A power supply system and method for a photovoltaic-storage integrated distribution substation

Abstract

The present invention discloses a power supply system and method for a photovoltaic-storage integrated distribution substation, which relates to the field of power supply for photovoltaic-storage distribution substations. It is used to solve the problems that due to the intermittency and volatility of photovoltaic power generation, after introducing the external power grid, the frequency regulation discharge mode causes the electric energy in the energy storage module battery to be unable to be released, reducing the utilization rate of resources, increasing the scheduling frequency, and wasting production resources. Logical regression calculation is performed according to the remaining capacity of the energy storage battery and the predicted value of the current power supply to determine the rationality of scheduling. And according to the scheduling rationality, the received intensity of solar radiation to the photovoltaic panel, the temperature difference between the current temperature and the surface temperature of the photovoltaic panel, and the short-term load prediction value are obtained. Prediction is carried out by establishing a time series model. According to the prediction results, a scheduling adjustment plan is determined. Based on the similarity of the predicted scheduling adjustment plan of similar history and the realization rate of the historical predicted scheduling adjustment plan, a simple scheduling strategy is obtained to reduce excessive scheduling and ensure the maximization of photovoltaic power generation resources.

Description

Technical Field

[0001] The present invention relates to the field of power supply for photovoltaic energy storage distribution stations. More specifically, the present invention relates to a power supply system and method for a photovoltaic energy storage integrated distribution station. Background Art

[0002] With the rapid development of renewable energy, photovoltaic power generation systems are gradually applied to distribution stations to achieve the efficient utilization of green energy. Photovoltaic energy storage integration refers to the energy storage technology through photovoltaic power generation, that is, converting solar energy into electrical energy and then performing battery energy storage operations to achieve efficient and stable power supply, which can effectively improve the power supply reliability and energy utilization rate of the system.

[0003] The prior art has the following deficiencies:

[0004] Currently, due to the intermittency and volatility of photovoltaic power generation, a single photovoltaic system is difficult to meet the stable demand of the load. After introducing the external power grid to ensure the stability of the power supply system, it is often necessary to frequently adjust the discharge mode, which increases the dispatching frequency and has the potential risk of energy loss. Obviously, when adjusting to discharge to the external power grid, the photovoltaic continues to store energy, resulting in the inability to release the electrical energy in the energy storage module battery, reducing the utilization rate of resources, increasing the dispatching frequency, and wasting production resources. Therefore, a power supply system and method for a photovoltaic energy storage integrated distribution station are proposed.

[0005] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure. Therefore, it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a power supply system and method for a photovoltaic energy storage integrated distribution station, and solve the problems raised in the above background art by using different product inspection methods.

[0007] To achieve the above object, the present invention provides the following technical solution. A power supply method for a photovoltaic energy storage integrated distribution station includes S1: collecting the remaining capacity of the energy storage battery and the predicted value of the current power supply, performing normalization processing, performing logistic regression calculation on the normalized data to obtain a reasonable probability, comparing it with a preset reasonable threshold, and determining the current dispatching unreasonable result;

[0008] S2: obtaining the current dispatching unreasonable result, collecting weather prediction information and load information to obtain the received intensity of solar radiation on the photovoltaic panel, the temperature difference between the current temperature and the surface temperature of the photovoltaic panel, and the short-term load prediction value;

[0009] S3: According to the solar radiation received by the photovoltaic panel, the difference between the current temperature and the surface temperature of the photovoltaic panel, and the short-term load prediction value, perform photovoltaic power generation satisfaction prediction analysis through a time series model, and further refine and optimize the dispatching adjustment plan;

[0010] S4: According to the dispatching adjustment plan, calculate the similarity with the historical predicted dispatching adjustment plan, and select the implementation rate of the historical predicted dispatching adjustment plan and the similarity of the corresponding similar historical predicted dispatching adjustment plan based on the similarity calculation;

[0011] S5: Obtain the similarity of the similar historical predicted dispatching adjustment plan and the implementation rate of the historical predicted dispatching adjustment plan for clustering analysis, determine the implementation rate of the dispatching adjustment plan, and obtain a simple dispatching strategy.

[0012] In a preferred embodiment, by real-time monitoring the voltage and current of the battery, and combining the discharge characteristics and capacity curve of the battery, the remaining capacity of the energy storage battery is obtained;

[0013] By taking the load data, as well as the temperature, humidity, and sunshine within the current total sampling time for linear regression, combining the historical load data and meteorological data, a historical load curve is obtained, and the weighted average calculation is performed based on the load data corresponding to the sampling time on the curve to obtain the current power supply prediction value;

[0014] Substitute the remaining capacity of the energy storage battery and the current power supply prediction value into the specific formula of the logistic regression calculation as follows:

[0015] ;

[0016] In the formula, L is the result of the logistic regression calculation, that is, the reasonable probability, e is the natural base, y is the linear combination term of the logistic regression model, and specifically y can be set as:

[0017] ;

[0018] In the formula, is the bias term, and are the regression coefficients of the remaining capacity of the energy storage battery and the current power supply prediction value respectively, is the remaining capacity of the energy storage battery, is the current power supply prediction value;

[0019] Compare the reasonable probability with the preset reasonable threshold, specifically:

[0020] If the reasonable probability is greater than or equal to the reasonable threshold, the current dispatching is evaluated as reasonable, the current dispatching rationality result is obtained, and it is returned to the start stage. If the reasonable probability is less than the reasonable threshold, the current dispatching is evaluated as unreasonable, the current dispatching unreasonable result is obtained, and it is transmitted to the next layer.

