Day-ahead source-load cooperative operation optimization method for optical storage charging station

Abstract

The invention discloses a day-ahead source-load collaborative operation optimization method for an optical storage charging station, and mainly solves the problems that the real-time performance of a source-load collaborative operation optimization strategy is insufficient, the source-load collaborative operation optimization strategy is not matched with applicable working conditions, and an optimal solution is not easy to find. The method comprises the following steps: S1, clustering historical meteorological data by taking irradiance as a clustering index, and clustering historical traffic flow data by taking daily traffic flow as a clustering index; s2, acquiring characteristic parameters of a photovoltaic power generation prediction model and characteristic parameters of a charging pile load prediction model; s3, training to obtain a photovoltaic power generation deep learning prediction model and a charging pile load deep learning prediction model; s4, establishing a source load operation optimization model by taking the lowest operation cost and the lowest carbon emission as optimization objectives; and S5, obtaining an energy storage charging and discharging operation strategy and an electricity purchasing strategy with the lowest source-load cooperative operation cost. The invention provides a day-ahead operation strategy which can timely guide an energy storage charging and discharging strategy and an electricity purchasing strategy of the next day and accords with an actual operation scene.

Description

Technical Field

[0001] The present invention belongs to the technical field of the operation of photovoltaic-storage charging stations, and specifically relates to a method for optimizing the day-ahead source-load collaborative operation of a photovoltaic-storage charging station. Background Art

[0002] The optimization of source-load collaborative operation is of great significance for improving the comprehensive operation efficiency of the photovoltaic-storage charging system, reducing the operation cost, and promoting the consumption of renewable energy.

[0003] Chinese Patent with Application No. 202010683118.4 discloses a method and device for collaborative optimization scheduling of a photovoltaic-storage charging and swapping integrated charging station based on PSO. The scheme is roughly as follows: 1) Obtain data: Obtain the data of the power generation module, energy storage module, and charging module of the photovoltaic-storage charging station. 2) Establish a collaborative scheduling strategy model based on the obtained information, with the lowest operation cost as the optimization goal and the power balance constraint, energy storage charge and discharge power, battery SOC constraint, charger power constraint, and grid input power as the constraints. 3) Solve the model based on the PSO particle swarm algorithm to find the optimal operation strategy plan. 4) Formulate the grid power input and energy storage output plan of the photovoltaic-storage charging station according to the obtained optimal solution. The main disadvantages of the existing technology are the lack of the ability to provide real-time optimization strategies and the inapplicability to source-load scenarios such as different weather conditions and holiday conditions. The specific technical disadvantages are as follows:

[0004] 1) The existing technology forms a fixed energy consumption scenario through the historical data of the power generation module, energy storage module, and charging module, and further proposes an operation optimization strategy. The flexibility of the operation optimization strategy proposed by this method is relatively low, and there is a situation of mismatch with the actual working conditions, making it difficult to effectively reduce the system operation cost.

[0005] 2) The photovoltaic power supply and vehicle charging load of the photovoltaic-storage charging station are greatly affected by meteorological factors and holidays, and meteorological and holiday factors need to be fully considered.

[0006] 3) The traditional PSO algorithm is prone to falling into local optimal solutions, and the differences in the results of each operation are relatively large. Summary of the Invention

[0007] The purpose of the present invention is to provide a method for optimizing the day-ahead source-load collaborative operation of a photovoltaic-storage charging station, mainly solving the problems of insufficient real-time performance of the source-load collaborative operation optimization strategy, mismatch with applicable working conditions, and difficulty in finding the optimal solution.

