, the light time of the distributed solar power station environment and the real-time power generation of the distributed solar power station are used as the first data set, which are divided into a training set and a test set in a ratio of 8:2, and the data in the first data set are used as input, and the solar power generation impact index is used as output; Through the training set data and the multivariate linear regression algorithm, the intercept term, the regression coefficient of the ambient light intensity of the distributed solar power station, the regression coefficient of the ambient light time of the distributed solar power station, and the regression coefficient of the real-time power generation of the distributed solar power station are adjusted respectively, and the linear relationship between the ambient light intensity of the distributed solar power station, the ambient light time of the distributed solar power station, the real-time power generation of the distributed solar power station and the solar power generation impact index is learned to train the solar power generation monitoring model; Input the test set data into the solar power generation monitoring model, evaluate the performance of the solar power generation monitoring model, adjust the parameters of the solar power generation monitoring model, optimize the solar power generation monitoring model, and deploy the optimized solar power generation monitoring model to the system.

5. According to claim 4, a complementary station distribution system for a new energy decentralized power station is characterized in that: The construction process of the new energy power generation monitoring module and the wind power generation monitoring model includes: The wind speed and wind direction of the distributed wind power station environment and the real-time power generation of the distributed wind power station are used as the second data set, which is divided into a training set and a test set in a ratio of 8:2, and the data in the second data set is used as input, and the wind power generation impact index is used as output; Through the training set data and the multivariate linear regression algorithm, the intercept term, the regression coefficient of the wind speed of the distributed wind power station environment, the regression coefficient of the wind direction angle of the distributed wind power station environment, and the regression coefficient of the real-time power generation of the distributed wind power station are adjusted respectively, and the linear relationship between the wind speed of the distributed wind power station environment, the wind direction angle of the distributed wind power station environment, the real-time power generation of the distributed wind power station and the wind power generation impact index is learned to train the wind power generation monitoring model; The test set data is input into the wind power generation monitoring model, the performance of the wind power generation monitoring model is evaluated, the parameters of the wind power generation monitoring model are adjusted, the wind power generation monitoring model is optimized, and the optimized wind power generation monitoring model is deployed into the system.

6. The complementary station distribution system of a new energy decentralized power station according to claim 5 is characterized by: The monitoring optimization module optimizes the solar power generation monitoring model in the following process: Using genetic algorithm, the regression coefficient of solar power generation monitoring model is encoded by real numbers to obtain a real vector of length 4, which is expressed as [ρ0, ρ1, ρ2, ρ3]. The individuals are represented by real vectors, N individuals are randomly generated, and the N individuals are combined to obtain the initial population. The inverse of the mean square error is used as the fitness function, and the fitness threshold and the number of iterations are set. According to the fitness value of each individual in the population, the roulette wheel selection method is adopted. The larger the fitness value of each individual, the greater the probability of the individual being selected, and the individuals in the population are selected to participate in the reproduction of the next generation; the arithmetic crossover method is adopted to exchange genes for the selected individuals; the regression coefficient of the solar power generation monitoring model in the individuals after gene exchange is randomly selected, and the regression coefficient of the solar power generation monitoring model is randomly changed; the iterative operations of selection, crossover and mutation are repeated until the fitness value of the individual in the population reaches the fitness threshold and the number of repeated iterative operations reaches the set number of iterations, the repeated iterative operation is terminated, and after the repeated iterative operation is terminated, the regression coefficient corresponding to the individual with the highest fitness value in the population is used to update the regression coefficient in the solar power generation monitoring model to obtain the optimized solar power generation monitoring model.

7. The complementary station distribution system of a new energy decentralized power station according to claim 6 is characterized by: The monitoring optimization module optimizes the wind power generation monitoring model in the following process: Using genetic algorithm, the regression coefficient of wind power generation monitoring model is encoded by real numbers to obtain a real vector of length 4, which is expressed as [ρ0, ρ1, ρ2, ρ3]. The real vector represents the individual, M individuals are randomly generated, and the M individuals are combined to obtain the initial population. The inverse of the mean square error is used as the fitness function, and the fitness threshold and the number of iterations are set. According to the fitness value of each individual in the population, the roulette selection method is adopted. The larger the fitness value of each individual, the greater the probability of the individual being selected, and the individuals in the population are selected to participate in the reproduction of the next generation; the arithmetic crossover method is adopted to perform gene exchange on the selected individuals; the regression coefficient of the wind power generation monitoring model in the individuals after gene exchange is randomly selected, and the regression coefficient of the wind power generation monitoring model is randomly changed, and the iterative operations of selection, crossover and mutation are repeated until the fitness value of the individual in the population reaches the fitness threshold and the number of repeated iteration operations reaches the set number of iterations, the repeated iteration operation is terminated, and after the repeated iteration operation is terminated, the regression coefficient corresponding to the individual with the highest fitness value in the population is used to update the regression coefficient in the wind power generation monitoring model to obtain the optimized wind power generation monitoring model.

