New energy consumption scheduling method considering electric vehicle

Abstract

The invention discloses a new energy consumption scheduling method considering an electric vehicle. The method comprises the following steps: establishing a day-ahead optimization model; performing optimization solution on the day-ahead scheduling by adopting a distributed algorithm to obtain a day-ahead scheduling scheme; constructing an intra-day optimization model; and according to the load and the photovoltaic prediction error value, an intra-day optimization scheduling model strategy is implemented, and rapid response is carried out by adjusting regional power grid interaction, an electric vehicle and energy storage so as to eliminate the error to the maximum extent. By means of the energy storage system and full play of the flexibility of the electric vehicle cluster, the power between different connecting lines can be flexibly adjusted, and it is ensured that new energy can be stably consumed. Meanwhile, load prediction errors need to be considered, and in order to enhance the reliability and stability of the system, the operation of closely tracking the scheduling plan is executed. The rolling optimization strategy can effectively cope with the uncertainty of the day-ahead stage, ensure the smooth consumption of new energy, and reduce the influence of prediction errors on the operation of the system.

Description

Technical Field

[0001] The present invention relates to a scheduling method, and in particular, to a scheduling method for new energy consumption taking electric vehicles into consideration. Background Art

[0002] Faced with the instability and volatility of new energy, ensuring the safe operation of the power grid has become challenging, and formulating effective dispatching plans to increase the consumption of new energy has become an important research topic. Monitoring and controlling the power grid and innovating new energy equipment are the key to solving this problem. Through technological innovation and system optimization, the challenges brought by the intermittent and volatile nature of new energy can be solved and the sustainable development of new energy can be achieved. This is of great significance for promoting the transformation of clean energy and maintaining the safety and stability of the power grid.

[0003] In order to promote the consumption of high proportion of new energy and realize the economic dispatch of micro energy network, there are technical solutions to analyze the specific composition of micro energy network and the role of various resources and equipment in the consumption of new energy. First, the technical solution models the internal resources of the system such as electric boilers, gas turbines, cogeneration units, electric vehicles, and demand response. Among them, in terms of electric vehicles, the solution analyzes the dispatch conditions of a single electric vehicle, constructs the constraints of vehicle charging and discharging power, charging and discharging state, and charge state, and builds a model around the charging and discharging process of electric vehicles. Secondly, combined with the relevant theories of mixed integer linear programming, the economic dispatch model of micro energy network is established, and GUROB I is used to solve it, so as to analyze the high proportion of new energy consumption in the energy network, explore the role of resources such as electric vehicles and demand response in the consumption of new energy, and lay the foundation for further improving the energy utilization efficiency of the system and ensuring the stability of the system.

[0004] In order to solve the problems caused by the randomness of distributed power generation and the superposition of various types of load fluctuations to the operation and control of the distribution network and improve the operating efficiency of the integrated energy system, existing technical solutions have established a multi-time scale day-ahead and intraday optimization scheduling model, predicted the output and load of new energy through historical data and weather forecasts, and established a day-ahead scheduling model based on constraints such as electric vehicles, voltage, flexible loads, and equipment output boundaries. The scheduling model is optimized with operational economy and environmental protection of new energy consumption as the two aspects, and intelligent optimization algorithms such as GA genetic algorithm, cuckoo algorithm, and imperial competition algorithm are used to solve the scheduling model, thereby improving the synergy of electric vehicles, energy storage and other equipment participating in the integrated scheduling of source, grid, load and storage, and improving the new energy consumption capacity.

[0005] However, due to the randomness of new energy, the traditional active distribution network distributed dispatching strategy usually adopts day-ahead planning, but this will lead to a large day-ahead prediction error, thereby weakening the robustness of the system and making it difficult to apply the established dispatching strategy in practice; and the intraday dispatching model has a large amount of calculation to solve, making it difficult to achieve a faster response. Therefore, in the improvement research, it is necessary to fully consider the role of electric vehicles in the mutual assistance of electric energy, and develop more accurate and robust dispatching strategies to solve the randomness problem of new energy, and use distributed algorithms to improve computing efficiency. Through this measure, the effect of mutual assistance of electric energy between distribution network areas can be effectively improved, and the level of consumption of new energy can be further promoted. Summary of the invention

