Method for Phase-Separated Guidance of Electric Vehicles to Improve New Energy Consumption Based on Evolutionary Game

Abstract

The present invention belongs to the technical field of electric vehicle charging, and particularly relates to a method for guiding the split-phase charging of electric vehicles, specifically a method for guiding the split-phase of electric vehicles based on evolutionary game to improve the consumption of new energy. The method for guiding the split-phase charging of electric vehicles based on evolutionary game to improve the consumption of new energy in the present invention is specifically a three-layer model. The upper-layer optimization model establishes a split-phase and time-sharing charging electricity price model for the charging network operator with the goal of maximizing the net income of the charging network operator. The middle-layer optimization model establishes an evolutionary game scheduling model for electric vehicles with the goal of minimizing the total charging cost during the scheduling period of electric vehicles. The lower-layer optimization model establishes a nodal marginal price clearing model considering the consumption of new energy with the goal of minimizing the sum of the operation cost of the distribution network and the penalty for wind and light curtailment. The present invention proposes a method for guiding the split-phase of electric vehicles based on evolutionary game to improve the consumption of new energy, so as to maximize the income of the charging network operator, minimize the charging cost of electric vehicles and the total cost of the distribution network, and achieve a win-win situation for all three parties.

Description

Technical Field

[0001] The present invention relates to a method for guiding the split-phase charging of electric vehicles, specifically a method for guiding the split-phase charging of electric vehicles based on evolutionary game to improve the consumption of new energy. Background Art

[0002] In the process of the development of smart cities, environmental problems such as carbon emission pollution and oil consumption have become the main factors restricting sustainable development. In recent years, as a clean energy transportation vehicle, the penetration rate of electric vehicles in cities has been continuously increasing, and gradually replacing fossil fuel vehicles. And the charging price will affect the charging behavior of electric vehicles. Setting a reasonable charging price and making the charging behavior of electric vehicles associated with the output of new energy can improve the consumption capacity of the distribution network for new energy.

[0003] Currently, the research usually uses the time-of-use electricity price of the nodes in the distribution network as the charging electricity price of electric vehicles, ignoring the revenue problem of the charging network operator. In addition, the single-phase access of new energy makes a large difference between the phases of the distribution network, and currently many electric vehicles are also single-phase access. By reasonably guiding the charging of electric vehicles, the difference problem caused by the single-phase access of new energy can be solved.

[0004] Therefore, in order to guide electric vehicles to better consume new energy, when determining the charging price, it is necessary to consider the new energy generation and the operation conditions of each phase of the distribution network, determine the charging price of each phase of each charging station, and then affect the charging cost of electric vehicle users and the operation cost of the distribution network. The relationship between electric vehicle users and the charging network operator is a master-slave game relationship, and the process of guiding electric vehicle users is a gradually evolving process. Therefore, an evolutionary game is introduced to describe the change of the charging strategy of electric vehicle users during the change of electricity price and the change of new energy output, so as to determine the method for guiding the split-phase charging of electric vehicles to improve the consumption of new energy. Summary of the Invention

[0005] Aiming at the problems of large differences between phases caused by the single-phase access of new energy to the distribution network and ignoring the revenue of the charging network operator, the present invention provides a method for guiding the split-phase charging of electric vehicles based on evolutionary game to improve the consumption of new energy.

