A new energy power station digital intelligence integrated grid connection cooperation method and system

Abstract

The application provides a new energy power station digital integrated grid-connected cooperation method and system, and relates to the technical field of energy storage charging and discharging control. The method comprises the following steps: collecting the environmental temperature, irradiance and output power of the power station; determining the photoelectric conversion promotion degree based on the current environmental temperature and the optimal working temperature, extracting the high-frequency detail component of the irradiance time sequence and determining the amplitude discreteness thereof, combining the photoelectric conversion promotion degree to determine the high-frequency cooperation suppression degree; extracting the energy spectrum of the output power time sequence and constructing the high-frequency energy change sequence, determining the energy attenuation deviation degree based on the fitting function fitting point change slope Shannon entropy, the high-frequency energy and the corresponding fitting point frequency domain energy, and combining the high-frequency cooperation suppression degree to determine the cooperation bad interference degree; obtaining the predicted output power through the power prediction model based on the current environmental temperature, irradiance signal and cooperation bad interference degree; and controlling the energy storage unit charging and discharging based on the prediction error. The application can realize the charging and discharging cooperation control of the energy storage unit.

Description

Technical Field

[0001] This application relates to the field of energy storage charging and discharging control technology, specifically to a digital and intelligent integrated grid-connected collaborative method and system for new energy power plants. Background Technology

[0002] As the global energy structure transitions towards cleaner and lower-carbon energy, the scale of new energy power plant construction and the number of grid-connected plants in the region continue to grow, and the proportion of new energy power generation in the power system is increasing year by year. However, due to environmental factors, the output of new energy is subject to random fluctuations and intermittentity. When a large number of new energy power plants are connected to the grid, this output instability will be directly transmitted to the large-scale power grid, causing problems such as grid frequency fluctuations and voltage deviations, seriously threatening the safety and stability of grid operation, and becoming a key bottleneck restricting the large-scale consumption of new energy and the construction of smart grids.

[0003] The relevant technologies improve the reliability of grid-connected operation of new energy power plants by introducing a digital and intelligent integrated platform. Specifically, the platform's perception layer collects power-related data from new energy power plants; the collected power-related data is input into a data prediction model to predict the output status of new energy; and intelligent scheduling of power resources under grid connection is achieved based on the prediction results to ensure grid stability.

[0004] However, in practical applications, the above-mentioned integrated digital and intelligent solutions have the following problems: the output of new energy sources is not only affected by temperature and irradiance alone, but also by the synergistic effect of the two, which exacerbates the fluctuations. Existing technologies cannot accurately measure this synergistic suppression effect, resulting in a large deviation between the predicted results and the actual output. Existing technologies lack sufficient analysis of the high-frequency energy characteristics of the output power of new energy power plants. The abnormal high-frequency energy attenuation caused by external noise interference is difficult to effectively identify and quantify, further reducing the accuracy of output prediction. When the predicted output deviates too much from the actual output, it is impossible to accurately coordinate the charging and discharging status of energy storage in new energy power plants under grid-connected conditions. This is not conducive to the safety and stability of large-scale power grid operation and restricts the construction and implementation of large-scale power grid security and defense systems and intelligent dispatching systems. Summary of the Invention

[0005] To address the shortcomings of existing technologies in accurately measuring the synergistic suppression effect of temperature and irradiance, and inadequate high-frequency energy characteristic analysis of the output power of renewable energy power plants, which leads to significant discrepancies between predicted and actual power output and hinders accurate coordinated control of energy storage charging and discharging states under grid-connected conditions, thus compromising the safety and stability of large-scale power grid operation, this application provides a digital and intelligent integrated grid-connected coordination method for renewable energy power plants. The specific technical solution adopted is as follows: Collect ambient temperature, irradiance, and output power of new energy power plants that are connected to the grid; The photoelectric conversion enhancement is determined based on the current ambient temperature and the optimal operating temperature of the photovoltaic module. The irradiance time series sequence within a preset time period before the current sampling time is obtained. The high-frequency detail components of each decomposition layer of the irradiance time series sequence are obtained by wavelet decomposition algorithm and their first-order difference value variation coefficient is calculated. The high-frequency cooperative suppression degree is determined based on the photoelectric conversion enhancement and the first-order difference value variation coefficient. The output power time series sequence within a preset time period before the current sampling time is obtained, the energy spectrum of the output power time series sequence is extracted by Fourier transform, a high-frequency energy change sequence is constructed based on the energy spectrum and its fitting function is determined, and the energy attenuation deviation is determined based on the Shannon entropy of the slope of all fitting points of the fitting function, each high-frequency energy in the high-frequency energy change sequence and the frequency domain energy of the corresponding fitting point. Based on the high-frequency cooperative suppression degree and the energy attenuation deviation degree, the cooperative poor interference degree is determined. The current ambient temperature, the current irradiance signal and the current cooperative poor interference degree are input into the pre-trained power prediction model to obtain the predicted output power. The charging and discharging control of the energy storage unit is achieved based on the prediction error between the current output power and the predicted output power.

[0006] For example, determining the photoelectric conversion improvement based on the current ambient temperature and the optimal operating temperature of the photovoltaic module includes: calculating the temperature ratio between the current ambient temperature and the optimal operating temperature; calculating the absolute value of the difference between the temperature ratio and 1, and negatively mapping it through an exponential function with the natural constant as the base, to obtain the photoelectric conversion improvement.

[0007] For example, the step of obtaining the irradiance time series sequence within a preset time period before the current sampling time, and using a wavelet decomposition algorithm to obtain the high-frequency detail components of each decomposition layer of the irradiance time series sequence and calculate their first-order difference value variation coefficient, includes: obtaining the historical irradiance of each historical sampling time within the preset time period before the current sampling time, and arranging each of the historical irradiance in chronological order to obtain the irradiance time series sequence; decomposing the irradiance time series sequence using a wavelet decomposition algorithm to extract the high-frequency detail components of each decomposition layer; calculating the difference between all adjacent data points of each high-frequency detail component to construct the first-order difference value sequence of the high-frequency detail components; calculating the mean and standard deviation of the first-order difference value sequence, and calculating the absolute value of their ratio, denoted as the first-order difference value variation coefficient.

