Optimized scheduling method and device for optical storage, charging and conversion integrated electric bus station

Abstract

The invention discloses an optimal scheduling method and device for an optical storage, charging and conversion integrated electric bus station, and relates to the technical field of electric power market bidding, and the method comprises the steps: obtaining the operation related data of the optical storage, charging and conversion integrated electric bus station, and carrying out the preprocessing of the operation related data of the optical storage, charging and conversion integrated electric bus station, the method comprises the following steps: processing related operation data of an optical storage, charging and conversion integrated electric bus station to obtain processed related operation data of the optical storage, charging and conversion integrated electric bus station, pre-constructing a day-ahead and day-ahead multi-market multi-stage bidding optimization model, and carrying out multi-stage bidding optimization on the basis of the processed related operation data of the optical storage, charging and conversion integrated electric bus station by using the day-ahead and day-ahead multi-market multi-node optimization model. Constructing a day-ahead and intra-day electric energy auxiliary market settlement rule; and solving the day-ahead and intra-day multi-market multi-stage bidding optimization model based on a multi-stage dynamic optimization solving method, and outputting to obtain an optimal scheduling result of the optical storage, charging and conversion integrated electric bus station.

Description

Technical Field

[0001] The present invention relates to the technical field of power market bidding, and in particular to an optimization scheduling method and device for an integrated photovoltaic, storage, charging and swapping electric vehicle station. Background Art

[0002] With the rapid growth of electric vehicles and the increasing diversity of electricity market transactions, the impact of electric vehicle charging and swapping on distribution networks is becoming increasingly significant. On the one hand, the peak load caused by large-scale centralized charging of electric vehicles increases the burden on grid dispatch and peak management. On the other hand, the increasing penetration of renewable energy, coupled with the uncertainty and difficulty of accurate forecasting of output, has significantly increased the complexity of system operation and dispatch. Summary of the Invention

[0003] In order to solve the deficiencies mentioned in the above background technology, the purpose of the present invention is to provide an optimized scheduling method and device for an integrated photovoltaic, storage, charging and swapping electric vehicle station.

[0004] In a first aspect, the purpose of the present invention can be achieved by the following technical solution: an optimization scheduling method for an integrated photovoltaic storage charging and swapping electric vehicle station, the method comprising the following steps:

[0005] Obtaining operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station, preprocessing the operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station to obtain processed operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station, wherein the operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station includes photovoltaic power generation power, energy storage equipment parameters, charging and swapping demand characteristics of electric vehicle users, and electricity market prices;

[0006] Pre-building a multi-market, multi-stage bidding optimization model for the day-ahead, where the multi-market, multi-node optimization model is based on processed operational data related to the integrated photovoltaic, storage, charging, and swapping electric vehicle station and settlement rules for the day-ahead electric energy auxiliary market;

[0007] Based on the multi-stage dynamic optimization solution method, the multi-market and multi-stage bidding optimization model of the day-ahead is solved, and the optimized scheduling results of the integrated photovoltaic, storage, charging and swapping electric vehicle station are output.

[0008] In conjunction with the first aspect, in certain implementations of the first aspect, the method further includes: the process of pre-building the multi-market multi-stage bidding optimization model for the day before the deadline includes:

[0009] Assume that the system state is S t Represents, where t=1,2,...,T is the discrete time step, S t A combination of multiple state variables including photovoltaic power generation, electricity price, and electric vehicle aggregate power;

[0010] Introducing a sample screening mechanism based on price fluctuation sensitivity, for each sample point S t , calculate the local fluctuation degree of the price change rate before and after, and assign high clustering weights:

[0011]

[0012] For all sample points, according to the threshold θ, the points above the price sensitivity threshold are selected to enter the enhanced clustering sample set:

[0013] s enh ={S t |γ t >θ}

[0014] The density-based clustering algorithm DBSCAN is used to enh ∪s is discretized, BSCAN can automatically identify clusters by setting the neighborhood radius and the minimum number of samples, and the number of identified clusters is recorded as K, each cluster k represents a typical state; define the cluster center M k,t , then all sample states are assigned to the nearest cluster according to the minimum Euclidean distance principle:

[0015]

[0016] Where d(·) is the Euclidean distance metric function;

[0017] Introducing a weighted state transfer memory mechanism: when estimating the state from S t ∈Ω t,k When considering the probability of transferring to , we comprehensively consider the historical statistical frequency and the price fluctuation amplitude γ when the state occurs. t , the following weighted probability estimation formula is used:

[0018]

[0019] in, Represents the observation value of the i-th sample in the historical data at time t; based on the statistics of historical data, the state is obtained from S t Migrate to S t+1 Probability of:

[0020]

[0021] Among them, the conditional probability can be obtained through historical data statistics:

[0022]

[0023] P ij Represents the probability of state transition from i to j, satisfying:

[0024]

[0025] Furthermore, the state transition is expressed as:

[0026]

[0027] Among them, P(S t+1 |S t ) is obtained from statistical historical data.