[0006] The goal of the present invention is to provide a distributed algorithm that considers electric vehicles to improve the new energy consumption strategy, so as to make full use of the role of electric vehicles. The method is divided into two stages. In the first stage, i.e., the day-ahead stage, by comprehensively considering the electric vehicles and energy-consuming equipment in different regions, the energy mutual assistance and complementarity between the distribution network areas are achieved, thereby improving the system's ability to absorb new energy. By referring to historical data, a day-ahead optimization model is established to predict energy data for a period of time in the future. Subsequently, in the second stage, the strategy takes into account the source-load prediction error, and will make appropriate fine-tuning to the energy storage system of the electric vehicle, and will ensure that it is implemented according to the day-ahead plan in the future stage, and complete the second stage of new energy vehicle consumption; the second stage uses a synchronous ADMM algorithm to preliminarily optimize the model for the distributed day-ahead scheduling, and then uses the intraday optimization model for further optimization, and accurately adjusts the energy consumption of regional power grid interaction, electric vehicles, and energy storage. Through these technologies, the advantages of electric vehicles can be effectively utilized, the dependence on day-ahead predictions can be reduced, and the stability of the system and the efficiency of new energy consumption can be improved. The technical scheme is as follows:

[0007] A new energy consumption scheduling method considering electric vehicles is characterized by comprising the following steps:

[0008] Step 1: Establish a day-ahead optimization model;

[0009] Step 2: Use the synchronous ADMM algorithm to optimize and solve the model for the distributed day-ahead scheduling;

[0010] Step 3: Construct an intraday optimization model;

[0011] Step 4: Based on the load and PV forecast error values, implement the intraday optimization dispatch model strategy to quickly respond by adjusting regional grid interaction, electric vehicles, and energy storage to minimize errors.

[0012] The present invention also discloses a new energy consumption scheduling device considering electric vehicles, which is characterized by:

[0013] Establish a day-ahead optimization model module based on load and photovoltaic forecast values;

[0014] The distributed algorithm is used to optimize and solve the day-ahead scheduling, and the day-ahead scheduling solution module is obtained;

[0015] Construct intraday optimization model module;

[0016] According to the load and photovoltaic forecast error values, the intraday optimization scheduling model strategy is implemented to quickly respond by adjusting regional power grid interaction, electric vehicles and energy storage to eliminate the error module to the greatest extent.

[0017] The present invention also discloses a non-volatile storage medium, characterized in that the non-volatile storage medium includes a stored program, wherein when the program is running, the device where the non-volatile storage medium is located is controlled to execute the above method.

[0018] The present invention also discloses an electronic device that takes into account the new energy consumption scheduling of electric vehicles, characterized in that it includes a processor and a memory; the memory stores computer-readable instructions, and the processor is used to run the computer-readable instructions, wherein the computer-readable instructions execute the above method when running.

[0019] Beneficial effects:

[0020] (1) It can simultaneously ensure the economy and reliability of the distribution network and better meet the needs of the energy system;

[0021] (2) Efficiently handle large-scale consumption models and improve computing speed and efficiency, thereby effectively alleviating the computing pressure brought by centralized optimization;

[0022] (3) With the help of energy storage systems and the flexibility of electric vehicle clusters, the power between different connection lines can be flexibly adjusted to ensure that new energy can be smoothly absorbed. At the same time, the error of load forecasting must also be considered, and in order to enhance the reliability and stability of the system, the operation of closely tracking the dispatch plan is performed. This rolling optimization strategy can effectively deal with the uncertainty of the day-ahead stage, ensure the smooth absorption of new energy, and reduce the impact of forecast errors on system operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a diagram of the optimization scheduling framework of the present invention;

[0024] Figure 2 Schematic diagram of the operating domain of a single electric vehicle of the present invention

[0025] Figure 3 It is a curve diagram of the power generation of new energy in the area of ​​the present invention;

[0026] Figure 4It is a curve diagram of load setting values ​​of each area of ​​the present invention;

[0027] Figure 5 It is the interactive power curve diagram of each area of ​​the present invention. DETAILED DESCRIPTION

[0028] A new energy consumption scheduling method considering electric vehicles includes the following steps:

[0029] Step 1: Build a day-ahead optimization model

[0030] 1) Objective function:

[0031] In order to promote the consumption of new energy in different regions, reduce the demand for electricity purchase from the upper power grid, and take into account its own operating costs, the objective function of the design day optimization model is optimized. The objective function aims to achieve two goals: one is to reduce the purchase of electricity, that is, to reduce the demand for purchasing electricity from the upper power grid as much as possible and reduce the load pressure; the second is to consider its own operating costs, that is, to reduce operating costs and improve economic benefits through the optimal scheduling strategy. By setting a reasonable objective function, the best scheduling plan can be found, which can effectively promote the consumption of new energy and ensure the optimization of the system in terms of economy and operating costs. The objective function of the forward optimization model is:

[0032]

[0033] Where: P gird (t) is the power of interaction with the power grid; P DG (t) is the power of the diesel generator; c is the power generation cost of the diesel generator; C Grid+ , C Grid- (t) are the interaction prices with the superior power grid, where Grid+ means purchasing electricity from the superior power grid, and Grid- means selling electricity to the superior power grid; γ1 and γ2 are the corresponding target coefficients, and N is an integer, indicating that there are N regional power grids.

[0034] 2) Energy storage equipment constraints:

[0035] In order to improve the stability of the system, energy storage equipment is introduced to use its fast response characteristics to better track the day-ahead scheduling plan. According to the energy storage requirements of the system, a corresponding energy storage model is established. Through this constraint, the reliability and flexibility of the system are improved to a new level to meet the scheduling requirements of the daily stage and ensure the normal operation of all aspects of the system. The application of energy storage equipment will have a positive impact on the performance of the system and further optimize the operation effect of the system. The established energy storage model is as follows:

[0036]

[0037] P min,si ≤P si (t)≤Pmax,si (3)

[0038] S SOC,min ≤S SOC (t)≤S SOC，max (4)

[0039] Where: P max,si , P min,si are the upper and lower limits of the energy storage system’s charging and discharging power, respectively; η c , η disc are the charging and discharging efficiency of the energy storage system; P si (t) is the charging and discharging power of the energy storage system at time t; S SOC，max , S SOC,min They are the upper and lower limits of the energy storage system charge, SOC is the energy storage battery, t is the time, c represents charging, and disc represents discharging.

[0040] 3) Electric vehicle model:

[0041] (1) Electric vehicle aggregation model

[0042] Electric vehicles are huge in number, have great dispatching potential, and are fast and flexible. For better illustration, take the mileage of an i-type electric vehicle ( Figure 2 ) is used as an example to explain in detail. By dispatching and managing electric vehicle groups, the flexibility and response speed of electric vehicles are fully utilized to achieve the optimal operation of the system.

[0043] from Figure 2 The observation results show that at time t in , electric vehicle i is connected to the grid and charges at maximum power along the AB path, while discharging at maximum power along the AD path. Therefore, the constraints for charging and discharging power are established as follows:

[0044] P EV,min ≤P EV,i (t)≤P EV，max (5)

[0045] Where: P EV,i (t) is the power of the electric vehicle; P EV，max , P EV,min They are the upper and lower limits of electric vehicle power respectively.

[0046] exist Figure 2 In the above, it is found that the BC end limits the maximum value of the electric vehicle power, the DE end limits the minimum value of the electric vehicle power, the CF end represents the acceptable power range when the electric vehicle is charging, and the EF end represents the situation where the electric vehicle forced charging curve reaches the CF boundary. Based on these observations, the following constraints are established:

[0047]

[0048] S SOC,EV,i,min ≤S SOC,EV,i (t)≤S SOC,EV,i，max (7)

[0049] 0.95C i ≤S SOC,EV,last,i ≤C i (8)

[0050] Where: C i is the power of the i-th electric car; S SOC,EV,last,i The power of the electric vehicle when charging is finished; P EV,i (t) is the charging and discharging power of the i-th electric vehicle; S SOC,EV,i,min , S SOC,EV,i,max are the upper and lower limits of the power of the i-th electric vehicle.

[0051] (2) Electric vehicle travel pattern model:

[0052] After collecting statistics, the U.S. Department of Transportation standardized the driving characteristics of American family cars. Using the maximum likelihood estimation method, it is approximately assumed that the user's final return time conforms to the normal distribution, and the corresponding distribution function is given. In order to simulate the uncertain travel mode of electric vehicles, a random driving mode simulation is performed. Based on the Monte Carlo method, office areas and residential areas in different areas of the distribution network are sampled to obtain the distribution function of electric vehicles. Using these sample data can better understand the mobility pattern of electric vehicles.

[0053]

[0054]

[0055] The distribution function of the return time of users in residential areas is expressed by formula (9), where: μ t is the expected value, take 17.6; σ t is the standard deviation, take 3.4; t h is the return time of users in the residential area. The distribution function of the arrival time of users in the office area is expressed by formula (8), where: μ b is the expected value, take 11; t start ,t end are the start and end time, which are 4 and 20 respectively; t b Return time for users in the office area.

