Control method of transformer area side energy storage type electric energy quality comprehensive treatment device

Abstract

The invention relates to the technical field of power supply and consumption, in particular to a control method of a transformer area side energy storage type electric energy quality comprehensive treatment device. The method comprises the following steps: S1, setting a forward power lower limit, dynamically setting a power reverse sending judgment threshold value based on a dynamic threshold value adjustment algorithm, and generating a charging and discharging adjustment instruction when it is monitored that the output power of a transformer area reaches the power reverse sending judgment threshold value; s2, setting an out-of-limit threshold value according to the power supply capacity and operation requirements of a distribution transformer in the transformer area, entering a power out-of-limit control mode when monitoring that the power of the transformer area reaches the power out-of-limit threshold value, and generating an energy storage charging and discharging regulation instruction; and S3, establishing a source network load storage coordinated optimization model, and generating a control strategy according to economic operation and a photovoltaic maximization consumption target. By realizing reverse power transmission prevention, out-of-limit control and source-network-load-storage coordination optimization, the distributed photovoltaic bearing capacity and the access capacity of the alternating current charging pile of the electric vehicle can be remarkably improved, and powerful support is provided for large-scale application of distributed energy.

Description

Technical Field

[0001] The present invention relates to the technical field of power supply and consumption, and specifically, to a control method for a comprehensive energy quality management device with energy storage on the substation side. Background Art

[0002] The comprehensive energy quality management device with energy storage on the substation side is mainly used to address the growing energy quality problems in the power system. The large-scale access of low-voltage distributed photovoltaics to the distribution substation has significantly affected the safe and high-quality power supply of the distribution substation. Energy quality problems such as overvoltage, three-phase imbalance, and harmonics, as well as problems such as power reverse injection and reverse heavy overload, are prominent.

[0003] For single technical means for single types of problems, there are problems such as equipment redundancy, high investment costs, and complex operation and maintenance management. Currently, there is a lack of comprehensive and integrated efficient solutions and devices. The existing energy on the substation side does not have the functions of power flow control and coordinated optimization of the source-network-load-storage, and thus cannot support the development and application of distributed photovoltaics and electric vehicles. In view of this, a control method for a comprehensive energy quality management device with energy storage on the substation side is provided. Summary of the Invention

[0004] For single technical means for single types of problems, there are problems such as equipment redundancy, high investment costs, and complex operation and maintenance management. Currently, there is a lack of comprehensive and integrated efficient solutions and devices. The existing energy on the substation side does not have the functions of power flow control and coordinated optimization of the source-network-load-storage, and thus cannot support the development and application of distributed photovoltaics and electric vehicles.

[0005] To achieve the above object, the present invention aims to provide a control method for a comprehensive energy quality management device with energy storage on the substation side, including the following steps:

[0006] S1. Set a lower limit for the forward power. Combine the real-time photovoltaic output data, and dynamically set the power reverse injection judgment threshold based on the dynamic threshold adjustment algorithm. When it is monitored that the output power of the substation reaches the power reverse injection judgment threshold, enter the power reverse injection control mode and generate a charge-discharge adjustment instruction;

[0007] S2. Set an overlimit threshold according to the power supply capacity and operation requirements of the distribution transformer in the substation. When it is monitored that the power of the substation reaches the power overlimit threshold, enter the power overlimit control mode and generate a charge-discharge adjustment instruction for the energy storage;

[0008] S3. Establish a coordinated optimization model for the source-network-load-storage. Comprehensively consider the power generation of distributed photovoltaics, the power supply capacity of the substation, the real-time demand of the load, and the charge-discharge state of the energy storage. According to the set economic operation target and the maximum photovoltaic consumption target, optimize the power distribution and generate a control strategy for maximum photovoltaic consumption.

[0009] As a further improvement of this technical solution, in S1, the detailed operation steps of power reverse feed prevention control on the substation side are as follows:

[0010] S1.1. Use smart meters and sensors to continuously monitor the power generation of distributed energy in the substation area and the actual electricity consumption demand of users in real time;

[0011] S1.2. According to the power supply capacity and operation requirements of the distribution transformer in the substation area, preset the lower limit of forward power, and at the same time, combine the real-time photovoltaic output data to dynamically set the power reverse feed judgment threshold;

[0012] S1.3. When it is monitored that the output power of the substation area reaches the power reverse feed judgment threshold, the AC / DC converter control unit immediately enters the power reverse feed control mode, generates a charge-discharge adjustment instruction, and adjusts the current energy storage system.