[0028] In conjunction with the first aspect, in certain implementations of the first aspect, the method further includes: the step of pre-building a day-ahead intraday multi-market multi-stage bidding optimization model and solving it using the Markov chain-based MC-SDDiP algorithm includes:

[0029] In the forward sampling phase, under the current value function approximation, we sequentially roll the solution along the random scenario from the first stage to obtain the feasible solution and benefit of each stage. The problem of stages τ = 0 and τ = 1, 2, ..., T can be decomposed as follows:

[0030]

[0031] stΑ0x0≥b0

[0032] Among them, V0(ξ0) is the initial stage profit, represents the set of random scenarios that appear in the next stage, p ω,0 is its occurrence probability, V1(·) is the optimal value function in the subsequent stage;

[0033] For phases τ = 1, 2, ..., T:

[0034]

[0035] stΑ t x t +B t y t ≥b t (ξ t )

[0036] The multi-stage structure comes from: {V0,V1,...,V T} are nested together to form a dynamic programming. In the forward pass, MC-SDDiP solves the subproblems for each stage from t = 0 to t = T in turn, and uses the convex lower approximation of the current expectation function to perform dynamic programming backwash. For the stage 0 problem of the i-th iteration, the calculation method is as follows:

[0037]

[0038] stΑ0x0≥b0(ξ0)

[0039] For the t-th stage problem, the calculation method is as follows:

[0040]

[0041] stΑ τ x τ +B τ y τ +C τ z τ ≥b τ (ξ τ )

[0042]

[0043] x τ ∈{0,1}

[0044] z τ ∈[0,1]

[0045] in, is the expectation function of the truth The current convex lower approximation of the previous stage; the constraint is to The value of is used as the local variable z in this stage τ , the remaining continuous variables need to be discretized through binary decomposition; after the forward push is completed, the statistical upper bound is calculated by the sample mean and variance, and the calculation method is as follows:

[0046]

[0047] in, N in this iteration m The average cost in each scenario, is the variance, and α / 2 is the z score that guarantees a given confidence interval;

[0048] By performing Lagrangian relaxation on the problem, the dual solution of the subproblem is used to generate the Lagrangian tangent surface, which is a convex lower approximation of the cost function of the next stage. The Lagrangian relaxation of the subproblem is as follows:

[0049]

[0050] in, is a Lagrange multiplier, and:

[0051]

[0052] The repeated variable constraint is dualized, and any solution violating the constraint is penalized by the Lagrange multiplier. The penalty bundle method is used to solve the Lagrange multiplier formula, and the Lagrange section is Q τ+1The convex lower approximation of (·) yields the solution to the dual problem; the lower bound can be obtained by solving the relaxation problem of the root node, i.e.

[0053] In conjunction with the first aspect, in certain implementations of the first aspect, the method further includes: the day-ahead and intraday bidding models of the pre-built day-ahead and intraday multi-market multi-stage bidding optimization model include the following objective functions:

[0054]

[0055] in and They represent the electricity sold and purchased in the tth period respectively, and are the bid amounts for uplink and downlink backup of auxiliary services, and They correspond to the market prices of day-ahead electricity and ancillary services respectively, and T is the total number of day-ahead time periods; the function must meet the constraints of photovoltaic output power, energy storage power threshold, and the maximum adjustable range of electric vehicle polymer batteries.

[0056] In conjunction with the first aspect, in certain implementations of the first aspect, the method further includes: the rolling optimization objective based on the Markov chain scenario in the intraday stage can be expressed as:

[0057]

[0058] Where k and i represent the scene stage and the specific scene number respectively, N k is the number of scenes in the kth stage, P k,i is the probability of the scene occurring, and is the electricity purchase and sale decision under scenario i and time period τ, and denote the aggregated charging and swapping power of the energy storage system and electric vehicles, respectively. F(·) represents the sum of the profit function considering the default penalty and the reserve call income. The bid is gradually revised within the rolling period.

[0059] In conjunction with the first aspect, in certain implementations of the first aspect, the method further includes: the day-ahead bidding model includes the following constraints:

[0060]

[0061] in, is the upper limit of the energy storage system discharge power, the upper limit of the photovoltaic unit output power and the upper limit of the electric vehicle discharge power, P es , P pv , P evIt is the upper limit of charging power of energy storage system, the lower limit of output power of photovoltaic unit and the upper limit of charging power of electric vehicle.

