Mobile energy storage auxiliary scheduling method, system and equipment based on multi-region power utilization requirements

Abstract

The invention provides a mobile energy storage auxiliary scheduling method, system and equipment based on multi-area power consumption demands, and the method comprises the steps: firstly predicting the power consumption of a plurality of power consumption areas in an emergency scene, then constructing a mobile energy storage vehicle optimization scheduling model with the maximum overall income in a power supply process as a target based on the power consumption of the plurality of power consumption areas, and carrying out the optimization scheduling of the mobile energy storage vehicle. And finally, solving the mobile energy storage vehicle optimization scheduling model to obtain a mobile energy storage vehicle optimization scheduling scheme. According to the invention, the satisfaction rate of regional power demand can be improved, the power consumption and power supply balance of each region is maintained, and the overall benefit of power supply is improved.

Description

Technical Field

[0001] The present invention belongs to the field of mobile energy storage scheduling, and specifically relates to a mobile energy storage auxiliary scheduling method, system and equipment based on multi-regional electricity demand. Background Art

[0002] In view of the increasing demand for electricity and the further increase in electric vehicles and other electrical equipment, the power transmission network of the power system has insufficient transmission capacity. For this reason, mobile energy storage technology has been proposed. With its flexible spatial transfer capability and stable energy storage and charging characteristics, the advantages of mobile energy storage systems in emergency power supply are gradually highlighted. In particular, truck-type mobile energy storage systems have been applied and promoted in post-disaster load recovery at home and abroad.

[0003] The existing scheduling plans are all based on scenarios that have already occurred, and cannot effectively solve the problem of untimely power supply caused by insufficient power supply of mobile energy storage vehicles, and fixed scheduling plans cannot maximize overall benefits. Summary of the invention

[0004] The purpose of the present invention is to provide a mobile energy storage auxiliary scheduling method, system and equipment based on multi-regional electricity demand in order to solve the above problems in the prior art.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows:

[0006] In a first aspect, the present invention proposes a mobile energy storage auxiliary scheduling method based on multi-regional electricity demand, comprising:

[0007] S1. In an emergency scenario, predicting the power consumption of multiple power consumption areas, wherein the multiple power consumption areas include commercial areas and residential areas;

[0008] S2. Based on the power consumption of multiple power consumption areas, a mobile energy storage vehicle optimization scheduling model is constructed with the goal of maximizing the overall benefit of the power supply process;

[0009] S3. Solve the optimal dispatch model of the mobile energy storage vehicle and obtain the optimal dispatch plan of the mobile energy storage vehicle.

[0010] In S2, the mobile energy storage vehicle optimization scheduling model includes:

[0011]

[0012] In the above formula, Ear is the overall benefit of the power supply process, Y ind,i , are the electricity consumption of the i-th commercial area and the j-th residential area, z is the dispatching plan, N s is the number of mobile energy storage vehicles in the sth power consumption area, track is the driving route of the mobile energy storage vehicle, P1,i , P 2,j are the electricity prices of the i-th commercial area and the j-th residential area, respectively. E and Δ are the electricity consumption and electricity loss per unit distance of the mobile energy storage vehicle, dis is the one-way distance of the mobile energy storage vehicle, and P ini The cost of electricity for charging the charging station, M 1 、M 2 are the number of commercial areas and residential areas respectively, Y s is the electricity consumption of the sth electricity consumption area, Aviable k is the power supply of the kth mobile energy storage vehicle, K s is the number of mobile energy storage vehicles in each power consumption area.