[0006] The technical solution adopted by the present invention is: an electric vehicle phase-separated guidance method for improving new energy consumption based on evolutionary game, which method includes a three-layer optimization model. The upper-layer optimization model is a phase-separated and time-sharing charging price model for the charging network operator. The middle-layer optimization model is an electric vehicle phase-separated charging scheduling model based on evolutionary game. The lower-layer optimization model is a nodal marginal price clearing model considering new energy consumption. In the upper-layer optimization model, the charging network operator formulates different phase-separated and time-sharing charging prices for the electric vehicle charging stations under its management and transmits them to the middle-layer optimization model. In the middle-layer optimization model, the electric vehicle users select the optimal charging strategy with the minimum payment function through evolutionary game in combination with the phase-separated and time-sharing charging prices transmitted by the upper-layer optimization model, and at the same time obtain the phase-separated and time-sharing charging loads and transmit them to the lower-layer optimization model and the upper-layer optimization model respectively. According to the phase-separated and time-sharing charging loads transmitted by the middle-layer optimization model, the lower-layer optimization model establishes a nodal marginal price clearing model considering new energy consumption with the minimum total cost of distribution network operation cost and wind and light abandonment penalties to determine the phase-separated and time-sharing power purchase prices of the charging network operator and transmits them to the upper-layer optimization model. The upper-layer optimization model adjusts the phase-separated and time-sharing charging prices based on the phase-separated and time-sharing charging loads obtained from the middle-layer optimization model and the phase-separated and time-sharing power purchase prices obtained from the lower-layer optimization model to maximize the net income of the charging network operator.

[0007] For the above-mentioned electric vehicle phase-separated guidance method for improving new energy consumption based on evolutionary game, the specific process of establishing the three-layer optimization model is as follows:

[0008] The upper-layer optimization model establishes a phase-separated and time-sharing charging price model for the charging network operator with the goal of maximizing the net income of the charging network operator: In the formula: U CNO is the income of the charging network operator, I is the total number of all charging stations within the jurisdiction of the charging network operator, is the charging price of the y-phase of the i-th charging station at time t, that is, the phase-separated and time-sharing charging price, is the power purchase price of the y-phase of the i-th charging station at time t, that is, the phase-separated and time-sharing power purchase price, is the charging load of the y-phase of the i-th charging station at time t, that is, the phase-separated and time-sharing charging load, and Δt is the time interval;

[0009] The specific steps for solving the evolutionary game of the middle-layer optimization model are as follows:

[0010] Step 1: Assume that there are I charging stations and Z electric vehicles, and each vehicle has 3I charging strategies, each of which is described as s i-y , where i is the charging station, i = 1, 2,..., m,..., I, y is the charging phase, y = A, B, C; the proportion of the number of electric vehicles selecting s i-y at time t to the total number of electric vehicles is x i-y (t), 0 ≤ xi-y If \(n(t)\leq1\), the state of the electric vehicle population can be expressed as \(X = [x^{(1)}(t),\cdots,x^{(i)}(t),\cdots,x^{(n)}(t)]\); if an electric vehicle selects a charging strategy \(s^{(i)}(t)\) at time \(t\), then the state of this electric vehicle is \(s^{(i)}(t)=1\), otherwise \(s^{(i)}(t)=0\); 1-y (t),…,x m-y (t),…,x I-y (t)]; if the electric vehicle selects the charging strategy \(s^{(i)}\) at time \(t\) i-y , then the state of this electric vehicle is \(s^{(i)}\) i-y (t)=1, otherwise \(s^{(i)}\) i-y (t)=0;

[0011] Step 2: Establish a payment function with the goal of minimizing the sum of the costs of all electric vehicles that select the charging strategy \(s^{(i)}\) at time \(t\) i-y where: In the formula: is the total cost of all electric vehicles that select the charging strategy \(s^{(i)}\) at time \(t\), i-y is the charging price of the \(y\)-phase of the \(i\)-th charging station at time \(t\), and \(P^{(i)}\) is the charging power of the electric vehicle; 0

[0012] Step 3: The kinetic equation of the dynamic evolution process of the electric vehicle population state is where, is the conditional transition probability from the charging strategy \(s^{(i)}\) at time \(t\) to the charging strategy \(s^{(j)}\), i-y to the charging strategy \(s^{(j)}\) m-y , is the conditional transition probability from the charging strategy \(s^{(j)}\) at time \(t\) to the charging strategy \(s^{(k)}\); m-y to the charging strategy \(s^{(k)}\) i-y ;