[0056] 4) Diesel generator constraints:

[0057] The diesel generator model includes elements of power limitation and speed constraint:

[0058] P DG,min≤P DG (t)≤P DG，max (11)

[0059] P DG,min ≤P DG (t)-P DG (t-1)≤ΔP DG，max (12)

[0060] Where: P DG (t) is the output of the diesel generator set at time t; P DG，max , P DG,min They are the upper and lower limits of diesel generator set output respectively; ΔP DG，max , ΔP DG,min They are the uphill and downhill climbing power of the diesel generator set output respectively.

[0061] 5) Constraints on new energy output:

[0062]

[0063] P grid,min ≤P grid (t)≤P grid,max (14)

[0064] Where: P PV (t), P WT (t) are the output of photovoltaic and wind turbine at time t, P grid,max , P grid,min They are the upper and lower limits of the power interaction with the upper power grid.

[0065] Step 2: Use the synchronous ADMM algorithm to optimize the model for the distributed scheduling of the day-ahead

[0066] 1) Optimization problem decomposition:

[0067] Considering that the direct solution may bring a huge computational burden to the dispatch center, a distributed solution based on synchronous ADMM is adopted to ensure that the coupling variables in adjacent areas are equal, thereby ensuring the consistency of solving the problem. Establish a control model:

[0068]

[0069]

[0070] Where: f i (x i ) represents the target subfunction of region i; h i (x i ) represents the power equality constraint of region i; g i (x i) represents the inequality constraint of the device in region i; X i represents the coupling variable between system i and the adjacent region; j and k are the regions adjacent to i.

[0071] 2) Synchronous ADMM distributed solution:

[0072] As a distributed solution with fast convergence speed and strong robustness, ADMM can simplify complex optimization problems by decomposing the problem. It is suitable for solving large-scale optimization problems. Therefore, it has been widely used in recent years. For the constructed convex optimization model, this task involves separable objective functions and linear boundary coupling constraints. In order to solve this problem, we can use the synchronous ADMM method. By using a distributed solution method, we can solve each area independently and process the Lagrangian function related to the objective function separately. This method enables us to better understand and solve the problem. The Lagrangian function corresponding to the objective function is as follows:

[0073]

[0074]

[0075] Where: t is the number of iterations; is the penalty parameter of ADMM; i and λ j are the quantities composed of the dual variables of regions i and j respectively.

[0076] Assume that regions i, j, k are coupled, and regions j, l, i are also coupled. At the t+1th iteration, the reference values ​​of regions i and j are X Ki,t With X Kj,t In the tth iteration, we will get the average of the adjacent coupled branch states from regions i and j and use it as the basis for the calculation:

[0077]

[0078] The controller in each region performs a distributed solution to the optimization problem as follows:

[0079] x i,t+1 =argminL i (x i ,X Ki,t ,λ i,t ) (19)

[0080] x j,t+1 =argminL j (x j ,X Kj,t ,λ j,t ) (20)

[0081] (1) In the t+1th iteration, the objective functions of regions i and j are solved independently, and L is calculated in parallel. i (x i ,X Ki,t ,λ i,t ) and L j (x j ,X Kj,t ,λ j,t ), and obtain the coupling branch state x of each region i,t+1 and x j,t+1 ;

[0082] (2) Based on the coupling variable states of region i and region j, the average value of the coupling variable is calculated using equation (18) to obtain X Ki,t and X Kj,t , and use it as the reference value for the next iteration;

[0083] (3) Perform a loop update operation for a dual variable in each region:

[0084] λ i,t+1 =λ i,t +(X i,t+1 -X Ki,t+1 ) (twenty one)

[0085] λ j,t+1 =λ j,t +(X j,t+1 -X Kj,t+1 ) (twenty two)

[0086] (4) In order to verify the convergence of the algorithm, the following situation can be considered: when the regional residual value gradually approaches zero and reaches the preset number of iterations, or the Euclidean norm square of the adjacent coupling factor is less than or equal to the preset residual test threshold, then:

[0087]

[0088] Jump out of the iteration and the algorithm ends.

