Optimal configuration and power scheduling algorithm of distributed power system with large range power flow transfer

The distributed power system is composed of multiple distributed energy sources, loads and energy storage devices, which have the characteristics of decentralization and complexity. Therefore, a distributed power system optimal allocation and power dispatching algorithm considering large-scale power flow transfer is proposed. Calculate the power flow values of different nodes, and construct an optimization model according to the results to complete power flow transmission. In the inner loop, the current dispatching is set as the objective, and in the outer loop, the voltage dispatching is set as the objective. Distributed power system scheduling is completed by double closed-loop vectors. The experimental results show that the power oscillation amplitude of this method is small and the duration is short.


Introduction
With the transformation of global energy structure of renewable energy, distributed power system has the advantages of high efficiency, environmental protection and flexibility, it can ensure energy security and promoting sustainable energy development.However, the wide access of power supply and large-scale power flow transfer have brought new challenges to the power system [1].First of all, the access and integration of distributed energy resources increase the complexity of the system, which requires reasonable planning and optimal configuration of various components and resources in the system.Secondly, variable loads and unstable energy inputs in distributed power systems make power scheduling more difficult.Effective scheduling algorithms are needed to achieve supply and demand balance and optimal energy allocation [2].In addition, the distributed power system also involves the storage and management of energy, the expansion of the scale of the grid and the interconnection with the traditional power system.
Reference [3] proposes an optimization configuration of distributed generation based on interval power flow.It introduces interval power flow into the combined optimization problem, which is responsible for the size and allocation of DG units in the power grid.The voltage is adjusted through the objective function to complete the optimal configuration of distributed generation.However, this method is easy to generate redundant data in the process of combinatorial optimization.Reference [4] puts forward the optimal power flow technology of distribution system, establishes the linear constraint equations of current and voltage, and constructs the electrical model by using the nonlinear objective function to complete the optimization of distribution system.This method overcomes the nonlinearity of distributed energy resources and adapts to power demand.However, this method has large power oscillations.Reference [5] proposes an integrated transmission and distribution network considering uncertain renewable energy.The SVSM interval distributed model is established, and the model is solved by the multiplier alternation method to complete the optimal power flow calculation.However, the stability and dispatching of this method still need to be further improved.
An optimal allocation and power scheduling algorithm for large-scale power flow transmission distributed power system is proposed.

Large-scale power flow transfer estimation of distribution network
In the distributed power system, the optimal allocation mainly focuses on the reasonable allocation of the scale and location of various power sources under the conditions of meeting the system requirements and constraints, so as to achieve the optimal economic and environmental benefits of the whole system.Power scheduling is to formulate a reasonable scheduling strategy according to the realtime monitoring system state and load demand in the operation stage to ensure the stable and economical operation of the system.In the aspect of optimal allocation, it is very important to consider the influence of large-scale power flow transfer When determining the configuration scheme, the interaction between each component and the dynamic change of power flow transmission must be fully considered to ensure the stability and reliability of the whole system.Power flow transmission is described as [6]: In Formula (1), Z represents the admittance matrix of distribution network nodes; E represents the voltage complex vector; I represents the current complex vector.
A node is a connection point in a power system, where current and power can enter and exit.Grouping the complex vectors of voltage and current according to nodes can better understand the power relationship between different nodes in the system, and carry out various power calculations and analysis [7][8].Group voltage and current: In Formula (2), subscript v represents the part corresponding to the equilibrium node; q represents the part corresponding to the decomposed node.For different types of nodes, the complex vector characteristics of voltage and current will be different, which can be divided into power supply nodes, transformer nodes, feeder nodes and load nodes.
In the excitation process of the distribution network, the load impedance is analyzed according to the line capacitance.5 nodes and 7 lines are calculated.The overall equivalent impedance calculation results are shown in Figure 1.By using compute nodes as reference data, the calculation method of independent nodes is presented as follows: The resulting algebraic form of the power flow transfer matrix is: In the above formula: z represents the node admittance matrix.
In the distributed power system, there are multiple distributed energy nodes and load nodes, and the power flow transmission becomes more scattered and complex.By comparing the data between different nodes, the running state of the whole system is evaluated, and corresponding measures are taken to optimize it as needed.