[0013] Step 4: The charging strategy correction protocol of the evolutionary game adopts the logit protocol, and the conditional transition probability Then the time-continuous dynamic evolution equation of the electric vehicle population state is: Discretize the above time-continuous dynamic evolution equation to obtain the time-discretized dynamic evolution equation of the electric vehicle population state where: \(w\) is the number of iterations of the evolutionary game; \(\lambda\) is the iteration step of the evolutionary game;

[0014] Step 5: Judge whether the payment functions of all charging strategies reach the evolutionary convergence condition. When \(|x^{(i)}(w + 1)-x^{(i)}(w)|\leq\sigma\), where \(\sigma\) is the set convergence error, it is considered that the evolutionary convergence condition is reached, and the solution is terminated. The electric vehicle population will no longer change the charging strategy; m-y (w + 1)-x m-y (w)|≤σ, σ is the set convergence error, it is considered that the evolutionary convergence condition is reached, the solution ends, and the electric vehicle population will no longer change the charging strategy;

[0015] Step 6: Output the final charging strategy of the electric vehicle population, and the phase-separated and time-sharing charging loads of each station and each phase can be obtained;

[0016] Through the evolutionary game process of the above middle-level optimization model, the phase-separated and time-sharing charging electricity price can be realized To the mapping process of the phased and time-sharing charging load ;

[0017] The lower-layer optimization model establishes a nodal marginal price clearing model considering new energy consumption with the minimum sum of the operation cost of the distribution network and the penalty for wind and light curtailment:

[0018]

[0019] In the formula, f pri is the total cost of the distribution network, a g , b g and b 0 are the generator cost coefficients, is the active power output of the distributed power source at the generator node h at time t, and L is the set of generator nodes; is the electricity quantity purchased from the external power grid K i at time t, and Π(K) is the set connected to the external power grid; is the power generated by the wind power and photovoltaic units connected to the y-phase of the distribution network node; is the actual power of the wind power and photovoltaic units connected to the y-phase of the distribution network node; c loss is the penalty coefficient;

[0020] For the optimal power flow model of the distribution network, the active power equality constraint and the reactive power equality constraint are expressed as follows: In the formula, are the active power and reactive power of the y-phase branch l sj at time t respectively; is the reactive power output of the distributed power source at the generator node h at time t; are the active load and reactive load at the y-phase node j at time t respectively; is the current flowing through the branch l sj in the y-phase at time t; r sj , x sj are the resistance and reactance of the branch l sj respectively; w j is the set of the end nodes of all branches with the head end at j; are the active power and reactive power of the y-phase branch l jk at time t respectively; is the charging load connected to the distribution network node in the y-phase at time t, and is also the charging load of the i-th charging station in the y-phase at time t; J L is the set of distribution network nodes;

[0021] Solving the Lagrange multipliers of the active power equality constraints in the optimal power flow model of the distribution network is equivalent to solving the phased nodal marginal price of the distribution network. The phased time-of-use electricity purchase price of the charging station is the phased nodal marginal price of the node where it is located, and thus the mapping process from the phased time-of-use charging load to the phased time-of-use electricity purchase price can be realized.

[0022] The present invention proposes an electric vehicle phased guidance method based on evolutionary game to improve new energy consumption, so as to maximize the revenue of the charging network operator, minimize the charging cost of electric vehicles and the total cost of the distribution network, and achieve a win-win situation for all three parties. Description of the Drawings

[0023] Figure 1 It is a framework diagram of electric vehicle phased charging guidance based on evolutionary game to improve new energy consumption. Detailed Embodiment

[0024] An electric vehicle phased guidance method based on evolutionary game to improve new energy consumption, the method includes a three-layer optimization model. The upper-layer optimization model is the phased time-of-use charging electricity price model of the charging network operator. The middle-layer optimization model is the electric vehicle phased charging scheduling model based on evolutionary game. The lower-layer optimization model is the nodal marginal price clearing model considering new energy consumption.