[0008] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0009] A method for optimizing the day-ahead source-load collaborative operation of a photovoltaic-storage charging station includes the following steps:

[0010] S1. Based on the FCM fuzzy clustering algorithm, cluster the historical meteorological data with irradiance as the clustering index, and cluster the historical traffic flow data with the daily traffic flow size as the clustering index;

[0011] S2. Based on the Spearman correlation coefficient, screen the factors affecting the power generation power and the charging pile load power, and obtain the characteristic parameters of the photovoltaic power generation prediction model and the characteristic parameters of the charging pile load prediction model;

[0012] S3. Use the LSTM deep learning model to train and obtain the photovoltaic power generation deep learning prediction model and the charging pile load deep learning prediction model;

[0013] S4. Based on the photovoltaic power generation deep learning prediction model and the charging pile load deep learning prediction model, establish a source-load operation optimization model with the lowest operating cost and the lowest carbon emissions as the optimization objectives;

[0014] S5. Determine the boundary conditions of the source-load operation optimization model, and use the hippopotamus optimization algorithm to solve the source-load operation optimization model to obtain the energy storage charge-discharge operation strategy and power purchase strategy with the lowest source-load collaborative operation cost.

[0015] Furthermore, in the step S1, the historical meteorological data is clustered into three weather conditions: sunny, cloudy, and rainy, and the historical traffic flow data is clustered into holidays and non-holidays.

[0016] Furthermore, in the step S1, the objective function of the FCM fuzzy clustering algorithm is:

[0017]

[0018] In the formula, u ij is the membership degree of the sample point x i and the clustering center v j ; m is the fuzzy index (m>1), which determines the fuzziness of the clustering; d ij is the distance between the sample point x i and the clustering center v j ; k is the number of clusters; n is the number of samples;

[0019] The constraint condition of the FCM fuzzy clustering algorithm objective function is: ensure that the sum of the membership degrees of each data point to all clusters is 1; that is:

[0020]

[0021] In the formula, the value range of u ij is [0,1].

[0022] Furthermore, in the step S2, the expressions for obtaining the characteristic parameters of the photovoltaic power generation prediction model and the characteristic parameters of the charging pile load prediction model are:

[0023]

[0024] Wherein, X i is the influencing factor magnitude of the i-th sample point; Y i is the power generation power magnitude or the charging pile load power magnitude of the i-th sample point; X and Y are respectively the average values of the influencing factor and the power generation power among N sample points.

[0025] Furthermore, in the step S4, the operating cost includes the electricity purchase cost of the system, the unit operating cost of the photovoltaic system, the unit operating cost of the energy storage system, and the unit operating cost of the charging pile, and its calculation formula is as follows:

[0026]

[0027] Wherein, is the price of purchasing unit electricity from the Internet, yuan / kW; is the power transmitted by the power grid, kW; is the unit operating cost of the photovoltaic system, yuan / kW; is the photovoltaic power generation; is the energy storage operation cost, yuan / kW; is the energy storage charge and discharge power, kW; is the unit operating cost of the charging pile, yuan / kW; is the charging pile power, kW;

[0028] The carbon emission is caused by the consumption of mains electricity and the loss of the energy storage, and its calculation formula is as follows:

[0029]

[0030] Wherein, is the carbon emission generated by the power grid electricity purchase, t; E is the carbon emission coefficient of the local power grid, t / kW; P loss is the energy storage charge and discharge loss power, kW.

[0031] Furthermore, in the step S5, the boundary conditions of the source-load operation optimization model include:

[0032] Power balance constraint, and its expression is:

[0033]

[0034] Wherein, is the electricity load of the charging pile, kW.h; is the energy storage charge and discharge amount, kW.h, when it is greater than 0, it is discharging, and when it is less than 0, it is charging;

[0035] Energy storage state of charge constraint, and its expression is:

[0036]

[0037] In the formula, is the remaining power of the energy storage at time t + 1; γ is the self-discharge rate of the energy storage battery, taking 0.001; and are the charging and discharging efficiencies of the energy storage respectively; E b is the rated capacity of the energy storage system; is the discharge amount of the system at time t.

