Active power distribution network coordination control method

Abstract

The invention relates to the technical field of power distribution network coordination control, in particular to an active power distribution network coordination control method. The method comprises the following steps: S1, collecting power grid operation data; s2, constructing a power grid model based on the collected data and an extended Kalman filtering algorithm, predicting the power grid state in a period of time in the future based on the power grid model, and determining the current state of the power grid by using the extended Kalman filtering algorithm; s3, making a control strategy according to the current and predicted states of the power grid; and S4, issuing the formulated control strategy to the control node, monitoring the execution effect of the control node, and adjusting the control strategy in real time to ensure that the power grid operates in an optimal state. According to the design, the power grid operation data is collected in real time, and the state of the power grid is accurately estimated and predicted by using the extended Kalman filtering algorithm, so that the power distribution network can quickly respond to changes of various operation conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of coordinated control of distribution networks, and in particular to an active coordinated control method of distribution networks. Background Art

[0002] With the popularity of renewable energy (such as solar photovoltaic panels and wind power generation), a large number of distributed power sources have been connected to the distribution network. The randomness and intermittency of these power sources pose challenges to the stable operation of the power grid; the access of distributed power sources may cause local voltage increases, especially during peak hours or when DG generation exceeds local load demand. In addition, the unpredictability of DG may also cause power quality problems; with the addition of more controllable resources, such as energy storage systems and controllable loads, the structure of the distribution network becomes more complex; in summary, an active distribution network coordinated control method is provided. Summary of the invention

[0003] The object of the present invention is to provide an active distribution network coordinated control method to solve the problems of distributed power source growth, network complexity, voltage fluctuation and quality proposed in the above background technology.

[0004] To achieve the above object, the present invention provides an active power distribution network coordinated control method, comprising the following steps:

[0005] S1. Collect power grid operation data;

[0006] S2. Based on the collected data, a power grid model is constructed based on the extended Kalman filter algorithm, the power grid state in the future is predicted based on the power grid model, and the extended Kalman filter algorithm is used to determine the current state of the power grid;

[0007] S3, formulate control strategies based on the current and predicted status of the power grid;

[0008] S4. Send the formulated control strategy to each execution unit, monitor its execution effect and adjust the control strategy in real time to ensure that the power grid operates in the optimal state.

[0009] As a further improvement of the technical solution, in S1, the power grid operation data includes:

[0010] Voltage, current, frequency, power factor, and power generation of distributed generation.

[0011] As a further improvement of the technical solution, in S2, the current state of the power grid is determined by using a Kalman filter algorithm, including the following steps:

[0012] S2.1. Collect historical data, current network topology and load conditions to make a preliminary estimate of the grid status;

[0013] S2.2. Use the extended Kalman filter algorithm to solve the state estimation problem and find the power grid state through the optimization process.

[0014] S2.3. A model of the power grid is constructed based on the extended Kalman filter algorithm. The model includes the topological structure and component parameters of the power grid, establishes the relationship between each measurement point, and forms a complete power grid model for predicting the future power grid status.

[0015] As a further improvement of the technical solution, in S2.3, the extended Kalman filter algorithm is specifically:

[0016] Prediction stage:

[0017] ;

[0018] ;

[0019] Update phase:

[0020] ;

[0021] ;

[0022] ;

[0023] in, Indicates at time The predicted value of the power grid status; Indicates time The optimal grid state estimation value; Indicates at time Control input of represents the state transition function; Indicates time The state prediction error covariance matrix of Represents the state transfer matrix exist The Jacobian matrix at ; At the moment The optimal estimation error covariance matrix of ; Representation Matrix The transpose of represents the covariance matrix of the process noise; represents the Kalman gain; Represents the measurement matrix exist The Jacobian matrix at ; Representation Matrix The transpose of represents the covariance matrix of the measurement noise; Indicates at time The best estimate of ; Indicates at time The measured value of represents the measurement function; Indicates at time The optimal estimation error covariance matrix of .

[0024] As a further improvement of the technical solution, in S3, the formulating of the control strategy includes:

[0025] Adjust the output of distributed generation to meet load demand;

[0026] Optimize grid topology to reduce losses;

[0027] Controlling the charging and discharging of energy storage devices is used to smooth the load curve.