Distributed power energy system configuration optimization model
The optimization goal of distributed power system is defined and the mathematical model is established.Based on the zero phase angle difference of node voltage, the power transmission and voltage law in power system are described.Assuming that the Lagrange multiplier under the whole cycle investment cost is , sr m and the capacity configuration of the gas turbine is

Power system power scheduling
Power scheduling needs to comprehensively consider the overall characteristics of the system, the needs of users and the distribution of energy resources, so as to determine the power output strategy of each node in the system.Double closed-loop vector control method realizes load balance and power balance of power system through the coordination of torque control loop and flux control loop.The working state of the motor is dynamically adjusted to maximize the energy utilization efficiency.Through the regulation of the outer loop, adverse effects of excessive or insufficient voltage on the system are avoided, ensuring the reliability and stability of power supply.
Then the DC voltage-current ( )

Experimental part
Power oscillation refers to the fluctuation of power output in distributed power system.The existence of power oscillation may lead to instability of power system, increase energy consumption and affect the quality of power supply.Therefore, by testing the power oscillation level, the stability and control effect of the system can be evaluated.A smaller power oscillation represents a more stable power output.Reference [4] and reference [5] are selected as comparison methods, and the results are shown in Figure 2. From Figure 2, the method in reference [4] has good stability before 4s, but it gradually oscillates after 4s.The method in reference [5] exhibits large oscillations throughout the testing period.However, the power oscillation of the method in this paper is smaller throughout the entire testing period.This is because the method in this paper estimates the power flow transmission conditions at different nodes and optimizes the configuration of the distributed power system.
Power oscillation under different DC sizes refers to the capacity of DC components or energy storage devices in distributed power systems.Different sizes of DC have significant impacts on power regulation and stability of the system.By testing power oscillation under different DC sizes, the relationship between DC capacity and stability can be observed, which helps to determine the appropriate DC size to maintain stable power output under different load conditions.The result is shown in Figure 3.  3, after the power scheduling of the proposed method, no matter how the DC in the power grid changes, its power remains unchanged, which proves that the stability of the method is extremely good.The power of the other two methods always changes with the change of the DC.
The oscillation time measurement under different power conditions measures the time required for the distributed power system to recover stability after disturbances or changes occur.In practical operation, a fast recovery time for oscillation is an important performance indicator.By testing the oscillation time under different power conditions, the system's resistance to external disturbances and response speed can be evaluated.A shorter oscillation time means better stability and robustness of the system.The result is shown in Figure 4.According to Figure 4, with the increase of power level, the oscillation time of the proposed method is prolonged less.This is because the method uses the double closed-loop vector regulation method to dynamically adjust and control the current and voltage based on the system's operating status and requirements.By cooperating with the inner and outer control loops, the voltage stability is ensured, further optimizing the power dispatching.

Conclusion
Under the background of increasing energy demand and serious environmental problems, distributed power system, as an efficient and environmentally friendly way of energy utilization, has gradually attracted widespread attention.Based on this, the optimal allocation and power scheduling algorithm of distributed power system is proposed.The algorithm comprehensively considers the requirements of economy, safety and environmental protection of the system, and achieves the optimal overall performance of the system by optimizing the scale and location of distributed power sources and formulating reasonable power scheduling strategies.Through the mutual cooperation between the inner and outer loops, we can effectively improve the overall performance of the distributed power system.In the future research, with the continuous progress of technology and the increase of application demand, machine learning is introduced to the further development of sustainable power system.

S
are a total of n typical scenarios under the operating cost.The capacity configurations of each energy component under scenario s and the previous scenario 1 − .The iteration times of the two objectives are r times.The penalty factor is  .Then the augmented Lagrange function of the distributed electric power energy system is defined [9-10]:

C
represents all the capital costs invested in the whole investment cycle [11]., op s C represents the operating cost of scenario s .
real-time value of DC voltage, _ dc G I represents the real-time value of DC measured current and G k represents the coefficient of the sag curve.

Figure 2 .
Figure 2. Power scheduling PV array curves of the three methods.

Figure 4 .
Figure 4. Oscillation time of different methods.
Power scheduling of the three methods.