[0038] Furthermore, in the step S5, the specific process of the hippopotamus optimization algorithm is as follows:

[0039] S51, population initialization:

[0040] X ij = lb j + r·(ub j - lb j )

[0041] where X ij represents the position information of the i-th hippopotamus under the j-th decision variable; r is a random number in the range of 0 to 1, used to introduce randomness; lb j and ub j represent the lower and upper limits of the j-th decision variable respectively; N represents the number of hippopotamuses in the population; m represents the number of decision variables in the problem;

[0042] S52, update the position of the hippopotamus in the river or pond:

[0043] X P1 (i,:) = X(i,:) + r·(X best - I1·X(i,:))

[0044] where X P1 (i,:) represents the position vector of the i-th hippopotamus after the first position update; X(i,:) represents the current position vector of the i-th hippopotamus; X best represents the position vector of the optimal hippopotamus in the current population; I1 is a random integer, usually taking 1 or 2, used to introduce randomness in the position update;

[0045] S53, hippopotamus defends against predators:

[0046]

[0047] Among them, p represents the position of the predator; distancep represents the distance between the hippopotamus and the predator; b, c, d, and l are randomly generated parameters; RL(i,:) represents the random vector generated by the Levy flight of the i-th hippopotamus;

[0048] S54, Hippopotamus escaping from the predator:

[0049] X P2 (i,:) = X(i,:) + A·(X best -I2·MeanGroup)

[0050] Among them, I2 is a random integer, usually taking 1 or 2, which is used to introduce randomness in position update; MeanGroup represents the average position of a randomly selected group of hippopotamuses; A is a randomly generated parameter.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] (1) The present invention classifies the source-load scenarios according to the meteorological factors affecting the photovoltaic power generation side, as well as the meteorological factors and holiday factors affecting the charging pile load side, and can give corresponding operation strategies for holiday working conditions, non-holiday working conditions, as well as sunny, cloudy, and rainy working conditions.

[0053] (2) The present invention provides a day-ahead operation strategy, which can timely guide the energy storage charge and discharge strategy and power purchase strategy for the next day, and conforms to the actual operation scenario.

[0054] (3) The optimization algorithm adopted by the present invention is the hippopotamus algorithm, which has many advantages such as fast convergence, high-precision solution, good global search ability, and avoiding falling into local optima compared with the traditional PSO algorithm when searching for the optimal solution. Brief Description of the Drawings

[0055] Figure 1 It is a flowchart of the method of the present invention.

[0056] Figure 2 It is a schematic diagram of the clustering of photovoltaic power generation working conditions in the embodiment of the present invention.

[0057] Figure 3 It is a schematic diagram of the clustering of charging pile load working conditions in the embodiment of the present invention.

[0058] Figure 4 It is a schematic diagram of the screening of characteristic parameters of the photovoltaic power generation prediction model in the embodiment of the present invention.

[0059] Figure 5 It is a schematic diagram of the screening of characteristic parameters of the charging pile load prediction model in the embodiment of the present invention.

[0060] Figure 6Schematic diagram for generating the deep learning prediction model of photovoltaic power generation in the embodiments of the present invention.

[0061] Figure 7 Schematic diagram for generating the deep learning prediction model of the charging pile load in the embodiments of the present invention.

[0062] Figure 8 Comparison diagram of the system operation strategies after optimization by the optimization method of the present invention. Detailed implementation manners

[0063] The present invention will be further described below in conjunction with the accompanying drawings and embodiments. The implementation manners of the present invention include but are not limited to the following embodiments.

[0064] Embodiment

[0065] As Figure 1 shown, a day-ahead source-load coordinated operation optimization method for a photovoltaic energy storage charging station disclosed by the present invention realizes a day-ahead source-load coordinated operation strategy applicable to different weather and holiday conditions. This technology plays a key role in improving the operation and management efficiency of photovoltaic energy storage charging stations, ensuring grid stability, enhancing the competitiveness of the power market, and promoting the consumption of renewable energy. The core content of this technology includes energy consumption scenario clustering, construction of a deep learning model for source-load prediction, and construction of a source-load operation optimization model.