[0062] In conjunction with the first aspect, in certain implementations of the first aspect, the method further includes: the intraday bidding model includes the following physical constraints:

[0063]

[0064] in, is the amount of energy stored, is the energy storage charging and discharging power, η es is the energy storage charge and discharge efficiency, C es , are the upper and lower limits of energy storage capacity, is the upper limit of energy storage charging and discharging power, A Boolean variable that limits the energy storage from being charged and discharged simultaneously. is the amount of electricity contained in the battery of the electric vehicle after polymerization, is the charging and discharging power of the electric vehicle after aggregation, η ev is the charging and discharging efficiency of the electric vehicle after polymerization, The battery power change of the aggregated electric vehicle due to charging, discharging and battery replacement, are the upper and lower limits of the battery capacity of the electric vehicle after aggregation, The upper and lower limits of battery power changes in electric vehicles after polymerization due to charging and discharging, A Boolean variable that restricts the simultaneous charging and discharging of the electric vehicle battery after polymerization. is the output power of the photovoltaic unit, The upper and lower limits of the photovoltaic unit output power;

[0065] The intraday bidding model includes the following market constraints:

[0066]

[0067] in, It is the power used by the upstream and downstream auxiliary service markets during the day. are the probabilities of the intraday upstream and downstream auxiliary service market call volume compared to the intraday bidding volume, are the bid amounts for electricity sales in the intraday and day-ahead energy markets, is the physical quantity used for bidding in the intraday electrical energy market, The amount of electricity purchase bids in the intraday electricity energy market.

[0068] In a second aspect, in order to achieve the above-mentioned purpose, the present invention discloses an optimization scheduling device for an integrated photovoltaic, storage, charging and swapping electric vehicle station, comprising:

[0069] a data receiving module for acquiring operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station, preprocessing the operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station, and obtaining processed operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station, wherein the operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station includes photovoltaic power generation power, energy storage equipment parameters, charging and swapping demand characteristics of electric vehicle users, and electricity market prices;

[0070] A model pre-building module is used to pre-build a multi-market, multi-stage bidding optimization model for the day before the deadline. The multi-market, multi-node optimization model for the day before the deadline is built based on the processed operation data of the integrated photovoltaic, storage, charging and swapping electric vehicle station and the day before the deadline electric energy auxiliary market settlement rules.

[0071] The model solving module is used to solve the multi-market and multi-stage bidding optimization model within the day before based on the multi-stage dynamic optimization solution method, and output the optimized scheduling results of the integrated photovoltaic storage and charging and swapping electric vehicle station.

[0072] In another aspect of the present invention, in order to achieve the above-mentioned purpose, a terminal device is disclosed, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. The memory stores a computer program capable of running on the processor, and when the processor loads and executes the computer program, the above-mentioned optimization scheduling method for the integrated photovoltaic storage and charging electric vehicle station is adopted.

[0073] In another aspect of the present invention, in order to achieve the above-mentioned purpose, a computer-readable storage medium is disclosed, in which a computer program is stored. When the computer program is loaded and executed by a processor, the above-mentioned optimization scheduling method for an integrated photovoltaic, storage, charging and swapping electric vehicle station is adopted.

[0074] Beneficial effects of the present invention:

[0075] The present invention can achieve coordinated scheduling of electric and hydrogen energy equipment, effectively improve the on-site absorption capacity of renewable energy, alleviate peak and valley load fluctuations in the power system, and enhance the operational flexibility and stability of the energy system. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, those skilled in the art can derive other drawings based on these drawings without inventive effort.

[0077] Figure 1 It is a schematic flow chart of the method of the present invention;

[0078] Figure 2 This is a schematic diagram of the charging station scheduling of the present invention;

[0079] Figure 3 Schematic diagram of the bidding results of this embodiment;

[0080] Figure 4 It is a schematic diagram of the structure of the device of the present invention. DETAILED DESCRIPTION

[0081] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0082] Example 1:

[0083] like Figure 1 As shown, the optimization scheduling method for an integrated photovoltaic storage charging and swapping electric vehicle station includes the following steps:

[0084] S101: Acquire operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station, pre-process the operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station, and obtain processed operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station, wherein the operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station includes photovoltaic power generation power, energy storage equipment parameters, charging and swapping demand characteristics of electric vehicle users, and electricity market prices;

[0085] S102: Pre-constructing a multi-market, multi-stage bidding optimization model for the day before the current day, wherein the multi-market, multi-node optimization model for the day before the current day is constructed based on the processed operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station and the day before the current day electric energy auxiliary market settlement rules;

[0086] The process of constructing the pre-built multi-market multi-stage bidding optimization model for the day before includes:

[0087] Assume that the system state is represented by S t Represents, where t=1,2,...,T is the discrete time step, S t It includes a combination of multiple state variables such as photovoltaic power generation, electricity price, and electric vehicle aggregate power.

[0088] In order to improve the ability to respond to changes in key price signals of the system, a sample screening mechanism based on price fluctuation sensitivity is introduced before state clustering. Specifically, for each sample point S t, calculate the local fluctuation degree of the price change rate before and after, to identify abnormal high-frequency change points and assign higher clustering weights:

[0089]

[0090] For all sample points, according to the threshold θ, the points above the price sensitivity threshold are selected to enter the enhanced clustering sample set:

[0091] s enh ={S t |γ t >θ}

[0092] This enhanced set adopts a weighted density strategy in subsequent clustering to improve the clustering ability to identify key state boundaries, thereby enhancing the sensitivity and generalization ability of the model at key market nodes.