[0013] As a further improvement of this technical solution, in S1.2, the arithmetic expression of the dynamic threshold adjustment algorithm is:

[0014] P reverse_threshold (t) = max(0, min(P available (t), P forward_min ));

[0015] In the formula, P reverse_threshold (t) represents the power reverse feed judgment threshold at time point t, P forward_min represents the preset lower limit of forward power, and P available (t) represents the available power supply capacity of the distribution transformer at time point t.

[0016] As a further improvement of this technical solution, in S1.3, the adjustment steps of the energy storage system are as follows:

[0017] S1.3.1. When both the photovoltaic output and the power reverse feed judgment threshold are 0, the energy storage system adjusts according to the current state: it reduces to 95% of the original discharge power in the discharge state and increases to 105% of the original charge power in the charge state;

[0018] S1.3.2. When the photovoltaic output is not zero and the power reverse feed judgment threshold is set to the lower limit of forward power, accurately control the charge-discharge power of the energy storage system according to the difference between the current output power of the substation area and the threshold to absorb the excess power.

[0019] As a further improvement of this technical solution, in S2, the specific steps of power overlimit control are as follows:

[0020] S2.1. By deploying smart meter devices on the low-voltage side of the distribution transformer in the substation area, continuously monitor the electrical parameters such as active power, reactive power, and apparent power in real time;

[0021] S2.2. Set the over-limit threshold according to the power supply capacity and operation requirements of the distribution transformer in the substation area. The forward over-limit threshold is usually the rated capacity of the distribution transformer;

[0022] S2.3. When it is monitored that the power in the substation area reaches the power over-limit threshold, the AC / DC converter control unit immediately enters the power over-limit control mode and generates an energy storage charge and discharge adjustment command.

[0023] As a further improvement of this technical solution, in S2.3, the operation steps for generating the energy storage charge and discharge adjustment command are as follows:

[0024] Calculate the amount of power that needs to be adjusted according to the current power level and the target power level. The expression is:

[0025] ΔP = P current - P target ;

[0026] In the formula, ΔP represents the power difference, P current represents the current power, and P target represents the target power; if ΔP>0, then perform the charging operation, and the charging power can be set to P charge = k·ΔP, where k is a coefficient used to adjust the charging rate; if ΔP<0, then perform the discharging operation, and the discharging power can be set to P discharge = |k·ΔP|.

[0027] As a further improvement of this technical solution, in S3, the specific steps for constructing the source-network-load-storage coordinated optimization model are as follows:

[0028] S3.1. Take economic operation and maximum utilization of photovoltaic power as the goals of the optimization model, and collect and sort out power parameters;

[0029] S3.2. Use the long short-term memory network deep learning algorithm to perform short-term and ultra-short-term predictions on the load demand and the generated power of distributed photovoltaics;

[0030] S3.3. Take the minimum total cost of one day and the maximum utilization of the generated power of distributed photovoltaics as the objective function, define a mathematical model with the safe operation and power balance of the substation area, distributed photovoltaics, energy storage, and load as constraints, and use the interior point method in linear programming to optimize and calculate the model to obtain the substation area power, photovoltaic power generation, energy storage charge and discharge power, load adjustment amount, and photovoltaic curtailment power;

[0031] S3.4. Analyze the effectiveness and feasibility of the optimization results, check whether all constraints are met, and evaluate the actual effect of the solution;

[0032] S3.5. Apply the finally determined optimization solution to the actual scenario, and continuously adjust and improve the model according to the feedback after implementation.

[0033] As a further improvement of this technical solution, in S3.2, the specific steps of using the long short-term memory network deep learning algorithm to perform short-term and ultra-short-term predictions on the load demand and the power generation power of distributed photovoltaics are as follows:

[0034] S3.2.1. Construct a long short-term memory network deep learning model architecture, and define the input layer, LSTM layer, and output layer; among them, the input layer designs the input dimension according to the selected time step, and the input feature vector usually includes historical load data and historical photovoltaic power generation data; the output layer determines the output dimension according to the prediction task.

[0035] S3.2.2. Use the collected power parameter data to train the model, and at the same time use the validation set to adjust the hyperparameters to prevent overfitting, and automatically stop training when the validation loss no longer improves.

[0036] S3.2.3. Evaluate the model performance based on the mean absolute error algorithm.