[0066] The power generation power of a photovoltaic power generation system varies significantly under different weather conditions, and the load of a charging pile also varies significantly between holidays and non-holidays. To improve the prediction accuracy for different conditions, based on the FCM fuzzy clustering algorithm, with irradiance as the clustering index, historical meteorological data is clustered into three weather conditions: sunny, cloudy, and rainy. And with the daily traffic flow size as the clustering index, historical traffic flow data is clustered into holidays and non-holidays. The implementation path of the scenario clustering analysis of the photovoltaic energy storage charging station is as Figure 2 、 Figure 3 shown.

[0067] The FCM fuzzy clustering algorithm allows data points to belong to multiple clusters with different membership degrees. The core idea is to determine the relationship between data points and cluster centers by iteratively optimizing the objective function, so as to achieve the purpose of clustering. The objective and constraint functions in its iterative optimization process are as follows:

[0068] (1) Objective function:

[0069]

[0070] In the formula, u ij is the membership degree of the sample point x i and the cluster center v j ; m is the fuzzy index (m>1), which determines the fuzziness of the clustering; d ij is the sample point xi The distance to the clustering center v j ; k is the number of clusters; n is the number of samples.

[0071] (2) The constraints are as follows: To ensure that the sum of the membership degrees of each data point to all clusters is 1:

[0072]

[0073] In the formula, u ij has a numerical range of [0, 1].

[0074] The power generation power of the photovoltaic system is mainly affected by weather factors and the backplane temperature, and the load power of the charging pile is mainly affected by weather factors and traffic flow. Based on the Spearman correlation coefficient, the main factors affecting the power generation power and the load power of the charging pile are screened. The implementation path for screening the characteristic parameters of the prediction model is as Figure 4 、 Figure 5 shown.

[0075] The calculation result ρ of the Spearman correlation coefficient of each influencing factor s The closer it is to 1, the greater the correlation. The influencing factors with large correlations are selected as the characteristic parameters of the prediction model. Its calculation formula is:

[0076]

[0077] In the formula, X i is the magnitude of the influencing factor of the i-th sample point. In the present invention, the influencing factors affecting photovoltaic power generation or the load of the charging pile are total radiation, temperature, humidity, air pressure, wind speed, wind direction, etc.; Y i is the magnitude of the power generation power or the load power of the charging pile of the i-th sample point; X and Y are the average values of the influencing factor and the power generation power among N sample points, respectively.

[0078] Taking the screened influencing factors of photovoltaic power generation power, the influencing factors of charging pile load power, as well as the historical photovoltaic power generation power and historical charging pile load power, as the characteristic parameters of the LSTM deep learning prediction models of the photovoltaic power generation prediction model and the charging pile load prediction model respectively. Data is collected at a frequency of 15 min / time, and the first 50 data points are used as parameters to predict the next 1 data point, and so on, to realize the source-load power of the next day. Its implementation path is as Figure 6 、 Figure 7 shown.

[0079] The long short-term memory network is a special recurrent neural network architecture. Its core principle is to control the flow of information by introducing a complex gating mechanism, which enables the network to effectively learn and remember information in long sequences. The following are the key components and working principles of the LSTM network:

[0080] (1) Input gate: The input gate determines which new information will be allowed to enter the cell state at each time step. It is calculated by the following formula:

[0081] i t = σ(W i ·|h i-1 , x t | + b i ) (4)

[0082] In the formula, i t is the activation value of the input gate, σ is the sigmoid activation function, W i is the weight, h i-1 is the hidden state of the previous time step, x t is the input of the current time step, b i is the input gate bias term.

[0083] (2) Forget gate: The forget gate determines which information should be forgotten or retained from the cell state at each time step. It is calculated by the following formula:

[0084] f t = σ(W f ·|h i-1 , x t | + b f ) (5)

[0085] In the formula, f t is the activation value of the forget gate, W f is the weight, b f is the forget gate bias term, h t-1 is the hidden state of the previous time step

[0086] (3) Cell state: The cell state is a key concept in LSTM and is responsible for storing long-term memories. At each time step, the cell state is updated according to the activation values of the input gate and the forget gate:

[0087] C t = f t * C t-1 + i t * C’ t (6)

[0088] In the formula, where C t is the cell state of the current time step, C t, ’ is the candidate cell state, * represents element-wise multiplication C t-1 is the memory cell state at time step t - 1.