[0028] As a further improvement of the technical solution, the step of adjusting the output of the distributed power source comprises the following steps:

[0029] S3.1. Analyze the current grid demand based on the current grid status data, including the demand for frequency stability, voltage level, and load balance;

[0030] S3.2. Evaluate the actual output of each distributed power source to understand whether the output is close to the maximum and minimum power limits;

[0031] S3.3. Calculate the optimal output power of each distributed power source.

[0032] As a further improvement of this technical solution, the following steps are included:

[0033] S3.4. Use matrix operations to represent the physical connections of the power grid and build a digital twin model of the power grid in the simulation software;

[0034] S3.5. Set optimization goals according to the specific needs of the power grid, including minimizing network losses, improving power supply reliability, and reducing operating costs, taking into account the actual limitations of power grid operation, including voltage deviation range and equipment capacity limitations;

[0035] S3.6. Select the particle swarm optimization algorithm to find the best topology configuration and generate a new power grid topology structure based on the results of the optimization algorithm.

[0036] As a further improvement of the technical solution, in S4.3, the particle swarm optimization algorithm is specifically:

[0037] ;

[0038] ;

[0039] in, Indicates the topology configuration status; Indicates the configuration of the existing topology The amount of change; represents the inertia weight; Represents particles How much tendency there is to return to a previously discovered good configuration; Represents particles how much tends to follow the best configuration found by the group; and Represents a random number, used to increase the randomness of the search process; Represents particles the best position achieved since the search began; represents the best position found so far by all particles in the entire particle swarm; Indicates the number of iterations.

[0040] As a further improvement of the technical solution, the control of charging and discharging of the energy storage device includes the following steps:

[0041] S3.7. Obtain load data for a period of time, including the distribution of daily electricity consumption, especially the load curves during peak and off-peak hours, and analyze wind and solar power generation data;

[0042] S3.8. Based on historical data, predict future load curves, especially at different times of the day, predict the power generation of renewable energy, especially its volatility, integrate load forecasting and power generation forecasting, and analyze and predict periods of supply and demand imbalance;

[0043] S3.9. Calculate the required energy storage capacity based on demand forecasts and store excess electricity during off-peak hours;

[0044] S3.10. Select the load off-peak period for charging, determine the charging rate based on the battery status SOC and the expected load demand, and set the SOC upper limit ; Select the peak load period for discharge, determine the discharge rate according to the load demand and the remaining battery capacity, and set the SOC lower limit .

[0045] As a further improvement of the technical solution, in S4, the formulated control strategy is sent to each control node, including the following steps:

[0046] S4.1. Convert the formulated control strategy into a control instruction format, where the control instructions include adjusting the output power of the distributed power source, changing the grid switch state to optimize the topology structure, and controlling the charging and discharging of the energy storage device;

[0047] S4.2, select the communication protocol Modbus to ensure that the command can be sent to the corresponding execution unit;

[0048] S4.3. Control instructions are sent to each control node through the automatic control system, and the instructions are executed to adjust the operating status of the power grid.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] 1. In this active distribution network coordinated control method, by collecting grid operation data in real time and using the extended Kalman filter algorithm to accurately estimate and predict the grid state, the distribution network can quickly respond to changes in various operating conditions. This not only helps to promptly discover and prevent potential problems, but also ensures that the grid can maintain efficient and stable operation under any circumstances by dynamically adjusting the output of distributed power sources, optimizing the grid topology, and controlling the charging and discharging strategies of energy storage devices, thereby improving the reliability and economy of the overall grid.

[0051] 2. In this active distribution network coordinated control method, by effectively managing distributed power sources, analyzing the power generation characteristics of renewable energy resources, and combining load forecasting to guide the charging and discharging decisions of the energy storage system, the supply and demand fluctuations caused by the intermittent nature of renewable energy can be effectively smoothed. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 The figure is a flow chart of the overall method of the present invention. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] Example:

[0055] See also Figure 1 As shown, this embodiment provides an active distribution network coordinated control method, comprising the following steps:

[0056] S1. Collect power grid operation data;

[0057] In this embodiment, the power grid operation data includes:

[0058] Voltage, current, frequency, power factor, and power generation of distributed generation.