[0093] Then, the density-based clustering algorithm DBSCAN is used to cluster s enh ∪s is discretized. BSCAN can automatically identify clusters by setting the neighborhood radius and the minimum number of samples, and the number of identified clusters is recorded as K. Each cluster k represents a typical state. In order to facilitate use in subsequent Markov chain modeling, the cluster center M can be defined k,t , then all sample states are assigned to the nearest cluster according to the minimum Euclidean distance principle:

[0094]

[0095] Where d(·) is the Euclidean distance metric function.

[0096] When constructing the Markov chain, in order to improve the ability to express the long-term state evolution trend, a weighted state transfer memory mechanism is introduced. t ∈Ω t,k When considering the probability of transferring to , we comprehensively consider the historical statistical frequency and the price fluctuation amplitude γ when the state occurs. t , the following weighted probability estimation formula is used:

[0097]

[0098] in, Represents the observation value of the i-th sample in the historical data at time t. Based on the statistics of historical data, the state is obtained from S t Migrate to S t+1 Probability of:

[0099]

[0100] Among them, the conditional probability can be obtained through historical data statistics:

[0101]

[0102] P ij Represents the probability of state transition from i to j, satisfying:

[0103]

[0104] Furthermore, the state transition can be expressed as:

[0105]

[0106] Among them, P(S t+1 |S t ) is obtained from statistical historical data to describe the evolution trend of the state.

[0107] Through this method, a finite state space is constructed based on historical data, and the transition relationships between states are characterized using Markov chains. This allows for an accurate description of the dynamic evolution of variables such as photovoltaic power generation, electricity prices, and the aggregated power of electric vehicles. This state transition matrix provides a theoretical basis for subsequent optimization decisions and can be further used to study optimal scheduling strategies based on Markov decision processes. This state transition matrix serves as the basic input for a multi-stage stochastic dual dynamic integer programming (MC-SDDiP) algorithm, supporting day-ahead and intraday joint market bidding for integrated photovoltaic, storage, charging, and swapping electric vehicle stations.

[0108] The steps of pre-building a multi-market multi-stage bidding optimization model for the day before and solving it using the Markov chain-based MC-SDDiP algorithm include:

[0109] First, in the forward sampling phase, under the current value function approximation, we sequentially solve the random scenarios from stage 1, obtaining feasible solutions and benefits for each stage, and evaluating the lower bound. The problem for stages τ = 0 and τ = 1, 2, ..., T can be decomposed as follows:

[0110]

[0111] stΑ0x0≥b0

[0112] Among them, V0(ξ0) is the initial stage profit, represents the set of random scenarios that may appear in the next stage, p ω,0 is its occurrence probability, and V1(·) is the optimal value function in the subsequent stage.

[0113] For phases τ = 1, 2, ..., T:

[0114]

[0115] stΑ t xt +B t y t ≥b t (ξ t )

[0116] The multi-stage structure of the whole problem comes from: {V0,V1,...,V T} are nested together to form a dynamic programming. In the forward pass, MC-SDDiP solves the subproblems for each stage from t = 0 to t = T in turn, and uses the convex lower approximation of the current expectation function to perform dynamic programming backwash. For the stage 0 problem in the i-th iteration, the calculation method is as follows:

[0117]

[0118] stΑ0x0≥b0(ξ0)

[0119] For the t-th stage problem, the calculation method is as follows:

[0120]

[0121] stΑ τ x τ +B τ y τ +C τ z τ ≥b τ (ξ τ )

[0122]

[0123] x τ ∈{0,1}

[0124] z τ ∈[0,1]

[0125] in, is the expectation function of the truth The current convex lower approximation of . The constraint is to change the state variables of the previous stage The value of is used as the local variable z in this stage τ , thereby establishing connections between different stages. Since some decision variables (such as the energy storage charge and discharge status) are binary variables, the Lagrangian duality has no duality gap. The remaining continuous variables need to be discretized through binary decomposition. After the forward push is completed, the statistical upper bound is calculated using the sample mean and variance. The calculation method is as follows:

[0126]

[0127] in, N in this iteration mThe average cost in each scenario, is the variance, and α / 2 is the z score that guarantees a given confidence interval.

[0128] Secondly, in the backward process, by performing Lagrangian relaxation on the problem, the dual solution of the subproblem generates Lagrangian cut surfaces, which are convex lower approximations of the cost function in the next stage. These cut surfaces are accumulated, gradually improving the lower bound, and eventually converge to the upper bound. When mixed integer decisions are involved, the Lagrangian relaxation of the subproblem is as follows:

[0129]

[0130] in, is a Lagrange multiplier, and:

[0131]

[0132] The repeated variable constraint is dualized and any solution that violates the constraint is penalized by the Lagrange multiplier. The closer the Lagrange multiplier is to the optimal value, the tighter the approximation is and the more accurate the generated section is. The penaltybundle method is used to solve the Lagrange multiplier formula, and the Lagrange section is Q τ+1 The convex lower approximation of (·) is the solution to the dual problem. The lower bound can be obtained by solving the relaxation problem of the root node, because the relaxation problem only contains part of the constraints of the original problem, that is,

[0133] S103: Solve the multi-market and multi-stage bidding optimization model for the day before based on a multi-stage dynamic optimization solution method, and output the optimized scheduling result of the integrated photovoltaic, storage, charging and swapping electric vehicle station.