[0008] For example, determining the high-frequency synergistic suppression degree based on the photoelectric conversion enhancement and the first-order difference value variation coefficient includes: for each decomposition layer, calculating the ratio of the first-order difference value variation coefficient of the high-frequency detail component to the photoelectric conversion enhancement, and recording it as the layer high-frequency synergistic suppression factor of the corresponding decomposition layer; calculating the sum of the high-frequency synergistic suppression factors of each layer and averaging them to obtain the high-frequency synergistic suppression degree.

[0009] For example, the step of obtaining the output power time series sequence within a preset time period before the current sampling time, extracting the energy spectrum of the output power time series sequence using Fourier transform, constructing a high-frequency energy change sequence based on the energy spectrum, and determining its fitting function includes: obtaining the historical output power of each historical sampling time within a preset time period before the current sampling time, and arranging each historical output power in chronological order to obtain the output power time series sequence; extracting the energy spectrum of the output power time series sequence using discrete Fourier transform, determining the frequency domain energy corresponding to each frequency higher than the fundamental frequency in the energy spectrum as high-frequency energy, sorting the high-frequency energy in ascending order of its corresponding frequency to obtain the high-frequency energy change sequence; and fitting the high-frequency energy change sequence using nonlinear least squares method to obtain the fitting function.

[0010] For example, determining the energy attenuation deviation based on the Shannon entropy of the slope of all fitting points of the fitting function, each high-frequency energy in the high-frequency energy change sequence, and the frequency domain energy of the corresponding fitting point includes: differentiating the fitting function to obtain the slope of all fitting points in the fitting function, constructing a corresponding fitting point slope sequence, converting the fitting point slope sequence into a discrete probability distribution, substituting the discrete probability distribution into the Shannon entropy formula to obtain the Shannon entropy of the slope of all fitting points of the fitting function, denoted as the fitting slope Shannon entropy; determining the number of elements in the high-frequency energy change sequence, calculating the ratio of the fitting slope Shannon entropy to the number of elements, denoted as the attenuation trend stabilization factor; determining the energy anomaly factor based on the degree of deviation between each high-frequency energy in the high-frequency energy change sequence and the frequency domain energy of its corresponding fitting point in the fitting function; and determining the energy attenuation deviation based on the attenuation trend stabilization factor and the energy anomaly factor of each high-frequency energy.

[0011] For example, determining the degree of uncooperative interference based on the high-frequency cooperative suppression degree and the energy attenuation deviation degree includes: calculating the sum of the high-frequency cooperative suppression degree and the energy attenuation deviation degree and averaging them, which is denoted as the degree of uncooperative interference.

[0012] For example, the power prediction model is trained by the following method: acquiring historical ambient temperature, historical irradiance and historical output power at multiple historical sampling times, and determining the historical cooperative interference degree at each of the historical sampling times; using the historical ambient temperature, historical irradiance and historical cooperative interference degree at each of the historical sampling times as training samples and the historical output power as training labels to train the LSTM neural network model to obtain the power prediction model.

[0013] For example, the method of controlling the charging and discharging of the energy storage unit based on the prediction error between the current output power and the predicted output power includes: calculating the difference between the current output power and the predicted output power, and recording it as the prediction error; if the prediction error is positive, controlling the energy storage unit to charge; if the prediction error is negative, controlling the energy storage unit to discharge.

[0014] Correspondingly, this application also provides a new energy power plant digital and intelligent integrated grid-connected collaborative system, including: The data acquisition module is used to collect ambient temperature, irradiance, and output power of new energy power plants in grid-connected status; The data processing module is used to determine the photoelectric conversion enhancement based on the current ambient temperature and the optimal operating temperature of the photovoltaic module, obtain the irradiance time series sequence within a preset time period before the current sampling time, use wavelet decomposition algorithm to obtain the high-frequency detail components of each decomposition layer of the irradiance time series sequence and calculate its first-order difference value variation coefficient, and determine the high-frequency cooperative suppression degree based on the photoelectric conversion enhancement and the first-order difference value variation coefficient. The data processing module is also used to obtain the output power time series sequence within a preset time period before the current sampling time, extract the energy spectrum of the output power time series sequence using Fourier transform, construct a high-frequency energy change sequence based on the energy spectrum and determine its fitting function, and determine the energy attenuation deviation based on the Shannon entropy of the slope of all fitting points of the fitting function, each high-frequency energy in the high-frequency energy change sequence and the frequency domain energy of the corresponding fitting point. The data processing module is also used to determine the degree of poor coordination interference based on the high-frequency coordination suppression degree and the energy attenuation deviation degree, and input the current ambient temperature, the current irradiance signal and the current degree of poor coordination interference into the pre-trained power prediction model to obtain the predicted output power. The collaborative control module is used to control the charging and discharging of the energy storage unit based on the prediction error between the current output power and the predicted output power.

[0015] This application may have some or all of the following beneficial effects: In the integrated digital and intelligent grid-connected collaborative method for new energy power plants provided in this application, the ambient temperature, irradiance, and output power of the new energy power plant are collected. The photoelectric conversion enhancement is determined by combining the current ambient temperature and the optimal operating temperature of the photovoltaic modules. High-frequency collaborative suppression is determined by extracting high-frequency detail components and the coefficient of variation of the first-order difference value from the irradiance time-series data using wavelet decomposition. This method can accurately measure the collaborative suppression effect of temperature and irradiance on the new energy output, solving the problem of large output prediction deviations caused by the inability to quantify this collaborative effect in related technologies. The method also extracts the energy spectrum of the output power time-series data through Fourier transform, constructs a high-frequency energy change sequence, and determines the fitting function. This is combined with the slope of the fitting point change, Shannon entropy, high-frequency energy, and the effect on... The energy attenuation deviation can be determined by fitting the frequency domain energy of the fitting point, which can effectively identify and quantify the high-frequency energy attenuation anomaly caused by external noise interference in the output power. This makes up for the deficiency of related technologies in the analysis of high-frequency energy characteristics of output power, and further improves the accuracy of new energy output prediction. Based on the high-frequency cooperative suppression degree and the energy attenuation deviation degree, the cooperative interference degree is determined. The current ambient temperature, irradiance and cooperative interference degree are input into the power prediction model to obtain the predicted output power. Thus, the charging and discharging of the energy storage unit can be controlled based on the prediction error between the predicted output power and the actual output power. This enables precise cooperative control of the charging and discharging state of the energy storage unit under the grid connection conditions of new energy power plants, and ensures the safety and stability of large-scale power grid operation.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0017] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating an exemplary embodiment of this application of a digital and intelligent integrated grid-connected collaborative method for a new energy power plant is shown. Figure 2 A schematic block diagram of a new energy power plant digital and intelligent integrated grid-connected collaborative system according to an exemplary embodiment of this application is shown. Detailed Implementation