[0013] The S1 includes:

[0014] S11. The following information interaction is performed on the historical power consumption sequence of each power consumption area through the self-attention mechanism:

[0015]

[0016] In the above formula, is the feature mapping result of the historical electricity consumption sequence of the i-th electricity consumption area. Embedding is an independent feature mapping operation. is the lth group of data in the historical electricity consumption sequence of the i-th electricity consumption area, and L is the length of the historical electricity consumption sequence;

[0017] S12. Use layer normalization to unify the feature channel distribution of different variables:

[0018]

[0019] In the above formula, LayerNorm is the layer normalization operation, H i After layer normalization Mean is the average value, var is the variance;

[0020] S13, perform fully connected feature encoding in the feedforward network, and predict the power consumption of each power consumption area through projection layer mapping:

[0021] Y i {L:L+m} =Projection(H i )

[0022] In the above formula, Y i {L:L+m} is the power consumption prediction result of the ith power consumption area with a sequence length of m, and Projection is a mapping operation.

[0023] The S3 adopts the DDQN algorithm model to solve the mobile energy storage vehicle optimization scheduling model, including:

[0024] S31, converting the mobile energy storage vehicle optimization scheduling model into an energy storage vehicle quantity optimization model and a path planning model;

[0025] S32. Construct the equivalent Markov state transition process of the path planning model:

[0026] max Qπ(s,a;θ)=E[R 1 +γR 2 +…|S,A,π;θ]

[0027] In the above formula, Q π (s, a; θ) is the cumulative expected reward under path planning, s, a, θ are the state, action, and network parameters respectively, R is the reward under each corresponding action, S and A are the state space and action space respectively, the action space represents the choice at each intersection, π is the formulated path planning strategy, γ is the attenuation coefficient, and E[·] is the expectation for x;

[0028] S33. Use the DDQN algorithm to iteratively solve the problem, thereby obtaining the path planning strategy π, which is the track in the objective function.

[0029] In a second aspect, the present invention proposes a mobile energy storage auxiliary dispatching system based on multi-regional electricity demand, including an electricity consumption prediction module, a dispatching model construction module, and a dispatching model solving module;

[0030] The power consumption prediction module is used to predict the power consumption of multiple power consumption areas in an emergency scenario, and the multiple power consumption areas include commercial areas and residential areas;

[0031] The scheduling model building module is used to build an optimization scheduling model for mobile energy storage vehicles with the goal of maximizing the overall benefit of the power supply process based on the power consumption of multiple power consumption areas;

[0032] The scheduling model solving module is used to solve the mobile energy storage vehicle optimization scheduling model to obtain the mobile energy storage vehicle optimization scheduling plan.

[0033] The mobile energy storage vehicle optimization scheduling model includes:

[0034]

[0035] In the above formula, Ear is the overall benefit of the power supply process, Y ind,i , are the electricity consumption of the i-th commercial area and the j-th residential area, z is the dispatching plan, N s is the number of mobile energy storage vehicles in the sth power consumption area, track is the driving route of the mobile energy storage vehicle, P1,i , P 2,j are the electricity prices of the i-th commercial area and the j-th residential area, respectively. E and Δ are the electricity consumption and electricity loss per unit distance of the mobile energy storage vehicle, dis is the one-way distance of the mobile energy storage vehicle, and P ini The cost of electricity for charging the charging station, M 1 、M 2 are the number of commercial areas and residential areas respectively, Y s is the electricity consumption of the sth electricity consumption area, Aviable k is the power supply of the kth mobile energy storage vehicle, K s is the number of mobile energy storage vehicles in each power consumption area.

[0036] The power consumption prediction module uses the following steps to predict the power consumption of multiple power consumption areas:

[0037] A. The following information interaction is performed on the historical power consumption sequence of each power consumption area through the self-attention mechanism:

[0038]

[0039] In the above formula, is the feature mapping result of the historical electricity consumption sequence of the i-th electricity consumption area. Embedding is an independent feature mapping operation. is the lth group of data in the historical electricity consumption sequence of the i-th electricity consumption area, and L is the length of the historical electricity consumption sequence;

[0040] B. Use layer normalization to unify the feature channel distribution of different variables:

[0041]

[0042] In the above formula, LayerNorm is the layer normalization operation, H i After layer normalization Mean is the average value, var is the variance;

[0043] C. Perform fully connected feature encoding in the feedforward network and predict the power consumption of each power consumption area through projection layer mapping:

[0044] Y i {L:L+m} =Projection(H i )

[0045] In the above formula, Y i {L:L+m} is the power consumption prediction result of the ith power consumption area with a sequence length of m, and Projection is a mapping operation.

