Multiscale dynamic time‐domain simulation of regional integrated energy considering seasonal load difference

To improve the multiscale dynamic dispatching and distribution capacity of regional integrated energy, a multiscale dynamic time‐domain dispatching model of regional integrated energy with seasonal load difference is proposed. The master–slave game coordination planning model of regional comprehensive energy multiscale dispatching is constructed, and the difference fusion model of regional comprehensive energy multiscale dynamic dispatching is established by using the active and reactive power coordination control method according to the influence factors of voltage deviation and line overload absorption. Maximize the weight factors of the positive contribution of distributed photovoltaic (PV) power to the active distribution network. On the basis of the quantitative regulation analysis method of demand response power based on the equivalent slope of the load curve, establish an energy absorption capacity analysis model of seasonal load differentiation. On the basis of the dynamic interaction between different systems, a multiscale dynamic time‐domain analysis and dynamic simulation model of regional integrated energy is constructed by using the method of PV absorptive capacity assessment on both sides of demand uncertainty, to achieve multiscale dynamic time‐domain feature extraction of integrated energy and integrated scheduling of seasonal load differences. The simulation results show that the multiscale dynamic time‐domain allocation of regional comprehensive energy using this method improves the energy dispatching and balanced allocation capability, and the time‐domain dynamic allocation capability is good. Considering the terminal load demand and the electricity price demand formulated by the distribution network investment operators, the seasonal load response and comprehensive dispatching capability are improved.

The integrated energy system uses energy production, conversion, and storage equipment to interconnect multiple energy systems.The comprehensive energy system can make full use of the complementary ability of different energy sources, and is an important means to promote the development of energy Internet technology.With the continuous development of regional integrated energy system management, a multiscale dynamic scheduling method is adopted to achieve regional integrated energy comprehensive scheduling and intelligent management design, which fully integrates traditional energy and renewable energy.As an independent interest subject, the integrated energy system has its own energy supply characteristics.In the integrated energy management system, the coupling utilization of multiple energy sources is analyzed according to the composition characteristics of the connected distribution network, which effectively reduces the dependence on the reliability of the external energy network.Under the influence of seasonal load difference, the power sales of the power grid are reduced, the economic benefits become worse, and the equipment utilization efficiency is greatly reduced.To improve the multiscale dynamic time-domain distribution capability of regional comprehensive energy, the power grid and the comprehensive energy system are coordinated and planned through multiple operation modes of supply, so as to achieve multiscale dynamic distribution of regional comprehensive energy.It is of great significance to study the multiscale dynamic time-domain scheduling model of regional integrated energy. 1 The main task of distribution network planning is to determine the optimal power supply or grid expansion construction scheme according to the load forecasting results and the original grid structure on the basis of meeting the load growth demand and various constraints in a certain planning period, so as to build a safe, reliable, cost-effective and flexible distribution network. 2 Different researchers analyzed the multiscale dynamic time domain of integrated energy in the distribution grid area.Yongjun et al. 3 propose a constant capacity planning method for wind optical storage access distribution network, constructs constraints, such as power balance and voltage deviation, and establishes a multiobjective optimization model for wind optical storage access distribution network considering economy and stability based on the consideration of the timing characteristics of distributed generation, the timing characteristics of different seasonal loads, and the real-time price.The nondominated sorting genetic algorithm is used to obtain the Pareto optimal solution.However, the electrical coupling degree of the multiobjective optimization model used in this method is low.Junjie et al. 4 establish a coordinated planning model considering the distribution network, distributed generation, and users realize multiscale dynamic scheduling and distribution of integrated energy from multiple aspects, such as system planning, operation, dispatching, safety, evaluation, and so forth, and optimizes the planning and operation of the integrated energy system at the same time to improve the ability of multiscale dynamic time-domain simulation of energy.However, this model only considers the support of comprehensive energy for electric load, ignoring the support of electric power for heat load.Wei et al. 5 analyze the load complementarity characteristics of different nodes in the distributed generation and combine the external energy network expansion planning and design of regional multienergy system to achieve multiscale scheduling of integrated energy.However, this method does not consider the energy absorptive capacity in the planning objective function, which leads to the uncertainty of various loads.
To solve these problems, this paper proposes a multiscale dynamic time-domain scheduling model for regional integrated energy with seasonal load differences.The master-slave game coordination planning model for multiscale dispatching of regional comprehensive energy is established, and the difference fusion model for multiscale dynamic dispatching of regional comprehensive energy is established.The active contribution weight factors of the active distribution network are obtained by using distributed photovoltaic (PV) power, and the quantitative regulation analysis method of demand response power based on the equivalent slope of the load curve.The energy absorptive capacity analysis model of seasonal load difference is established, and the port electricity and power quality factors are calculated by the PV absorptive capacity evaluation method of both sides of demand uncertainty.The multiscale dynamic timedomain analysis and dynamic simulation model of regional integrated energy are constructed to realize multiscale dynamic time-domain feature extraction of integrated energy and integrated scheduling of seasonal load difference.The simulation test shows that the method in this paper has superior performance in improving the multiscale dynamic time-domain scheduling and simulation capability of regional integrated energy.ENERGY MULTISCALE SCHEDULING PLANNING MODEL AND FACTOR ANALYSIS