[0019] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation methods, structures, features, and effects of the new energy power plant digital and intelligent integrated grid-connected collaborative method and system proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0021] The following, in conjunction with the accompanying drawings, details the specific scheme of the new energy power plant digital and intelligent integrated grid-connected collaborative method and system provided in this application.

[0022] Please see Figure 1 It illustrates a flowchart of a new energy power plant digital-intelligent integrated grid-connected collaborative method according to an embodiment of this application, such as... Figure 1 As shown, the digital and intelligent integrated grid-connected collaborative method for new energy power plants specifically includes the following steps: S110: Collects ambient temperature, irradiance, and output power of new energy power plants in grid-connected state; S120: Based on the current ambient temperature and the optimal operating temperature of the photovoltaic module, determine the photoelectric conversion enhancement degree, obtain the irradiance time series sequence within a preset time before the current sampling time, use the wavelet decomposition algorithm to obtain the high-frequency detail components of each decomposition layer of the irradiance time series sequence and calculate its first-order difference value variation coefficient, and determine the high-frequency cooperative suppression degree based on the photoelectric conversion enhancement degree and the first-order difference value variation coefficient. S130: Obtain the output power time series within a preset time period before the current sampling time, use Fourier transform to extract the energy spectrum of the output power time series, construct a high-frequency energy change sequence based on the energy spectrum and determine its fitting function, and determine the energy attenuation deviation based on the Shannon entropy of the slope of all fitting points of the fitting function, each high-frequency energy in the high-frequency energy change sequence and the frequency domain energy of the corresponding fitting point. S140: Based on the high-frequency cooperative suppression degree and energy attenuation deviation degree, the cooperative poor interference degree is determined. The current ambient temperature, current irradiance signal and current cooperative poor interference degree are input into the pre-trained power prediction model to obtain the predicted output power. S150: The charging and discharging control of the energy storage unit is realized based on the prediction error between the current output power and the predicted output power.

[0023] The following is a detailed explanation of each step in the above-mentioned integrated digital and intelligent grid-connected collaborative method for new energy power plants: In step S110, the ambient temperature, irradiance, and output power of the new energy power station in grid-connected state are collected.

[0024] In this embodiment of the application, the above-mentioned grid-connected state refers to the operating state in which the new energy power station connects to the large-scale power grid through equipment such as transmission lines and substations, transmits the electricity generated in the station to the power grid, and accepts the grid dispatch management.

[0025] In this embodiment of the application, the aforementioned new energy power station is a power generation station that uses non-fossil energy sources such as solar energy, wind energy, biomass energy, geothermal energy, and ocean energy as core power generation raw materials, converts the aforementioned renewable energy sources into electrical energy through professional power generation equipment, has independent power generation capabilities, can be connected to a large-scale power grid for grid operation, accepts grid dispatch management, and provides clean electricity to the power system; specifically, in this embodiment of the application, the new energy power station is a photovoltaic power station.

[0026] In this embodiment, the ambient temperature refers to the real-time air temperature in the photovoltaic module installation area of ​​the new energy power station, which is a key environmental factor affecting the photoelectric conversion efficiency of the photovoltaic module.

[0027] In the embodiments of this application, the above-mentioned irradiance is the solar radiation energy projected onto a unit area of ​​the photovoltaic module per unit time, which directly determines the power generation of the photovoltaic module. Its high-frequency fluctuations (such as those caused by cloud cover or water vapor changes) are the core cause of random fluctuations in the output of new energy sources.

[0028] In this embodiment of the application, the above-mentioned output power is the real-time active power transmitted by the new energy power plant to the grid, which is a core indicator reflecting the power output status of the power plant.

[0029] The new energy power plant digital and intelligent integrated grid-connected collaborative method provided in this application embodiment is realized based on the digital and intelligent integrated platform. The digital and intelligent integrated platform refers to a closed-loop system designed for the grid-connected operation of new energy power plants, with the core objective of ensuring the safety and stability of large-scale power grids, and integrating Internet of Things sensing, big data analysis and artificial intelligence technologies.

[0030] Specifically, the aforementioned integrated digital and intelligent platform includes a data perception layer, an intelligent analysis layer, and a collaborative control layer. The data perception layer is equipped with intelligent sensors, including temperature sensors, irradiance sensors, and power sensors, to collect ambient temperature, irradiance signals, and output power signals within the new energy power plant in real time. The intelligent analysis layer is used to predict and analyze the power output status of the new energy power plant through big data analysis and artificial intelligence processing. The collaborative control layer uses the predictive analysis results to achieve collaborative control of the energy storage charging and discharging status of the new energy power plant under grid-connected conditions.

[0031] In one specific implementation of this application embodiment, the above-mentioned collection of ambient temperature, irradiance and output power of the new energy power station in grid-connected state can be achieved as follows: the ambient temperature, irradiance and output power of the new energy power station in grid-connected state are collected in real time through the data sensing layer of the digital intelligence integration platform, wherein the new energy power station is specifically a photovoltaic power station, and the data collection frequency is 10Hz.

