Optimization scheduling method for multiple flexibility resources considering electric vehicle cluster

To unleash the scheduling potential of flexible resources, an enhanced flexibility approach is proposed, which takes into account various flexible resources including electric vehicle (EV) clusters. Initially, the dispatch potential of individual EV users is predicted based on historical data regarding travel habits and charging station parameters. Subsequently, the power and energy-feasible regions are constructed to quantify the flexibility of the EV clusters. To enhance the overall system’s operational efficiency and flexibility, various constraints related to flexibility resources are taken into consideration, including the source-grid-load-storage aspects. As a result, a multidimensional flexibility resource supply-demand model is established and the analysis target cascading method is applied to solve this sophisticated model. Finally, the modified system is used to test and analyze the proposed model and method. The results demonstrate the proposed model and method can enhance the operational flexibility of high-proportion new energy distribution networks.


Introduction
The rapid development of new energy grid integration technology and the large-scale integration of electric vehicles have led to a significant increase in the operational and scheduling challenges of the power system.In this context, there is an urgent need for abundant flexibility resources to mitigate the power imbalances caused by strong fluctuations and uncertainties in the supply and demand dynamics [1].
By 2030, the number of EVs is projected to exceed 80 million, which makes EVs a crucial and economically valuable flexibility resource for the new power system [2].The quantification of EV flexibility has been extensively researched by scholars in various fields, including transportation, architecture, energy, and collaborative efforts [3].In [4], it proposes an energy management optimization scheme for EVs, enhancing distribution network flexibility through storage and reducing losses.Additionally, In [5], it proposes a new model to explore orderly battery swapping, efficient management, and Battery-to-Grid (B2G) technology to increase the system's flexibility by transferring surplus energy to the power grid.Furthermore, In [6], it proposes a new model to address volatility from renewable energy generation and charging loads by identifying critical uncertain factors and optimally allocating flexible resources to ensure the secure operation of distribution networks amid new energy sources and EVs.In [7], it proposes a multi-time scale scheduling method for regional power grids, incorporating EV flexibility and wind power integration.Although successful in mitigating load fluctuations and mitigating wind power integration impact, it may not fully leverage the potential of EV flexibility.On the other hand, In [8], it proposes a new model to introduce EV virtual power plants in distribution network planning, enhancing economic efficiency and renewable energy integration.In [9], it proposes a new model to treat EVs as flexible load resources, calculating schedulable potential capacity using compression techniques.In [10], it proposes a new model to establish a two-stage flexibility enhancement optimization model involving EV scheduling, energy storage, and interruptible load dispatch.
This paper establishes a comprehensive model that considers EV clusters in the optimization and scheduling process to address the aforementioned issue.To effectively exploit the flexibility complementarity of diverse resources, we establish a collaborative relationship between the transmission network (TN) and the active distribution network (DN).We leverage linked decisions and flexible optimizations between the transmission and distribution networks, fully tapping into the potential of multi-dimensional flexibility resources.The results demonstrate that the integrated flexibility resources of generation, transmission, load, and storage significantly enhance the overall system's operational flexibility.

EV cluster flexibility model
(1) EV charging and discharging power constraints  (2) EV battery state of charge constraints where , v t E is the battery SOC of the v-th EV during time t.(3) Feasible region of EV battery capacity When the initial capacity of the v-th EV upon integration into the power grid exceeds the minimum SOC threshold, the energy feasible region of the EV can be expressed as When the initial capacity of the v-th EV upon integration into the power grid is less than the minimum SOC threshold, the energy-feasible region of the EV can be expressed as  (3) Grid Interconnections By connecting the TN with the DN through interconnection lines, power can be mutually regulated, expanding the system's balancing scope, and enhancing operational flexibility.E means the energy value of the energy storage during time t.