[0089] (4) Candidate unit state: The candidate unit state is a temporary value for updating the unit state, which is calculated by the activation function from the current input and the hidden state of the previous time step:

[0090] C’ t = tanh(w c ·|h i-1 ,x t | + b o ) (7)

[0091] (5) Output gate: The output gate determines the value of the next hidden state, which is based on the current unit state and the hidden state of the previous time step:

[0092] O t = σ(w o ·|h t-1 ,x t | + b o ) (8)

[0093] h t = O t *tanh(c t ) (9)

[0094] In the formula, O t is the activation value of the output gate, h t is the hidden state of the current time step, w o is the weight matrix of the output gate, b o is the bias term of the output gate; h t is the hidden state at time step t.

[0095] To improve the comprehensive benefits of the photovoltaic-storage charging station, the optimization objectives are the lowest operating cost and the lowest carbon emissions. The operating cost mainly considers the electricity purchase cost of the system, the unit operating cost of the photovoltaic system, the unit operating cost of the energy storage system, and the unit operating cost of the charging pile. The calculation formula is as follows:

[0096]

[0097] In the formula, is the price of purchasing unit electricity from the grid, yuan / kW; is the power transmitted by the grid, kW; is the unit operating cost of the photovoltaic system, yuan / kW; is the photovoltaic power generation; is the energy storage operating cost, yuan / kW; is the energy storage charge and discharge power, kW; is the unit operating cost of the charging pile, yuan / kW; is the charging pile power, kW.

[0098] The carbon emissions of the photovoltaic-storage-charging system are mainly caused by the consumption of mains electricity and the losses of energy storage. The calculation formula is as follows:

[0099]

[0100] In the formula, is the carbon emissions generated by purchasing electricity from the power grid, in t; E is the carbon emission coefficient of the local power grid, in t / kW; P loss is the loss power of energy storage charging and discharging, in kW.

[0101] The boundary conditions of the source-load operation optimization model include:

[0102] 1) Power constraint balance:

[0103] During the operation of the photovoltaic-storage-charging system, a balance relationship should be maintained among the photovoltaic power generation, the energy storage charging and discharging, the electricity purchased from the power grid, and the charging pile load, to avoid situations such as curtailment of photovoltaic power or insufficient load demand. The calculation formula is as follows:

[0104]

[0105] In the formula, is the electricity load of the charging pile, in kW·h; is the energy storage charging and discharging amount, in kW·h, when it is greater than 0, it is discharging; when it is less than 0, it is charging.

[0106] 2) Energy storage state of charge constraint:

[0107] To ensure the system operates within the safe temperature range, the charging and discharging power of the energy storage should not exceed the upper and lower limits of the rated power. And the remaining charge SOC of the energy storage does not exceed the upper limit value of 0.9 and is not lower than the lower limit value of 0.1 to avoid overcharging and over-discharging.

[0108]

[0109] In the formula, is the remaining charge of the energy storage at time t + 1; γ is the self-discharge rate of the energy storage battery, taking 0.001; and are the charging and discharging efficiencies of the energy storage respectively; E b is the rated capacity of the energy storage system; is the discharge amount of the system at time t.

[0110] In this embodiment, the process of the Hippopotamus optimization algorithm is as follows:

[0111] (1) Population initialization

[0112] X ij = lb j + r·(ub j - lbj ) (17)

[0113] Among them, X ij represents the position information of the \(i\)th hippopotamus under the \(j\)th decision variable; \(r\) is a random number within the range of 0 to 1, used to introduce randomness; \(lb\ j and \(ub\ j represent the lower and upper limits of the \(j\)th decision variable respectively; \(N\) represents the number of hippopotamuses in the population; \(m\) represents the number of decision variables in the problem.