[0037] S3.2.4. Apply the trained model to the actual scenario, obtain the input data in real time, and generate predictions.

[0038] As a further improvement of this technical solution, in S3.2, the objective function expression for the minimum total cost in a day is:

[0039]

[0040] In the formula, z represents the objective of minimizing the total cost in a day, T represents the total number of time steps in a day, C grid (t) represents the price of purchasing electricity from the power grid at time t, represents the amount of electricity purchased from the power grid at time t, C sell (t) represents the price of selling electricity to the power grid at time t, represents the amount of electricity sold to the power grid at time t.

[0041] As a further improvement of this technical solution, in S3.2, the arithmetic expressions for the photovoltaic power generation power, energy storage charge and discharge power, substation area power, load regulation amount, and power constraint conditions are:

[0042] The power balance constraint is:

[0043] P pv (t)+P es,discharge (t)-P es,charge (t)+P net (t) = L(t);

[0044] In the formula, P pv (t) represents the output power of the distributed photovoltaic system at time point t, P es,discharge (t) represents the discharge power of the energy storage system at time point t, P es,charge (t) represents the charging power of the energy storage system at time point t, P net (t) represents the net exchange power with the external power grid at time point t, and L(t) represents the load demand at time point t;

[0045] The photovoltaic output limit condition is:

[0046] 0 ≤ P pv (t) ≤ P pv,max ;

[0047] In the formula, P pv,max represents the maximum power generation of the photovoltaic system;

[0048] The operating range of the energy storage system is:

[0049] P es,min ≤ P es ≤ P es,max ;

[0050] In the formula, P es,min represents the minimum discharge power or the lowest charging rate that the energy storage system can perform, P es represents the charge and discharge power of the energy storage system, P es,max is the maximum charge or discharge power that the energy storage system can reach;

[0051] The grid interaction power limit is:

[0052] P net,min ≤ P net (t) ≤ P net,max ;

[0053] In the formula, P net,min represents the minimum power value purchased from the power grid, P net,max is the maximum power value purchased from the power grid;

[0054] The load demand limit condition is:

[0055] L min (t) ≤ L(t) ≤ L max (t);

[0056] In the formula, L min (t) represents the minimum load adjustment amount that can be performed, L max (t) represents the maximum load adjustment amount that can be performed.

[0057] Compared with the prior art, the beneficial effects of the present invention:

[0058] By implementing power reverse power feed prevention, over-limit control, and coordinated optimization of the power grid, load, and energy storage, the power flow control ability is effectively improved. This not only helps to solve the power quality problems brought by the access of distributed photovoltaics and electric vehicles, but also significantly improves the bearing capacity of distributed photovoltaics and the access ability of AC electric vehicle chargers, providing strong support for the large-scale application of distributed energy. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is the overall method flow chart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0060] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the 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 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.

[0061] Embodiment:

[0062] Please refer to Figure 1 As shown, this embodiment provides a control method for a power quality comprehensive management device with energy storage on the substation side, including the following steps:

[0063] S1. Set the lower limit of forward power. Combine the real-time photovoltaic output data, and dynamically set the power reverse power feed judgment threshold based on the dynamic threshold adjustment algorithm. When it is monitored that the output power of the substation reaches the power reverse power feed judgment threshold, enter the power reverse power feed control mode and generate a charge and discharge adjustment instruction;

[0064] S2. Set the over-limit threshold according to the power supply capacity and operation requirements of the distribution transformer of the substation. When it is monitored that the power of the substation reaches the power over-limit threshold, enter the power over-limit control mode and generate a charge and discharge adjustment instruction for the energy storage;

[0065] S3. Establish a coordinated optimization model of the power grid, load, and energy storage. Comprehensively consider the power generation power of distributed photovoltaics, the power supply capacity of the substation, the real-time demand of the load, and the charge and discharge status of the energy storage. According to the set economic operation target and the maximum photovoltaic consumption target, optimize the power distribution and generate a maximum photovoltaic consumption control strategy.

[0066] In S1 of this embodiment, the detailed operation steps of the power reverse power feed prevention control on the substation side are as follows:

[0067] S1.1. Use smart meters and sensors to monitor the power generation of distributed energy in the substation and the actual electricity consumption demand of users in real time;

[0068] S1.2. According to the power supply capacity and operation requirements of the distribution transformer in the substation area, preset the lower limit of the forward power, and at the same time, combine the real-time photovoltaic output data to dynamically set the power reverse transmission judgment threshold;

[0069] S1.3. When it is monitored that the output power of the substation area reaches the power reverse transmission judgment threshold, the AC / DC converter control unit immediately enters the power reverse transmission control mode, generates a charge-discharge adjustment instruction, and adjusts the current energy storage system.

