Direct-current bus two-transformer-area mutual aid scheduling method and system with photovoltaic energy storage function

Abstract

The invention discloses a direct current bus photovoltaic energy storage two-transformer-area mutual aid scheduling method and system, and relates to the technical field of transformer-area mutual aid scheduling, and the method comprises the steps: collecting transformer-area photovoltaic energy storage historical data, and constructing a prediction model to predict the future load and photovoltaic power generation; monitoring the real-time load power of the charging piles in the transformer area, and calculating the net load and excess power generation according to the prediction data and the real-time data; executing energy scheduling, including photovoltaic energy distribution and energy storage charge and discharge control, and processing the excess power generation; and setting a night energy storage discharge and loss compensation strategy, and continuously learning and optimizing a control strategy. Aiming at the problem of heavy overload of a transformer in a transformer area, the invention provides a direct-current bus light storage two-transformer-area mutual aid dispatching method, the problem of heavy overload of the transformer area is solved by coordinating dispatching transformer area and light storage energy, and stable and safe operation of the transformer area is guaranteed; and optical storage energy optimization scheduling is fully considered, so that the utilization efficiency of resources is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of mutual assistance dispatching between power substations, and in particular to a method and system for mutual assistance dispatching between two power substations with a direct current busbar and photovoltaic energy storage. Background Art

[0002] With the advancement of the construction of new power systems with new energy as the main body, distributed new energy represented by wind, solar and storage has poured in. However, since new energy is affected by weather and climate, its output is highly random and intermittent. Low-voltage and high-proportion access leads to insufficient absorption of new energy, causing the output of new energy to exceed the system load demand, resulting in many problems such as unbalanced feeder load, voltage over-limit and power reverse transmission. At the same time, due to peak electricity consumption caused by seasonal and special weather reasons, the distribution transformer is overloaded, causing safety problems such as transformer burning. Summary of the invention

[0003] In view of the problems existing in the existing method for mutual assistance scheduling of two power stations with a DC bus and photovoltaic energy storage, the present invention is proposed.

[0004] Therefore, the problem to be solved by the present invention is that new energy is affected by weather and climate, and its output is highly random and intermittent. Low-voltage and high-proportion access leads to insufficient absorption of new energy, causing the output of new energy to be greater than the system load demand.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] In the first aspect, an embodiment of the present invention provides a method for mutual assistance scheduling of two substations with a DC bus and photovoltaic energy storage, which includes collecting historical data of photovoltaic energy storage in the substation, building a prediction model to predict future load and photovoltaic power generation; monitoring the real-time load power of charging piles in the substation, and calculating the net load and excess power generation based on the predicted data and real-time data; executing energy scheduling, including photovoltaic energy distribution and energy storage charging and discharging control, and handling photovoltaic excess conditions; setting nighttime energy storage discharge and loss compensation strategies, and continuously learning and optimizing control strategies.

[0007] As a preferred solution of the mutual assistance dispatching method of two substations with a DC bus with photovoltaic energy storage according to the present invention, the photovoltaic energy storage historical data includes historical load data, historical photovoltaic power generation data and weather data; the collected data is standardized, missing values ​​are monitored, and abnormal values ​​are detected and processed to obtain processed valid data; the load prediction characteristics of the valid data are expressed as:

[0008] X l (t) = [L hist (t-24:t), T(t), T(t+1:t+24), W(t), W(t+1:t+24)]

[0009] Among them, L hist (t-24) represents the load history data of the past 24 hours, T(t) represents the temperature data at the current time t, T(t+1:t+24) represents the temperature forecast data for the next 24 hours, W(t) represents the weather conditions at the current time t, and W(t+1:t+24) represents the weather forecast data for the next 24 hours; and the photovoltaic power generation prediction characteristics are performed, which is expressed as:

[0010] X p (t) = [PV hist (t-24:t),I(t),I(t+1:t+24),W(t),W(t+1:t+24)]

[0011] Among them, X p (t) represents the input feature vector of photovoltaic power generation prediction at time t, PV hist (t-24) represents the historical data of photovoltaic power generation in the past 24 hours, I(t) represents the solar irradiance at the current time t, I(t+1:t+24) represents the predicted solar irradiance data for the next 24 hours, W(t) represents the weather conditions at the current time t, and W(t+1:t+24) represents the weather forecast data for the next 24 hours.