[0032] In step S120, the photoelectric conversion enhancement is determined based on the current ambient temperature and the optimal operating temperature of the photovoltaic module. The irradiance time series sequence within a preset time period before the current sampling time is obtained. The high-frequency detail components of each decomposition layer of the irradiance time series sequence are obtained by wavelet decomposition algorithm and their first-order difference value variation coefficient is calculated. The high-frequency cooperative suppression degree is determined based on the photoelectric conversion enhancement and the first-order difference value variation coefficient.

[0033] In this embodiment, the optimal operating temperature of the photovoltaic module is a specific temperature range in which the photovoltaic module can achieve the highest photoelectric conversion efficiency. When the ambient temperature of the photovoltaic module is close to the optimal operating temperature, the carrier migration efficiency of the internal semiconductor material of the photovoltaic module is the highest, the power loss is the lowest, the photoelectric conversion efficiency reaches its peak, and the absorbed solar radiation can be converted into electrical energy to the maximum extent. Its value range is usually 24-26℃, and in this embodiment, it is specifically taken as 25℃.

[0034] In the embodiments of this application, the aforementioned photoelectric conversion efficiency is a parameter used to quantify the influence of ambient temperature on the photoelectric conversion efficiency of photovoltaic modules; the closer its value is to 1, the closer the current ambient temperature is to the optimal operating temperature of the photovoltaic module, and the higher the photoelectric conversion efficiency.

[0035] For example, the above method of determining the photoelectric conversion improvement based on the current ambient temperature and the optimal operating temperature of the photovoltaic module can be achieved as follows: calculate the temperature ratio between the current ambient temperature and the optimal operating temperature; calculate the absolute value of the difference between the temperature ratio and 1, and negatively map it using an exponential function with the natural constant as the base to obtain the photoelectric conversion improvement.

[0036] Specifically, taking the current sampling time t as an example, the aforementioned photoelectric conversion improvement can be calculated using the following formula: in, The effect of ambient temperature at the new energy power station on the photoelectric conversion efficiency of photovoltaic modules at the current sampling time t; The ambient temperature of the new energy power station is collected at the current sampling time t; The optimal operating temperature for photovoltaic modules; The function is an exponential function with the natural constant as its base, which maps the calculation results to the range of (0,1] through negative mapping; the above formula eliminates the dimension of the temperature parameter by using a ratio; the photoelectric conversion efficiency reflects the effect of ambient temperature on the photoelectric conversion efficiency. The smaller the value, the more unfavorable it is for new energy power plants to generate photovoltaic power, that is, the stronger the inhibitory effect on photovoltaic power generation.

[0037] In this embodiment of the application, the above-mentioned irradiance time sequence is a sequence of all irradiance data arranged in chronological order within a preset time period before the current sampling time, which is used to reflect the dynamic change trend of irradiance in a short period of time.

[0038] In this embodiment, the high-frequency detail component is the signal component corresponding to the high frequency range obtained after wavelet decomposition, which can reflect the short-term rapid fluctuation characteristics in the original irradiance time series. The high-frequency detail component can directly reflect the high-frequency characteristic changes of solar irradiance caused by cloud layer changes and water vapor changes. The more violent the fluctuation, the more significant the random fluctuation of irradiance, and the stronger the interference to the output of new energy.

[0039] For example, the above-mentioned acquisition of the irradiance time series sequence within a preset time period before the current sampling time, and the acquisition of high-frequency detail components of each decomposition layer of the irradiance time series sequence using the wavelet decomposition algorithm and the calculation of its first-order difference value variation coefficient can be achieved as follows: acquire the historical irradiance of each historical sampling time within the preset time period before the current sampling time, and arrange each historical irradiance in chronological order to obtain the irradiance time series sequence; decompose the irradiance time series sequence using the wavelet decomposition algorithm to extract the high-frequency detail components of each decomposition layer; for each high-frequency detail component, calculate the difference between all its adjacent data points to construct the first-order difference value sequence of the high-frequency detail components; calculate the mean and standard deviation of the first-order difference value sequence, and calculate the absolute value of their ratio, which is denoted as the first-order difference value variation coefficient.

[0040] In one specific implementation of this application embodiment, taking the current sampling time t as an example, the process of determining the first-order difference value variation coefficient can be implemented as follows: One minute is traced back from the current sampling time t to obtain the time interval [t-60s, t]. All irradiance data collected within this time interval are extracted and arranged in chronological order of collection time to form the irradiance time series sequence corresponding to the current sampling time t. The irradiance time series sequence is used as the input of the wavelet decomposition algorithm. The wavelet basis is set to db4, and the decomposition level is 4. Four high-frequency detail components are obtained through the wavelet decomposition algorithm. The specific implementation of the wavelet decomposition algorithm is the same as that in the prior art and will not be described in detail here. The difference between all adjacent data points in each high-frequency detail component is calculated to obtain its corresponding first-order difference value sequence. The mean and standard deviation of all data points in the first-order difference value sequence are calculated, and the absolute value of their ratio is calculated, which is denoted as the first-order difference value variation coefficient of the corresponding high-frequency detail component.

[0041] If the dispersion of high-frequency component fluctuations in all high-frequency detail components of all decomposition layers is higher at a certain sampling time, it indicates that the solar irradiance is more significantly affected by the high-frequency characteristics of cloud and water vapor changes, and is more likely to cause random fluctuations in the output of new energy sources. At the same time, if the photoelectric conversion improvement is smaller at this time, it can more comprehensively indicate that the output of new energy sources is more affected by the high-frequency synergistic inhibition effect between environmental factors, the output of new energy power plants is worse, and it is less conducive to ensuring the safety and stability of large-scale power grids.

[0042] In the embodiments of this application, the above-mentioned high-frequency synergistic suppression degree is a parameter that comprehensively quantifies the suppressive effect of ambient temperature on photoelectric conversion and the interference effect of high-frequency fluctuations in irradiance on power output. The larger the value, the more severe the synergistic suppression effect of the new energy power output caused by the decrease in conversion efficiency due to temperature deviation and the instability of power output caused by high-frequency fluctuations in irradiance is. At this time, the power output condition of the power station is worse and more unfavorable to the safe and stable operation of large-scale power grids.