[0070] In S1.2 of this embodiment, the arithmetic expression of the dynamic threshold adjustment algorithm is:

[0071] P reverse_threshold (t) = max(0, min(P available (t), P forward_min ));

[0072] In the formula, P reverse_threshold (t) represents the power reverse transmission judgment threshold at time point t, P forward_min represents the preset lower limit of the forward power, and P available (t) represents the available power supply capacity of the distribution transformer at time point t.

[0073] In S1.3 of this embodiment, the steps for adjusting the energy storage system are as follows:

[0074] S1.3.1. When both the photovoltaic output and the power reverse transmission judgment threshold are 0, the energy storage system adjusts according to the current state: it reduces to 95% of the original discharge power in the discharge state and increases to 105% of the original charge power in the charge state;

[0075] S1.3.2. When the photovoltaic output is not zero and the power reverse transmission judgment threshold is set to the lower limit of the forward power, accurately control the charge-discharge power of the energy storage system according to the difference between the current output power of the substation area and the threshold to absorb the excess power.

[0076] In S2 of this embodiment, the specific steps of power overlimit control are as follows:

[0077] S2.1. By deploying intelligent meter devices on the low-voltage side of the distribution transformer in the substation area, continuously monitor the power parameters such as active power, reactive power, and apparent power in real time;

[0078] S2.2. According to the power supply capacity and operation requirements of the distribution transformer in the substation area, set the overlimit threshold, and the forward overlimit threshold is usually the rated capacity of the distribution transformer;

[0079] S2.3. When it is monitored that the power of the substation area reaches the power overlimit threshold, the AC / DC converter control unit immediately enters the power overlimit control mode and generates an energy storage charge-discharge adjustment instruction.

[0080] In S2.3 of this embodiment, the operation steps for generating the energy storage charge and discharge adjustment instruction are as follows:

[0081] According to the current power level and the target power level, calculate the power amount that needs to be adjusted. The expression is:

[0082] ΔP = P current -P target ;

[0083] In the formula, ΔP represents the power difference, P current represents the current power, and P target represents the target power; if ΔP>0, then perform the charging operation, and the charging power can be set to P charge = k·ΔP, where k is a coefficient used to adjust the charging rate; if ΔP<0, then perform the discharging operation, and the discharging power can be set to P discharge = |k·ΔP|.

[0084] In S3 of this embodiment, the specific steps for constructing the source-network-load-storage coordinated optimization model are as follows:

[0085] S3.1. Take economic operation and maximum PV power consumption as the objectives of the optimization model, and collect and sort out power parameters;

[0086] S3.2. Use the long short-term memory network deep learning algorithm to perform short-term and ultra-short-term predictions on the load demand and the power generation power of distributed PV;

[0087] S3.3. Take the minimum total cost of one day and the maximum consumption of the power generation power of distributed PV as the objective function, define a mathematical model with the safe operation and power balance of the substation area, distributed PV, energy storage, and load as constraints, and use the interior point method in linear programming to optimize and calculate the model to obtain the substation area power, PV power generation power, energy storage charge and discharge power, load adjustment amount, and PV curtailment power;

[0088] S3.4. Analyze the effectiveness and feasibility of the optimization results, check whether all constraints are met, and evaluate the actual effect of the scheme;

[0089] S3.5. Apply the finally determined optimization scheme to the actual scenario, and continuously adjust and improve the model according to the feedback after implementation.

[0090] In S3.2 of this embodiment, the specific steps for using the long short-term memory network deep learning algorithm to perform short-term and ultra-short-term predictions on the load demand and the power generation power of distributed PV are as follows:

[0091] S3.2.1. Construct the deep learning model architecture of the long short-term memory network, and define the input layer, LSTM layer, and output layer. Among them, the input layer designs the input dimension according to the selected time step, and the input feature vector usually includes historical load data and historical photovoltaic power generation data. The LSTM layer determines the number of hidden units to control the model complexity. The output layer determines the output dimension according to the prediction task. When making a single-step prediction, one neuron is output, and when making a multi-step prediction, the corresponding number of neurons is output.