[0012] As a preferred solution of the mutual assistance scheduling method of two-station area with DC bus and photovoltaic energy storage described in the present invention, the prediction model uses a stacked LSTM model, and the structure includes: input layer, LSTM layer 1, Dropout layer, LSTM layer 2, Dropout layer, fully connected layer and output layer.

[0013] As a preferred solution of the mutual assistance dispatching method of two substations with a DC bus with photovoltaic energy storage in the present invention, the net load and excess power generation are obtained by calculating the load forecast output by the prediction model and the photovoltaic power generation forecast, which is used to realize the energy storage dispatching of the two substations; the formula of the load forecast is expressed as:

[0014] P load (t:t+24)=LSTM l (X l (t), θ l )

[0015] Among them, LSTM l LSTM network for load forecasting, X l (t) represents the load forecast input feature vector at time t, θ l LSTM model l Parameters.

[0016] As a preferred solution of the mutual assistance dispatching method of two substations with a DC bus with photovoltaic energy storage according to the present invention, the photovoltaic power generation prediction formula is expressed as:

[0017] Ppv(t:t+24)=LSTM p (X p (t), θ p )

[0018] Among them, LSTM p LSTM network for photovoltaic power generation prediction, X p (t) represents the input feature vector of photovoltaic power generation prediction at time t, θ p LSTM model p Parameters used to control the behavior of the photovoltaic power generation prediction model.

[0019] As a preferred solution of the mutual assistance dispatching method of two substations with a DC bus with photovoltaic energy storage described in the present invention, the energy dispatching includes photovoltaic energy distribution and energy storage charging and discharging control. According to the results of the prediction model, the difference between the load prediction and the photovoltaic power generation prediction is calculated to obtain the predicted net load and excess power generation; if there is excess power generation, the load power of the charging pile in substation 2 is monitored; when it is judged that the charging load power is greater than the first threshold, photovoltaic energy is extracted to substation 2, and the extracted power size is the load power minus the first threshold; when the photovoltaic power is insufficient, the energy storage is set to discharge, and the discharge power size is the charging load power minus the first threshold and then minus the photovoltaic power; if the photovoltaic and energy storage power generation powers are both 0, at this time, the charging load is fully supplemented by substation 2; when it is judged that the charging load power is less than the first threshold, the energy storage is set to charge according to the photovoltaic power size, if the energy storage SOC is fully charged, the energy storage power is set to 0, and the energy storage device is shut down; in addition, considering the existence of faults, a fault clearing is performed each time the machine is turned on to avoid startup failure; if the charging load is 0 and the energy storage is fully charged, all photovoltaics are connected to the Internet through substation 1.

[0020] As a preferred solution of the mutual assistance dispatching method of two stations with DC bus and photovoltaic energy storage in the present invention, the nighttime energy storage discharge is achieved by setting a constant discharge power P constant , providing stable power output for the substation equipment at night to ensure that all system parameters are initialized normally before nighttime operation; dynamically adjusting the energy storage compensation and the power compensation of the power grid substation 1 according to the available energy of the energy storage equipment and the equipment loss, with energy storage giving priority to compensation. When the energy storage is insufficient, it automatically switches to the power grid; when the current power percentage of the energy storage device SOC current Close to the set minimum power SOC minWhen the energy storage device is in a state of emergency, the discharge power of the energy storage device is gradually reduced, the energy output is reduced, and warnings are issued at different power thresholds to ensure that managers or automatic systems have enough time to make adjustments and emergency responses.

[0021] In the second aspect, an embodiment of the present invention provides a two-station mutual assistance dispatching system with a DC bus and photovoltaic energy storage, which includes: a prediction module for collecting historical data of photovoltaic energy storage in the station area, and constructing a prediction model to predict future load and photovoltaic power generation; a prediction module for monitoring the real-time load power of charging piles in the station area, and calculating the net load and excess power generation based on the predicted data and real-time data; a dispatching module for performing energy dispatching, including photovoltaic energy distribution and energy storage charging and discharging control, and processing the excess power generation; a nighttime control module for setting nighttime energy storage discharge and loss compensation strategies, and continuously learning and optimizing control strategies.

[0022] In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the processor executes the computer program, any step of the above-mentioned method for mutual assistance scheduling of two stations with a DC bus and photovoltaic energy storage is implemented.

[0023] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the above-mentioned method for mutual assistance scheduling of two stations with a DC bus and photovoltaic energy storage is implemented.