[0025] In the upper-layer optimization model, the charging network operator formulates different phased time-of-use charging electricity prices for the electric vehicle charging stations it manages and transmits them to the middle-layer optimization model. In the middle-layer optimization model, the electric vehicle users select the optimal charging strategy with the minimum payment function through evolutionary game in combination with the phased time-of-use charging electricity prices transmitted by the upper-layer optimization model, and at the same time obtain the phased time-of-use charging load and transmit it to the lower-layer optimization model and the upper-layer optimization model. According to the phased time-of-use charging load transmitted by the middle-layer optimization model, the lower-layer optimization model establishes a nodal marginal price clearing model considering new energy consumption with the minimum total cost of the distribution network operation cost and the penalty for wind and light abandonment to determine the phased time-of-use electricity purchase price of the charging network operator and transmit it to the upper-layer optimization model. The upper-layer optimization model adjusts the phased time-of-use charging electricity price based on the phased time-of-use charging load obtained from the middle-layer optimization model and the phased time-of-use electricity purchase price obtained from the lower-layer optimization model to achieve the maximum net revenue of the charging network operator.

[0026] For the upper-layer optimization model:

[0027] The charging network operator formulates the phased time-of-use charging electricity price, and the electric vehicle (EV) users change their charging behaviors according to the electricity price. The feedback of the users on the electricity price will also affect the revenue of the charging network operator.

[0028] Charging network operators benefit from the price difference by purchasing electricity from distribution network operators and selling it to electric vehicle users. A charging network operator (CNO) phase - time - of - use charging price model is established with the goal of maximizing the net income of the charging network operator:

[0029]

[0030] In the formula: U CNO is the income of the charging network operator, I is the total number of all charging stations within the jurisdiction of the charging network operator, is the charging price of the y - phase of the i - th charging station at time t, that is, the phase - time - of - use charging price, is the electricity purchase price of the y - phase of the i - th charging station at time t, that is, the phase - time - of - use electricity purchase price, is the charging load of the y - phase of the i - th charging station at time t, that is, the phase - time - of - use charging load, and Δt is the time interval.

[0031] Phase - time - of - use electricity price constraint

[0032]

[0033] In the formula: α min is the lower limit of the charging price, and α max is the upper limit of the charging price.

[0034] For the middle - layer optimization model, the specific steps of evolutionary game solution are as follows:

[0035] Step 1: Assume there are I charging stations and Z electric vehicles. Each vehicle has 3I charging strategies, each described as s i-y , where i is the charging station (i = 1, 2, … m …, I) and y is the charging phase (y = A, B, C). At time t, the proportion of the number of electric vehicles choosing s i-y to the total number of electric vehicles is x i-y (t), 0 ≤ x i-y (t) ≤ 1, then the state of the electric vehicle population can be expressed as X = [x 1-y (t), …, x m-y (t), …, x I-y (t)].

[0036] For an electric vehicle, if it chooses the charging strategy s i-y at time t, then the state of this electric vehicle is s i-y (t) = 1, otherwise s i-y (t) = 0.

[0037] Step 2: Calculate the probability that an electric vehicle chooses the charging strategy s i-yThe total payment function;

[0038] The total cost includes the charging cost of electric vehicles. Taking the minimum sum of the costs of all electric vehicles that choose the charging strategy s at time t as the goal, a payment function is established as follows: i-y The total cost of all electric vehicles that choose the charging strategy s at time t is the lowest. As follows:

[0039]

[0040] Where: is the total cost of all electric vehicles that choose the charging strategy s at time t. i-y is the charging price of the y-phase of the i-th charging station at time t, which is determined by the charging network operator in the upper-level optimization model. P is the charging power of the electric vehicle. 0