[0059] S2. Based on the collected data, a power grid model is constructed based on the extended Kalman filter algorithm, the power grid state in the future is predicted based on the power grid model, and the extended Kalman filter algorithm is used to determine the current state of the power grid;

[0060] Among them, the grid status includes node voltage, branch current, system frequency, power distribution, load conditions, and power generation of distributed generation;

[0061] In this embodiment, the Kalman filter algorithm is used to determine the current state of the power grid, including the following steps:

[0062] S2.1. Collect historical data, current network topology and load conditions to make a preliminary estimate of the grid status;

[0063] S2.2, use the extended Kalman filter algorithm to solve the state estimation problem, and find the power grid state through the optimization process, which should satisfy all known measurement values ​​and is the solution that minimizes the residual sum of squares;

[0064] S2.3. Build a grid model based on the extended Kalman filter algorithm. The model includes the topology of the grid (i.e., the nodes in the grid and the lines connecting them), component parameters (resistance, reactance, etc.), and establish the relationship between each measurement point (voltage, current sensor) to form a complete grid model for predicting the future grid status.

[0065] Among them, in active distribution networks, system models are often nonlinear, especially when it comes to the dynamic behavior of distributed power sources (such as photovoltaics, wind power generation) and energy storage devices. The extended Kalman filter algorithm enables the Kalman filter to be applied to these nonlinear systems by locally linearizing the nonlinear function at each moment; the extended Kalman filter algorithm provides a recursive update mechanism that can quickly update the state estimate after receiving new measurement data, which is particularly important for systems that require real-time control; the real-time nature of the extended Kalman filter algorithm makes it very suitable for real-time state estimation and control of active distribution networks, and can timely adjust the output of distributed power sources, optimize the grid topology, etc.; the extended Kalman filter algorithm is specifically:

[0066] Prediction stage:

[0067] ;

[0068] ;

[0069] Update phase:

[0070] ;

[0071] ;

[0072] ;

[0073] in, Indicates at time The predicted value of the power grid status; Indicates time The optimal grid state estimation value; Indicates at time Control input of It represents the state transfer function, which describes the change of the grid state over time, including node voltage, current, and distributed power output; Indicates time The state prediction error covariance matrix of Represents the state transfer matrix exist The Jacobian matrix at ; At the moment The optimal estimation error covariance matrix of ; Representation Matrix The transpose of The covariance matrix representing the process noise represents the uncertainty and randomness in the changes of the power grid state; represents the Kalman gain; Represents the measurement matrix exist The Jacobian matrix at ; Representation Matrix The transpose of represents the covariance matrix of the measurement noise; Indicates at time The best estimate of ; Indicates at time The measured values ​​include the actual measured data such as voltage and current of each node in the power grid; represents the measurement function, which describes how the state vector is mapped to the measurement value; Indicates at time The optimal estimation error covariance matrix of .

[0074] S3, formulate control strategies based on the current and predicted status of the power grid;

[0075] In this embodiment, the formulation of the control strategy includes:

[0076] Adjust the output of distributed generation to meet load demand;

[0077] Optimize grid topology to reduce losses;

[0078] Controlling the charging and discharging of energy storage devices is used to smooth the load curve.

[0079] Wherein, adjusting the output of the distributed power supply includes the following steps:

[0080] S3.1. Analyze the current grid demand based on the current grid status data, including the demand for frequency stability, voltage level, and load balance;

[0081] S3.2. Evaluate the actual output of each distributed power source to understand whether the output is close to the maximum and minimum power limits;

[0082] S3.3. Calculate the optimal output power of each distributed power source to meet load demand and improve system stability and efficiency;

[0083] The optimal output power of each distributed power source is calculated including load demand, frequency and voltage stability. If there are h distributed power units, each unit has its own cost function , which describes the power generation cost as a function of output power changes, and at the same time, the total load demand of the power grid needs to be met , as well as any technical constraints such as voltage levels and frequency stability;

[0084] S3.31. Define the objective function: minimize the total cost

[0085] ;

[0086] in is a quadratic function;

[0087] S3.32, the total load demand is:

[0088] ;

[0089] The power output limits are:

[0090] ;

[0091] The voltage and frequency are maintained within the allowable range:

[0092] ;

[0093] .