[0089] Step 3: Construct an intraday optimization model;

[0090] 1) Intraday forecasting model:

[0091] The deviation between the actual operation plan and the advance scheduling plan was corrected by using model predictive control (MPC). Taking the closed-loop network formed by paired connections of regions i, j and k as an example, the following state variables are selected: interactive power with the main power grid, interactive power of adjacent connecting lines between regions i and j, interactive power of adjacent connecting lines between regions j and k, solar power generation, load consumption capacity, air conditioning demand capacity, and electric vehicle charging power. To ensure rapid response capability, energy storage solutions and electric vehicle systems are introduced and allowed to interact with adjacent connecting lines between regions i and j, and between regions j and k. In order to establish an intraday prediction model, taking into account intraday power fluctuations, the daily varying energy output is used as the control variable, and the solar power generation and load prediction errors are used as external interference signals. The intraday prediction model is:

[0092]

[0093] 2) Intraday objective function:

[0094] The goal of intraday rolling optimization is to correct the difference between the actual intraday operation and the previously planned values, so as to give full play to the role of the energy storage system in peak load shaving and valley filling. The objective function is:

[0095]

[0096] Where: P grid,ref (t) is the day-ahead grid interaction power; P MGi_j,ref (t), P MGj_k,ref (t) are the day-ahead interaction powers of adjacent interconnection lines between regions i and j, and between regions j and k, respectively.

[0097] 3) Constraints:

[0098] (5) Power balance conditions:

[0099] P PV (t)+P grid (t)+P MGi_j (t)+P MGj_k (t)+P si (t) = P load (t)+P sum,EV (t) (26)

[0100] Where: P PV (t), P WT (t), P load (t) are the predicted output of photovoltaic power, wind turbine and load in a day; P sum,EV (t) is the air conditioning aggregate power.

[0101] (6) Aggregate electric vehicle load constraints:

[0102] The load regulation of electric vehicles is expressed in the following way:

[0103] ΔP sum,EV,min (t)≤ΔP sum,EV (t)≤ΔP sum,EV,max (t) (27)

[0104] Where: ΔP sum,EV,max (t), ΔP sum,EV,min (t) is the upper and lower limits of load regulation of aggregated electric vehicles.

[0105] (7) Interaction power constraints of adjacent grid tie lines:

[0106] In order to enhance the power complementarity and mutual assistance between different regions within a day, the following constraints are introduced:

[0107] ΔP MGi_j,EV,min ≤ΔP MGi_j (k)≤ΔP MGi_j,EV,max (28)

[0108] ΔP MGj_k,EV,min ≤ΔP MGj_k (k)≤ΔP MGj_k,EV,max (29) To ensure the interaction of communication lines between regions:

[0109] Where: ΔP MGj_k (k), ΔP MGi_j (k) are the day-ahead adjustment of the interactive power between adjacent interconnection lines between area j and area k, and between area i and area j.

[0110] (8) Energy storage system constraints:

[0111] In order to better adapt to the fluctuation of the remaining new energy and achieve smooth absorption, the intraday energy storage system adds constraints, which can be seen in equations (2) to (4).

[0112] Example 1

[0113] Taking an active distribution network with three regions as the object, in the power system networking, the rated power of the MG1 region is 1500 kW, the installed capacity of the wind turbine is 2500 kW, the capacity of the energy storage system is 750 kWh, and the rated power of the diesel generator is 500 kW. In addition, the installed capacity of solar power generation in the MG2 region is 1300 kW, while the installed capacity of wind turbines in the MG3 region is 1200 kW. At the same time, the upper limit of the interaction power of the contact line between regions is limited to 300kW. The detailed information of the equipment is shown in Table 1:

[0114] Table 1 System resource configuration in each region

[0115]

[0116] Typical output curves of wind turbine and photovoltaic equipment in various regions are as follows: Figure 3 shown.

[0117] In order to set the load value of each area, the load curve representing typical industries and office residents was selected as a reference in this case. The typical load curve is as follows: Figure 4 shown.

[0118] To deal with the current problem, a distributed solution method based on the synchronized alternating direction method of multipliers (S-ADMM) is used to solve it. The interval time for each solution task is set to 1 hour. During the solution process, the penalty factor ρ is set to 0.05, and the convergence parameter δ is set to 1e-3. In addition, the maximum number of iterations is limited to 50 times. After about 10 iterations, satisfactory results are obtained and the solution results reach a stable state. After about 20 iterations, it can be observed that the residual has a slight fluctuation, but then it stabilizes. This result proves the effectiveness of the target decomposition optimization method used previously, and shows that the variables in adjacent coupling regions are very close. By adopting a distributed solution method, while obtaining the same results as the centralized solution method, the computational pressure is greatly reduced.