[0041] The payment functions of other charging strategies are similar to the above formula. The payment functions of all charging strategies are

[0042] Step 3: Calculate the conditional switching rate of the charging strategy;

[0043] In the dynamic evolution process of electric vehicles choosing charging strategies, every time there is an opportunity to modify the charging strategy, the electric vehicle will compare the average revenue of choosing the current charging strategy with the average revenue of choosing other charging strategies, and correct the charging strategy at the conditional conversion rate ρ. The utility of the electric vehicle population is continuously optimized as the charging strategy changes, and finally an evolutionary equilibrium is achieved. All electric vehicles correct their own charging strategies, and the kinetic equation of the dynamic evolution process of the electric vehicle population state is as follows.

[0044]

[0045] In the formula, is the conditional conversion probability from the charging strategy s i-y to the charging strategy s m-y at time t. Its meaning is the proportion of electric vehicles that convert from the charging strategy s i-y to the charging strategy s m-y at time t, which is determined by the revision protocol of the selected charging strategy; is the conditional conversion probability from the charging strategy s m-y to the charging strategy s i-y at time t. Its meaning is the proportion of electric vehicles that convert from the charging strategy s m-y to the charging strategy s i-y at time t, which is determined by the revision protocol of the selected charging strategy;

[0046] Step 4: Update the proportion of individuals adopting strategies;

[0047] Due to the incomplete information received by each individual in the electric vehicle population, there will inevitably be random deviations in the assessment of charging costs, resulting in a certain degree of randomness in the selection of charging strategies. Logit dynamically accounts for the perception errors of individuals, considers the individual preferences and information incompleteness of decision-making agents, and makes decisions based on the principle of maximizing individual utility.

[0048] The charging strategy correction protocol of evolutionary game adopts the logit protocol, and the conditional transition probability

[0049]

[0050] Then the time-continuous dynamic evolution equation of the electric vehicle population is:

[0051]

[0052] The above formula reveals the evolution law of the proportion of individuals choosing strategy s m-y in the electric vehicle population, and depicts the game behavior of electric vehicle users in the actual decision-making scenario.

[0053] Discretize the time-continuous dynamic evolution equation given above to obtain the final time-discretized dynamic evolution equation of the electric vehicle population.

[0054] In the formula: w is the number of iterations of the evolutionary game; λ is the step size of the evolutionary game iteration.

[0055] Step 5: Judge whether the payoff functions of all charging strategies reach the evolutionary convergence condition. When σ is the set convergence error, it is considered that the evolutionary convergence condition is reached, and the solution ends. The electric vehicle population will no longer change the charging strategy. This means that the evolutionary game of the electric vehicle charging strategy arrangement reaches an equilibrium state, and at this time the cost of the electric vehicle population reaches the minimum.

[0056] Step 6: Output the final charging strategy of the electric vehicle population, and the phase-separated and time-sharing charging loads of each phase of each station can be obtained.

[0057] Through the evolutionary game process of the above middle-layer optimization model, the mapping process from the phase-separated and time-sharing charging electricity price to the phase-separated and time-sharing charging load can be realized.

[0058] For the lower-layer optimization model, a nodal marginal price clearing model considering new energy accommodation is established with the sum of the operation cost of the distribution network and the penalty for abandoning wind and light minimized:

[0059]

[0060] In the formula, f pri is the total cost of the distribution network, a g and bg and b 0 is the generator cost coefficient, is the active power output of the distributed power source at the generator node h at time t, and L is the set of generator nodes; is the power purchased from the external power grid K at time t i Π(K) is the set connected to the external power grid; is the power generated by the wind power and photovoltaic units connected to the distribution network node in phase y; is the actual power of the wind power and photovoltaic units connected to the distribution network node in phase y; c loss is the penalty coefficient.