[0046] The scheduling model solving module adopts the DDQN algorithm model to solve the mobile energy storage vehicle optimization scheduling model, including:

[0047] a. Convert the mobile energy storage vehicle optimization scheduling model into an energy storage vehicle quantity optimization model and a path planning model;

[0048] b. Construct the equivalent Markov state transition process of the path planning model:

[0049] max Qπ(s,a;θ)=E[R 1 +γR 2 +…|S,A,π;θ]

[0050] In the above formula, Q π (s, a; θ) is the cumulative expected reward under path planning, s, a, θ are the state, action, and network parameters respectively, R is the reward under each corresponding action, S and A are the state space and action space respectively, the action space represents the choice at each intersection, π is the formulated path planning strategy, γ is the attenuation coefficient, and E[·] is the expectation for x;

[0051] c. Use the DDQN algorithm to iteratively solve the problem and obtain the path planning strategy π, which is the track in the objective function.

[0052] In a third aspect, the present invention provides a mobile energy storage auxiliary dispatching device based on multi-regional electricity demand, including a memory and a processor;

[0053] The memory is used to store computer program code and transmit the computer program code to the processor;

[0054] The processor is used to execute the aforementioned method according to the instructions in the computer program code.

[0055] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the aforementioned method when executed by a processor.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] 1. The present invention provides a mobile energy storage auxiliary scheduling method based on multi-regional electricity demand. The power consumption of multiple power consumption areas is first predicted in an emergency scenario. Then, based on the power consumption of multiple power consumption areas, a mobile energy storage vehicle optimization scheduling model is constructed with the goal of maximizing the overall benefit of the power supply process. Finally, the mobile energy storage vehicle optimization scheduling model is solved to obtain the mobile energy storage vehicle optimization scheduling plan. Starting from the perspective of the actual power demand of the region, this method optimizes the scheduling plan of the mobile energy storage vehicle while ensuring the power demand of each region. It can not only improve the satisfaction rate of regional power demand and maintain the power supply balance of each region, but also improve the overall benefit of power supply.

[0058] 2. The mobile energy storage auxiliary scheduling method based on multi-regional electricity demand of the present invention predicts the electricity consumption of each region by using the self-attention mechanism to exchange information between the historical electricity consumption of four regions, using layer normalization operation to unify the feature distribution of different variables, and performing full-connection feature encoding in the feedforward network, and directly predicting through the projection layer. This method effectively improves the accuracy of regional electricity consumption prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is a diagram of the regional network model used in Example 1.

[0060] Figure 2 This is a flowchart of the method described in Example 1.

[0061] Figure 3 This is a comparison chart of the overall benefits of the power supply process of the method described in the present invention and the fixed driving method.

[0062] Figure 4 This is a structural block diagram of the system described in Example 2.

[0063] Figure 5 This is a structural block diagram of the device described in Example 3. DETAILED DESCRIPTION

[0064] The present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0065] Embodiment 1:

[0066] This embodiment is aimed at Figure 1 The regional network model diagram shown in FIG. 1 (including four power consumption areas, namely, industrial park, residential area 1, residential area 2 and residential area 3) implements a mobile energy storage auxiliary scheduling method based on multi-regional power demand, such as Figure 2 As shown, the specific steps are as follows:

[0067] 1. In the case of a power outage, obtain the historical power consumption data D of the four power consumption areas recorded in the power grid system ind={d ind,1 ,…,d ind,L}、

[0068] 2. The following information interaction is performed on the historical power consumption sequence of each power consumption area through the self-attention mechanism:

[0069]

[0070] In the above formula, They are the feature mapping results of the historical electricity consumption series of the industrial park, residential area 1, residential area 2, and residential area 3. Embedding is an independent feature mapping operation. They are the lth group of data in the historical electricity consumption series of the industrial park, residential area 1, residential area 2 and residential area 3 respectively, and L is the length of the historical electricity consumption series.