[0043] For example, the above determination of high-frequency synergistic suppression degree based on photoelectric conversion enhancement and first-order difference value variation coefficient can be achieved as follows: For each decomposition layer, calculate the ratio of the first-order difference value variation coefficient of the high-frequency detail component to the photoelectric conversion enhancement degree, and record it as the high-frequency synergistic suppression factor of the corresponding decomposition layer; calculate the sum of the high-frequency synergistic suppression factors of each layer and take the average to obtain the high-frequency synergistic suppression degree.

[0044] Specifically, taking the current sampling time t as an example, the above-mentioned high-frequency cooperative suppression degree can be calculated using the following formula: in, The high-frequency synergistic suppression degree of the new energy power station under the current sampling time t, which is simultaneously affected by high-frequency fluctuations in ambient temperature and irradiance. This represents the number of decomposition levels in the wavelet decomposition algorithm. The first difference value of the high-frequency detail component of the wavelet decomposition algorithm at the current sampling time t is the coefficient of variation. The larger the value, the higher the dispersion of the high-frequency component fluctuation in the high-frequency detail component, which is less conducive to photovoltaic power generation in new energy power plants. The above formula represents the effect of ambient temperature at the current sampling time t on the photovoltaic module's photoelectric conversion efficiency improvement. It combines the high-frequency oscillation suppression generated by the irradiance signal with photoelectric conversion suppression to measure the high-frequency synergistic suppression of renewable energy output by environmental factors. Specifically, the fluctuation and dispersion of the high-frequency detail components reflect the intensity of high-frequency interference from irradiance, while the photoelectric conversion efficiency improvement reflects the degree of suppression of conversion efficiency by temperature. The ratio of these two factors quantifies the synergistic suppression effect of environmental factors. A larger ratio indicates a more severe impact of the high-frequency synergistic suppression effect between environmental factors on renewable energy output, making it more unfavorable for photovoltaic power generation and resulting in a worse power output condition for the renewable energy power plant.

[0045] In step S130, the output power time series within a preset time period before the current sampling time is obtained, the energy spectrum of the output power time series is extracted by Fourier transform, a high-frequency energy change sequence is constructed based on the energy spectrum and its fitting function is determined, and the energy attenuation deviation is determined based on the Shannon entropy of the slope of all fitting points of the fitting function, each high-frequency energy in the high-frequency energy change sequence and the frequency domain energy of the corresponding fitting point.

[0046] In this embodiment of the application, the above-mentioned output power time sequence is a sequence of all output power data arranged in chronological order within a preset time period before the current sampling time, which is used to reflect the dynamic changes in the real-time active power transmitted from the new energy power plant to the grid in a short period of time.

[0047] In the embodiments of this application, the above-mentioned energy spectrum is a spectrum diagram reflecting the correspondence between frequency and energy after Fourier transform. Its horizontal axis is frequency, and its vertical axis is the frequency domain energy corresponding to the frequency. Among them, the high-frequency energy in the energy spectrum shows a decay trend, which is the key basis for judging external noise interference, while the low-frequency energy reflects the overall trend of power change.

[0048] In this embodiment of the application, the above-mentioned high-frequency energy change sequence is a sequence obtained by arranging high-frequency energy in ascending order of its corresponding frequency; wherein, high-frequency energy is the frequency domain energy of the corresponding frequency in the energy spectrum that is higher than the fundamental frequency; the weaker the decay trend of high-frequency energy in the high-frequency energy change sequence, the more likely the output power of the new energy power station is to be adversely affected by external noise interference, the more likely the output power of the new energy power station is to experience severe random fluctuations, the worse the power output condition, and the more detrimental it is to the safety and stability of the large-scale power grid.

[0049] For example, the above-mentioned acquisition of the output power time series sequence within a preset time period before the current sampling time, extraction of the energy spectrum of the output power time series sequence using Fourier transform, construction of a high-frequency energy change sequence based on the energy spectrum, and determination of its fitting function can be achieved as follows: The historical output power of each historical sampling time within the preset time period before the current sampling time is acquired, and each historical output power is arranged in chronological order to obtain the output power time series sequence; the energy spectrum of the output power time series sequence is extracted using discrete Fourier transform, and the frequency domain energy corresponding to each frequency higher than the fundamental frequency in the energy spectrum is determined as high-frequency energy; the high-frequency energy is sorted in ascending order of its corresponding frequency to obtain the high-frequency energy change sequence; the high-frequency energy change sequence is fitted using the nonlinear least squares method to obtain the fitting function.

[0050] In one specific implementation of this application embodiment, taking the current sampling time t as an example, the process of determining the fitting function for the high-frequency energy change sequence can be implemented as follows: Starting from the current sampling time t, trace back 1 minute to obtain the time interval [t-60s, t]. Extract all output power collected within this time interval and arrange them in chronological order of collection time to form the output power time series sequence corresponding to the current sampling time t. Obtain the energy spectrum of this output power time series sequence through Fourier transform (either discrete Fourier transform or fast Fourier transform). The specific implementation of the Fourier transform is the same as in the prior art and will not be repeated here. Determine the fundamental frequency in the energy spectrum and select all frequencies higher than the fundamental frequency and their corresponding frequency domain energies (i.e., the aforementioned high-frequency energies). Arrange the high-frequency energies in ascending order according to their corresponding high-frequency frequencies to form a high-frequency energy change sequence. Using the frequency of the high-frequency energy change sequence as the independent variable and the frequency domain energy as the dependent variable, substitute them into the nonlinear least squares method for curve fitting to obtain the fitting function.

[0051] In the embodiments of this application, the slope of the above-mentioned fitting function can reflect the energy decay trend. For example, the sign and magnitude of the slope represent the direction and intensity of the decay trend (a negative slope indicates energy decay, and the larger the absolute value of the negative slope, the faster the decay); the degree of disorder of the slope (such as fluctuating between positive and negative, or drastic fluctuations in absolute value) can reflect the stability of the ideal decay trend.

[0052] In this embodiment, the Shannon entropy of the slope of all fitting points of the above fitting function is an entropy value based on information theory, used to quantify the disorder of the slope of the fitting points; the larger the value, the more dispersed and disordered the distribution of the slope of the fitting points, the more unstable the ideal decay trend of high-frequency energy, and the greater the possibility that the output power is affected by external interference.