[0021] In a preferred embodiment, the weather prediction information includes the solar radiation intensity received by the photovoltaic panel and the difference between the current temperature and the surface temperature of the photovoltaic panel, and the load information includes the short-term load prediction value;

[0022] The solar radiation intensity received by the photovoltaic panel is obtained by connecting satellite remote sensing to obtain the radiation intensity of the atmosphere and the earth's surface, multiplying it by the incident angle between the photovoltaic panel and the sun's rays, and the photoelectric conversion efficiency of the photovoltaic panel;

[0023] The surface temperature of the photovoltaic panel and the external environment temperature are obtained by temperature sensors installed on the surface of the photovoltaic panel and temperature sensors installed in the external environment, and the difference between the current temperature and the surface temperature of the photovoltaic panel is calculated by subtraction;

[0024] The historical load duration, the external power grid output value, and the temperature change value are obtained by sensors installed on the surface of the substation, and are substituted into the decision tree to obtain the short-term load prediction value.

[0025] In a preferred embodiment, the photovoltaic power generation satisfaction prediction analysis is carried out through a time series model. The time series model used is the ARIMAX model. The specific steps for carrying out the photovoltaic power generation satisfaction prediction analysis through the time series model are as follows:

[0026] Step A1, obtain the data for prediction;

[0027] Step A2, establish an ARIMAX model;

[0028] Step A3, use the maximum likelihood estimation (MLE) method to estimate the parameters of the ARIMAX model;

[0029] Step A4, verify the effect of the fitting model, and check the goodness of fit of the model through the residual analysis method;

[0030] Step A5, use the fitted model to predict the future dispatching adjustment plan, and take the maximum value of the prediction result as the maximum probability value of photovoltaic power generation satisfaction within the future time period.

[0031] In a preferred embodiment, the exogenous variables in the ARIMAX model include the solar radiation intensity received by the photovoltaic panel, the difference between the current temperature and the surface temperature of the photovoltaic panel, and the short-term load prediction value. The ARIMAX model formula is:

[0032] ;

[0033] In the formula, is the probability of photovoltaic power generation satisfaction at the current sampling time, is the photovoltaic power generation volume at the i-th lag order, is the constant term, is the autoregressive parameter of the i-th order, is the order of the autoregressive term, is the moving average parameter of the j-th order, is the order of the moving average term, is the white noise term lagged by j periods, is the white noise term, representing random error, is the exogenous variable is the coefficient of, is the lag order of the exogenous variable, is the exogenous variable lagged by k periods;

[0034] In step A3, , , , , are calculated and obtained by the maximum likelihood estimation method.

[0035] In a preferred embodiment, the values corresponding to all sampling times of the short-term prediction are substituted into the formula for calculation to obtain the maximum probability value satisfied by the photovoltaic power generation;

[0036] The maximum probability value of the photovoltaic power generation is compared with a preset satisfaction threshold to determine the dispatching adjustment plan. If it is greater than or equal to the satisfaction threshold, it indicates that the photovoltaic power generation system can meet the future load demand, and it is adjusted to preferentially use the photovoltaic power generation to reduce the dependence on the external power grid. Otherwise, the output state of the external power grid is maintained.

[0037] In a preferred embodiment, the characteristics of the dispatching adjustment plan include the solar radiation to the photovoltaic panel reception intensity, the difference between the current temperature and the photovoltaic panel surface temperature, the short-term load prediction value, and the maximum probability value of the photovoltaic power generation being full;

[0038] The characteristics of the historical prediction dispatching adjustment plan include the historical solar radiation to the photovoltaic panel reception intensity, the difference between the current temperature and the photovoltaic panel surface temperature, the short-term load prediction value, and the maximum probability value of the photovoltaic power generation being full;

[0039] The similarity between the dispatching adjustment plan and the historical prediction dispatching adjustment plan is calculated through similarity calculation, and they are sorted from large to small in sequence. Several historical prediction dispatching adjustment plans with the top similarity are selected as the similarity of the similar historical prediction dispatching adjustment plans;

[0040] The relative error is calculated based on the photovoltaic power generation duration determined by the historical prediction dispatching adjustment plan and its corresponding actual photovoltaic power generation duration to obtain the realization rate of the historical prediction dispatching adjustment plan.

[0041] In a preferred embodiment, the k-means clustering analysis is used to obtain the realization rate of the dispatching adjustment plan;

[0042] Take the similarity of the similar historical prediction scheduling adjustment schemes and the implementation rate of the historical prediction scheduling adjustment schemes as input data to generate multiple data points;

[0043] Then use K-means clustering to cluster the multiple data points into k clusters. The specific steps are as follows:

[0044] Step B1: Randomly select k initial clustering centers;

[0045] Step B2: Assign each data point to the nearest initial clustering center;

[0046] Step B3: Calculate the mean of the data points in each cluster and update the clustering centers;

[0047] Step B4: Repeat the steps of assignment and updating the clustering centers until the clustering centers no longer change.

[0048] In a preferred embodiment, take the average implementation rate of the specified cluster as the implementation rate of the current scheduling adjustment scheme, and set a simple scheduling strategy based on the implementation rate of the current scheduling adjustment scheme.