[0134] The day-ahead and intraday bidding models of the pre-built day-ahead and intraday multi-market multi-stage bidding optimization model include the following objective functions:

[0135]

[0136] in and They represent the electricity sold and purchased in the tth period respectively, and are the bid amounts for uplink and downlink backup of auxiliary services, and The corresponding market prices for day-ahead electricity and ancillary services are T, respectively. This function must satisfy a series of constraints, including photovoltaic output power, energy storage power threshold, and the maximum adjustable range of electric vehicle polymer batteries.

[0137] The rolling optimization objective based on the Markov chain scenario in the intraday stage can be expressed as:

[0138]

[0139] Where k and i represent the scene stage and the specific scene number respectively, N k is the number of scenes in the kth stage, P k,i is the probability of the scene occurring, and is the electricity purchase and sale decision under scenario i and time period τ, and denote the aggregated charging and swapping power of the energy storage system and electric vehicles, respectively. F(·) represents the sum of the profit function considering the default penalty and the reserve call income. The bid is gradually revised within the rolling cycle to fit the actual situation.

[0140] The day-ahead bidding model includes the following constraints:

[0141]

[0142] in, is the upper limit of the energy storage system discharge power, the upper limit of the photovoltaic unit output power and the upper limit of the electric vehicle discharge power, P es , P pv , P ev It is the upper limit of charging power of energy storage system, the lower limit of output power of photovoltaic unit and the upper limit of charging power of electric vehicle.

[0143] The intraday bidding model includes the following physical constraints:

[0144]

[0145] in, is the amount of energy stored, is the energy storage charging and discharging power, η es is the energy storage charge and discharge efficiency, C es , are the upper and lower limits of energy storage capacity, is the upper limit of energy storage charging and discharging power, A Boolean variable that limits the energy storage from being charged and discharged simultaneously. is the amount of electricity contained in the battery of the electric vehicle after polymerization, is the charging and discharging power of the electric vehicle after aggregation, η ev is the charging and discharging efficiency of the electric vehicle after polymerization, The battery power change of the aggregated electric vehicle due to charging, discharging and battery replacement, are the upper and lower limits of the battery capacity of the electric vehicle after aggregation, The upper and lower limits of battery power changes in electric vehicles after polymerization due to charging and discharging, A Boolean variable that restricts the simultaneous charging and discharging of the electric vehicle battery after polymerization. is the output power of the photovoltaic unit, The upper and lower limits of the PV unit output power.

[0146] The intraday bidding model includes the following market constraints:

[0147]

[0148] in, It is the power used by the upstream and downstream auxiliary service markets during the day. are the probabilities of the intraday upstream and downstream auxiliary service market call volume compared to the intraday bidding volume, are the bid amounts for electricity sales in the intraday and day-ahead energy markets, is the physical quantity used for bidding in the intraday electrical energy market, The amount of electricity purchase bids in the intraday electricity energy market.

[0149] To verify the effectiveness of the proposed method, a case study was conducted in the simulation software. The bidding and station scheduling results are as follows:

[0150] (1) When integrated power plants jointly bid in the electricity energy and ancillary service market: A market-responsive energy dispatch strategy is proposed based on the operating characteristics of photovoltaic energy storage charging stations at different times of the day. This strategy dynamically coordinates the output and electricity purchase and sales behavior of the energy storage system (ESS), battery swap system, and photovoltaic system based on photovoltaic output, electric vehicle (EV) charging demand, electricity price changes, and ancillary service calls to achieve optimization of system economy and energy balance.