[0053] In the embodiments of this application, the above-mentioned energy attenuation deviation is a parameter that comprehensively quantifies the deviation between the actual value of high-frequency energy and the ideal fitting value, as well as the degree of disorder of the fitting slope. It is used to measure the intensity of external noise interference to the output power. The larger the value, the more significant the deviation between the actual change of high-frequency energy and the ideal attenuation trend, and the more unstable the attenuation trend. The more serious the random fluctuation of the output power, the worse the output of the new energy power station, and the greater the threat to the safe and stable operation of the large-scale power grid.

[0054] For example, the above-mentioned construction of a high-frequency energy change sequence based on the energy spectrum and determination of its fitting function, and determination of the energy attenuation deviation based on the Shannon entropy of the slope of all fitting points of the fitting function, each high-frequency energy in the high-frequency energy change sequence, and the frequency domain energy of the corresponding fitting point can be achieved as follows: Differentiate the fitting function to obtain the slope of all fitting points in the fitting function; construct the corresponding fitting point slope sequence; transform the fitting point slope sequence into a discrete probability distribution; substitute the discrete probability distribution into the Shannon entropy formula to obtain the Shannon entropy of the slope of all fitting points of the fitting function, denoted as the fitting slope Shannon entropy; determine the number of elements in the high-frequency energy change sequence; calculate the ratio of the fitting slope Shannon entropy to the number of elements, denoted as the attenuation trend stabilization factor; determine the energy anomaly factor based on the degree of deviation of each high-frequency energy in the high-frequency energy change sequence from the frequency domain energy of its corresponding fitting point in the fitting function; and determine the energy attenuation deviation based on the attenuation trend stabilization factor and the energy anomaly factor of each high-frequency energy.

[0055] In one specific implementation of this application embodiment, taking the current sampling time t as an example, the process of determining the energy attenuation deviation can be implemented as follows: Calculate the first derivative of the fitting function of the high-frequency energy change sequence to obtain the corresponding slope function; substitute each frequency point of the high-frequency energy change sequence into the slope function to obtain the slope of the corresponding fitting point, and construct the fitting point slope sequence; statistically analyze the value distribution of all slopes, calculate the probability of each slope appearing in all slopes, and substitute it into the Shannon entropy formula to obtain the Shannon entropy of the fitting slope; calculate the ratio of the Shannon entropy of the fitting slope to the number of elements in the high-frequency energy change sequence, denoted as the attenuation trend stabilization factor; determine the energy anomaly factor based on the degree of deviation between each high-frequency energy in the high-frequency energy change sequence and its corresponding frequency domain energy at the fitting point in the fitting function; calculate the energy attenuation deviation using the following formula: in, The energy attenuation deviation of the output power at the current sampling time t reflects the abnormal deviation of the high-frequency energy attenuation of the output power in the new energy power plant. The larger the value, the more likely the output power of the new energy power plant is to be adversely affected by external noise interference, the more likely the output power of the new energy power plant is to have serious random fluctuations, and the worse the output condition. The above fitting slope is the Shannon entropy; The number of elements in the high-frequency energy change sequence; The larger the value of the above-mentioned decay trend stabilization factor, the greater the disorder of the slope of the fitting function at all fitting points, indicating that the decay trend of high-frequency energy in the high-frequency energy change sequence is more unstable. This represents the k-th high-frequency energy value in the high-frequency energy change sequence. This represents the frequency domain energy of the fitting point corresponding to the k-th high-frequency energy value in the high-frequency energy change sequence on the fitting function. The larger the value of the aforementioned energy anomaly factor, the more significant the deviation of the actual energy from the ideal decay trend. This is an error parameter, used to avoid the denominator being 0. It takes a value within (0.001, 0.01), and its impact on the calculation result is small and can be ignored. In this embodiment, the value is 0.005.

[0056] In step S140, the degree of poor coordination interference is determined based on the high-frequency coordination suppression degree and the energy attenuation deviation degree. The current ambient temperature, the current irradiance signal and the current degree of poor coordination interference are input into the pre-trained power prediction model to obtain the predicted output power.

[0057] In this embodiment, the aforementioned cooperative adverse interference degree is a parameter that comprehensively quantifies the high-frequency cooperative suppression effect of environmental factors on the output of new energy and the abnormal deviation of high-frequency energy attenuation of output power. It is used to reflect the total intensity of interference from external adverse factors on the output of new energy power plants. The larger the value, the greater the cooperative interference impact on the output of new energy power plants during photovoltaic power generation. It can more comprehensively reflect the poor output status of new energy power plants and is conducive to accurate prediction and analysis of the output status of new energy power plants in the future.

[0058] For example, the determination of the cooperative interference degree based on the high-frequency cooperative suppression degree and the energy attenuation deviation degree can be achieved as follows: calculate the sum of the high-frequency cooperative suppression degree and the energy attenuation deviation degree and take the average, which is denoted as the cooperative interference degree.

[0059] In this embodiment of the application, the power prediction model is a model that has the ability to predict the output power of new energy power plants after being trained with historical data.

[0060] For example, the power prediction model described above can be trained by the following method: obtaining historical ambient temperature, historical irradiance, and historical output power at multiple historical sampling times; determining the historical cooperative interference degree at each historical sampling time using the above method; and training the LSTM neural network model using the historical ambient temperature, historical irradiance, and historical cooperative interference degree at each historical sampling time as training samples and the historical output power as training label to obtain the power prediction model.

[0061] In one specific implementation of this application, the training process of the power prediction model can be as follows: The ambient temperature, irradiance signals, and output power signals at N historical acquisition times are arranged in chronological order to obtain a temperature feature vector, an irradiance feature vector, and a power label vector. The cooperative adverse interference degree is calculated using a cooperative adverse interference degree calculation method at N historical acquisition times and then arranged in chronological order to obtain a cooperative feature vector. The lengths of the temperature feature vector, irradiance feature vector, cooperative feature vector, and power label vector are all N. In this specific implementation, N=10000. The temperature feature vector, irradiance feature vector, and cooperative feature vector are the training samples of the LSTM neural network model, and the power label vector is the training label of the LSTM neural network model. The LSTM neural network model is trained using the training samples and training labels to obtain the power prediction model. Specifically, during the training process, the ReLU function is used as the activation function, the Adam optimizer is used as the optimizer, and the MSE (mean squared error) function is used as the loss function.