[0049] A power supply system of a photovoltaic-storage integrated substation includes a reasonable analysis module, a data acquisition module, a scheduling prediction module, and a scheme evaluation module;

[0050] The reasonable analysis module is used to collect the remaining capacity of the energy storage battery and the predicted value of the current power supply, perform normalization processing, perform logistic regression calculation on the normalized data, compare it with a preset reasonable threshold, determine the current scheduling unreasonable result, and send it to the data acquisition module;

[0051] The data acquisition module is used to obtain the current scheduling unreasonable result, collect weather prediction information and load information, obtain the received intensity of solar radiation on the photovoltaic panel, the temperature difference between the current temperature and the surface temperature of the photovoltaic panel, and the short-term load prediction value, and send it to the scheduling prediction module;

[0052] The scheduling prediction module is used to perform photovoltaic power generation satisfaction prediction analysis through a time series model according to the received intensity of solar radiation on the photovoltaic panel, the temperature difference between the current temperature and the surface temperature of the photovoltaic panel, and the short-term load prediction value, further refine and optimize the scheduling adjustment scheme, and send it to the scheme evaluation module;

[0053] The scheme evaluation module is used to calculate the similarity with the historical prediction scheduling adjustment scheme according to the scheduling adjustment scheme, and perform clustering analysis based on the implementation rate of the selected historical prediction scheduling adjustment scheme and the similarity of the corresponding similar historical prediction scheduling adjustment scheme after similarity calculation, determine the implementation rate of the scheduling adjustment scheme, and obtain a simple scheduling strategy.

[0054] The technical effects and advantages of the present invention:

[0055] 1. The present invention first performs a logical regression calculation based on the remaining capacity of the energy storage battery and the predicted current power supply, determines the rationality of the scheduling, and obtains the received intensity of solar radiation on the photovoltaic panel, the temperature difference between the current temperature and the surface temperature of the photovoltaic panel, and the short-term load prediction value according to the scheduling rationality. A time series model is established for prediction. According to the prediction results, a scheduling adjustment plan is determined to reduce over-scheduling, improve power supply continuity, and ensure the maximization of photovoltaic power generation resources.

[0056] 2. The present invention calculates the similarity with the historical predicted scheduling adjustment plan according to the scheduling adjustment plan, and performs clustering analysis based on the implementation rate of the historical predicted scheduling adjustment plan selected according to the similarity calculation and the similarity between the corresponding scheduling adjustment plan and the historical predicted scheduling adjustment plan to determine the implementation rate of the scheduling adjustment plan, obtain a simple scheduling strategy, improve the accuracy and rationality of scheduling decisions, reduce resource waste in scheduling adjustments, and determine the feasibility of the scheduling adjustment plan. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a flowchart of the power supply method of a photovoltaic-storage integrated substation according to the present invention.

[0058] Figure 2 It is a schematic diagram of the modules of a power supply system of a photovoltaic-storage integrated substation according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0059] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0060] Embodiment 1

[0061] Please refer to Figure 1 , a power supply method for a photovoltaic-storage integrated substation, and the specific operation process is as follows:

[0062] S1: Collect the remaining capacity of the energy storage battery and the predicted current power supply, perform normalization processing, perform a logical regression calculation on the normalized data to obtain a reasonable probability, and compare it with a preset reasonable threshold to determine the current unreasonable scheduling result;

[0063] For the set start collection condition, the power supply method of the current photovoltaic-storage integrated substation is external grid power supply. When the photovoltaic-storage integrated substation stores electricity, the remaining capacity of the current energy storage battery and the current continuous power supply are collected;

[0064] It should be noted that there can be more than one way to set the start collection condition. The experimenter can limit it according to the remaining capacity of the current battery or the ratio of light energy converted into electrical energy, etc., and no limitation is made here;

[0065] The remaining capacity of the energy storage battery refers to the current remaining power of the energy storage battery, which is expressed as the battery charge state or the remaining available electrical energy. Its acquisition logic is to obtain the remaining capacity of the energy storage battery by real-time monitoring of the battery voltage and current, combined with the discharge characteristics and capacity curve of the battery;

[0066] Specifically, the calculation formula for the remaining capacity of the energy storage battery is expressed as follows:

[0067] ;

[0068] In the formula, is the remaining capacity of the energy storage battery, is the remaining capacity of the energy storage battery at the previous sampling time, is the current current of the battery, is the sampling time interval, is the rated capacity of the battery, where t = 1, 2, 3... T, and t is the sampling time, and T is the total sampling time;

[0069] Specifically, no limitation is made on the sampling time interval and the number of sampling times, but the experimenter sets them according to the charge and discharge frequency of the energy storage battery or the scheduling frequency of the power supply equipment, which will not be elaborated here;

[0070] The predicted value of the current power supply refers to the predicted power supply demand of the system within the current total sampling time. Its acquisition logic is to use the historical load curve for weighted average calculation to obtain the predicted value of the current power supply;

[0071] Specifically, the selection of the historical load curve includes taking the load data, temperature, humidity, and sunshine within the current total sampling time and using linear regression. Combining the historical load data with the meteorological data to obtain the historical load curve, and performing weighted average calculation based on the load data corresponding to the sampling time on the curve to obtain the predicted value of the current power supply ;

[0072] Normalize the remaining capacity of the energy storage battery and the predicted value of the current power supply. All input variables will be converted to the same range to ensure the balanced contribution of each input to the model. Specifically, the normalization method is to normalize through the [0,1] interval, and the specific formula is expressed as:

[0073] ;

[0074] In the formula, is the remaining capacity of the energy storage battery, is the normalized remaining capacity of the energy storage battery, is the minimum value of the remaining capacity of the energy storage battery, is the maximum value of the remaining capacity of the energy storage battery;

[0075] Among them, the above formula is also used for normalization processing of the estimated value of the current power supply, which will not be elaborated here;

[0076] After normalization processing, the values of the remaining capacity of the energy storage battery and the estimated value of the current power supply are both within the range of [0, 1]. The system can compare and make decisions on a unified scale, thereby improving the accuracy and efficiency of scheduling;

[0077] Substitute the remaining capacity of the energy storage battery and the estimated value of the current power supply into the specific formula of logistic regression calculation as follows:

[0078] ;

[0079] In the formula, L is the result of logistic regression calculation, that is, the reasonable probability, e is the natural logarithm base, y is the linear combination term of the logistic regression model, and specifically y can be set as:

[0080] ;

[0081] In the formula, is the bias term, and are the regression coefficients of the remaining capacity of the energy storage battery and the estimated value of the current power supply respectively;

[0082] Compare the result of logistic regression calculation, that is, the reasonable probability, with the preset reasonable threshold. Specifically:

[0083] If the reasonable probability is greater than or equal to the reasonable threshold, the current scheduling is evaluated as reasonable, and the current scheduling rationality result is obtained and returned to the start stage. If the reasonable probability is less than the reasonable threshold, the current scheduling is evaluated as unreasonable, and the current scheduling unreasonable result is obtained and transmitted to the next layer;

[0084] Specifically, the preset reasonable threshold is set by the experimenters based on historical data analysis and the tolerance of the system, which will not be elaborated here;

[0085] Among them, the start stage means returning to the system initialization operation stage, which requires manual confirmation and then performing the operation of step S1. Otherwise, the system will not run;

[0086] Specifically, the next layer refers to the operation method of the next step or the reference to transmitting the current information to the next module, which will not be elaborated here;

[0087] S2: Obtain the current unreasonable scheduling results, collect weather prediction information and load information, and obtain the solar radiation intensity received by the photovoltaic panel, the temperature difference between the current ambient temperature and the surface temperature of the photovoltaic panel, and the short-term load prediction value;

[0088] Among them, the weather prediction information includes the solar radiation intensity received by the photovoltaic panel and the temperature difference between the current ambient temperature and the surface temperature of the photovoltaic panel, and the load information includes the short-term load prediction value;

[0089] Specifically, both the weather prediction information and the load information are used for short-term prediction. The specific setting of the short-term prediction can be to collect data once every hour or every three hours, which will not be elaborated here;

[0090] The solar radiation intensity received by the photovoltaic panel refers to the intensity of the solar radiation energy received on the surface of the photovoltaic panel, which is usually directly related to the output power of the photovoltaic panel. The acquisition logic is to multiply the radiation intensity of the atmosphere and the earth's surface obtained by connecting satellite remote sensing with the incident angle between the photovoltaic panel and the sun's rays and the photoelectric conversion efficiency of the photovoltaic panel to obtain the solar radiation intensity received by the photovoltaic panel;

[0091] Specifically, the calculation formula for obtaining the solar radiation intensity received by the photovoltaic panel is as follows:

[0092] ;

[0093] In the formula, is the solar radiation intensity received by the photovoltaic panel, is the solar radiation intensity, is the incident angle between the photovoltaic panel and the sun's rays, is the photoelectric conversion efficiency of the photovoltaic panel;

[0094] It should be noted that the connected satellite remote sensing refers to the solar radiation data provided by meteorological satellites, which includes the satellite remote sensing technology using sensors carried by satellites to observe the earth's surface to obtain information about solar radiation, surface coverage, etc. Among them, the incident angle between the photovoltaic panel and the sun's rays is obtained based on the latitude of the installation position of the photovoltaic panel, the installation angle, and the solar zenith angle, and the photoelectric conversion efficiency of the photovoltaic panel is determined by the technical specification of the photovoltaic panel or the results obtained by the experimenters in the laboratory;

[0095] The temperature difference between the current ambient temperature and the surface temperature of the photovoltaic panel refers to the difference between the ambient temperature and the surface temperature of the photovoltaic panel. The acquisition logic is to obtain the surface temperature of the photovoltaic panel and the external ambient temperature through the temperature sensors installed on the surface of the photovoltaic panel and the temperature sensors installed in the external environment, and perform subtraction calculation to obtain the temperature difference between the current ambient temperature and the surface temperature of the photovoltaic panel;

[0096] Specifically, the installation position of the temperature sensor is not limited. Instead, it is set by the experimenter according to the position of the light energy absorption element corresponding to the photovoltaic panel, which will not be elaborated here;

[0097] The short-term load prediction value refers to the prediction of the future short-term load demand. Its acquisition logic is to substitute the historical load duration, the output value of the external power grid, and the temperature change value into the decision tree to obtain the short-term load prediction value;

[0098] Among them, the historical load duration, the output value of the external power grid, and the temperature change value are all obtained through the sensors installed on the surface of the substation;

[0099] Among them, the decision tree refers to a machine learning method. In load prediction, the decision tree is usually used for regression tasks, that is, to predict a continuous numerical value according to input features (such as historical load, seasonal factors, temperature, etc.), which is the short-term load prediction value. Specifically, as an existing technology, the steps for obtaining the short-term load prediction value will not be elaborated;

[0100] S3: According to the solar radiation intensity received by the photovoltaic panel, the difference between the current temperature and the surface temperature of the photovoltaic panel, and the short-term load prediction value, perform photovoltaic power generation satisfaction prediction analysis through a time series model, and further refine and optimize the scheduling adjustment plan;

[0101] As can be seen from the above, the solar radiation intensity received by the photovoltaic panel, the difference between the current temperature and the surface temperature of the photovoltaic panel, and the short-term load prediction value are continuous numerical values within a period of time. Therefore, for the time series model, the ARIMAX model is adopted in this embodiment. Refining and optimizing the current scheduling adjustment plan means that in the case of determining that the current scheduling is unreasonable, through further analysis and adjustment, the power supply distribution plan is made more scientific and reasonable, so as to realize the control of photovoltaic power supply;

[0102] Furthermore, the specific steps of performing photovoltaic power generation satisfaction prediction analysis through the time series model are as follows:

[0103] Step A1, obtain the data for prediction;

[0104] Step A2, establish an ARIMAX model;

[0105] Step A3, use the maximum likelihood estimation (MLE) method to estimate the parameters of the ARIMAX model;

[0106] Step A4, verify the effect of the fitted model, and check the goodness of fit of the model through the residual analysis method;

[0107] Step A5, use the fitted model to predict the future scheduling adjustment plan, and take the maximum value of the prediction result as the maximum probability value of photovoltaic power generation satisfaction within the future time period;

[0108] The photovoltaic power generation satisfaction probability refers to the historical photovoltaic power generation satisfaction probability, and its data serves as the main variable of the time series;

[0109] Furthermore, the basic form of the ARIMAX model is:

[0110] ;

[0111] In the formula, is the photovoltaic power generation satisfaction probability at the current sampling time, is the photovoltaic power generation at the i-th lag order, is the constant term, is the i-th order autoregressive parameter, is the order of the autoregressive term, is the j-th order moving average parameter, is the order of the moving average term, is the white noise term at the j-th lag period, is the white noise term, representing the random error, is the exogenous variable of the coefficient, is the lag order of the exogenous variable, is the exogenous variable at the k-th lag period;

[0112] It should be noted that the exogenous variable part can integrate the influence of other relevant variables (including the solar radiation received by the photovoltaic panel, the difference between the current temperature and the surface temperature of the photovoltaic panel, and the short-term load prediction value) on the photovoltaic power generation satisfaction probability;

[0113] It should be noted that in step A3, , , , , are calculated and obtained by the maximum likelihood estimation (MLE) method. The specific steps are as follows:

[0114] The error term follows a normal distribution , then the likelihood function is:

[0115] ;

[0116] Take the logarithm of the likelihood function to obtain the log-likelihood function:

[0117] ;

[0118] By maximizing the log-likelihood function, the parameter estimates , , , , ;

[0119] Substitute the values corresponding to all sampling times of the short-term prediction into the formula for calculation to obtain the maximum probability value satisfied by the photovoltaic power generation.

[0120] Compare the maximum probability value of the photovoltaic power generation with a preset satisfaction threshold to determine the dispatching adjustment plan. If it is greater than or equal to the satisfaction threshold, it indicates that the photovoltaic power generation system can meet the future load demand, and adjust to give priority to using photovoltaic power generation to reduce the dependence on the external power grid. Otherwise, maintain the output state of the external power grid.

[0121] Among them, the satisfaction threshold is set by the experimenter according to the historical photovoltaic power generation duration and probability, which will not be elaborated here.

[0122] The present invention first performs a logical regression calculation according to the remaining capacity of the energy storage battery and the predicted value of the current power supply to determine the rationality of the dispatching, and obtains the received intensity of solar radiation to the photovoltaic panel, the difference between the current temperature and the surface temperature of the photovoltaic panel, and the short-term load prediction value according to the dispatching rationality. Then, a time series model is established for prediction. According to the prediction results, the dispatching adjustment plan is determined, reducing excessive dispatching, improving power supply continuity, and ensuring the maximization of photovoltaic power generation resources.

[0123] Embodiment 2

[0124] In Embodiment 1 of the present invention, it is mainly illustrated by examples that a logical regression calculation is first performed according to the remaining capacity of the energy storage battery and the predicted value of the current power supply to determine the rationality of the dispatching, and the received intensity of solar radiation to the photovoltaic panel, the difference between the current temperature and the surface temperature of the photovoltaic panel, and the short-term load prediction value are obtained according to the dispatching rationality. Then, a time series model is established for prediction. According to the prediction results, the dispatching adjustment plan is determined. However, in Embodiment 1, only starting from the adjustment plan, the feasibility of the plan is not analyzed. Obviously, since the probability of the photovoltaic power generation being satisfied is predicted after predicting the load, this will lead to the accumulation of data prediction errors and over-reliance on historical data and model assumptions, thus reducing the feasibility of the dispatching adjustment plan. For the above problems, Embodiment 2 of the present invention is further refined.

[0125] S4: Calculate the similarity with the historical predicted dispatching adjustment plan according to the dispatching adjustment plan, and select the realization rate of the historical predicted dispatching adjustment plan and the similarity of the corresponding similar historical predicted dispatching adjustment plan based on the similarity calculation.

[0126] Specifically, the dispatching adjustment plan includes the maximum probability value of the photovoltaic power generation being full.

[0127] Among them, the features of the scheduling adjustment plan include the solar radiation received by the photovoltaic panel, the difference between the current temperature and the surface temperature of the photovoltaic panel, the short-term load prediction value, and the maximum probability value of full photovoltaic power generation described in the above-mentioned Embodiment 1. The acquisition logic of specific data will not be elaborated here;

[0128] The historical predicted scheduling adjustment plan refers to the scheduling adjustment plan predicted by a time series model based on the solar radiation received by the photovoltaic panel, the difference between the current temperature and the surface temperature of the photovoltaic panel, and the short-term load prediction value. Its features include the solar radiation received by the photovoltaic panel, the difference between the current temperature and the surface temperature of the photovoltaic panel, the short-term load prediction value, and the maximum probability value of full photovoltaic power generation;

[0129] As can be seen from the above, the number of features of the scheduling adjustment plan is the same as that of the historical predicted scheduling adjustment plan, and the features correspond to each other;

[0130] Vectorize the features of the scheduling adjustment plan and the historical predicted scheduling adjustment plan. The specific vectorization representation is as follows:

[0131] and ;