[0094] Further, optimizing the grid topology includes the following steps:

[0095] S3.4. Use matrix operations to represent the physical connections of the power grid and build a digital twin model of the power grid in the simulation software for subsequent optimization tests;

[0096] S3.5. Set optimization goals according to the specific needs of the power grid, including minimizing network losses, improving power supply reliability, and reducing operating costs, taking into account the actual limitations of power grid operation, including voltage deviation range and equipment capacity limitations;

[0097] S3.6, select the particle swarm optimization algorithm to find the best topology configuration, and generate a new power grid topology structure according to the results of the optimization algorithm;

[0098] Among them, the particle swarm optimization algorithm has good adaptability to nonlinear and complex constraint problems. In the active distribution network, there are many variables such as distributed generation (DG), energy storage system (ESS) and renewable energy, which will increase the complexity and nonlinearity of the system; the particle swarm optimization algorithm can flexibly adjust its parameters to adapt to different problem characteristics, so that it can still effectively find the best solution when facing the variable constraints in the power grid; through the method of swarm intelligence, the particle swarm optimization algorithm can search in the entire solution space, which helps to avoid falling into the local optimal solution. This is particularly important for the optimization of active distribution networks, because the local optimum may not bring about the maximization of overall performance; the particle swarm optimization algorithm is specifically:

[0099] ;

[0100] ;

[0101] in, Indicates the topology configuration status; Indicates the configuration of the existing topology The amount of change; represents the inertia weight, which is used to balance the trust in the existing configuration and the need to explore new configurations; Represents particles How much tendency there is to return to a previously discovered good configuration; Represents particles how much tends to follow the best configuration found by the group; and Represents a random number, used to increase the randomness of the search process; Represents particles The best position reached since the start of the search, i.e. the best topological configuration; It represents the best position found so far by all particles in the entire particle swarm, and represents the optimal solution found in the entire search process; Indicates the number of iterations.

[0102] Further, controlling the charging and discharging of the energy storage device includes the following steps:

[0103] S3.7. Obtain load data over a period of time, including the distribution of daily electricity consumption, especially the load curves during peak and off-peak hours, and analyze wind and solar power generation data. For power grids connected to renewable energy (such as wind and solar energy), it is necessary to collect power generation data of these energy sources in order to understand their power generation characteristics and predictability;

[0104] S3.8. Based on historical data, predict future load curves, especially at different times of the day, predict the power generation of renewable energy, especially its volatility, integrate load forecasting and power generation forecasting, and analyze and predict periods of supply and demand imbalance;

[0105] S3.9. Calculate the required energy storage capacity based on demand forecasts to ensure that sufficient power can be provided during peak hours and excess power can be stored during off-peak hours;

[0106] S3.10. Select the off-peak period (cheap electricity price period) for charging, determine the charging rate according to the battery status SOC and the expected load demand, and set the SOC upper limit , to prevent battery damage caused by overcharging; select peak load period (when electricity is expensive) for discharge, determine the discharge rate based on load demand and remaining battery capacity, and set the SOC lower limit , to prevent excessive discharge from affecting battery life.

[0107] The specific steps of determining the charging rate include:

[0108] S3.101, obtain the current SOC and health status of the battery, and obtain the load demand at the current moment; define the maximum discharge power and minimum SOC threshold of the battery ( ) and the maximum SOC threshold ( );

[0109] S3.102, read the current SOC of the battery. If the current SOC is lower than the set minimum threshold ( ), then stop discharging and wait for charging; if the current SOC is higher than the set maximum threshold ( ), then stop discharging and enter standby mode; if the current SOC is within the allowed range ( ), then proceed to the next step;

[0110] S3.103. Read the current load demand. If the current load demand is greater than the normal power supply capacity of the power grid, calculate the additional power required. If the additional power required is greater than zero, the battery needs to be discharged to supplement this power. Otherwise, no discharge is required.