[0062] After training the power prediction model using the above method, this embodiment of the application can calculate the cooperative interference degree at the current acquisition time in real time, input the ambient temperature, irradiance signal, and cooperative interference degree at the current acquisition time into the trained power prediction model, and predict the output power of the new energy power station at the current time through the power prediction model to obtain the predicted output power of the new energy power station at the current time.

[0063] In step S150, the charging and discharging control of the energy storage unit is achieved based on the prediction error between the current output power and the predicted output power.

[0064] Furthermore, in this embodiment of the application, the charging and discharging state of the energy storage unit can be coordinated and controlled based on the predicted output power using a traditional energy storage control strategy. For example, this coordinated control can be implemented as follows: calculate the difference between the current output power and the predicted output power, denoted as the prediction error; if the prediction error is positive, control the energy storage unit to charge; if the prediction error is negative, control the energy storage unit to discharge.

[0065] Specifically, the prediction error between the real-time output power and the predicted output power at the current acquisition moment is calculated in real time. If the prediction error is greater than 0, it means that the actual output power of the new energy power station is greater than the predicted output power, and the energy storage unit is charging. Conversely, if the prediction error is less than 0, it means that the actual output power of the new energy power station is less than the predicted output power, and the energy storage unit is discharging. This ensures the safety and stability of the large-scale power grid under the condition of grid connection of new energy power stations.

[0066] The foregoing mainly describes the solutions provided by the embodiments of this application from a methodological perspective. To achieve the above functions, it includes corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0067] Correspondingly, this application also provides a new energy power plant digital and intelligent integrated grid-connected collaborative system, referencing... Figure 2 As shown, the new energy power plant's integrated digital and intelligent grid-connected collaborative system 200 may include a data acquisition module 210, a data processing module 220, and a collaborative control module 230, wherein: The data acquisition module is used to collect ambient temperature, irradiance, and output power of new energy power plants in grid-connected status; The data processing module is used to determine the photoelectric conversion enhancement based on the current ambient temperature and the optimal operating temperature of the photovoltaic module, obtain the irradiance time series sequence within a preset time period before the current sampling time, use the wavelet decomposition algorithm to obtain the high-frequency detail components of each decomposition layer of the irradiance time series sequence and calculate its first-order difference value variation coefficient, and determine the high-frequency cooperative suppression degree based on the photoelectric conversion enhancement and the first-order difference value variation coefficient. The data processing module is also used to obtain the output power time series sequence within a preset time period before the current sampling time, extract the energy spectrum of the output power time series sequence using Fourier transform, construct a high-frequency energy change sequence based on the energy spectrum and determine its fitting function, and determine the energy attenuation deviation based on the Shannon entropy of the slope of all fitting points of the fitting function, each high-frequency energy in the high-frequency energy change sequence and the frequency domain energy of the corresponding fitting point. The data processing module is also used to determine the degree of poor coordination interference based on the high-frequency coordination suppression degree and energy attenuation deviation degree. The current ambient temperature, current irradiance signal and current degree of poor coordination interference are input into the pre-trained power prediction model to obtain the predicted output power. The collaborative control module is used to control the charging and discharging of the energy storage unit based on the prediction error between the current output power and the predicted output power.

[0068] The specific implementation details of the aforementioned digital and intelligent integrated grid-connected collaborative system for new energy power plants have been explained in detail in the corresponding section of the digital and intelligent integrated grid-connected collaborative method for new energy power plants, and therefore will not be repeated here.

[0069] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0070] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A digital and intelligent integrated grid-connected collaborative method for new energy power plants, characterized in that, The method includes: Collect ambient temperature, irradiance, and output power of new energy power plants that are connected to the grid; The photoelectric conversion enhancement is determined based on the current ambient temperature and the optimal operating temperature of the photovoltaic module. The irradiance time series sequence within a preset time period before the current sampling time is obtained. The high-frequency detail components of each decomposition layer of the irradiance time series sequence are obtained by wavelet decomposition algorithm and their first-order difference value variation coefficient is calculated. The high-frequency cooperative suppression degree is determined based on the photoelectric conversion enhancement and the first-order difference value variation coefficient. The output power time series sequence within a preset time period before the current sampling time is obtained, the energy spectrum of the output power time series sequence is extracted by Fourier transform, a high-frequency energy change sequence is constructed based on the energy spectrum and its fitting function is determined, and the energy attenuation deviation is determined based on the Shannon entropy of the slope of all fitting points of the fitting function, each high-frequency energy in the high-frequency energy change sequence and the frequency domain energy of the corresponding fitting point. Based on the high-frequency cooperative suppression degree and the energy attenuation deviation degree, the cooperative poor interference degree is determined. The current ambient temperature, the current irradiance signal and the current cooperative poor interference degree are input into the pre-trained power prediction model to obtain the predicted output power. The charging and discharging control of the energy storage unit is achieved based on the prediction error between the current output power and the predicted output power.

2. The method for integrated digital and intelligent grid connection and coordination of new energy power plants according to claim 1, characterized in that, The determination of photoelectric conversion efficiency based on the current ambient temperature and the optimal operating temperature of the photovoltaic module includes: Calculate the temperature ratio between the current ambient temperature and the optimal operating temperature; The absolute value of the difference between the temperature ratio and 1 is calculated, and the photoelectric conversion enhancement is obtained by negatively mapping it through an exponential function with the natural constant as the base.