[0151] During periods 0–20, the PV system outputs no power. During periods 12–15, electricity prices are at their lowest in the morning, and the station prioritizes purchasing electricity from the day-ahead market to meet EV charging loads. During periods with higher electricity prices, the station discharges electricity through the ESS to support charging demand. Furthermore, if downward ancillary service calls are made during the mid-day period, the station can utilize the downward adjustment instructions submitted in the day-ahead ancillary service market to both meet internal load demand and generate ancillary service revenue, effectively reducing the system's overall electricity purchase cost. During periods 21–32, the PV system begins generating power, and electricity prices enter a secondary peak phase. During this period, the PV system and ESS operate in tandem, with PV output prioritized for EV charging. The ESS flexibly adjusts discharge power based on electricity price signals to ensure that the battery swap system and EV loads avoid using high-priced electricity, thereby reducing electricity purchase costs and increasing the local PV consumption rate. During periods 33–48, EV charging demand continues to rise, PV output gradually increases and approaches its peak, while electricity prices gradually decline to a trough. To cope with the significant increase in charging load, the ESS and battery swapping system fully absorb low-priced electricity and surplus PV output for charging, reserving energy for subsequent periods of high electricity prices and low PV output. If low-priced electricity purchases are still insufficient to support charging demand, appropriate additional purchases can be made from the real-time electricity market to maintain energy balance. During periods 49–70, EV charging demand remains high, PV output gradually decreases to zero, and electricity prices rise to peak levels. During this period, the ESS and battery swapping system jointly release previously stored energy, minimizing reliance on high-priced electricity. In particular, during periods 60–63, system energy storage remains sufficient, and the PV system also has some output. EV load can be met through PV output and a small amount of day-ahead electricity purchases, while the battery swapping system continues to perform partial charging operations. Given the high probability of ancillary service upscaling during this period, stations can submit bids for these services. If successful, they can earn ancillary service revenue based on opportunity costs. If internal power is still insufficient, a small amount of electricity can be purchased from the real-time market. During periods 70–76, the PV system is essentially idle, but EV charging still requires some charging, despite high electricity prices. During this period, the station can appropriately delay EV charging, waiting for electricity prices to fall before proceeding. Simultaneously, the ESS and battery swap systems can sell stored energy to the energy market at a high price to generate economic benefits. Due to the frequent invocation of downgraded ancillary services during this period, the station can still participate in bidding to meet internal demand and increase revenue. Given the high risk of default, the station's day-ahead power sales strategy tends to be conservative, but dynamic sales based on remaining power can be used in real-time to further increase revenue. Finally, during periods 76–96, the PV system ceases power generation, and EV charging demand gradually decreases. During this period, downgraded ancillary services can be used to simultaneously meet the power needs of EVs, the ESS, and the battery swap system. If internal power is still insufficient, appropriate amounts can be purchased from the day-ahead or real-time electricity markets. Furthermore, during periods 68–92, the ESS and battery swap systems maintain high energy storage levels, and EV charging is largely complete.Since electricity prices are relatively high during this period, the station can raise its bid in the ancillary service market in the near future to obtain ancillary service revenue based on opportunity costs, thereby further improving the overall economic efficiency of the system.

[0152] (2) When the integrated power station bids in a single electricity market: The on-site dispatch and electricity market bidding strategies of an integrated charging station that only participates in the electricity market are analyzed, and compared with the joint bidding strategy that also participates in the ancillary service market. In terms of on-site dispatch, the overall operating characteristics are similar to those of the typical scenario 1. The main difference is reflected in the 8th to 11th period: Since the station does not participate in the ancillary service market, it cannot respond to the downward adjustment of the ancillary service call during the day's operation, and therefore cannot use the service to charge electric vehicles. As a result, the charging demand in this period is met by the energy storage system discharge, and a small amount of electricity is purchased during the low electricity price period to maintain the on-site energy balance. In terms of electricity market bidding, since the downward adjustment of the ancillary service is no longer provided in the 72nd to 91st period, the day-ahead and intraday electricity sales in the 72nd to 75th period have decreased. In order to ensure the charging demand within the station, the day-ahead electricity purchases in the 76th to 91st period have increased accordingly. The above dispatch and bidding results are as follows Figure 2 、 Figure 3 shown.

[0153] In this embodiment, it is further explained how to utilize the dispatchable characteristics of the replaced batteries together with the energy storage device for intraday optimization when a centralized battery replacement area is also provided in the integrated photovoltaic storage charging and replacement electric vehicle station. Consistent with the first embodiment, the photovoltaic power generation level and the electric vehicle entry and exit conditions are also modeled in time series by a Markov chain, and intraday scheduling is performed based on a rolling cycle. The difference is that if the vehicle chooses the battery replacement mode after arriving at the station, the removed batteries are stored centrally in the station and can be regarded as a group of potential energy storage units. When the price is high or there is a shortage of electricity, the fully charged batteries are resupplied to the vehicle or the electricity is sold through auxiliary service clearing. In this way, the flexibility of power dispatch can be significantly improved without extending the waiting time of the vehicle. When constructing the intraday optimization model, this embodiment only needs to add the capacity limit and charge / discharge efficiency of the battery replacement battery to the aggregated energy constraint to ensure that the battery replacement process is physically feasible.

[0154] Example 2: In order to achieve the above purpose, Figure 4 As shown, based on the first embodiment, the present invention discloses an optimization scheduling device for an integrated photovoltaic storage charging and swapping electric vehicle station, which is characterized by including:

[0155] The data receiving module 11 is used to obtain operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station, pre-process the operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station, and obtain the processed operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station, wherein the operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station includes photovoltaic power generation power, energy storage equipment parameters, charging and swapping demand characteristics of electric vehicle users, and electricity market prices;

[0156] A model pre-building module 12 is used to pre-build a multi-market, multi-stage bidding optimization model for the day before the current day. The multi-market, multi-node optimization model for the day before the current day is built based on the processed operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station and the day before the current day electric energy auxiliary market settlement rules.

[0157] The model solving module 13 is used to solve the multi-market and multi-stage bidding optimization model within the day before based on the multi-stage dynamic optimization solution method, and output the optimized scheduling result of the integrated photovoltaic storage charging and swapping electric vehicle station.

[0158] Based on the same inventive concept, the present invention also provides a computer device, which includes: one or more processors and a memory for storing one or more computer programs; the program includes program instructions, and the processor is used to execute the program instructions stored in the memory. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is used to implement one or more instructions, specifically for loading and executing one or more instructions in a computer storage medium to implement the above method.