[0111] S3.104. If the additional power required is greater than the maximum discharge power of the battery, the actual discharge power is set to the maximum discharge power; otherwise, the actual discharge power is the additional power required; based on the actual power discharged from the battery and the time length of the load demand, estimate the time the battery can continue to discharge;

[0112] S3.105, update the battery SOC according to the actual discharge power and discharge time, ;in, Indicates the discharge time interval, Indicates the rated capacity of the battery. Indicates the new SOC, Indicates the actual discharge power, Indicates the discharge efficiency of the battery.

[0113] The specific steps to determine the discharge rate include:

[0114] S3.1011. Obtain the load forecast data for the current time and the future period, obtain the current SOC of the battery, and define the maximum charging power minimum SOC threshold of the battery ( ) and the maximum SOC threshold ( );

[0115] S3.1021, read the current SOC of the battery. If the current SOC is higher than the set maximum threshold ( ), then stop charging and wait for discharge; if the current SOC is lower than the set minimum threshold ( ), you need to start charging as soon as possible; if the current SOC is within the allowed range ( ), then it can be decided whether charging is needed according to the load demand;

[0116] S3.1031. Based on the load forecast data, analyze the load change trend in the future period. If the load is expected to increase, it may be necessary to charge in advance so that the battery can be used for power supply during peak hours. If the load is expected to decrease, emergency charging may not be required, or charging can be selected when the load is lower.

[0117] S3.1041. Calculate the additional energy required to meet future load demand based on future load demand and current SOC. , ;

[0118] S3.1051. Calculate the required charging power based on the required additional energy and the expected charging time window , ; If the required charging power is greater than the maximum charging power of the battery, the actual charging power is set to the maximum charging power of the battery; otherwise, the actual charging power is the required charging power;

[0119] S3.1061, update the battery SOC according to the actual charging power and charging time, ;in, Indicates the charging time interval, Indicates the rated capacity of the battery. Indicates the actual power input to the battery.

[0120] S4. Send the formulated control strategy to each execution unit, monitor its execution effect and adjust the control strategy in real time to ensure that the power grid operates in the optimal state;

[0121] In this embodiment, the formulated control strategy is sent to each control node, including the following steps:

[0122] S4.1. Convert the formulated control strategy into a control instruction format, where the control instructions include adjusting the output power of the distributed power source, changing the grid switch state to optimize the topology structure, and controlling the charging and discharging of the energy storage device;

[0123] S4.2, select the communication protocol Modbus to ensure that the command can be sent to the corresponding execution unit;

[0124] S4.3. Control instructions are sent to various control nodes through the automated control system. The control nodes include intelligent switches, distributed energy management systems, etc., and the instructions are executed to adjust the operation status of the power grid.

[0125] The above shows and describes 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 descriptions are only preferred examples of the present invention, and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. An active distribution network coordinated control method, characterized in that: The following steps are involved: S1. Collect power grid operation data; S2. Based on the collected data, a power grid model is constructed based on the extended Kalman filter algorithm, the power grid state in the future is predicted based on the power grid model, and the extended Kalman filter algorithm is used to determine the current state of the power grid; S3, formulate control strategies based on the current and predicted status of the power grid; S4. Send the formulated control strategy to the control node, monitor its execution effect and adjust the control strategy in real time to ensure that the power grid operates in the optimal state.

2. The active distribution network coordinated control method according to claim 1, characterized in that: In S1, the power grid operation data includes: Voltage, current, frequency, power factor, and power generation of distributed generation.

3. The active distribution network coordinated control method according to claim 1, characterized in that: In S2, the current state of the power grid is determined using a Kalman filter algorithm, including the following steps: S2.

1. Collect historical data, current network topology and load conditions to make a preliminary estimate of the grid status; S2.2, use the extended Kalman filter algorithm to solve the state estimation problem and find the power grid state through the optimization process; S2.

3. A model of the power grid is constructed based on the extended Kalman filter algorithm. The model includes the topological structure and component parameters of the power grid, establishes the relationship between each measurement point, and forms a complete power grid model for predicting the future power grid status.