| Regional comprehensive energy multiscale dispatching planning model
The topology model of the multisource microgrid cluster system is established by using the method of active and reactive power coordinated control, combined with the existence of multiple sourcesnetworks-loads-storage in multisource microgrids.Considering the functions and dynamic response characteristics 6 of various scheduling resources, according to the multiscale scheduling conditions and scheduling requirements of energy, combined with the situation of various virtual energy storage equipment, the joint dynamic regulation law of transferable load and controllable load is defined as Among them, f M H ( , ) a is the coupling coefficient of new energy, const is the price coupling coefficient, m f control quantity of power supply for the system, n L is to output new energy power, A e is the control quantity of joint power supply intensity, and T * 4 is the assign results to output prices.In the case of steady state control, the load parameters of the power supply control in the transferable loads within the multisource microgrid change according to the specified law, and the structure of the optimization model of its distribution and coordination layer is shown in Figure 1.
It can be seen from Figure 1 that, without decoupling, the joint control parameters of the cloud platform integrated multisource microgrids are n Lref , T * ref 4 when generating unit step, n Lref is 0, the survey output is n L and T * 4 .In case of m f change, the control instruction of multi-interest balance will be issued to each multisource microgrid, and the integrated energy system will be real-time paired T * 4 the coordinated control is generated, and the external energy network connecting the regional multienergy system is planned and designed for expansion.The system state equation is obtained through the cyclic iteration of the economic optimization model and the reliability verification model: Considering the uncertainty of energy purchase price and new energy, a two-level planning and operation model 7 is established.Considering the multiagent interaction behavior of the integrated energy system optimization problem, it is concluded that the dynamic characteristics of regional integrated energy dispatching depend on the dynamic balance characteristics between operators and users. 8A user master-slave game model is constructed to realize the operation optimization equation of the integrated energy microgrid as follows: Among them, J H and J L are decision momentum for each independent subject, M Δ H and M Δ L are the model parameters of the interaction between energy supply and demand.The residual torque that can be converted internally in the integrated energy system using different types of energy such as electricity and natural gas is defined as (5) Among them, M T is the energy management parameters for multiple virtual power plants, and M C is the optimize model parameters for the operation of the microelectric network group.The electricity is formulated with the goal of maximizing profits, and the state equation is obtained according to the master-slave game equation model between multiple energy stations and users in the integrated energy system: Among them, is the weighted coefficient of on-grid price and wind storage dispatching, Collaborative layer optimization model structure of the transregional multienergy system.
the fuzzy weighting coefficient of the feedback error.Through the closed-loop system, the continuous convergence can reduce the output error and finally achieve the optimal fuzzy control effect.