[0119] Depend on Figure 5 It can be seen that: during the period of 05:00-06:00, the photovoltaic power generation in the MG2 area is insufficient, while the wind turbines in the MG3 area have a higher output during this period, and can be transmitted to the MG2 area while meeting the load. The MG1 area also transmits electricity to the MG2 area via the interconnection line. During the period of 12:00-13:00, the output of new energy in the MG3 area is relatively small. At this time, the MG2 area transmits electricity to the MG3 area through the adjacent power grid interconnection line.

[0120] While obtaining the interaction power of the interconnection lines between adjacent regions, the comparison of the interconnection lines of the power grid within a few days is as follows: through the connecting lines, the adjacent regions have achieved interaction, successfully controlled the energy consumption in the first stage of the system, and ensured the rational use of new energy in each system. When the new energy generation capacity of a region is not enough to meet the load demand, the excess power of the region will be transmitted to the area that needs to supplement power through the adjacent grid-connected lines, thereby achieving power balance.

[0121] The new energy consumption strategy adopts the daytime energy consumption strategy, taking MG2 as an example. Based on the day-ahead load, a strategy of introducing random disturbance is adopted to implement the scheme to simulate the output of renewable energy and load on that day.

[0122] Based on the established prediction model (22), a rolling optimization solution is performed. Through this process, the comparison between the power grid and the upstream power grid is successfully obtained, and the power status of each control variable in response to a day is also obtained.

[0123] In most time periods, the power of the grid's interconnection lines is related to the dispatch time within the day. On this basis, the government will uniformly dispatch basic state variables such as electric vehicles, energy storage systems, MG1 area, MG2 area and MG area, so as to complete the timely tracking of the day-ahead dispatch plan, thereby completing the effective consumption of the newly added electricity within the day. Considering the particularity of the operation rules of electric vehicles in residential areas, most electric vehicles respond in the afternoon or at night. With the increase in photovoltaic power generation, it is necessary to adjust the charging strategy of electric vehicles and energy storage equipment, and the excess power is transmitted to the neighboring power grid through the grid-connected line. However, there is still excess power from photovoltaic power generation that cannot be fully absorbed, resulting in a power deviation in the interconnection line with the upper power grid, so the way of selling electricity to the grid is added to maintain the power balance.

[0124] Example 2

[0125] A new energy consumption dispatching device considering electric vehicles is characterized by:

[0126] Establish a day-ahead optimization model module based on load and photovoltaic forecast values;

[0127] The distributed algorithm is used to optimize and solve the day-ahead scheduling, and the day-ahead scheduling solution module is obtained;

[0128] Construct intraday optimization model module;

[0129] According to the load and photovoltaic forecast error values, the intraday optimization scheduling model strategy is implemented to quickly respond by adjusting regional power grid interaction, electric vehicles and energy storage to eliminate the error module to the greatest extent.

[0130] Example 3

[0131] A non-volatile storage medium, characterized in that the non-volatile storage medium includes a stored program, wherein when the program is run, the device where the non-volatile storage medium is located is controlled to execute the above method.

[0132] Example 4

[0133] An electronic device that considers the new energy consumption scheduling of electric vehicles is characterized by comprising a processor and a memory; the memory stores computer-readable instructions, and the processor is used to run the computer-readable instructions, wherein the computer-readable instructions execute the above method when running.

[0134] In order to increase the consumption of new energy in the region, the present invention proposes an optimization model that takes its own interests into consideration, with the goal of maximizing the consumption of new energy and minimizing the number of interactions with the large power grid; in the first stage of actively utilizing new energy, electric vehicles are regarded as a dispatchable resource, and the unique travel patterns of each region are combined to coordinate and dispatch the internal electrical facilities. At the same time, on the basis of considering the power interaction between adjacent regions, the goal of complementary effects between regions is successfully achieved, the absorption of new energy in the region is increased, and the stability of the entire system is improved; by decomposing the centralized optimization problem into multiple sub-problems and solving them in a distributed manner, the amount of calculation is successfully reduced. This method makes complex problems more feasible and technically achieves efficient step-by-step processing; in the second stage, the error of the day-ahead source power forecast is considered, and the characteristics of rapid response of energy storage systems and electric vehicles are introduced in detail. By coordinating the two, the transmission power of the interconnection lines between regions can be accurately tracked, the new energy consumption in the second stage can be achieved, and accurate tracking can be performed according to the scheduling plan of the previous day.