[0024] The beneficial effect of the present invention is that, in response to the problem of heavy overload of transformers in substations, the present invention proposes a mutual assistance scheduling method for two substations with a DC bus and photovoltaic storage, which solves the problem of heavy overload in substations and ensures stable and safe operation of substations by coordinating the scheduling between substations and the photovoltaic storage energy; and fully considers the optimal scheduling of photovoltaic storage energy to improve the utilization efficiency of resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0026] Figure 1 Scenario diagram of the mutual assistance dispatching method for two zones with a DC bus and photovoltaic energy storage. DETAILED DESCRIPTION

[0027] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0028] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0029] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0030] Example 1

[0031] Reference Figure 1 , which is the first embodiment of the present invention, and provides a method for mutual assistance scheduling of two stations with a DC bus and photovoltaic energy storage, including:

[0032] S1: Collect historical data on photovoltaic energy storage in the substation area and build a model to predict future load and photovoltaic power generation.

[0033] The two substations are connected through a mutual aid device. One mutual aid device builds a 750V DC bus, and photovoltaic and energy storage equipment are connected to the DC bus. In addition, the mutual aid, photovoltaic, and energy storage capacities are 250kVA, 100kW, and 100kW / 150kWh, respectively.

[0034] Substation 1 is directly connected to the transformer through mutual aid equipment and has no AC load, while substation 2 has charging piles and AC loads on the AC side.

[0035] Photovoltaic power is consumed first in substation 2. When the charging pile is under load, photovoltaic power is supplied. When photovoltaic power generation is greater than the charging load or there is no charging load, the excess photovoltaic power is absorbed by the energy storage. When the energy storage is fully charged and cannot absorb photovoltaic power, it is connected to the grid through substation 1.

[0036] It is necessary to predict the photovoltaic power generation load and collect historical data on photovoltaic energy storage, including historical load data, historical photovoltaic power generation data and weather data;

[0037] The collected data are standardized, missing values ​​are handled using interpolation or forward filling methods, and outliers are detected and handled using the IQR method or the Z-score method.

[0038] The load forecasting characteristics of the effective data are expressed as:

[0039] X l (t) = [L hist (t-24:t),T(t),T(t+1:t+24),W(t),W(t+1:t+24)]

[0040] Among them, L hist (t-24) represents the load history data of the past 24 hours, T(t) represents the temperature data at the current time t, T(t+1:t+24) represents the temperature forecast data for the next 24 hours, W(t) represents the weather conditions at the current time t, and W(t+1:t+24) represents the weather forecast data for the next 24 hours;

[0041] At the same time, the photovoltaic power generation prediction characteristics are performed, which are expressed as:

[0042] X p (t) = [PV hist (t-24:t),I(t),I(t+1:t+24),W(t),W(t+1:t+24)]

[0043] Among them, X p (t) represents the input feature vector of photovoltaic power generation prediction at time t, PV hist (t-24) represents the photovoltaic power generation history data for the past 24 hours. Photovoltaic power generation has a strong intraday periodicity, so past power generation can help predict future power generation. I(t) represents the solar irradiance (or light intensity) at the current time t. Solar irradiance is the most direct factor affecting photovoltaic power generation.

[0044] I(t+1:t+24) represents the forecast data of solar irradiance in the next 24 hours. This helps predict the future changes in photovoltaic power generation. W(t) represents the weather conditions at the current time t. In addition to light intensity, other weather factors (such as temperature, cloud cover, wind speed, etc.) will also affect the efficiency of photovoltaic power generation.

[0045] W(t+1:t+24) represents the weather forecast data for the next 24 hours. Future changes in weather conditions can help predict future photovoltaic power generation.

[0046] A model is constructed based on the prediction features to predict future load and photovoltaic power generation. A stacked LSTM model is used, and its structure includes: input layer, LSTM layer 1 (128 units), Dropout layer (rate 0.2), LSTM layer 2 (64 units), Dropout layer (rate 0.2), fully connected layer, and output layer.

[0047] S2: Monitor the real-time load power of charging piles in the substation, and calculate the net load and excess power generation based on the predicted data and real-time data.