| Analysis of factors affecting absorption
With the goal of real-time distributed regulation, the optimization constraint of the centralized regulation layer is built as the multisource micronetwork control model, and the transmission power limit of the tie line between the multisource micro and other units is analyzed.According to the electricity price formulated by the distribution network on the electricity demand 9,10 that affects the integrated energy system, a mutual competition model is established between the participants to determine: Among them, u t ( ) is making decisions for all parties involved and has a time sequence, K p is the degree to which each participant has mastered the decision-making information of other participants, e t ( ) is the control error, T I is the time constant of integration, and T D is the time constant of differential.
A master-slave game model is used to analyze the dynamic parameters 11 of distribution network investment and operation by leaders.On the basis of the model of mutual competition between participants, the definition of reliable electricity price strategy distribution law is obtained: Among them, k is the sampling point serial number of the absorption influence factor and T is the sampling period.
The strategies of integrated energy system investment operators include planning strategies and operation strategies e i ( ) carry out summation and accumulation, 12 calculate the strategy space control parameters of distribution network investment operators and integrated energy system investment operators, and obtain the equilibrium solution output of the master-slave game as follows: When the integrated energy system investment operator is connected to the same distribution system at the same time, the parameters of the equilibrium solution optimization model of the master-slave game planning model are derived as follows: Considering the costs of investment, maintenance, energy purchase, and network loss, the revenue is obtained through the sale of electricity, and the annual values such as annual profit, annual electricity sales revenue, investment cost, annual maintenance cost, annual energy purchase cost, and annual network loss cost after the distribution network expansion planning are the constrained independent variables.13 For node i, assume N i 1 is a hop adjacent node, and the cost of capacity expansion or new construction is the following: Among them, N j 1 is time slot allocation.Calculate the electricity quantity purchased by the grid from the superior grid and the integrated energy system at the nodes: Among them, M is the dimension, x mi is a node i at the branch road endpoint of m, and N is the demand vector.With the line, interconnection switch and substation investment between nodes as variables, the network radial constraints of two hop nodes are Considering the seasonal load difference, the expanded planned distribution network needs to be able to ensure the increase of power supply to all users in its power supply area.At this time, the scheduling success rate is defined as For each cell, the node power balance constraint index allocation model is constructed.The load dispatching efficiency of all users in the power supply area is calculated as follows: By analyzing the node set within the power supply range of the substation, and according to the electricity price, planning, and operation optimization dispatching effect, the substation capacity scale is obtained as follows: Among them, x p M < < , x allocates the number of nodes for the decision variable, and x nodes need to send the packet of slot p.The total annual cost of the integrated energy system is Develop operation strategy with the objective of minimizing the sum of operation and maintenance and energy purchase costs.In power automatic dispatching, assuming that the management node and the managed node are in relative change, the management information feedback model of the node target is Among them, f t ( ) is the control signal from the investment and operation management node to the managed node of the integrated energy system, s is a discrete equipment investment constraint parameter, representing the expansion and contraction of the managed node and the managed node, and τ is the set management time interval for PV and HP in the x comprehensive energy system.
Taking the operation constraints and thermoelectric coupling constraints of gas turbine equipment as indicators, 14 and considering the impact of distributed PV power generation on solar radiation intensity, the square-integrable function is obtained from the continuous wavelet transform of y t ( ) towards ψ t ( ): Among them, a is the normalized factor of the cost of investment, operation and maintenance, power purchase, and gas purchase, and b is the energy distribution factor.Establish the utility function of the distribution network investment operator who is the leader of the master-slave game: With the help of a PV inverter, the power quality of the distribution network can be improved.Under the rated power of active output of the PV inverter, the time management scale of active distribution network voltage can be obtained: Among them, s is the electrical load of the client.According to the above analysis, the PV absorptive capacity of the active distribution network is evaluated under the time management scale of the distributed PV power supply. 15