[0135] 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 to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A new energy consumption dispatching method considering electric vehicles, characterized by: The following steps are involved: Step 1: Establish a day-ahead optimization model based on load and photovoltaic forecast values; Step 2: Use a distributed algorithm to optimize and solve the day-ahead scheduling to obtain the day-ahead scheduling solution; Step 3: Construct an intraday optimization model; Step 4: Based on the load and PV forecast error values, implement the intraday optimization dispatch model strategy to quickly respond by adjusting regional grid interaction, electric vehicles, and energy storage to minimize errors.

2. The new energy consumption scheduling method considering electric vehicles according to claim 1 is characterized in that: the step 1 further includes the following contents; (1) Establish the objective function of the forward optimization model; (2) Establish energy storage equipment constraint model; (3) Establish an electric vehicle model; (4) Establish a diesel generator constraint model.

3. The new energy consumption scheduling method considering electric vehicles according to claim 2 is characterized by: The objective function of the forward optimization model is: Where: P gird (t) is the power of interaction with the power grid; P DG (t) is the power of the diesel generator; c is the power generation cost of the diesel generator; C Grid+ , C Grid- (t) are the interaction prices with the upper-level power grid; γ1 and γ2 are the corresponding target coefficients. N is an integer, indicating that there are N regional power grids, Grid+ means purchasing electricity from the upper-level power grid, and Grid- means selling electricity to the upper-level power grid.

4. The new energy consumption dispatching method considering electric vehicles according to claim 2 is characterized by: The energy storage model is as follows: SOC is the energy storage battery, t is the time, c represents charging, and disc represents discharging. P min,si ≤P si (t)≤P max,si (3) S SOC,min ≤S SOC (t)≤S SOC，max (4) Where: P max,si , P min,si are the upper and lower limits of the energy storage system’s charging and discharging power, respectively; η c , η disc are the charging and discharging efficiency of the energy storage system; S SOC，max , S SOC,min They are the upper and lower limits of the energy storage system charge respectively.

5. The new energy consumption dispatching method considering electric vehicles according to claim 2 is characterized by: The electric vehicle aggregation model is as follows: by analyzing the charging and discharging operation domain of a single electric vehicle, according to the initial SOC situation and the acceptable power range of the owner when leaving, combined with the charging and discharging power of the electric vehicle battery, the charging and discharging operation constraints of the electric vehicle are established, and the charging and discharging SOC aggregation model of the electric vehicle is proposed; The travel pattern model of electric vehicles is as follows: based on the statistics of electric vehicle user data, random sampling is performed on the travel distribution functions of electric vehicles in office and residential areas in different regions to simulate the random travel pattern of electric vehicles and the normal distribution of user travel load. The actual travel probability data of users is substituted to obtain specific parameters.

6. The new energy consumption scheduling method considering electric vehicles according to claim 2 is characterized in that: the step 2 further includes the following contents: 1) Optimization problem decomposition: Build the control model: Where: f i (x i ) represents the target subfunction of region i; h i (x i ) represents the power equality constraint of region i; g i (x i ) represents the inequality constraint of the device in region i; X i represents the coupling variable between system i and the adjacent region; j and k are the regions adjacent to i; 2) Synchronous ADMM distributed solution: The Lagrangian function corresponding to the objective function is as follows: Where: t is the number of iterations; is the penalty parameter of ADMM; i and λ j are the quantities composed of the dual variables of regions i and j respectively; Assume that regions i, j, k are coupled, and regions j, l, i are also coupled; at the t+1th iteration, the reference values ​​of regions i and j are X Ki,t With X Kj,t In the tth iteration, we will get the average of the adjacent coupled branch states from regions i and j and use it as the basis for the calculation: The controller in each region performs a distributed solution to the optimization problem as follows: x i,t+1 =arg minL i (x) i ,X Ki,t ,λ i,t ) (19) x j,t+1 =arg minL j (x) j ,X Kj,t ,λ j,t ) (20) (1) In the t+1th iteration, the objective functions of regions i and j are solved independently, and L is calculated in parallel. i (x i ,X Ki,t ,λ i,t ) and L j (x j ,X Kj,t ,λ j,t ), and obtain the coupling branch state x of each region i,t+1 and x j,t+1 ; (2) Based on the coupling variable states of region i and region j, the average value of the coupling variable is calculated using equation (18) to obtain X Ki,t and X Kj,t , and use it as the reference value for the next iteration; (3) Perform a loop update operation for a dual variable in each region: l i,t+1 =λ i,t +(X i,t+1 -X Ki,t+1 ) (21) l j,t+1 =λ j,t +(X j,t+1 -X Kj,t+1 ) (22) (4) In order to verify the convergence of the algorithm, the following situations can be considered: when the regional residual value gradually approaches zero and reaches the preset number of iterations, or the Euclidean norm square of the adjacent coupling factor is less than or equal to the pre-set residual test threshold, then: jump out of the iteration and the algorithm ends.