8. The complementary station distribution system of a new energy decentralized power station according to claim 7 is characterized by: The construction process of the power distribution monitoring module and the new energy power generation monitoring model includes: Input the solar power generation monitoring data and the real-time power generation of the distributed solar power station into the solar power generation monitoring model to obtain the solar power generation impact index; input the wind power generation monitoring data and the real-time power generation of the distributed wind power station into the wind power generation monitoring model to obtain the wind power generation impact index; The solar power generation monitoring data, the wind power generation monitoring data, the solar power generation impact index and the wind power generation impact index are used as the third data set, which are divided into a training set and a test set in a ratio of 7:3, the solar power generation monitoring data and the wind power generation monitoring data are used as input, and the solar power generation impact index and the wind power generation impact index are used as output; Using the training set data and the neural network algorithm, the nonlinear relationship between the solar power generation monitoring data and the solar power generation impact index and the nonlinear relationship between the wind power generation monitoring data and the wind power generation impact index are learned through repeated iterative training until the set number of iterative training times is reached, and the new energy power generation monitoring model is trained; The test set data is input into the renewable energy power generation monitoring model, and the MSE function is used to evaluate the errors between the solar power generation impact index and wind power generation impact index output by the renewable energy power generation monitoring model and the actual solar power generation impact index and wind power generation impact index. According to the evaluation results, the parameters of the renewable energy power generation monitoring model are adjusted to optimize the performance of the renewable energy power generation monitoring model, and the optimized renewable energy power generation monitoring model is deployed into the system.

9. The complementary station distribution system of a new energy decentralized power station according to claim 8 is characterized by: The power distribution control module, in combination with the output result of the new energy power generation monitoring model, takes corresponding measures to allocate the distributed solar power station and the distributed wind power station, and performs power distribution control according to the load demand, including: The solar power generation monitoring data and the wind power generation monitoring data are respectively input into the new energy power generation monitoring model, and the new energy power generation monitoring model outputs the solar power generation impact index and the wind power generation impact index respectively; When the solar power generation impact index is between 0 and 0.3, the impact of solar power generation monitoring data on the real-time power generation of distributed solar power stations is low; when the solar power generation impact index is between 0.3 and 0.7, the impact of solar power generation monitoring data on the real-time power generation of distributed solar power stations is medium; when the solar power generation impact index is between 0.7 and 1, the impact of solar power generation monitoring data on the real-time power generation of distributed solar power stations is high; When the wind power generation impact index is between 0 and 0.3, the impact of wind power generation monitoring data on the real-time power generation of distributed wind power stations is low; when the wind power generation impact index is between 0.3 and 0.7, the impact of wind power generation monitoring data on the real-time power generation of distributed wind power stations is medium; when the wind power generation impact index is between 0.7 and 1, the impact of wind power generation monitoring data on the real-time power generation of distributed wind power stations is high; When the impact of solar power generation monitoring data on the real-time power generation of decentralized solar power stations is greater than the impact of wind power generation monitoring data on the real-time power generation of decentralized wind power stations, the decentralized solar power stations are shut down for power generation, and the decentralized wind power stations are deployed to operate at full load; When the impact of solar power generation monitoring data on the real-time power generation of decentralized solar power stations is lower than the impact of wind power generation monitoring data on the real-time power generation of decentralized wind power stations, the decentralized wind power stations are shut down for power generation, and the decentralized solar power stations are deployed to operate at full load; When the impact of solar power generation monitoring data on the real-time power generation of distributed solar power stations is the same as the impact of wind power generation monitoring data on the real-time power generation of distributed wind power stations and is low, the distributed solar power stations and distributed wind power stations are deployed to operate at full load to maximize power generation; When the impact of solar power generation monitoring data on the real-time power generation of decentralized solar power stations is the same as the impact of wind power generation monitoring data on the real-time power generation of decentralized wind power stations and is medium, load operation is allocated to decentralized solar power stations and decentralized wind power stations in a 1:1 ratio to dynamically balance the supply and demand of power distribution; When the impact of the solar power generation monitoring data on the real-time power generation of the distributed solar power station is the same as the impact of the wind power generation monitoring data on the real-time power generation of the distributed wind power station and is high, the power generation process of the distributed solar power station and the distributed wind power station is shut down and the backup power supply is started; The power distribution plan is dynamically adjusted according to the total power generation of the decentralized solar power stations and decentralized wind power stations and the load demand. When the total power generation of the decentralized solar power stations and decentralized wind power stations is greater than the load demand, the total power generation of the decentralized solar power stations and decentralized wind power stations is allocated according to the load demand, and the excess power generation is stored; when the total power generation of the decentralized solar power stations and decentralized wind power stations is less than the load demand, the load demand is reduced according to the load priority, and the stored power generation is allocated.

10. A complementary power collection and distribution control method for a new energy decentralized power station, implemented based on a complementary power collection and distribution system for a new energy decentralized power station as described in any one of claims 1 to 9, characterized in that: It consists of the following steps: Step 1: Use a light intensity sensor, a sunshine meter, an ultrasonic anemometer, an ultrasonic wind direction sensor and a smart meter to collect solar power generation monitoring data, wind power generation monitoring data and real-time power generation of new energy decentralized power stations, and pre-process the collected data; Step 2: Based on the preprocessed data, combined with the multivariate linear regression algorithm, a solar power generation monitoring model and a wind power generation monitoring model are constructed respectively; Step 3: Optimize the solar power generation monitoring model and the wind power generation monitoring model; Step 4: Obtain a new energy power generation monitoring model through the output results of the solar power generation monitoring model and the wind power generation monitoring model; Step 5: According to the output results of the new energy power generation monitoring model, take corresponding measures to deploy decentralized solar power stations and decentralized wind power stations.

Images