| Fusion of seasonal load difference
On the basis of the quantitative regulation analysis method of demand response power based on the equivalent slope of the load curve, the PV absorptive capacity analysis model of seasonal load difference is established.For PV absorption on both sides of demand uncertainty, a cloud-side collaborative framework is established, and multisource microelectricity network scale scheduling is considered to integrate the original residual and dual residual.The schematic diagram of the power dispatching block area where the PV power supply is connected to the voltage of the active distribution network is shown in Figure 2.
On the basis of the PV absorptive capacity assessment method of the active distribution network, the impact of distributed PV power supply access on the voltage of the active distribution network is analyzed, and the load correlation parameter estimation model of voltage deviation, voltage fluctuation, short circuit capacity, and system loss is obtained.Set the system loss of the active distribution network as a one-dimensional vector X n based on the joint characteristics of voltage deviation, voltage fluctuation, short circuit capacity, and system network loss, the power generated by PV on the active distribution network is obtained: Among them, τ i is feeder branch impedance and δ is the modulation weight for voltage instability in the distribution network.The loss rate of sectional selective voltage regulation control of a parallel network of the active distribution network is X η n ( ) .The expected equation of the power load controlled by subsection selective voltage regulation at the grid connection point of the active distribution network is constructed: (23) Among them, c t ( ) l is a random vector of adjustable electrical load and χ is the cost per unit capacity.Assume that the dynamic distribution of PV absorption of the active distribution network is j cluster head nodes, and build the energy consumption factor matrix of PV sink nodes of the active distribution Among them, S j L × is the power matrix that can be scheduled in the functional area and T j×1 is the space-time distribution matrix.According to the characteristics of the potential reliability benefits under the coupling effect of the integrated energy system, a master-slave game model framework model is established.The master-slave game model is shown in Figure 3.
Through the analysis of the equilibrium solution of the master-slave game theory, the logarithmic likelihood function is obtained: Among them, y s ( ) refers to the proportion of the maximum allowable supply shortage to the annual load, R is the total investment cost, and E is the penalty cost for abandoning wind and light.Considering the influence of peak factors, an independent planning model of a distribution network multi-integrated energy system is constructed.Under the constraint of maximum power gain, the cost prediction objective function of the integrated energy system investment operator at each node is as follows: is the annual operation and maintenance cost, is the weighted sum of all cost changes of the integrated energy system at the new node, λ α Γ( ) is the total energy supply cost component of the investment operator, and w α is the proportion of the maximum value to the annual load.Through the analysis of the total energy supply cost of the integrated energy system with the user reliability requirements, the objective function is minimized, and the coordinated control between STATCOM and LCC is achieved.

| Multiscale dynamic time-domain feature extraction and integrated scheduling of integrated energy
On the basis of the master-slave game coordination planning method of multiple integrated energy systems, the dynamic multiscale decomposition is carried out, and the characteristic premeasurement of independent planning of distribution network multiple integrated energy systems is obtained as C″ n and measured values as C n : − is the operational decision variable, e n is a transferable load, MC n n p − +1 is a flexible load dispatching variable, and w n can reduce the load within the day.On the basis of the dynamic interaction between different power systems, the port power and power quality factors are analyzed by using the method of PV absorptive capacity assessment on both sides of demand uncertainty.The power price-guided reliability investment component is used to analyze the load difference and obtain the allocation capacity of the energy storage system.Analyze the replacement cost cycle caused by energy storage loss in the scheduling cycle, change the constraint limit of the feasible region of the decision variable, and obtain the detection objective functions A and M: ( ) The reliability of the optimized planning scheme is checked under K feasible scenarios.If the reliability index is not met, the weak link parameters need to be calculated.According to the feasible region constraints of decision variables, the matching strategy of time series power balance constraints is obtained P* i , and so on P* j , …, P* M .The lithium-ion energy storage battery is considered for the energy storage device, and its operating constraints include power balance constraints, capacity constraints, and charging and discharging state constraints.The adoption number and comprehensive scheduling function under the operating state of period t need to meet the following requirements: is the operational decision variable, e n is a transferable load, MC n n p − +1 is a flexible load dispatching variable, and w n can reduce the load within the day.According to the output power of regional comprehensive energy power supply is limited by the maximum and minimum output, the constraint function of energy and electricity proportion under the regional distribution mechanism of the seasonal load is obtained as follows: Among them, d G Za is the electricity sold for the grid, v Q is the electricity consumed by the power system itself, and g G Za is the power output of the production unit.At , linearize the equation by changing the constraint limit of the feasible region of the decision variable: Among them, W D is the flow index and σ is the efficiency coefficient.When the external applied voltage reaches the standard voltage, the interference node is suppressed to optimize the dispatching model of regional comprehensive energy optimization: Among them, B ll is the sum of the branch flows flowing in and out of the node, x B ( ) is the conjugate transpose matrix of the power load vector, y B ( ) is the conjugate transpose matrix of the power demand vector, and s l is the heat load distribution coefficient.According to the seasonal load difference scheduling, the above model is used to adopt power tracking control to ensure the maximum power output of PV modules, and realize multiscale dynamic timedomain feature extraction of integrated energy and seasonal load difference integrated scheduling.The implementation flow chart is shown in Figure 4.