[0071] 3. Use layer normalization to unify the feature channel distribution of different variables:

[0072]

[0073] In the above formula, LayerNorm is the layer normalization operation, H i After layer normalization Mean is the average value, var is the variance;

[0074] 4. Perform fully connected feature encoding in the feedforward network and predict the power consumption of each power consumption area through projection layer mapping:

[0075] Y i {L:L+m} =Projection(H i )

[0076] In the above formula, Y i {L:L+m} is the power consumption prediction result of the ith power consumption area with a sequence length of m, and Projection is a mapping operation.

[0077] 5. Combined with the regional network model, the overall benefit model of the power supply process is derived.

[0078] The electricity price in the commercial area is P 1 , the electricity price in the residential area is P 2 The mobile energy storage vehicle is a pure electric vehicle, and the power consumption per kilometer is E. However, the energy storage loss during the movement is not taken into account. The speed of the mobile energy storage vehicle on the trunk line is v 1 , the speed on the branch line is v 2The expenditure of using mobile vehicles to supply electricity to these four areas consists of two parts, namely the electric energy used by the electric vehicle during driving and the loss of the energy storage vehicle, which is expressed as follows:

[0079] consume=2·(E+Δ)·dis

[0080] In the above formula, consume is the overall expenditure, Δ is the loss of the energy storage vehicle per kilometer, and dis is the one-way distance traveled by the mobile energy storage vehicle.

[0081] The revenue is the electricity charges collected in these four areas, expressed as:

[0082]

[0083] In the above formula, Y ind , are the electricity consumption of the commercial area and the i-th residential area respectively.

[0084] Therefore, the overall benefits of the mobile energy storage vehicle in the power supply process are expressed as:

[0085]

[0086] In the above formula, P ini The cost of electricity to charge the charging station.

[0087] 6. Under the premise of ensuring the electricity demand in each region, the following mobile energy storage vehicle optimization scheduling model is constructed:

[0088]

[0089] In the above formula, z is the scheduling scheme, N i is the number of mobile energy storage vehicles in the i-th power consumption area, and track is the driving route of the mobile energy storage vehicle.

[0090] 7. Convert the above scheduling model into an energy storage vehicle quantity optimization model and a path planning model, which are:

[0091]

[0092] Since the number of energy storage vehicles N only meets the electricity demand of the industrial park, residential area 1, residential area 2 and residential area 3, this part can be converted into an equivalent constraint. Therefore, the optimal solution of the optimization model (P1) can be directly obtained, that is, N s =Y s / Aviable, Aviable is the power supply of each mobile energy storage vehicle. The energy storage vehicle driving route of the optimization model (P2) still needs further analysis, so the optimization model (OP) can be simplified as follows:

[0093]

[0094] 8. Construct the equivalent Markov state transition process of the path planning model (P3):

[0095] max Q π (s,a;θ)=E[R 1 +γR 2 +…|S,A,π;θ]

[0096] In the above formula, Q π (s, a; θ) is the cumulative expected reward under path planning, s, a, θ are the state, action, and network parameters respectively, R is the reward under each corresponding action, S and A are the state space and action space respectively, the action space represents the choice at each intersection, including four actions: up, down, left, and right, π is the formulated path planning strategy, γ is the attenuation coefficient, and E[·] is the expectation for x.

[0097] 9. Use the DDQN algorithm to iteratively solve the problem, and then obtain the path planning strategy π, which is the track in the objective function. The network selection and network update methods are as follows:

[0098] Y t Q =R t+1 +γQ(S t+1 ,argmaxQ(S t+1 ,a;θ t );θ t )

[0099] Y t DoubleQ =R t+1 +γQ(S t+1 ,argmaxQ(S t+1 ,a;θ t );θ t ′)

[0100] In the above formula, Y t Q , Y t DoubleQ is the optimization target of the target network and the estimated network at time t, R t+1 is the reward value obtained after executing the action at time t+1, S t+1 is the state at time t+1, Q(S t+1 ,a;θ t ) is the Q value of the target network, θ t ,θ t ′ are the parameters of the target network and the estimated network at time t respectively.