[0132] In the formula, the solar radiation received by the photovoltaic panel, the difference between the current temperature and the surface temperature of the photovoltaic panel, the short-term load prediction value, and the maximum probability value of full photovoltaic power generation of the scheduling adjustment plan and the historical predicted scheduling adjustment plan are respectively expressed as and , and , and and and ; where A is the vectorized representation of the scheduling adjustment plan, and B is the vectorized representation of the historical predicted scheduling adjustment plan;

[0133] Substitute the vectorized representation of the above data into the similarity formula. In this embodiment, the similarity between the scheduling adjustment plan and the historical predicted scheduling adjustment plan is expressed based on cosine similarity. The specific formula is as follows:

[0134] ;

[0135] In the formula, is the similarity between the scheduling adjustment plan and the historical predicted scheduling adjustment plan, is the dot product of the two vectors, and respectively represent the norms of vectors A and B; and i is the similarity comparison with the historical predicted scheduling adjustment plan for the i-th time; i = 1, 2, 3... n, and i is the number of comparison times, and n is the total number of comparison times;

[0136] Specifically, the calculation formula for the dot product of two vectors is:

[0137] ;

[0138] Specifically, and The calculation formulas of

[0139] and

[0140] Sort the similarities between the scheduling adjustment plan obtained through similarity calculation and the historical predicted scheduling adjustment plan from largest to smallest to obtain a data set as , and select several historical predicted scheduling adjustment plans with higher similarity as the similarities of the similar historical predicted scheduling adjustment plans;

[0141] For example, select the top ten historical predicted scheduling adjustment plans with higher similarity as the similar historical predicted scheduling adjustment plans;

[0142] It should be noted that the number of several historical predicted scheduling adjustment plans selected with higher similarity can be set by the experimenter, or the corresponding historical predicted scheduling adjustment plans above the specified similarity can be obtained by judging according to the set similarity threshold. The specific number selected and the implementation method are not limited, but are implemented by the experimenter according to the specific plan and will not be elaborated here;

[0143] Obtain the implementation rate corresponding to the similar historical predicted scheduling adjustment plan, that is, the implementation rate of the historical predicted scheduling adjustment plan refers to the final implementation degree of the historical predicted scheduling adjustment plan. Its acquisition logic is to calculate using the relative error between the photovoltaic power generation duration determined by the historical predicted scheduling adjustment plan and its corresponding actual photovoltaic power generation duration to obtain the implementation rate of the historical predicted scheduling adjustment plan;

[0144] S5: Obtain the similarities of the similar historical predicted scheduling adjustment plans and the implementation rate of the historical predicted scheduling adjustment plan for clustering analysis, determine the implementation rate of the scheduling adjustment plan, and obtain a simple scheduling strategy;

[0145] It should be noted that in this embodiment, k-means clustering analysis is used to obtain the implementation rate of the scheduling adjustment plan;

[0146] Use the similarities of the similar historical predicted scheduling adjustment plans and the implementation rate of the historical predicted scheduling adjustment plan as input data to generate multiple data points, and each data point consists of a pair of values (similarity and implementation rate);

[0147] Then perform data normalization processing on the input data. The specific normalization processing is prior art and is not limited here;

[0148] Use K-means clustering to cluster the data into k clusters. The specific steps of K-means clustering are as follows:

[0149] Step B1: Randomly select k initial cluster centers;

[0150] Step B2: Assign each data point to the nearest initial cluster center;

[0151] Step B3: Calculate the mean of the data points in each cluster and update the cluster centers;

[0152] Step B4: Repeat the steps of assignment and updating of cluster centers until the cluster centers no longer change;

[0153] Specifically, before starting the K-means clustering algorithm, determine the optimal number of clusters k through the elbow method to decide the number of clusters for the data structure;

[0154] When running K-means clustering, try different values of k, calculate the sum of squared errors of clustering for each k value, and then plot the graph of SSE changing with k. Find the "elbow" in the graph, that is, the inflection point, which is the point where the rate of decrease of the sum of squared errors slows down significantly. The k value corresponding to this point is usually the optimal number of clusters;

[0155] Step B5: Obtain the implementation rate of each cluster through K-means clustering analysis, and the cluster center represents the average implementation rate of this category. Then substitute the similarity of the current scheduling adjustment plan into the clustering model to determine the cluster it belongs to;

[0156] Step B6: The current scheduling adjustment plan is classified into the specified cluster according to the similarity value with the historical predicted scheduling adjustment plan, and the result is output;

[0157] Specifically, take the average implementation rate of the specified cluster as the implementation rate of the current scheduling adjustment plan;

[0158] Set a simple scheduling strategy based on the implementation rate of the current scheduling adjustment plan;

[0159] For example, when the implementation rate of the scheduling adjustment plan is 80%, keep the current photovoltaic power generation strategy without subsequent adjustment. When the implementation rate of the scheduling adjustment plan is 40%, change it to external power generation and make only one adjustment;

[0160] The specific simple scheduling strategy is not limited, but is set by the experimenters according to the size of the implementation rate of the scheduling adjustment plan and the number of times of changing the power supply mode in the historical predicted scheduling adjustment plan, which will not be elaborated here;

[0161] The present invention calculates the similarity with the historical predicted scheduling adjustment plan according to the scheduling adjustment plan, and performs clustering analysis based on the implementation rate of the selected historical predicted scheduling adjustment plan and the similarity between the corresponding scheduling adjustment plan and the historical predicted scheduling adjustment plan after the similarity calculation, determines the implementation rate of the scheduling adjustment plan, obtains a simple scheduling strategy, improves the accuracy and rationality of scheduling decisions, reduces resource waste in scheduling adjustments, and determines the feasibility of the scheduling adjustment plan.