| Simulation environment
To verify the application performance of this method in realizing multiscale dynamic time-domain feature extraction of integrated energy, experimental tests were carried out.In the simulation analysis, the regulation strategy of PV power generation capacity is introduced to make it match the actual load demand.This can include adjusting the amount of PV power generation based on weather forecasts or using grid feedback mechanisms to ensure that PV power generation does not exceed the actual load demand.The purpose of PV absorption capacity analysis is usually to determine the capacity that can accept PV power generation stably and reliably to meet the load demand of the region.According to the actual lighting conditions and the characteristics of PV cells, the actual output of PV power generation is calculated and simulated.This helps understand the volatility and contribution levels of PV power generation under different seasons.The actual PV power generation output is compared with the load demand, and the influence of PV power generation on the energy supply and demand balance of the system is analyzed.Taking into account the seasonal load differences, it is necessary to evaluate the absorption level of PV power generation and the remaining supply or demand gap, and conduct a comprehensive assessment combined with the actual PV power generation output and energy supply and demand balance to more accurately understand the role of PV in system operation.The peak power of the distributed PV F I G U R E 4 Realization process of multiscale time-domain scheduling of regional integrated energy.
T A B L E 1 List of control parameters for multiscale dynamic dispatching of regional comprehensive energy.power supply was set to 1000 kW, the absorption capacity control compensation of the PV power supply was 24, the peak power of PV absorption assessment of active distribution network was 2500 kW, the system network loss was 55 kW, the access capacity was 25 kW, and the algorithm ran 30 optimal individuals.The controller parameters and their corresponding objective functions are shown in Table 1.

| Multiscale dynamic time-domain simulation of regional integrated energy
The voltage of the power grid is determined by the voltage of each node in the power system.When the input power is large, the grid voltage will change.
According to the simulation parameters in Table 1, set the multiscale dynamic time-domain simulation of regional comprehensive energy, and set the access power as 620 and 700 kW.The specific node voltage variation diagram is shown in Figure 5.
According to the analysis of Figure 5, the voltage changes greatly with time.This is because the input power will cause the voltage of resistance, inductance, and other components to increase, resulting in voltage fluctuations.With the increase of the input power, the voltage fluctuations will gradually increase.The algorithm in this paper analyzes the PV absorptive capacity optimization model of multiple integrated energy sources, optimizes the load difference of the active distribution network, obtains the time-domain diagram of regional integrated energy distribution, further optimizes the network loss and PV absorptive capacity of the active distribution network, and provides a guarantee for the stable operation of the power grid.

| Cointegration vector test
Multiscale dynamic scheduling of multi-integrated energy refers to the dynamic scheduling of energy supply and demand under various loads of wind, solar, and hydro energy, so as to realize the efficient use of energy.Cointegration vector is a concept in multivariate time series analysis, which refers to the long-term relationship between different time series.In the multicomprehensive energy multiscale dynamic scheduling, the cointegration vector test calculates the mean and standard deviation by analyzing the relationship between different time series, and accurately predicts the energy supply and demand.On the basis of multiscale dynamic time-domain simulation of regional integrated energy, the cointegration vector of multiscale dynamic scheduling of multiple integrated energy is tested.The specific results are shown in Table 2.
According to the results in Table 2, the multiscale dynamic time-domain simulation of comprehensive energy using the method in this paper has improved the energy dispatching and balanced allocation capability, and the time-domain dynamic allocation capability is good.Considering the terminal load demand and the F I G U R E 5 Time-domain diagram of regional comprehensive energy distribution.
LI ET AL.
| 709 electricity price demand formulated by the distribution network investment operator, the seasonal load response and comprehensive dispatching capacity are improved.