[0048] The prediction model is trained using the mean square error (MSE) as the loss function, expressed as:

[0049]

[0050] Among them, N represents the total number of samples, indicating how many samples there are in the data set, y i Represents the true value of the i-th sample, that is, the actual value of the target variable to be predicted by the model. represents the predicted value of the i-th sample, that is, the prediction result calculated by the model based on the input features and parameters θ. Represents the square of the error of the i-th sample, which represents the difference between the true value and the predicted value of the sample. Squaring makes both positive and negative errors positive and amplifies larger errors.

[0051] The squared error is calculated for each sample to avoid the positive and negative errors from canceling each other out and to have a stronger penalty effect on large errors.

[0052] The prediction error is determined by the error between the model prediction and the actual value. The smaller the value of the loss function, the closer the model's prediction result is to the true value; conversely, the larger the loss, the larger the model's prediction error.

[0053] Using the parameter update rule of the Adam optimizer, the model parameters are updated by adaptively adjusting the learning rate to accelerate the optimization convergence, which is expressed as:

[0054]

[0055] Among them, θ t is the parameter vector of the model at time step t, and the Adam optimizer will gradually update these parameters to minimize the loss function. t+1 represents the updated parameter vector of the model at time step t+1. η is the learning rate, which is initially set to 0.001. The learning rate determines the step size of each parameter update and controls the speed of parameter update.

[0056] Represents the first-order moment estimate, which is the exponentially weighted moving average of the gradient of the loss function with respect to the parameter θ. Adam smoothes the moving average of the gradient to help avoid the influence of noise in the instantaneous gradient. It is the second-order moment estimate, that is, the exponentially weighted moving average of the square of the gradient, which is used to represent the variance (degree of change) of the gradient. Adam is calculated by The gradient is adjusted to ensure that the update step size is more balanced in each dimension.

[0057] The calculation method is:

[0058]

[0059] where v t is the exponentially weighted moving average of the squared gradient, and β2 is the decay rate.

[0060] ∈ is a constant used to prevent division by zero errors. When ∈ is very small or close to zero, the stability of numerical calculation is guaranteed.

[0061] Furthermore, the data set is divided into a training set (70%), a validation set (15%), and a test set (15%). The sliding window method is used to generate sequence samples, with a window size of 24 (corresponding to 24 hours), a batch size of 64, and 200 epochs of training. The early stopping method is used, and the training is stopped when the validation set loss has not improved for 10 consecutive epochs. The trained model is obtained to predict the load and photovoltaic power generation in the next 24 hours.

[0062] The formula for load forecasting is:

[0063] P load (t:t+24)=LSTM l (X l (t), θ l )

[0064] Among them, LSTM l LSTM network for load forecasting. l (t) represents the load forecast input feature vector at time t. According to the previous definition, X l (t) contains information such as load history, temperature, and weather, describing the current and future environment and historical data. l LSTM model l The parameters of the network include weights, biases, etc. Through the training process, these parameters are optimized so that the model can more accurately predict future loads.

[0065] The formula for predicting photovoltaic power generation is expressed as:

[0066] Ppv(t:t+24)=LSTM p (X p (t), θ p )

[0067] Among them, LSTM p LSTM network for photovoltaic power generation prediction, similar in structure to LSTM l , but focusing on photovoltaic power generation forecasting.

[0068] X p (t) represents the input feature vector of photovoltaic power generation prediction at time t. According to the previous definition, X p (t) contains information such as photovoltaic power generation history and weather, describing the current and future conditions of photovoltaic power generation.

[0069] θ p LSTM model p Parameters, similar to θ l , which is used to control the behavior of the photovoltaic power generation prediction model.

[0070] The difference between the two forecast results is used to estimate future net load and excess generation.

[0071] S3: Execute energy scheduling, including photovoltaic energy distribution and energy storage charging and discharging control, to handle the excess power generation.

[0072] If the price difference of the forecast results indicates that there may be excess power generation, the load power of the charging pile in area 2 is monitored;

[0073] When it is determined that the charging load power is greater than the first threshold, photovoltaic energy is extracted to area 2, and the extracted power is the load power minus the first threshold; when the photovoltaic power is insufficient, the energy storage is set to discharge, and the discharge power is the charging load power minus the first threshold minus the photovoltaic power; if the photovoltaic and energy storage power generation powers are both 0, at this time, the charging load is fully supplemented by area 2;

[0074] When it is determined that the charging load power is less than the first threshold, the energy storage is set to charge according to the photovoltaic power. If the energy storage SOC is fully charged, the energy storage power is set to 0 and the energy storage device is shut down. In addition, considering the existence of faults, a fault clearing is performed each time the device is turned on to avoid startup failure.