[0162] Embodiment 3

[0163] Please refer to Figure 2 , a power supply system of a photovoltaic-storage integrated substation, including a reasonable analysis module, a data acquisition module, a scheduling prediction module, and a scheme evaluation module;

[0164] The reasonable analysis module is used to collect the remaining capacity of the energy storage battery and the predicted value of the current power supply, perform normalization processing, perform logistic regression calculation on the normalized data, compare it with a preset reasonable threshold, determine the current unreasonable scheduling result, and send it to the data acquisition module;

[0165] The data acquisition module is used to obtain the current unreasonable scheduling result, collect weather prediction information and load information, obtain the received intensity of solar radiation on the photovoltaic panel, the temperature difference between the current temperature and the surface temperature of the photovoltaic panel, and the short-term load prediction value, and send it to the scheduling prediction module;

[0166] The scheduling prediction module is used to perform photovoltaic power generation satisfaction prediction analysis through a time series model according to the received intensity of solar radiation on the photovoltaic panel, the temperature difference between the current temperature and the surface temperature of the photovoltaic panel, and the short-term load prediction value, further refine and optimize the scheduling adjustment plan, and send it to the scheme evaluation module;

[0167] The scheme evaluation module is used to calculate the similarity with the historical predicted scheduling adjustment plan according to the scheduling adjustment plan, and perform clustering analysis based on the implementation rate of the selected historical predicted scheduling adjustment plan and the similarity of the corresponding similar historical predicted scheduling adjustment plan after the similarity calculation, determine the implementation rate of the scheduling adjustment plan, and obtain a simple scheduling strategy.

[0168] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to obtain a formula that is closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0169] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more collections of available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0170] It should be understood that in various embodiments of the present application, the order numbers of the above processes do not indicate the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0171] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0172] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0173] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0174] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0175] In addition, in each embodiment of the present application, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0176] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0177] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A power supply method for a photovoltaic and energy storage fusion distribution station, characterized in that: include: S1: Collect the remaining capacity of the energy storage battery and the estimated value of the current power supply, and perform normalization processing, perform logistic regression calculation on the normalized data to obtain a reasonable probability, and compare it with the preset reasonable threshold to determine the unreasonable result of the current scheduling; S2: Obtain the unreasonable results of the current scheduling, collect weather forecast information and load information, obtain the intensity of solar radiation received by the photovoltaic panel, the difference between the current temperature and the surface temperature of the photovoltaic panel, and the short-term load forecast value; S3: Based on the intensity of solar radiation received by the photovoltaic panels, the difference between the current air temperature and the surface temperature of the photovoltaic panels, and the short-term load forecast value, the photovoltaic power generation is predicted and analyzed through the time series model to further refine and optimize the scheduling adjustment plan; S4: Based on the scheduling adjustment plan, similarity calculation is performed with the historical prediction scheduling adjustment plan, and based on the similarity calculation, the realization rate of the selected historical prediction scheduling adjustment plan and the similarity of the corresponding similar historical prediction scheduling adjustment plan are calculated; S5: Obtain the similarity of similar historical prediction scheduling adjustment plans and the realization rate of historical prediction scheduling adjustment plans for cluster analysis, determine the realization rate of the scheduling adjustment plan, and obtain a simple scheduling strategy; By real-time monitoring of the battery voltage and current, combined with the battery discharge characteristics and capacity curve, the remaining amount of energy storage battery can be obtained; By taking load data within the current total sampling time and using linear regression with temperature, humidity and sunshine, and combining historical load data with meteorological data, a historical load curve is obtained, and a weighted average calculation is performed based on the load data at the corresponding sampling time on the curve to obtain the current power supply estimate; Substituting the estimated value of the remaining energy storage battery capacity and the current power supply into the logistic regression calculation, the specific formula is expressed as follows: In the formula, L is the result of logistic regression calculation, that is, reasonable probability, e is the natural base, and y is the linear combination term of the logistic regression model. The specific setting of y is: y=β0+β1·S(t)+β2·Z(t); Where β0 is the bias term, β1 and β2 are the regression coefficients of the energy storage battery free capacity and the current power supply estimated value, S(t) is the energy storage battery free capacity, and Z(t) is the current power supply estimated value; Compare the reasonable probability with the preset reasonable threshold, specifically: If the reasonable probability is greater than or equal to the reasonable threshold, the current schedule is evaluated as reasonable, the current schedule rationality result is obtained, and it returns to the starting stage; if the reasonable probability is less than the reasonable threshold, the current schedule is evaluated as unreasonable, the current schedule unreasonable result is obtained, and it is transmitted to the next layer; Weather forecast information includes the intensity of solar radiation received by the photovoltaic panel and the difference between the current air temperature and the surface temperature of the photovoltaic panel. Load information includes short-term load forecast values. The radiation intensity of the atmosphere and the ground is obtained by connecting to satellite remote sensing and multiplying it with the incident angle between the photovoltaic panel and the sun's rays and the photoelectric conversion efficiency of the photovoltaic panel to obtain the intensity of solar radiation received by the photovoltaic panel; The surface temperature of the photovoltaic panel and the external environment temperature are obtained by the temperature sensor installed on the surface of the photovoltaic panel and the temperature sensor installed in the external environment, and the difference between the current air temperature and the surface temperature of the photovoltaic panel is obtained by subtracting the temperature. Through the sensors installed on the surface of the distribution station, the historical load duration, external power grid output value, and temperature change value are obtained and substituted into the decision tree to obtain the short-term load forecast value.