[0101] In order to verify the effectiveness of the method of the present invention, the method of Example 1 is compared with the fixed driving method, and the overall benefits of the power supply process of the two methods are calculated respectively. The results are as follows: Figure 3 It can be seen that the overall benefit of the method of the present invention is significantly higher than that of the fixed driving method.

[0102] Embodiment 2:

[0103] A mobile energy storage auxiliary dispatching system based on multi-regional electricity demand, such as Figure 4 As shown, it includes a power consumption prediction module, a scheduling model construction module, and a scheduling model solving module.

[0104] The power consumption prediction module is used to predict the power consumption of multiple power consumption areas in an emergency scenario by using the following steps:

[0105] A. The following information interaction is performed on the historical power consumption sequence of each power consumption area through the self-attention mechanism:

[0106]

[0107] In the above formula, is the feature mapping result of the historical electricity consumption sequence of the i-th electricity consumption area, Embedding is an independent feature mapping operation, D hi,l is the lth group of data in the historical electricity consumption sequence of the i-th electricity consumption area, and L is the length of the historical electricity consumption sequence;

[0108] B. Use layer normalization to unify the feature channel distribution of different variables:

[0109]

[0110] In the above formula, LayerNorm is the layer normalization operation, H i After layer normalization Mean is the average value, var is the variance;

[0111] C. Perform fully connected feature encoding in the feedforward network and predict the power consumption of each power consumption area through projection layer mapping:

[0112] Y i {L:L+m} =Projection(H i )

[0113] In the above formula, Y i {L:L+m} is the power consumption prediction result of the ith power consumption area with a sequence length of m, and Projection is a mapping operation.

[0114] The scheduling model building module is used to build the following mobile energy storage vehicle optimization scheduling model with the goal of maximizing the overall benefit of the power supply process based on the power consumption of multiple power consumption areas:

[0115]

[0116] In the above formula, Ear is the overall benefit of the power supply process, Y ind,i , are the electricity consumption of the i-th commercial area and the j-th residential area, z is the dispatching plan, N s is the number of mobile energy storage vehicles in the sth power consumption area, track is the driving route of the mobile energy storage vehicle, P 1,i , P 2,j are the electricity prices of the i-th commercial area and the j-th residential area, respectively. E and Δ are the electricity consumption and electricity loss per unit distance of the mobile energy storage vehicle, dis is the one-way distance of the mobile energy storage vehicle, and P ini The cost of electricity for charging the charging station, M 1 、M 2 are the number of commercial areas and residential areas respectively, Y s is the electricity consumption of the sth electricity consumption area, Aviable k is the power supply of the kth mobile energy storage vehicle, K s is the number of mobile energy storage vehicles in each power consumption area.

[0117] The scheduling model solving module uses the DDQN algorithm model to solve the mobile energy storage vehicle optimization scheduling model to obtain the mobile energy storage vehicle optimization scheduling plan. The solving process includes:

[0118] a. Convert the mobile energy storage vehicle optimization scheduling model into an energy storage vehicle quantity optimization model and a path planning model;

[0119] b. Construct the equivalent Markov state transition process of the path planning model:

[0120] max Qπ(s,a;θ)=E[R 1 +γR 2 +…|S,A,π;θ]

[0121] In the above formula, Q π (s, a; θ) is the cumulative expected reward under path planning, s, a, θ are state, action, and network parameters respectively, R is the reward under each corresponding action, S and A are state space and action space respectively, the action space represents the choice at each intersection, π is the formulated path planning strategy, γ is the attenuation coefficient, and E[x] is the expectation for x;

[0122] c. Use the DDQN algorithm to iteratively solve the problem and obtain the path planning strategy π, which is the track in the objective function.