[0075] If the charging load is 0 and the energy storage is fully charged, all photovoltaics will be connected to the grid through area 1.

[0076] S4: Set nighttime energy storage discharge and loss compensation strategies, and continuously learn and optimize control strategies.

[0077] The night mode is activated after sunset. The system continuously monitors the light intensity of the photovoltaic panels and automatically activates the night mode when the light falls below the set threshold.

[0078] Set the constant discharge power P constant This power is used by the energy storage equipment to provide stable power output for the substation equipment at night, ensuring that all system parameters are initialized normally before nighttime operation, including energy storage status, grid status, etc.

[0079] SOC current It is the percentage of the current power of the energy storage device. It continuously monitors the current SOC status to ensure that the energy storage power is sufficient.

[0080] The available energy of the energy storage determines the total amount that can be discharged at night. Based on the current SOC, the available energy is calculated as:

[0081] E available =(SOC current -SOC min )·E storage

[0082] Among them, SOC current is the current energy storage capacity, SOC min It is the minimum safe power of the energy storage device (to avoid over-discharge), E storage is the total energy capacity of the energy storage device.

[0083] Start the constant power discharge process, the discharge power is the preset P constant , providing stable power output and compensating for the energy loss of equipment at night. Check the status of energy storage equipment at regular intervals and adjust the discharge power to ensure constant power output.

[0084] According to the available energy of the energy storage equipment and the equipment loss, the energy storage compensation and the power compensation of the grid station area 1 are dynamically adjusted. The energy storage is compensated first, and when the energy storage is insufficient, it automatically switches to the grid power supply.

[0085] When SOC current Close to the set minimum power SOC min When the energy storage device is in use, the discharge power of the energy storage device is gradually reduced, the energy output is reduced, and the energy storage operation time is extended. Warnings are issued at different power thresholds to ensure that managers or automatic systems have enough time to make adjustments and emergency responses.

[0086] If there is a sudden high-power demand, the system will switch to the power supply mode of Area 1. That is, when the energy storage equipment can no longer supply power or a major failure occurs, the system will switch to power supply in Area 1 in an emergency, and the power grid will fully supplement the power demand.

[0087] Assess the remaining power of the nighttime energy storage and formulate an initial charging plan for the next day based on the power, ensuring that the energy storage system can be effectively charged through photovoltaics during the day and prepare for the next night cycle.

[0088] In summary, the present invention proposes a mutual assistance dispatching method for two substations with a DC bus and photovoltaic storage to solve the problem of heavy overload of transformers in the substation area, and solves the problem of heavy overload in the substation area by coordinating the dispatching between substations and the photovoltaic storage energy, thereby ensuring the stable and safe operation of the substation area; and fully considers the optimal dispatching of photovoltaic storage energy to improve the utilization efficiency of resources.

[0089] Example 2

[0090] Reference Figure 1 Based on the first embodiment, this embodiment further provides a two-station mutual assistance dispatching system with a DC bus and photovoltaic energy storage, including:

[0091] The prediction module is used to collect historical data on photovoltaic energy storage in the substation area and build a prediction model to predict future load and photovoltaic power generation;

[0092] The prediction module is used to monitor the real-time load power of charging piles in the substation area and calculate the net load and excess power generation based on the predicted data and real-time data;

[0093] A dispatching module, used to perform energy dispatching, including photovoltaic energy distribution and energy storage charging and discharging control, and handle the excess power generation;

[0094] The night control module is used to set the nighttime energy storage discharge and loss compensation strategy, and continuously learn and optimize the control strategy.

[0095] This embodiment also provides a computer device, which is suitable for the mutual assistance scheduling method of two power stations with a DC bus and photovoltaic energy storage, including a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the mutual assistance scheduling method of two power stations with a DC bus and photovoltaic energy storage as proposed in the above embodiment.

[0096] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0097] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for realizing mutual assistance scheduling of two stations with a DC bus and photovoltaic energy storage as proposed in the above embodiment is implemented.

[0098] The storage medium proposed in this embodiment and the data storage method proposed in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be found in the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0099] Example 3

[0100] On the basis of the previous two embodiments, this embodiment provides a method for mutual assistance scheduling of two power stations with a DC bus and photovoltaic energy storage. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.