[0061] For the optimal power flow of the distribution network, first, any node should satisfy the balance of the active power equality constraint and the reactive power equality constraint, which can be expressed by the following formula

[0062]

[0063] In the formula, are the active power and reactive power of the branch l in phase y at time t respectively sj ; is the reactive power output of the distributed power source at the generator node h at time t; are the active load and reactive load at the node j in phase y at time t respectively; is the current flowing through the branch l in phase y at time t sj ; r sj and x sj are the resistance and reactance of the branch l respectively sj ; w j is the set of the end nodes of all branches with the head end being j; are the active power and reactive power of the branch l in phase y at time t respectively jk ; is the charging load connected to the node j in the distribution network in phase y at time t, and is also the charging load of the i-th charging station in phase y at time t; J L is the set of distribution network nodes.

[0064] Secondly, the lines of the radial distribution network should satisfy the voltage drop equality constraint, which can be expressed as

[0065]

[0066] In the formula, are the voltages of the distribution network node s and the node j in phase y at time t respectively.

[0067] The power flow related variables should also satisfy the following boundary constraint conditions:

[0068]

[0069]

[0070] where I max,sj is the maximum current amplitude of the line; U max,j , U min,j are the upper limit and lower limit of the amplitude of the node voltage respectively;

[0071] Meanwhile, the constraints among the line power, voltage and current amplitude are described by second-order cone inequalities, and the power flow problem is transformed into a second-order cone programming problem as follows.

[0072]

[0073] Solving the Lagrange multipliers of the active power equality constraint in the optimal power flow model of the distribution network is equivalent to solving the phase-separated nodal marginal price of the distribution network, and the phase-separated time-of-use electricity purchase price of the charging station is the phase-separated nodal marginal price of the corresponding node, thus realizing the mapping process from the phase-separated time-of-use charging load to the phase-separated time-of-use electricity purchase price