[0123] Embodiment 3:

[0124] A mobile energy storage auxiliary dispatching device based on multi-regional electricity demand, such as Figure 5 As shown, including a memory and a processor;

[0125] The memory is used to store computer program code and transmit the computer program code to the processor;

[0126] The processor is used to execute the method described in embodiment 1 according to the instructions in the computer program code.

[0127] Embodiment 4:

[0128] A computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the method described in Embodiment 1 when executed by a processor.

Claims

1. A mobile energy storage auxiliary scheduling method based on multi-regional electricity demand, characterized in that: The method comprises: S1. In an emergency scenario, predicting the power consumption of multiple power consumption areas, wherein the multiple power consumption areas include commercial areas and residential areas; S2. Based on the power consumption of multiple power consumption areas, a mobile energy storage vehicle optimization scheduling model is constructed with the goal of maximizing the overall benefit of the power supply process; S3. Solve the optimal dispatch model of the mobile energy storage vehicle and obtain the optimal dispatch plan of the mobile energy storage vehicle.

2. A mobile energy storage auxiliary scheduling method based on multi-regional electricity demand according to claim 1, characterized in that: In S2, the mobile energy storage vehicle optimization scheduling model includes: In the above formula, Ear is the overall benefit of the power supply process, Y ind,i , are the electricity consumption of the i-th commercial area and the j-th residential area, z is the dispatching plan, N s is the number of mobile energy storage vehicles in the sth power consumption area, track is the driving route of the mobile energy storage vehicle, P 1,i , P 2,j are the electricity prices of the i-th commercial area and the j-th residential area, respectively. E and Δ are the electricity consumption and electricity loss per unit distance of the mobile energy storage vehicle, dis is the one-way distance of the mobile energy storage vehicle, and P ini is the cost of electricity for charging at the charging station, M1 and M2 are the number of commercial and residential areas respectively, and Y s is the electricity consumption of the sth electricity consumption area, Aviable k is the power supply of the kth mobile energy storage vehicle, K s is the number of mobile energy storage vehicles in each power consumption area.

3. A mobile energy storage auxiliary scheduling method based on multi-regional electricity demand according to claim 1 or 2, characterized in that: The S1 includes: S11. The following information interaction is performed on the historical power consumption sequence of each power consumption area through the self-attention mechanism: In the above formula, is the feature mapping result of the historical electricity consumption sequence of the i-th electricity consumption area. Embedding is an independent feature mapping operation. is the lth group of data in the historical electricity consumption sequence of the i-th electricity consumption area, and L is the length of the historical electricity consumption sequence; S12. Use layer normalization to unify the feature channel distribution of different variables: In the above formula, LayerNorm is the layer normalization operation, H i After layer normalization Mean is the average value, and var is the variance; S13, perform fully connected feature encoding in the feedforward network, and predict the power consumption of each power consumption area through projection layer mapping: Y i { L:L+m }=Projection(H i ) In the above formula, Y i { L:L+m } is the power consumption prediction result of the i-th power consumption area with a sequence length of m, and Projection is the mapping operation.

4. A mobile energy storage auxiliary scheduling method based on multi-regional electricity demand according to claim 2, characterized in that: The S3 adopts the DDQN algorithm model to solve the mobile energy storage vehicle optimization scheduling model, including: S31, converting the mobile energy storage vehicle optimization scheduling model into an energy storage vehicle quantity optimization model and a path planning model; S32. Construct the equivalent Markov state transition process of the path planning model: max Qπ(s,a;θ)=E[R1+γR2+…|S,A,π;θ] In the above formula, Q π (s, a; θ) is the cumulative expected reward under path planning, s, a, θ are state, action, and network parameters respectively, R is the reward under each corresponding action, S and A are state space and action space respectively, the action space represents the choice at each intersection, π is the formulated path planning strategy, γ is the attenuation coefficient, and E[x] is the expectation for x; S33. Use the DDQN algorithm to iteratively solve the problem, thereby obtaining the path planning strategy π, which is the track in the objective function.