3. The method for integrated digital and intelligent grid connection and coordination of new energy power plants according to claim 2, characterized in that, The process of obtaining the irradiance time series sequence within a preset time period prior to the current sampling time, and using a wavelet decomposition algorithm to obtain the high-frequency detail components of each decomposition layer of the irradiance time series sequence and calculate their first-order difference coefficient of variation, includes: The historical irradiance of each historical sampling moment within a preset time period before the current sampling moment is obtained, and the historical irradiance is arranged in chronological order to obtain the irradiance time sequence. The irradiance time series is decomposed using a wavelet decomposition algorithm to extract the high-frequency detail components of each decomposition layer; For each of the high-frequency detail components, the difference between all its adjacent data points is calculated to construct a first-order difference value sequence for the high-frequency detail components; Calculate the mean and standard deviation of the first-order difference value sequence, and calculate the absolute value of their ratio, which is denoted as the coefficient of variation of the first-order difference value.

4. The new energy power plant digital and intelligent integrated grid-connected collaborative method according to claim 3, characterized in that, The determination of high-frequency cooperative suppression degree based on the photoelectric conversion enhancement and the first-order difference coefficient of variation includes: For each decomposition layer, the ratio of the coefficient of variation of the first difference value of the high-frequency detail component to the photoelectric conversion enhancement is calculated and denoted as the layer high-frequency synergistic suppression factor of the corresponding decomposition layer. The high-frequency synergistic inhibition degree is obtained by calculating the sum of the high-frequency synergistic inhibition factors of each layer and averaging them.

5. The method for integrated digital and intelligent grid connection and coordination of new energy power plants according to claim 1, characterized in that, The process of obtaining the output power time-series sequence within a preset time period prior to the current sampling time, extracting the energy spectrum of the output power time-series sequence using Fourier transform, constructing a high-frequency energy change sequence based on the energy spectrum, and determining its fitting function includes: Obtain the historical output power of each historical sampling moment within a preset time period before the current sampling moment, and arrange the historical output power in chronological order to obtain the output power time sequence; The energy spectrum of the output power time series is extracted using discrete Fourier transform. The frequency domain energy corresponding to each frequency higher than the fundamental frequency in the energy spectrum is determined as high-frequency energy. The high-frequency energy is sorted in ascending order of its corresponding frequency to obtain the high-frequency energy change sequence. The high-frequency energy change sequence is fitted using the nonlinear least squares method to obtain the fitting function.

6. The new energy power plant digital and intelligent integrated grid-connected collaborative method according to claim 5, characterized in that, The determination of energy attenuation deviation based on the Shannon entropy of the slope of all fitting points of the fitting function, each high-frequency energy in the high-frequency energy change sequence, and the frequency domain energy of the corresponding fitting point includes: Differentiate the fitting function to obtain the slope of all fitting points in the fitting function, construct the corresponding slope sequence of fitting points, transform the slope sequence of fitting points into a discrete probability distribution, substitute the discrete probability distribution into the Shannon entropy formula, and obtain the Shannon entropy of the slope of all fitting points of the fitting function, denoted as the fitting slope Shannon entropy. The number of elements in the high-frequency energy change sequence is determined, and the ratio of the fitting slope Shannon entropy to the number of elements is calculated and denoted as the decay trend stabilization factor. The energy anomaly factor is determined based on the degree of deviation between each high-frequency energy in the high-frequency energy change sequence and the frequency domain energy of its corresponding fitting point in the fitting function; The energy attenuation deviation is determined based on the attenuation trend stabilization factor and the energy anomaly factor of each of the high-frequency energies.

7. The method for integrated digital and intelligent grid connection and coordination of new energy power plants according to claim 1, characterized in that, The determination of the cooperative interference degree based on the high-frequency cooperative suppression degree and the energy attenuation deviation degree includes: The sum of the high-frequency cooperative suppression degree and the energy attenuation deviation degree is calculated and averaged, and denoted as the cooperative poor interference degree.

8. The method for integrated digital and intelligent grid connection and coordination of new energy power plants according to claim 1, characterized in that, The power prediction model is trained using the following methods: Acquire historical ambient temperature, historical irradiance, and historical output power at multiple historical sampling times, and determine the historical cooperative adverse interference degree at each of the historical sampling times; Using the historical ambient temperature, historical irradiance, and historical cooperative interference at each historical sampling time as training samples, and the historical output power as training labels, the LSTM neural network model is trained to obtain the power prediction model.

9. The method for integrated digital and intelligent grid connection and coordination of new energy power plants according to claim 1, characterized in that, The method of controlling the charging and discharging of the energy storage unit based on the prediction error between the current output power and the predicted output power includes: Calculate the difference between the current output power and the predicted output power, and denot it as the prediction error; If the prediction error is positive, the energy storage unit is controlled to charge; if the prediction error is negative, the energy storage unit is controlled to discharge.

10. A new energy power plant digital-intelligent integrated grid-connected collaborative system, characterized in that, The system includes: The data acquisition module is used to collect ambient temperature, irradiance, and output power of new energy power plants in grid-connected status; The data processing module is used to determine the photoelectric conversion enhancement based on the current ambient temperature and the optimal operating temperature of the photovoltaic module, obtain the irradiance time series sequence within a preset time period before the current sampling time, use wavelet decomposition algorithm to obtain the high-frequency detail components of each decomposition layer of the irradiance time series sequence and calculate its first-order difference value variation coefficient, and determine the high-frequency cooperative suppression degree based on the photoelectric conversion enhancement and the first-order difference value variation coefficient. The data processing module is also used to obtain the output power time series sequence within a preset time period before the current sampling time, extract the energy spectrum of the output power time series sequence using Fourier transform, construct a high-frequency energy change sequence based on the energy spectrum and determine its fitting function, and determine the energy attenuation deviation based on the Shannon entropy of the slope of all fitting points of the fitting function, each high-frequency energy in the high-frequency energy change sequence and the frequency domain energy of the corresponding fitting point. The data processing module is also used to determine the degree of poor coordination interference based on the high-frequency coordination suppression degree and the energy attenuation deviation degree, and input the current ambient temperature, the current irradiance signal and the current degree of poor coordination interference into the pre-trained power prediction model to obtain the predicted output power. The collaborative control module is used to control the charging and discharging of the energy storage unit based on the prediction error between the current output power and the predicted output power.

Images