Claims

1. An electric vehicle phase guidance method for improving new energy consumption based on evolutionary game theory is characterized by: The method includes three-layer optimization models, the upper optimization model is the phase-time charging price model of the charging network operator, the middle optimization model is the phase-time charging scheduling model of the electric vehicle based on evolutionary game, and the lower optimization model is the node marginal price clearing model considering the consumption of new energy; in the upper optimization model, the charging network operator formulates different phase-time charging prices for the electric vehicle charging stations it manages and passes them to the middle optimization model; in the middle optimization model, the electric vehicle user selects the optimal charging strategy through evolutionary game with the minimum payment function in combination with the phase-time charging prices passed by the upper optimization model. At the same time, the phase-by-phase time-sharing charging load is obtained and passed to the lower optimization model and the upper optimization model; according to the phase-by-phase time-sharing charging load passed by the middle optimization model, the lower optimization model establishes a node marginal price clearing model considering the consumption of new energy with the minimum total cost of the distribution network operation cost and the penalty for wind and solar power abandonment to determine the phase-by-phase time-sharing electricity purchase price of the charging network operator, and passes it to the upper optimization model; the upper optimization model adjusts the phase-by-phase time-sharing charging price based on the phase-by-phase time-sharing charging load obtained by the middle optimization model and the phase-by-phase time-sharing electricity purchase price obtained by the lower optimization model to maximize the net profit of the charging network operator; The specific process of establishing the three-layer optimization model is as follows: The upper optimization model establishes a phase-based and time-based charging price model for charging network operators with the goal of maximizing the net profit of charging network operators: Where: U CNO is the revenue of the charging network operator, I is the total number of charging stations within the jurisdiction of the charging network operator, is the charging price of phase y at the i-th charging station at time t, that is, the phase-by-phase and time-by-phase charging price, is the purchase price of electricity for phase y of the i-th charging station at time t, that is, the phase-by-phase and time-by-phase purchase price, is the charging load of phase y of the i-th charging station at time t, that is, the phase-by-phase and time-by-phase charging load, and Δt is the time interval; The specific steps for solving the evolutionary game of the middle-level optimization model are as follows: Step 1: Assume that there are I charging stations, Z electric vehicles, and each vehicle has 3I charging strategies, each described as s i-y , where i is the charging station and y is the charging phase; at time t, select s i-y The ratio of the number of electric vehicles to the total number of electric vehicles is x i-y (t), Then the electric vehicle population state can be expressed as X = [x 1-y (t),…,x m-y (t),…,x I-y (t)]; if the electric vehicle chooses charging strategy s at time t i-y , then the state of the electric car is s i-y (t)=1, otherwise s i-y (t) = 0; Step 2: All selected charging strategies s at time t i-y The payment function is established with the goal of minimizing the sum of the costs of electric vehicles. Where: Select charging strategy s for time t i-y The total cost of all electric vehicles, is the charging price of the y-phase at the i-th charging station at time t, and P0 is the charging power of the electric vehicle; Step 3: The dynamic equation of the dynamic evolution process of the electric vehicle population state is: In the formula, is the charging strategy s at time t i-y To charging strategies m-y The conditional switching probability, is the charging strategy s at time t m-y To charging strategies i-y The conditional switching probability of Step 4: The charging strategy modification protocol of the evolutionary game adopts the logit protocol, and the conditional conversion probability Then the time-continuous dynamic evolution equation of the electric vehicle population state is: Discretize the above time-continuous dynamic evolution equation to obtain the time-discrete dynamic evolution equation of the electric vehicle population state: Where: w is the number of evolutionary game iterations; λ is the evolutionary game iteration step length; Step 5: Determine whether the payment functions of all charging strategies have reached the evolution convergence condition. m-y (w+1)-x m-y (w)|≤σ, σ is the set convergence error, it is considered that the evolution convergence condition has been met, the solution is completed, and the electric vehicle population will no longer change the charging strategy; Step 6: Output the final charging strategy of the electric vehicle population to obtain the phase-by-phase and time-by-phase charging load of each phase at each station; Through the evolutionary game process of the above middle-level optimization model, the phase-by-phase and time-by-phase charging price can be realized. Phase-by-phase and time-by-time charging load The mapping process; The lower optimization model is to establish a node marginal electricity price clearing model considering the consumption of new energy by minimizing the sum of the distribution network operation cost and the penalty for wind and solar abandonment: In the formula, f pri is the total cost of the distribution network, a g 、b g and b0 is the generator cost coefficient, is the active output of the distributed generation at the generator node h at time t, and L is the set of generator nodes; is the power from the external power grid K at time t i The amount of electricity purchased, Π(K) is the set connected to the external grid; The power generated by the wind power and photovoltaic units connected to the distribution network node in phase y; is the actual power of the wind power and photovoltaic units connected to the distribution network node in phase y; c loss is the penalty coefficient; For the optimal power flow model of the distribution network, the active power equality constraint and the reactive power equality constraint are expressed as follows: In the formula, They are respectively the y-phase branch l at time t sj Active power and reactive power; is the reactive power output of the distributed generation at the generator node h at time t; are the active load and reactive load at node j of phase y at time t respectively; is the branch l at time t sj The current flowing in the y phase; r sj 、x sj Branch l sj Resistance and reactance; w j is the set of end nodes of all branches whose head end is j; They are respectively the y-phase branch l at time t jk Active power and reactive power; is the charging load on the y phase connected to the node j in the distribution network at time t, and is also the charging load on the y phase of the i-th charging station at time t; J L is a set of distribution network nodes; Solving the Lagrange multiplier of the active power equation constraint in the optimal power flow model of the distribution network is equivalent to solving the phase-by-phase node marginal electricity price of the distribution network. The phase-by-phase time-of-use electricity purchase price of the charging station is the phase-by-phase node marginal electricity price of the node where it is located, which can realize the phase-by-phase time-of-use charging load Phase-by-phase and time-of-use electricity purchase price The mapping process.

Images