5. A mobile energy storage auxiliary dispatching system based on multi-regional electricity demand, characterized in that: The system includes a power consumption prediction module, a scheduling model construction module, and a scheduling model solution module; The power consumption prediction module is used to predict the power consumption of multiple power consumption areas in an emergency scenario, and the multiple power consumption areas include commercial areas and residential areas; The scheduling model building module is used to build an optimization scheduling model for mobile energy storage vehicles with the goal of maximizing the overall benefit of the power supply process based on the power consumption of multiple power consumption areas; The scheduling model solving module is used to solve the mobile energy storage vehicle optimization scheduling model to obtain the mobile energy storage vehicle optimization scheduling plan.

6. A mobile energy storage auxiliary dispatching system based on multi-regional electricity demand according to claim 5, characterized in that: The mobile energy storage vehicle optimization scheduling model includes: In the above formula, Ear is the overall benefit of the power supply process, Y ind,i , are the electricity consumption of the i-th commercial area and the j-th residential area, z is the dispatching plan, N s is the number of mobile energy storage vehicles in the sth power consumption area, track is the driving route of the mobile energy storage vehicle, P 1,i , P 2,j are the electricity prices of the i-th commercial area and the j-th residential area, respectively. E and Δ are the electricity consumption and electricity loss per unit distance of the mobile energy storage vehicle, dis is the one-way distance of the mobile energy storage vehicle, and P ini is the cost of electricity for charging at the charging station, M1 and M2 are the number of commercial and residential areas respectively, and Y s is the electricity consumption of the sth electricity consumption area, Aviable k is the power supply of the kth mobile energy storage vehicle, K s is the number of mobile energy storage vehicles in each power consumption area.

7. A mobile energy storage auxiliary dispatching system based on multi-regional electricity demand according to claim 5 or 6, characterized in that: The power consumption prediction module uses the following steps to predict the power consumption of multiple power consumption areas: A. The following information interaction is performed on the historical power consumption sequence of each power consumption area through the self-attention mechanism: In the above formula, is the feature mapping result of the historical electricity consumption sequence of the i-th electricity consumption area. Embedding is an independent feature mapping operation. is the lth group of data in the historical electricity consumption sequence of the i-th electricity consumption area, and L is the length of the historical electricity consumption sequence; B. Use layer normalization to unify the feature channel distribution of different variables: In the above formula, LayerNorm is the layer normalization operation, H i After layer normalization Mean is the average value, and var is the variance; C. Perform fully connected feature encoding in the feedforward network and predict the power consumption of each power consumption area through projection layer mapping: Y i { L:L+m }=Projection(H i ) In the above formula, Y i { L:L+m } is the power consumption prediction result of the i-th power consumption area with a sequence length of m, and Projection is the mapping operation.

8. A mobile energy storage auxiliary dispatching system based on multi-regional electricity demand according to claim 5 or 6, characterized in that: The scheduling model solving module uses the DDQN algorithm model to solve the mobile energy storage vehicle optimization scheduling model. The solution process includes: a. Convert the mobile energy storage vehicle optimization scheduling model into an energy storage vehicle quantity optimization model and a path planning model; b. Construct the equivalent Markov state transition process of the path planning model: max Qπ(s,a;θ)=E[R1+γR2+…|S,A,π;θ] In the above formula, Q π (s, a; θ) is the cumulative expected reward under path planning, s, a, θ are the state, action, and network parameters respectively, R is the reward under each corresponding action, S and A are the state space and action space respectively, the action space represents the choice at each intersection, π is the formulated path planning strategy, γ is the attenuation coefficient, and E[·] is the expectation for x; c. Use the DDQN algorithm to iteratively solve the problem and obtain the path planning strategy π, which is the track in the objective function.

9. A mobile energy storage auxiliary dispatching device based on multi-regional electricity demand, characterized in that: The device comprises a memory and a processor; The memory is used to store computer program code and transmit the computer program code to the processor; The processor is configured to execute the method according to any one of claims 1 to 4 according to instructions in the computer program code.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images