[0101] The experimental conditions were set as follows:

[0102] Area 1: No AC load, connected to photovoltaic and energy storage equipment.

[0103] Area 2: There is AC load, the charging pile is connected, and the AC load fluctuates.

[0104] Equipment configuration: Huji equipment: 750V DC bus; photovoltaic equipment: 100kW; energy storage equipment: 100kW / 150kWh.

[0105] By comparing the prediction results and actual consumption conditions under different weather and load fluctuation scenarios, the scheduling optimization effect of the system is analyzed, and the key performance indicators of the system (such as load scheduling efficiency, photovoltaic consumption efficiency, energy storage utilization rate, system stability, etc.) are recorded. The experimental results are shown in Table 1:

[0106] Table 1 Comparison of key indicators

[0107] Experimental conditions Experimental indicators Traditional scheduling method The present invention Small load fluctuation Energy scheduling response time 2 seconds 1 second Photovoltaic power generation consumption rate 85％ 95％ Energy storage utilization 70％ 85％ System stability (trouble-free operation) 98％ 99.5％ Large load fluctuation Energy scheduling response time 3 seconds 1.5 seconds Photovoltaic power generation consumption rate 75％ 92％ Energy storage utilization 60％ 80％ System stability (trouble-free operation) 95％ 99％ Photovoltaic power generation surplus Photovoltaic grid-connected ratio 40％ 25％ Energy storage charging efficiency 80％ 90％ Nighttime energy storage and discharge mode Energy storage discharge duration 6 hours 8 hours Grid power supply ratio 30％ 15％ Comprehensive performance System energy scheduling efficiency 75％ 90％ Overall system stability 96％ 99％

[0108] By setting experimental conditions: small load fluctuation, large load fluctuation, excess photovoltaic power generation, night-time energy storage discharge mode, etc., the performance of the traditional dispatching system and the dispatching system of the present invention are compared. The experimental indicators include energy dispatching response time, photovoltaic power generation consumption rate, energy storage utilization rate, system stability, etc.

[0109] The present invention can maintain rapid dispatch when the load fluctuates greatly. Even if the load fluctuates greatly, photovoltaic power can still be effectively absorbed, and energy storage can be better utilized to adjust the load. Moreover, the system is more stable under complex circumstances.

[0110] The comparison results show that the present invention has obvious advantages in many aspects such as shorter response time, higher resource utilization, and more stable system.

[0111] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for mutual assistance dispatching of two stations with a DC bus and photovoltaic energy storage, characterized by: include, Collect historical data on photovoltaic energy storage in the substation area and build a prediction model to predict future load and photovoltaic power generation; Monitor the real-time load power of charging piles in the substation area, and calculate the net load and excess power generation based on the forecast data and real-time data; Perform energy dispatch, including photovoltaic energy allocation and energy storage charging and discharging control, to handle the excess power generation; Set nighttime energy storage discharge and loss compensation strategies, and continuously learn and optimize control strategies.

2. The mutual assistance dispatching method of two stations with a DC bus with photovoltaic energy storage as claimed in claim 1 is characterized by: The photovoltaic energy storage historical data includes historical load data, historical photovoltaic power generation data and weather data; Standardize the collected data, monitor missing values, detect and process outliers, and obtain valid data after processing; The load forecasting characteristics of the effective data are expressed as: X l (t)=[L hist (t-24:t),T(t),T(t+1:t+24),W(t),W(t+1:t+24)] Among them, L hist (t-24) represents the load history data of the past 24 hours, T(t) represents the temperature data at the current time t, T(t+1:t+24) represents the temperature forecast data for the next 24 hours, W(t) represents the weather conditions at the current time t, and W(t+1:t+24) represents the weather forecast data for the next 24 hours; And the photovoltaic power generation prediction characteristics are expressed as: X p (t)=[PV hist (t-24:t),I(t),I(t+1:t+24),W(t),W(t+1:t+24)] Among them, X p (t) represents the input feature vector of photovoltaic power generation prediction at time t, PV hist (t-24) represents the historical data of photovoltaic power generation in the past 24 hours, I(t) represents the solar irradiance at the current time t, I(t+1:t+24) represents the predicted solar irradiance data for the next 24 hours, W(t) represents the weather conditions at the current time t, and W(t+1:t+24) represents the weather forecast data for the next 24 hours.