[0092] S3.2.2. Use the collected power parameter data to train the model, and at the same time use the validation set to adjust the hyperparameters to prevent overfitting. When the validation loss no longer improves, the training is automatically stopped.

[0093] S3.2.3. Evaluate the model performance based on the mean absolute error algorithm.

[0094] S3.2.4. Apply the trained model to the actual scenario, obtain the input data in real time, and generate predictions.

[0095] In S3.3 of this embodiment, the objective function expression for minimizing the total cost of one day is:

[0096]

[0097] In the formula, z represents the objective of minimizing the total cost within one day, T represents the total number of time steps in one day, C grid (t) represents the price of purchasing electricity from the power grid at time t, represents the amount of electricity purchased from the power grid at time t, C sell (t) represents the price of selling electricity to the power grid at time t, represents the amount of electricity sold to the power grid at time t.

[0098] In S3.2 of this embodiment, the arithmetic expressions for photovoltaic power generation, energy storage charge and discharge power, substation area power, load regulation amount, and power constraint conditions are:

[0099] The power balance constraint is:

[0100] P pv (t)+P es,discharge (t)-P es,charge (t)+P net (t)=L(t);

[0101] In the formula, P pv (t) represents the output power of the distributed photovoltaic system at time point t, P es,discharge (t) represents the discharge power of the energy storage system at time point t, P es,charge (t) represents the charging power of the energy storage system at time point t, P net(t) represents the net exchange power with the external power grid at time point t, and L(t) represents the load demand at time point t;

[0102] The photovoltaic output limit condition is:

[0103] 0 ≤ P pv (t) ≤ P pv,max ;

[0104] In the formula, P pv,max represents the maximum power generation of the photovoltaic system;

[0105] The operating range of the energy storage system is:

[0106] P es,min ≤ P es ≤ P es,max ;

[0107] In the formula, P es,min represents the minimum discharge power or the lowest charging rate that the energy storage system can perform, P es represents the charge-discharge power of the energy storage system, P es,max is the maximum charge or discharge power that the energy storage system can reach;

[0108] The grid interaction power limit is:

[0109] P net,min ≤ P net (t) ≤ P net,max ;

[0110] In the formula, P net,min represents the minimum power value purchased from the grid, P net,max is the maximum power value purchased from the grid;

[0111] The load demand limit condition is:

[0112] L min (t) ≤ L(t) ≤ L max (t);

[0113] In the formula, L min (t) represents the minimum load adjustment amount that can be performed, L max (t) represents the maximum load adjustment amount that can be performed.

[0114] In this example S3.1, with the goal of maximizing the consumption of the generated power of distributed photovoltaics, that is, the objective function of minimizing the unused photovoltaic power generation can be expressed as:

[0115] min∑ t (P PV,t -P load,t -P ch,t +P disch,t) + ;

[0116] Wherein, P PV,t represents the photovoltaic output power at time t, P load,t represents the load demand at time t, P ch,t represents the charging power of the energy storage system at time t, P disch,t represents the discharging power of the energy storage system at time t, () + represents taking the positive part, that is, when the photovoltaic output exceeds the current load demand and the energy storage absorption capacity, the excess part is calculated.

[0117] In this example S3.1, with the goal of maximizing the consumption of the distributed photovoltaic power generation, that is, minimizing the constraint condition of the unused photovoltaic power generation amount is:

[0118] The power balance constraint is:

[0119] P PV,t = P load,t + P ch,t - P disch,t + Loss t ;

[0120] Wherein, Loss t represents the system energy loss amount at a specific time point t;

[0121] The constraint conditions of the energy storage system are:

[0122] 0 ≤ P ch,t ≤ P ch,max ;

[0123] SOC min ≤ SOC(t) ≤ SOC max ;

[0124] Wherein, P ch,max represents the maximum charging power limit of the energy storage device, SOC min is the minimum allowable state of charge of the energy storage system, SOC(t) is the actual state of charge of the energy storage system at time point t, SOC max is the maximum allowable state of charge of the energy storage system;

[0125] The constraint condition of the substation area power supply capacity is:

[0126] P grid,t ≤ P grid,max ;

[0127] Wherein, P grid,t represents the power obtained from or transmitted to the power grid at time point t, P grid,max represents the maximum limit of the power grid transmission capacity.