4. The active distribution network coordinated control method according to claim 3 is characterized in that: In S2.3, the extended Kalman filter algorithm is specifically: Prediction stage: ； ； Update phase: ； ； ； in, Indicates at time The predicted value of the power grid status; Indicates time The optimal grid state estimation value; Indicates at time Control input of represents the state transition function; Indicates time The state prediction error covariance matrix of Represents the state transfer matrix exist The Jacobian matrix at ; At the moment The optimal estimation error covariance matrix of ; Representation Matrix The transpose of represents the covariance matrix of the process noise; represents the Kalman gain; Represents the measurement matrix exist The Jacobian matrix at ; Representation Matrix The transpose of represents the covariance matrix of the measurement noise; Indicates at time The best estimate of ; Indicates at time The measured value of represents the measurement function; Indicates at time The optimal estimation error covariance matrix of .

5. The active distribution network coordinated control method according to claim 1, characterized in that: In S3, the formulating of the control strategy includes: Adjust the output of distributed generation to meet load demand; Optimize grid topology to reduce losses; Controlling the charging and discharging of energy storage devices is used to smooth the load curve.

6. The active distribution network coordinated control method according to claim 5, characterized in that: The method of adjusting the output of the distributed power supply comprises the following steps: S3.

1. Analyze the current grid demand based on the current grid status data, including the demand for frequency stability, voltage level, and load balance; S3.

2. Evaluate the actual output of each distributed power source to understand whether the output is close to the maximum and minimum power limits; S3.

3. Calculate the optimal output power of each distributed power source.

7. The active distribution network coordinated control method according to claim 5, characterized in that: The optimization of the power grid topology structure comprises the following steps: S3.

4. Use matrix operations to represent the physical connections of the power grid and build a digital twin model of the power grid in the simulation software; S3.

5. Set optimization goals according to the specific needs of the power grid, including minimizing network losses, improving power supply reliability, and reducing operating costs, taking into account the actual limitations of power grid operation, including voltage deviation range and equipment capacity limitations; S3.

6. Select the particle swarm optimization algorithm to find the best topology configuration and generate a new power grid topology structure based on the results of the optimization algorithm.

8. The active distribution network coordinated control method according to claim 7, characterized in that: In S3.6, the particle swarm optimization algorithm is specifically: ； ； in, Indicates the topology configuration status; Indicates the configuration of the existing topology The amount of change; represents the inertia weight; Represents particles How much tendency there is to return to a previously discovered good configuration; Represents particles how much tends to follow the best configuration found by the group; and Represents a random number, used to increase the randomness of the search process; Represents particles the best position achieved since the search began; represents the best position found so far by all particles in the entire particle swarm; Indicates the number of iterations.

9. The active distribution network coordinated control method according to claim 5, characterized in that: The control of charging and discharging of the energy storage device comprises the following steps: S3.

7. Obtain load data for a period of time, including the distribution of daily electricity consumption, especially the load curves during peak and off-peak hours, and analyze wind and solar power generation data; S3.

8. Based on historical data, predict future load curves, especially at different times of the day, predict the power generation of renewable energy, especially its volatility, integrate load forecasting and power generation forecasting, and analyze and predict periods of supply and demand imbalance; S3.

9. Calculate the required energy storage capacity based on demand forecasts and store excess electricity during off-peak hours; S3.

10. Select the load off-peak period for charging, determine the charging rate based on the battery status SOC and the expected load demand, and set the SOC upper limit ; Select the peak load period for discharge, determine the discharge rate according to the load demand and the remaining battery capacity, and set the SOC lower limit .

10. The active distribution network coordinated control method according to claim 1, characterized in that: In S4, the formulated control strategy is sent to each control node, including the following steps: S4.

1. Convert the formulated control strategy into a control instruction format, where the control instructions include adjusting the output power of the distributed power source, changing the grid switch state to optimize the topology structure, and controlling the charging and discharging of the energy storage device; S4.2, select the communication protocol Modbus to ensure that the command can be sent to the corresponding execution unit; S4.

3. Control instructions are sent to each control node through the automatic control system, and the instructions are executed to adjust the operating status of the power grid.

Images