[0114] (2) Position update of hippopotamuses in rivers or ponds (exploration stage)

[0115] X P1 (i, :) = X(i, :) + r·(X best - I1·X(i, :)) (18)

[0116] Among them, X P1 (i, :) represents the position vector of the \(i\)th hippopotamus after the first position update; X(i, :) represents the current position vector of the \(i\)th hippopotamus; X best represents the position vector of the optimal hippopotamus in the current population; I1 is a random integer, usually taking 1 or 2, used to introduce randomness in position update.

[0117] (3) Hippopotamuses defend against predators (exploration stage)

[0118]

[0119] Among them, \(p\) represents the position of the predator; \(distancep\) represents the distance between the hippopotamus and the predator; \(b\), \(c\), \(d\), and \(l\) are randomly generated parameters; \(RL(i, :)\) represents the random vector generated by the Levy flight of the \(i\)th hippopotamus.

[0120] (4) Hippopotamuses escape from predators (exploitation stage)

[0121] X P2 (i, :) = X(i, :) + A·(X best - I2·MeanGroup) (20)

[0122] Among them, I2 is a random integer, usually taking 1 or 2, used to introduce randomness in position update; MeanGroup represents the average position of a randomly selected group of hippopotamuses; A is a randomly generated parameter.

[0123] Taking the optical storage charging station on a certain highway as an example, such as Figure 8As shown in the figure, the hourly photovoltaic power generation (yellow) and charging pile load (red) on a certain day are predicted based on a deep learning model. According to the hippopotamus optimization algorithm, the energy storage charge and discharge operation strategy with the lowest system operation cost is analyzed (orange for energy storage charging and green for energy storage discharging), and the power purchase strategy (blue for power purchase). The system operates within the safe operation range, the SOC is between 0.1 and 0.9, and there is no frequent switching of the charge and discharge actions of the energy storage device.

[0124] Through the above design, the present invention classifies and divides the source-load scenarios according to the meteorological factors affecting the photovoltaic power generation side, as well as the meteorological factors and holiday factors affecting the charging pile load side, and can give corresponding operation strategies for holiday conditions, non-holiday conditions, sunny, cloudy, and rainy conditions.

[0125] The above embodiments are only one of the preferred embodiments of the present invention and should not be used to limit the protection scope of the present invention. Any meaningless changes or polish made on the main design concept and spirit of the present invention, as long as the technical problems solved are still consistent with the present invention, should be included in the protection scope of the present invention.

Claims

1. A method for optimizing the coordinated operation of a photovoltaic and energy storage charging station, characterized in that: The following steps are involved: S1, based on the FCM fuzzy clustering algorithm, the historical meteorological data are clustered with irradiance as the clustering index, and the historical traffic flow data are clustered with daily traffic flow as the clustering index; S2, screening the factors affecting the power generation and charging pile load power based on the Spearman correlation coefficient, and obtaining the characteristic parameters of the photovoltaic power generation prediction model and the characteristic parameters of the charging pile load prediction model; S3, using LSTM deep learning model training to obtain photovoltaic power generation deep learning prediction model and charging pile load deep learning prediction model; S4, based on the deep learning prediction model of photovoltaic power generation and the deep learning prediction model of charging pile load, establish a source-load operation optimization model with the lowest operating cost and the lowest carbon emission as the optimization goals; S5, determine the boundary conditions of the source-load operation optimization model, use the Hippo optimization algorithm to solve the source-load operation optimization model, and obtain the energy storage charging and discharging operation strategy and power purchase strategy with the lowest source-load coordinated operation cost.

2. The method for optimizing the coordinated operation of the photovoltaic and energy storage charging station according to claim 1, characterized in that: In step S1, historical meteorological data are clustered into three weather conditions: sunny, cloudy and rainy, and historical traffic flow data are clustered into holidays and non-holidays.