[0159] It should be further explained that, based on the same inventive concept, the present invention also provides a computer storage medium having a computer program stored thereon, which executes the above method when executed by a processor. The storage medium can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electrical, magnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or component.

[0160] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present disclosure. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0161] The above shows and describes the basic principles, main features and advantages of the present disclosure. Those skilled in the art should understand that the present disclosure is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present disclosure. Various changes and improvements may be made to the present disclosure without departing from the spirit and scope of the present disclosure, and such changes and improvements shall fall within the scope of the present disclosure.

Claims

1. An optimization scheduling method for an integrated photovoltaic, storage, charging and swapping electric vehicle station, characterized in that: The method comprises the following steps: Obtaining operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station, preprocessing the operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station to obtain processed operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station, wherein the operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station includes photovoltaic power generation power, energy storage equipment parameters, charging and swapping demand characteristics of electric vehicle users, and electricity market prices; Pre-building a multi-market, multi-stage bidding optimization model for the day-ahead, where the multi-market, multi-node optimization model is based on processed operational data related to the integrated photovoltaic, storage, charging, and swapping electric vehicle station and settlement rules for the day-ahead electric energy auxiliary market; Based on the multi-stage dynamic optimization solution method, the multi-market and multi-stage bidding optimization model of the day-ahead is solved, and the optimized scheduling results of the integrated photovoltaic, storage, charging and swapping electric vehicle station are output.

2. The optimization scheduling method for an integrated photovoltaic, storage, charging and swapping electric vehicle station according to claim 1 is characterized in that: The process of constructing the pre-built multi-market multi-stage bidding optimization model for the day before includes: Assume that the system state is S t Represents, where t=1,2,...,T is the discrete time step, S t A combination of multiple state variables including photovoltaic power generation, electricity price, and electric vehicle aggregate power; Introducing a sample screening mechanism based on price fluctuation sensitivity, for each sample point S t , calculate the local fluctuation degree of the price change rate before and after, and assign high clustering weights: For all sample points, according to the threshold θ, the points above the price sensitivity threshold are selected to enter the enhanced clustering sample set: s enh ={S t |c t >θ} The density-based clustering algorithm DBSCAN is used to enh ∪s is discretized, BSCAN can automatically identify clusters by setting the neighborhood radius and the minimum number of samples, and the number of identified clusters is recorded as K, each cluster k represents a typical state; define the cluster center M k,t , then all sample states are assigned to the nearest cluster according to the minimum Euclidean distance principle: Where d(·) is the Euclidean distance metric function; Introducing a weighted state transfer memory mechanism: when estimating the state from S t ∈Ω t,k When considering the probability of transferring to , we comprehensively consider the historical statistical frequency and the price fluctuation amplitude γ when the state occurs. t , the following weighted probability estimation formula is used: in, Represents the observation value of the i-th sample in the historical data at time t; based on the statistics of historical data, the state is obtained from S t Migrate to S t+1 Probability of: Among them, the conditional probability can be obtained through historical data statistics: P ij Represents the probability of state transition from i to j, satisfying: Furthermore, the state transition is expressed as: Among them, P(S t+1 |S t ) is obtained from statistical historical data.

3. The optimization scheduling method for an integrated photovoltaic, storage, charging and swapping electric vehicle station according to claim 2 is characterized in that: The steps of pre-building a multi-market multi-stage bidding optimization model for the day before and solving it using the Markov chain-based MC-SDDiP algorithm include: In the forward sampling phase, under the current value function approximation, we sequentially roll the solution along the random scenario from the first stage to obtain the feasible solution and benefit of each stage. The problem of stages τ = 0 and τ = 1, 2, ..., T can be decomposed as follows: Among them, V0(ξ0) is the initial stage profit, represents the set of random scenarios that appear in the next stage, p ω,0 is its occurrence probability, V1(·) is the optimal value function in the subsequent stage; For phases τ = 1, 2, ..., T: The multi-stage structure comes from: {V0,V1,...,V T } are nested together to form a dynamic programming. In the forward pass, MC-SDDiP solves the subproblems for each stage from t = 0 to t = T in turn, and uses the convex lower approximation of the current expectation function to perform dynamic programming backwash. For the stage 0 problem of the i-th iteration, the calculation method is as follows: For the t-th stage problem, the calculation method is as follows: in, is the expectation function of the truth The current convex lower approximation of the previous stage; the constraint is to The value of is used as the local variable z in this stage τ , the remaining continuous variables need to be discretized through binary decomposition; after the forward push is completed, the statistical upper bound is calculated by the sample mean and variance, and the calculation method is as follows: in, N in this iteration m The average cost in each scenario, is the variance, and α / 2 is the z score that guarantees a given confidence interval; By performing Lagrangian relaxation on the problem, the dual solution of the subproblem is used to generate the Lagrangian tangent surface, which is a convex lower approximation of the cost function of the next stage. The Lagrangian relaxation of the subproblem is as follows: in, is a Lagrange multiplier, and: The repeated variable constraint is dualized, and any solution violating the constraint is penalized by the Lagrange multiplier. The penalty bundle method is used to solve the Lagrange multiplier formula, and the Lagrange section is Q τ+1 The convex lower approximation of (·) yields the solution to the dual problem; the lower bound can be obtained by solving the relaxation problem of the root node, i.e.