[0128] The foregoing has shown and described the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and the above embodiments and the descriptions in the specification are only preferred examples of the present invention, and are not used to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements fall within the scope of the present invention claimed. The scope of the present invention claimed is defined by the appended claims and their equivalents.

Claims

1. A control method for a comprehensive power quality governance device with energy storage on the substation side, characterized in that, It includes the following steps: S1. Set a lower limit for forward power. Combine real-time photovoltaic output data and dynamically set the power reverse transmission judgment threshold based on the dynamic threshold adjustment algorithm. When it is monitored that the output power of the distribution area reaches the power reverse transmission judgment threshold, enter the power reverse transmission control mode and generate a charge-discharge adjustment instruction; S2. Set an overlimit threshold according to the power supply capacity and operation requirements of the distribution transformer in the distribution area. When it is monitored that the power of the distribution area reaches the power overlimit threshold, enter the power overlimit control mode and generate a charge-discharge adjustment instruction for the energy storage; S3. Establish a coordinated optimization model for the source, grid, load, and energy storage. Comprehensively consider the power generation of distributed photovoltaics, the power supply capacity of the distribution area, the real-time demand of the load, and the charge-discharge state of the energy storage. According to the set economic operation target and the maximum photovoltaic consumption target, optimize the power distribution and generate a maximum photovoltaic consumption control strategy.

2. The control method of the energy storage type power quality comprehensive governance device on the substation side according to claim 1, wherein: In the above S1, the detailed operation steps for power reverse transmission prevention control on the distribution area side are as follows: S1.

1. Use smart meters and sensors to monitor the power generation of distributed energy and the actual electricity consumption demand of users in the distribution area in real time; S1.

2. According to the power supply capacity and operation requirements of the distribution transformer in the distribution area, preset a lower limit for forward power, and at the same time combine real-time photovoltaic output data to dynamically set the power reverse transmission judgment threshold; S1.

3. When it is monitored that the output power of the distribution area reaches the power reverse transmission judgment threshold, the AC / DC converter control unit immediately enters the power reverse transmission control mode, generates a charge-discharge adjustment instruction, and adjusts the current energy storage system.

3. The control method of the energy storage type power quality comprehensive management device on the substation side according to claim 2, characterized in that: In the above S1.2, the arithmetic expression of the dynamic threshold adjustment algorithm is: P reverse_threshold (t) = max(0, min(P available (t), P forward_min )); Where P reverse_threshold (t) represents the power reverse transmission judgment threshold at time point t, and P forward_min represents the set lower limit of the forward power, and P available (t) represents the available power supply capacity of the distribution transformer at time point t.

4. The control method of the energy storage type power quality comprehensive management device on the substation side according to claim 2, wherein: In the above S1.3, the steps for adjusting the energy storage system are as follows: S1.3.

1. When both the photovoltaic output and the power reverse transmission judgment threshold are 0, the energy storage system adjusts according to the current state: reduce to 95% of the original discharge power in the discharge state and increase to 105% of the original charge power in the charge state; S1.3.

2. When the photovoltaic output is not zero and the power reverse transmission judgment threshold is set to the lower limit of forward power, accurately control the charge-discharge power of the energy storage system according to the difference between the current output power of the distribution area and the threshold to absorb the excess power.

5. The control method of the energy storage type power quality comprehensive management device on the substation side according to claim 1, characterized in that: In the above S2, the specific steps for power overlimit control are as follows: S2.

1. Deploy smart meter devices on the low-voltage side of the distribution transformer in the distribution area to continuously monitor power parameters such as active power, reactive power, and apparent power in real time; S2.

2. Set an overlimit threshold according to the power supply capacity and operation requirements of the distribution transformer in the distribution area. The forward overlimit threshold is usually the rated capacity of the distribution transformer; S2.

3. When it is monitored that the power of the distribution area reaches the power overlimit threshold, the AC / DC converter control unit immediately enters the power overlimit control mode and generates a charge-discharge adjustment instruction for the energy storage.

6. The control method of the energy storage type power quality comprehensive management device on the substation side according to claim 5, characterized in that: In the above S2.3, the operation steps for generating a charge-discharge adjustment instruction for the energy storage are as follows: Calculate the amount of power that needs to be adjusted according to the current power level and the target power level. The expression is: ΔP = P current - P target ; Wherein, ΔP represents the power difference, P current represents the current power, and P target represents the target power; if ΔP > 0, a charging operation is performed, and the charging power can be set to P charge = k·ΔP, where k is a coefficient for adjusting the charging rate; if ΔP < 0, a discharging operation is performed, and the discharging power can be set to P discharge = |k·ΔP|.