2. The power supply method of the photovoltaic energy storage fusion distribution station according to claim 1 is characterized in that: The photovoltaic power generation is predicted and analyzed by the time series model. The time series model adopts the ARIMAX model. The specific steps of the photovoltaic power generation prediction and analysis by the time series model are as follows: Step A1, obtaining prediction data; Step A2, establishing an ARIMAX model; Step A3, using the maximum likelihood estimation (MLE) method to estimate the ARIMAX model parameters; Step A4, verify the effect of the fitting model and check the goodness of fit of the model by residual analysis; Step A5, using the fitted model to predict the future scheduling adjustment plan, and taking the maximum value of the prediction result as the maximum probability value of photovoltaic power generation in the future time period.

3. The power supply method of the photovoltaic energy storage fusion distribution station according to claim 2 is characterized in that: The exogenous variables in the ARIMAX model include the intensity of solar radiation received by the photovoltaic panel, the difference between the current air temperature and the surface temperature of the photovoltaic panel, and the short-term load forecast value. The ARIMAX model formula is: In the formula, y t is the probability of photovoltaic power generation meeting the current sampling time, y t-i is the photovoltaic power generation with lag i, α is a constant term, is the i-th order autoregressive parameter, p is the order of the autoregressive term, θ j is the jth order moving average parameter, q is the order of the moving average term, ∈ t-j is the white noise term with lag j, ∈ t is a white noise term, representing random error, β k is an exogenous variable X t-k The coefficient of m is the lag order of the exogenous variable, X t-k is an exogenous variable lagged by k periods; In step A3, α, θ1, β1, and β2 are calculated using the maximum likelihood estimation method.

4. The power supply method of the photovoltaic energy storage fusion distribution station according to claim 3 is characterized in that: Substitute all sampling times corresponding to the short-term forecast into the formula to calculate and obtain the maximum probability value of photovoltaic power generation; The maximum probability value of photovoltaic power generation is compared with the preset satisfaction threshold to determine the scheduling adjustment plan. If it is greater than or equal to the satisfaction threshold, it means that the photovoltaic power generation system can meet future load demand and is adjusted to give priority to photovoltaic power generation to reduce dependence on the external power grid. Otherwise, the output status of the external power grid is maintained.

5. The power supply method of the photovoltaic energy storage fusion distribution station according to claim 4 is characterized in that: The characteristics of the scheduling adjustment plan include the intensity of solar radiation received by the photovoltaic panel, the difference between the current air temperature and the surface temperature of the photovoltaic panel, the short-term load forecast value, and the maximum probability value of photovoltaic power generation; The characteristics of the historical prediction scheduling adjustment plan include the historical basis of the intensity of solar radiation received by the photovoltaic panel, the difference between the current air temperature and the surface temperature of the photovoltaic panel, the short-term load forecast value, and the maximum probability value of photovoltaic power generation; The similarity between the scheduling adjustment plan and the historical prediction scheduling adjustment plan is obtained by similarity calculation, and they are sorted in order from large to small, and several historical prediction scheduling adjustment plans with the highest similarity are selected as the similarity of similar historical prediction scheduling adjustment plans; The photovoltaic power generation duration determined according to the historical forecast scheduling adjustment plan and its corresponding actual photovoltaic power generation duration are calculated using relative error to obtain the realization rate of the historical forecast scheduling adjustment plan.

6. The power supply method of the photovoltaic energy storage fusion distribution station according to claim 5 is characterized in that: Use k-means cluster analysis to obtain the scheduling adjustment plan realization rate; The similarity of similar historical forecast scheduling adjustment plans and the realization rate of historical forecast scheduling adjustment plans are used as input data to generate multiple data points; Then use K-means clustering to cluster multiple data points into k clusters. The specific steps are: Step B1: Randomly select k initial cluster centers; Step B2: Assign each data point to the nearest initial cluster center; Step B3: Calculate the mean of the data points in each cluster and update the cluster center; Step B4: Repeat the steps of assigning and updating cluster centers until the cluster centers no longer change.

7. The power supply method of the photovoltaic energy storage fusion distribution station according to claim 6 is characterized in that: The average realization rate of the specified cluster is used as the realization rate of the current scheduling adjustment plan, and a simple scheduling strategy is set according to the realization rate of the current scheduling adjustment plan.

8. A power supply system for a photovoltaic-storage fusion distribution station, used to implement the power supply method for a photovoltaic-storage fusion distribution station according to any one of claims 1 to 7, characterized in that: It includes rational analysis module, data collection module, scheduling prediction module and scheme evaluation module; The rational analysis module is used to collect the remaining capacity of the energy storage battery and the estimated current power supply, and perform normalization processing, perform logistic regression calculation on the normalized data, and compare it with the preset reasonable threshold value to determine the unreasonable result of the current scheduling and send it to the data acquisition module; The data acquisition module is used to obtain unreasonable results of the current scheduling, collect weather forecast information and load information, obtain the intensity of solar radiation received by the photovoltaic panel, the difference between the current temperature and the surface temperature of the photovoltaic panel, and the short-term load forecast value, and send them to the scheduling prediction module; The scheduling prediction module is used to perform prediction analysis on photovoltaic power generation based on the intensity of solar radiation received by the photovoltaic panels, the difference between the current temperature and the surface temperature of the photovoltaic panels, and the short-term load forecast value, and further refine and optimize the scheduling adjustment plan, and send it to the plan evaluation module; The scheme evaluation module is used to calculate the similarity between the scheduling adjustment scheme and the historical prediction scheduling adjustment scheme, and to perform cluster analysis based on the realization rate of the selected historical prediction scheduling adjustment scheme and the similarity of the corresponding similar historical prediction scheduling adjustment schemes, so as to determine the realization rate of the scheduling adjustment scheme and obtain a simple scheduling strategy.

Images