| Dynamic decomposition test
Multiscale dynamic decomposition of regional comprehensive energy can obtain energy consumption and generation at different time scales.On the basis of the amplitude, the amplitude fluctuation is controlled between 0 and 1, which indicates that the multiscale dynamic decomposition effect of regional comprehensive energy is better and can improve the energy utilization efficiency of the power system.The multiscale dynamic decomposition results of regional comprehensive energy are shown in Figure 6.
According to the analysis of Figure 6, the multiscale decomposition of comprehensive energy in this method has a good balance, which can optimize the energy dispatching strategy, improve energy utilization efficiency, reduce energy consumption costs, and achieve sustainable development.

| Deviation test
To further verify the multiscale dynamic effect of regional comprehensive energy of the method in this paper, the method in Junjie et al. 4 and the method in Wei et al. 5 are selected as the comparison method, and the iteration number is set as 120.In the multiscale dynamic time-domain analysis, the time management scale, long time scale, medium time scale, and short time scale are considered.By analyzing the deviation of the method to the multiscale scheduling of regional comprehensive energy, the effectiveness of the method in this paper is verified.The deviation results of regional integrated energy multiscale dispatching by different methods are shown in Figure 7.
According to the analysis of Figure 7, the energy dispatching deviation of this method is lower than that of the methods in Junjie et al. 4 and Wei et al. 5 When completing the iteration, the energy dispatching deviation of the method in this paper is 0.042, while the energy dispatching deviation of the method in Junjie et al. 4 is 0.107, and the energy dispatching deviation of the method in Wei et al. 5 is 0.112.This is because the method in this paper takes into account the difference in F I G U R E 7 Comparison of multiscale decomposition deviation of regional comprehensive energy.
seasonal load, and the deviation of multiscale dynamic time-domain decomposition of regional comprehensive energy is small.This proves that the method in this paper has high accuracy in the multiscale dynamic timedomain scheduling process of regional integrated energy.

| Robustness testing
To further verify the multiscale dynamic effect of the regional integrated energy of this method, the method 4 in the literature and the method 5 in the literature are selected as the comparison method, and the number of iterations is set to 120.Robustness under different energy systems is assessed to determine the reliability of the system and its ability to adapt to seasonal load changes.The robustness results of integrated energy multiscale scheduling in different methods are shown in Figure 8.
According to the analysis in Figure 8, the robustness of the proposed method is greater than that of the methods in Junjie et al. 4 and Wei et al. 5 After the completion of the iteration, the robustness of the method in this paper is up to 97%, that of the method in Junjie et al. 4 is up to 94%, and that of the method in Wei et al. 5 is up to 85%.This is because the method in this paper considers the difference in seasonal load, and the multiscale scheduling of regional integrated energy is relatively robust.

| CONCLUSION
To improve the multiscale dynamic time-domain distribution capability of regional comprehensive energy, the power grid and comprehensive energy system are coordinated and planned through multiple operation modes of supply, so as to realize the multiscale dynamic distribution of regional comprehensive energy.This paper proposes a multiscale dynamic time-domain scheduling model of regional integrated energy with seasonal load difference, and constructs a master-slave game coordination planning model of regional integrated energy multiscale scheduling.On the basis of the coordinated control of active and reactive power, the proportion of energy and electricity under the regional distribution mechanism of seasonal load is obtained according to the maximum and minimum output limits of the regional comprehensive energy power supply.According to seasonal load differential scheduling, power tracking control measures are taken to ensure the maximum power output of PV modules, and a multiscale dynamic time-domain analysis and dynamic simulation model of regional integrated energy is constructed to achieve multiscale dynamic time-domain feature extraction of integrated energy and seasonal load differential integrated scheduling.The simulation test shows that the multiscale dynamic time-domain simulation of integrated energy is carried out by the method in this paper, which improves the energy dispatching and balanced allocation capability, the time-domain dynamic allocation capability is good, and the scheduling deviation is small.

F I G U R E 2
Blocked area of power dispatching of photovoltaic power supply access to active distribution network voltage.F I G U R E 3 Schematic diagram of master-slave game model.
T A B L E 2 Cointegration vector estimation table for multiscale dynamic scheduling of multi-integrated energy.