Solution method
The ATC algorithm is used to solve the proposed model in this paper involves the following specific steps.
(1) Set initial values for penalty function multipliers and contact line power variables.
(2) Perform parallel optimization decisions for the resources (source, load, and storage sides) of the DN layer to obtain the contact line exchange power, which is defined as the forward contact line power.Transmit this result to the TN.
(3) The TN receives the contact line exchange power from the DN and uses it as a known parameter to invoke the optimization of flexibility units in the TN, obtaining the contact line exchange power defined as the backward contact line power.Then, this result is transmitted to each DN as a known parameter for the next parallel optimization.
(4) The difference between the contact line exchange power of the current iteration and the result of the previous iteration for both directions of transmission is calculated.Check whether this difference satisfies the termination tolerance.If it does, proceed to Step 5. Otherwise, continue the inner loop until the convergence condition is met.
(5) Calculate the difference in contact line exchange power for both directions in the current iteration and check if it meets the termination tolerance.Additionally, calculate the difference between the total system operating cost obtained in this iteration and the result of the previous iteration and check if it meets the termination tolerance.If both convergence criteria are met, the optimization process ends.
Otherwise, update the penalty factor and continue with the iteration.

Case Studies
The effectiveness of the proposed model is validated using the IEEE 30-node transmission network and two IEEE 33-node distribution networks as a test system.Two contrasting scenarios are constructed in this case study.Scenario 1 considers the flexibility of electric vehicle clusters.Scenario 2 does not consider the flexibility of electric vehicle clusters.Table 1 presents the operating costs for both scenarios, while Figure.As evident from Table 1, Scenario 2 enables the full mobilization of the EV cluster's flexibility supply capacity to meet the system's flexibility demands.Consequently, the discharge compensation cost of the EV cluster increases in Scenario 2, leading to a slight decline in its economic viability compared to Scenario 1.Although considering the flexibility constraints of the EV cluster may result in higher overall system operating costs, the lack of economic efficiency is not reflected in the flexibility deficiency cost.On the other hand, Scenario 1, which does not consider the flexibility constraints of the EV cluster, incurs higher flexibility deficiency penalty costs, highlighting the system's potential to utilize the EV cluster's flexibility resources for enhancing the system's flexibility supply capacity, thereby ensuring the fulfillment of the system's flexibility requirements.

Case Studies
The present study proposes an optimized schedule approach considering the diverse flexibility resources of EV clusters, with the following conclusions.By incorporating the flexibility constraints of EV clusters under fluctuating scenarios into the optimization schedue, the study avoids the penalty costs resulting from the flexibility deficiencies.The study maximizes the potential of diverse flexibility resources from the source, grid, and storage sides.Coordinating the output levels of various generating units, effectively mitigates the power imbalance between generation and load caused by uncertainties, providing the system ample flexibility reserves.In the future, further research is needed to combine the flexible power market and auxiliary services to enhance the interactive flexibility between EVs and distribution networks.

p
are the charging and discharging power of the v-th EV during the time t.
limits of the charging and discharging power of the v-th EV.E V v T is the grid connection period of the v-th EV.

E
is the initial SOC at the time of grid connection, and out v E is the final SOC at the end of the disconnection period.and lower boundaries of the battery SOC of the v-th EV.

E
are the energy upper and lower limits of the v-th EV during time t. in v T and out v T are the grid connection and disconnection times of the v-th EV. in v E and out v time of grid connection and disconnection, respectively.and discharging power upper and lower limits of the v-th EV.
upward and downward ramp rates of the unit.
charging and discharging power of the energy storage, respectively.and lower limits of the ES capacity., e t 1 and Figure.2illustrate the flexibility balance.

Figure 1 Figure 2
Figure 1 Flexibility deficiency in Scenario 1.In Scenario 1, there are instances where flexibility deficiency occurs, indicating an insufficient provision of flexibility supply.The upward and downward flexibility supply and demand relationships of Scenario 2 are shown in Figure.2(a)and Figure.2(b),respectively.In Scenario 2, by considering the flexibility supply of the EV cluster and fully leveraging the diverse flexibility resources of source, grid, and storage sides, the system effectively adjusts the output of various generating units, mitigating the uncertainties of generation and load.As a result, the flexibility supply capacity in Scenario 2 can meet the system's flexibility demands at all times, eliminating the occurrence of flexibility deficiencies and

Table 1 .
Cost Comparison in Different Scenarios.