7. The new energy consumption scheduling method considering electric vehicles according to claim 3 is characterized in that: the step 3 further includes the following contents: 1) Intraday forecasting model: Model predictive control (MPC) is used to correct the deviation between the actual operation plan and the advance scheduling plan; taking the closed-loop network formed by the paired connection of areas i, j and k as an example, the following state variables are selected: the interactive power with the main power grid, the interactive power of the adjacent connecting lines between areas i and j, the interactive power of the adjacent connecting lines between areas j and k, the solar power generation, the load consumption capacity, the air conditioning demand capacity and the electric vehicle charging power; it is allowed to interact with the adjacent connecting lines between areas i and j, and between areas j and k; in order to establish an intraday prediction model, taking into account the intraday power fluctuation, the daily changing energy output is used as the control variable, and the solar power generation and load prediction errors are used as external interference signals. The intraday prediction model is: 2) Intraday objective function: The goal of intraday rolling optimization is to correct the difference between the actual intraday operation and the previously planned values, so as to give full play to the role of the energy storage system in peak load shaving and valley filling. The objective function is: Where: P grid,ref (t) is the day-ahead grid interaction power; P MGi_j,ref (t), P MGj_k,ref (t) are the day-ahead interaction powers of adjacent tie lines between regions i and j, and between regions j and k, respectively; 3) Constraints: (1) Power balance conditions: P PV (t)+P grid (t)+P MGi_j (t)+P MGj_k (t)+P si (t)=P load (t)+P sum,EV (t) (26) Where: P PV (t), P WT (t), P load (t) are the predicted output of photovoltaic power, wind turbine and load in a day; P sum,EV (t) is the air conditioning aggregate power; (2) Aggregate electric vehicle load constraints: The load regulation of electric vehicles is expressed in the following way: ΔP sum,EV,min (t)≤ΔP sum,EV (t)≤ΔP sum,EV,max (t) (27) Where: ΔP sum,EV,max (t), ΔP sum,EV,min (t) is the upper and lower limits of load regulation of aggregated electric vehicles; (3) Interaction power constraints of adjacent grid tie lines: In order to enhance the power complementarity and mutual assistance between different regions within a day, the following constraints are introduced: ΔP MGi_j,EV,min ≤ΔP MGi_j (k)≤ΔP MGi_j,EV,max (28) ΔP MGj_k,EV,min ≤ΔP MGj_k (k)≤ΔP MGj_k,EV,max (29) To ensure the interaction of communication lines between regions: Where: ΔP MGj_k (k), ΔP MGi_j (k) are the day-ahead adjustment of the interactive power between adjacent tie lines between area j and area k, and between area i and area j; (4) Energy storage system constraints: In order to better adapt to the fluctuation of remaining renewable energy and achieve smooth absorption, constraints are added to the intraday energy storage system. These constraints can be found in equations (2) to (4).

8. A new energy consumption dispatching device considering electric vehicles, characterized by: Establish a day-ahead optimization model module based on load and photovoltaic forecast values; The distributed algorithm is used to optimize and solve the day-ahead scheduling, and the day-ahead scheduling solution module is obtained; Construct intraday optimization model module; According to the load and photovoltaic forecast error values, the intraday optimization scheduling model strategy is implemented to quickly respond by adjusting regional power grid interaction, electric vehicles and energy storage to eliminate the error module to the greatest extent.

9. A non-volatile storage medium, characterized in that: The non-volatile storage medium includes a stored program, wherein when the program is executed, the device where the non-volatile storage medium is located is controlled to execute the method according to any one of claims 1 to 7.

10. An electronic device for considering the new energy consumption and scheduling of electric vehicles, characterized in that: It comprises a processor and a memory; the memory stores computer-readable instructions, and the processor is used to execute the computer-readable instructions, wherein the computer-readable instructions execute the method described in any one of claims 1 to 7 when executed.

Images