3. The method for optimizing the coordinated operation of the photovoltaic and energy storage charging station according to claim 2, characterized in that: In step S1, the objective function of the FCM fuzzy clustering algorithm is: In the formula, u ij is the sample point x i With cluster center v j membership degree; m is the fuzzy index (m>1), which determines the fuzziness of the clustering; d ij is the sample point x i With cluster center v j The distance between them; k is the number of clusters; n is the number of samples; The constraints of the objective function of the FCM fuzzy clustering algorithm are: ensure that the sum of the membership of each data point to all clusters is 1; Right now: In the formula, u ij The value range is [0,1].

4. The method for optimizing the day-ahead source-load coordinated operation of a photovoltaic energy storage charging station according to claim 3 is characterized in that: In step S2, the expressions for obtaining the characteristic parameters of the photovoltaic power generation prediction model and the characteristic parameters of the charging pile load prediction model are: Where, X i is the influencing factor size of the i-th sample point; Y i is the power generation size or charging pile load power size of the i-th sample point; X and Y are the average values ​​of the influencing factors and power generation in N sample points respectively.

5. The method for optimizing the coordinated operation of the photovoltaic and energy storage charging station according to claim 4 is characterized in that: In step S4, the operating cost includes the system's electricity purchase cost, the photovoltaic system unit operating cost, the energy storage system unit operating cost, and the charging pile unit operating cost, and the calculation formula is as follows: In the formula, is the price of electricity purchased online, RMB / kW; Power delivered to the grid kW; is the unit operating cost of the photovoltaic system, RMB / kW; is the photovoltaic power generation power; is the energy storage operation cost, RMB / kW; is the energy storage charging and discharging power kW; is the unit operating cost of the charging pile, RMB / kW; is the charging pile power kW; Carbon emissions are caused by the consumption of mains electricity and the loss of energy storage, and the calculation formula is as follows: In the formula, is the carbon emissions generated by purchasing electricity from the power grid, t; E is the carbon emission coefficient of the local power grid, t / kW; P loss It is the energy storage charging and discharging loss power kW.

6. The method for optimizing the day-ahead source-load coordinated operation of a photovoltaic energy storage charging station according to claim 5, characterized in that: In step S5, the boundary conditions of the source-load operation optimization model include: The power balance constraint is expressed as: In the formula, is the power load of the charging pile, kW.h; is the energy storage charge and discharge capacity kW.h, When it is greater than 0, it is discharging, and when it is less than 0, it is charging; Energy storage charge state constraint, its expression is: In the formula, is the remaining energy storage capacity at time t+1; γ is the self-discharge rate of the energy storage battery, which is 0.001; and are the efficiency of energy storage charging and discharging; E b is the rated capacity of the energy storage system; is the discharge amount of the system at time t.

7. The method for optimizing the day-ahead source-load coordinated operation of a photovoltaic energy storage charging station according to claim 6, characterized in that: In step S5, the specific process of the Hippo optimization algorithm is as follows: S51, population initialization: X ij =lb j +r·(ub j -lb j ) Among them, X ij represents the position information of the i-th hippopotamus under the j-th decision variable; r is a random number between 0 and 1 to introduce randomness; lb j andub j They represent the lower and upper limits of the j-th decision variable, respectively; N represents the number of hippos in the population; m represents the number of decision variables in the problem; S52, Update the location of hippos in rivers or ponds: X P1 (i,:)=X(i,:)+r·(X best -I1·X(i,:)) Among them, X P1 (i,:) represents the position vector of the i-th hippopotamus after the first position update; X(i,:) represents the current position vector of the i-th hippopotamus; X best Represents the position vector of the best hippo in the current population; I1 is a random integer, usually 1 or 2, used to introduce randomness in position updates; S53, Hippopotamus defending against predators: Where p represents the position of the predator; distancep represents the distance between the hippopotamus and the predator; b, c, d and l are randomly generated parameters; RL(i,:) represents the random vector generated by the Levy flight of the i-th hippopotamus; S54, Hippopotamus escapes from predator: X P2 (i,:)=X(i,:)+A·(X best -I2·MeanGroup) Among them, I2 is a random integer, usually 1 or 2, used to introduce randomness in position updates; MeanGroup represents the average position of a randomly selected hippo group; A is a randomly generated parameter.

Images