4. The optimization scheduling method for an integrated photovoltaic, storage, charging and swapping electric vehicle station according to claim 3 is characterized in that: The day-ahead and intraday bidding models of the pre-built day-ahead and intraday multi-market multi-stage bidding optimization model include the following objective functions: in and They represent the electricity sold and purchased in the tth period respectively, and are the bid amounts for uplink and downlink backup of auxiliary services, and They correspond to the market prices of day-ahead electricity and ancillary services respectively, and T is the total number of day-ahead time periods; the function must meet the constraints of photovoltaic output power, energy storage power threshold, and the maximum adjustable range of electric vehicle polymer batteries.

5. The optimization scheduling method for an integrated photovoltaic, storage, charging and swapping electric vehicle station according to claim 4 is characterized in that: The rolling optimization objective based on the Markov chain scenario in the intraday stage can be expressed as: Where k and i represent the scene stage and the specific scene number respectively, N k is the number of scenes in the kth stage, P k,i is the probability of the scene occurring, and is the electricity purchase and sale decision under scenario i and time period τ, and denote the aggregated charging and swapping power of the energy storage system and electric vehicles, respectively. F(·) represents the sum of the profit function considering the default penalty and the reserve call income. The bid is gradually revised within the rolling period.

6. The optimization scheduling method for an integrated photovoltaic, storage, charging and swapping electric vehicle station according to claim 5 is characterized in that: The day-ahead bidding model includes the following constraints: in, is the upper limit of the energy storage system discharge power, the upper limit of the photovoltaic unit output power and the upper limit of the electric vehicle discharge power, P es , P pv , P ev It is the upper limit of charging power of energy storage system, the lower limit of output power of photovoltaic unit and the upper limit of charging power of electric vehicle.

7. The optimization scheduling method for an integrated photovoltaic, storage, charging and swapping electric vehicle station according to claim 6 is characterized in that: The intraday bidding model includes the following physical constraints: in, is the amount of energy stored, is the energy storage charging and discharging power, η es is the energy storage charge and discharge efficiency, C es are the upper and lower limits of energy storage capacity, is the upper limit of energy storage charging and discharging power, A Boolean variable that limits the energy storage from being charged and discharged simultaneously. is the amount of electricity contained in the battery of the electric vehicle after polymerization, is the charging and discharging power of the electric vehicle after aggregation, η ev is the charging and discharging efficiency of the electric vehicle after polymerization, The battery power change of the aggregated electric vehicle due to charging, discharging and battery replacement, are the upper and lower limits of the battery capacity of the electric vehicle after aggregation, The upper and lower limits of battery power changes in electric vehicles after polymerization due to charging and discharging, A Boolean variable that restricts the simultaneous charging and discharging of the electric vehicle battery after polymerization. is the output power of the photovoltaic unit, The upper and lower limits of the photovoltaic unit output power; The intraday bidding model includes the following market constraints: in, It is the power used by the upstream and downstream auxiliary service markets during the day. are the probabilities of the intraday upstream and downstream auxiliary service market call volume compared to the intraday bidding volume, are the bid amounts for electricity sales in the intraday and day-ahead energy markets, is the physical quantity used for bidding in the intraday electrical energy market, The amount of electricity purchase bids in the intraday electricity energy market.

8. An optimized dispatching device for an integrated photovoltaic, storage, charging and swapping electric vehicle station, characterized in that: include: a data receiving module for acquiring operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station, preprocessing the operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station, and obtaining processed operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station, wherein the operation-related data of the integrated photovoltaic, storage, charging and swapping electric vehicle station includes photovoltaic power generation power, energy storage equipment parameters, charging and swapping demand characteristics of electric vehicle users, and electricity market prices; A model pre-building module is used to pre-build a multi-market, multi-stage bidding optimization model for the day before the deadline. The multi-market, multi-node optimization model for the day before the deadline is built based on the processed operation data of the integrated photovoltaic, storage, charging and swapping electric vehicle station and the day before the deadline electric energy auxiliary market settlement rules. The model solving module is used to solve the multi-market and multi-stage bidding optimization model within the day before based on the multi-stage dynamic optimization solution method, and output the optimized scheduling results of the integrated photovoltaic storage and charging and swapping electric vehicle station.

9. A terminal device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that: The memory stores a computer program that can be run on the processor. When the processor loads and executes the computer program, the optimization scheduling method for an integrated photovoltaic, storage, charging and swapping electric vehicle station according to any one of claims 1 to 7 is adopted.

10. A computer-readable storage medium storing a computer program, wherein: When the computer program is loaded and executed by the processor, the optimization scheduling method for an integrated photovoltaic, storage, charging and swapping electric vehicle station according to any one of claims 1 to 7 is adopted.

Images