7. The control method of the energy storage type power quality comprehensive management device on the substation side according to claim 1, characterized in that: In the above S3, the specific steps for constructing a coordinated optimization model for the source, grid, load, and energy storage are as follows: S3.

1. Take economic operation and maximum photovoltaic consumption as the goals of the optimization model, and collect and sort out power parameters; S3.

2. Use the long short-term memory network deep learning algorithm to perform short-term and ultra-short-term predictions on the load demand and the power generation of distributed photovoltaic; S3.

3. Take the minimum total cost of one day and the maximum consumption of the power generation of distributed photovoltaic as the objective function, define a mathematical model with the safe operation and power balance of the substation area, distributed photovoltaic, energy storage, and load as constraints, and use the interior point method in linear programming to optimize and calculate the model to obtain the substation area power, photovoltaic power generation, energy storage charge and discharge power, load regulation amount, and photovoltaic curtailment power; S3.

4. Analyze the effectiveness and feasibility of the optimization results, check whether all constraints are met, and evaluate the actual effect of the plan; S3.

5. Apply the finally determined optimization plan to the actual scenario, and continuously adjust and improve the model according to the feedback after implementation.

8. A control method for a comprehensive power quality improvement device with energy storage on the substation side according to claim 7, characterized in that: In the above S3.2, the specific steps of using the long short-term memory network deep learning algorithm to perform short-term and ultra-short-term predictions on the load demand and the power generation of distributed photovoltaic are as follows: S3.2.

1. Build the architecture of the long short-term memory network deep learning model, and define the input layer, LSTM layer, and output layer; among them, the input layer designs the input dimension according to the selected time step, and the input feature vector usually includes historical load data and historical photovoltaic power generation data; the output layer determines the output dimension according to the prediction task; S3.2.

2. Use the collected power parameter data to train the model, and at the same time use the validation set to adjust the hyperparameters to prevent overfitting, and automatically stop training when the validation loss no longer improves; S3.2.

3. Evaluate the model performance based on the mean absolute error algorithm; S3.2.

4. Apply the trained model to the actual scenario, obtain the input data in real time, and generate predictions.

9. The control method of the power quality comprehensive governance device with energy storage on the substation side according to claim 7, characterized in that: In the above S3.3, the expression of the objective function for the minimum total cost of one day is: where z represents the goal of minimizing the total cost within a day, T represents the total number of time steps in a day, C grid (t) represents the price of purchasing electricity from the power grid at time t, represents the amount of electricity purchased from the power grid at time t, C sell (t) represents the price of selling electricity to the power grid at time t, represents the amount of electricity sold to the power grid at time t.

10. The control method of the energy storage type power quality comprehensive management device on the substation side according to claim 7, characterized in that: In the above S3.2, the arithmetic expressions of the photovoltaic power generation, energy storage charge and discharge power, substation area power, load regulation amount, and power constraint conditions are: The power balance constraint is: P pv (t) + P es,discharge (t) - P es,charge (t) + P net (t) = L(t); wherein, P pv (t) represents the output power of the distributed PV system at time point t, P es,discharge (t) represents the discharge power of the energy storage system at time point t, P es,charge (t) represents the charging power of the energy storage system at time point t, P net (t) represents the net exchange power with the external power grid at time point t, and L(t) represents the load demand at time point t; The photovoltaic output limit condition is: 0 ≤ P pv (t) ≤ P pv,max ; Wherein, P pv,max represents the maximum power generation of the photovoltaic system; The operating range of the energy storage system is: P es,min ≤P es ≤P es,max ; Wherein, P es,min represents the minimum discharge power or the lowest charging rate that the energy storage system can perform, P es represents the charge-discharge power of the energy storage system, and P es,max is the maximum charge or discharge power that the energy storage system can reach; The grid interaction power limit is: P net,min ≤P net (t)≤P net,max ; Wherein, P net,min represents the minimum power value purchased from the power grid, and P net,max is the maximum power value purchased from the power grid; The load demand limit condition is: L min L(t) ≤ L(t) ≤ L max (t); wherein, L min (t) represents the minimum load adjustment amount that can be made, and L max (t) represents the maximum load adjustment amount that can be made.

Images