3. The mutual assistance dispatching method of two stations with a DC bus with photovoltaic energy storage as claimed in claim 2 is characterized by: The prediction model uses a stacked LSTM model, and the structure includes: input layer, LSTM layer 1, Dropout layer, LSTM layer 2, Dropout layer, fully connected layer and output layer.

4. The mutual assistance dispatching method of two stations with a DC bus with photovoltaic energy storage as claimed in claim 3 is characterized by: The net load and excess power generation are calculated by outputting load forecast and photovoltaic power generation forecast of the forecast model, and are used to realize energy storage scheduling of the two substations; The load forecasting formula is expressed as: P load (t:t+24)=LSTM l (X l (t),θ l ) Among them, LSTM l LSTM network for load forecasting, X l (t) represents the load forecast input feature vector at time t, θ l LSTM model l Parameters.

5. The mutual assistance dispatching method of two stations with a DC bus with photovoltaic energy storage as claimed in claim 4 is characterized by: The photovoltaic power generation prediction formula is expressed as: Ppυ(t:t+24)=LSTM p (X p (t),θ p ) Among them, LSTM p LSTM network for photovoltaic power generation prediction, X p (t) represents the input feature vector of photovoltaic power generation prediction at time t, and θp is the parameter of the LSTM model LSTMp, which is used to control the behavior of the photovoltaic power generation prediction model.

6. The mutual assistance dispatching method of two stations with a DC bus with photovoltaic energy storage as claimed in claim 5 is characterized by: The energy dispatching includes photovoltaic energy distribution and energy storage charging and discharging control. Based on the results of the forecasting model, the difference between the load forecast and the photovoltaic power generation forecast is calculated to obtain the predicted net load and excess power generation; If there is excess power generation, the load power of the charging pile in area 2 is monitored; When it is determined that the charging load power is greater than the first threshold, photovoltaic energy is extracted to area 2, and the extracted power is the load power minus the first threshold; when the photovoltaic power is insufficient, the energy storage is set to discharge, and the discharge power is the charging load power minus the first threshold minus the photovoltaic power; if the photovoltaic and energy storage power generation powers are both 0, at this time, the charging load is fully supplemented by area 2; When it is determined that the charging load power is less than the first threshold, the energy storage is set to charge according to the photovoltaic power. If the energy storage SOC is fully charged, the energy storage power is set to 0 and the energy storage device is shut down. In addition, considering the existence of faults, a fault clearing is performed each time the device is turned on to avoid startup failure. If the charging load is 0 and the energy storage is fully charged, all photovoltaics will be connected to the grid through area 1.

7. The mutual assistance dispatching method of two stations with a DC bus with photovoltaic energy storage as claimed in claim 6 is characterized by: The nighttime energy storage discharge is performed by setting a constant discharge power P constant , provide stable power output for the substation equipment at night, and ensure that all system parameters are initialized normally before night operation; According to the available energy and equipment loss of the energy storage equipment, the energy storage compensation and the power compensation of the power grid area 1 are dynamically adjusted, and the energy storage is given priority. When the energy storage is insufficient, it automatically switches to the power grid for power supply; When the current power percentage of the energy storage device SOC current Close to the set minimum power SOC min When the energy storage device is in a state of emergency, the discharge power of the energy storage device is gradually reduced, the energy output is reduced, and warnings are issued at different power thresholds to ensure that managers or automatic systems have enough time to make adjustments and emergency responses.

8. A mutual assistance dispatching system for two substations with a DC bus and photovoltaic energy storage, based on the mutual assistance dispatching method for two substations with a DC bus and photovoltaic energy storage as claimed in any one of claims 1 to 7, characterized in that: include, The prediction module is used to collect historical data on photovoltaic energy storage in the substation area and build a prediction model to predict future load and photovoltaic power generation; The prediction module is used to monitor the real-time load power of charging piles in the substation area and calculate the net load and excess power generation based on the predicted data and real-time data; A dispatching module, used to perform energy dispatching, including photovoltaic energy distribution and energy storage charging and discharging control, and handle the excess power generation; The night control module is used to set the nighttime energy storage discharge and loss compensation strategy, and continuously learn and optimize the control strategy.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the two-station mutual assistance scheduling method of a DC bus with photovoltaic energy storage as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for mutual assistance scheduling of two stations with a DC bus and photovoltaic energy storage as described in any one of claims 1 to 7 are implemented.

Images