Dynamic economic and emission optimization scheduling method for plug-in electric vehicle power generation units

Abstract

The present invention discloses a method for dynamic economic and emission optimization scheduling in combination with a plug-in vehicle power generation unit, which belongs to the field of microgrid scheduling and includes the following steps: using the power demand in each time period and the number of plug-in vehicles that can be connected to the power grid as input parameters of the model; constructing a power generation unit optimization scheduling model; using a peak shaving and valley filling algorithm to allocate the charging and discharging of plug-in vehicles, and reasonably allocating the charging and discharging behaviors of plug-in vehicles to different time periods according to the power demand during peak and valley periods; using a multi-objective genetic algorithm to optimize the power generation of the power generation unit; and obtaining the power generation of the power generation unit at each moment and the number of plug-in vehicles connected to the power grid through optimization and solution. The present invention can effectively balance the power load and significantly improve the safety and stability of the power grid operation by coordinating the charging and discharging behaviors of plug-in vehicles with power generation units, discharging plug-in vehicles during peak periods and charging during valley periods through the coordinated scheduling of the charging and discharging behaviors of plug-in vehicles and power generation units.

Description

Technical Field

[0001] The present invention relates to the field of microgrid dispatching, and in particular to a dynamic economic and emission optimization dispatching method combined with a plug-in vehicle power generation unit. Background Art

[0002] Economic dispatch is an important task in the energy system. In recent years, with the concern for the environment, emissions have also been taken into consideration. Nowadays, in the energy system, economic and emission optimization dispatch is extremely important. However, economic emission optimization dispatch can only consider a specific moment. Considering the electricity demand of the city, the use of dynamic economic emission optimization dispatch has become an important consideration for the urban power supply system.

[0003] Plug-in vehicles are an important part of the future power system. A large number of plug-in vehicles are charged in cities, which leads to a large increase in the city's power demand. In peak hours, the power generation of the power supply system is not enough to meet the city's power demand. Adding more power stations will generate more costs and emissions. If no measures are taken, the city's power demand will be overloaded. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a dynamic economic and emission optimization scheduling method combining a plug-in vehicle power generation unit with a simple algorithm, safety, stability and reliability.

[0005] The technical solution of the present invention to solve the above technical problems is: a dynamic economic and emission optimization scheduling method combined with a plug-in vehicle power generation unit, comprising the following steps:

[0006] S1: Collect the electricity demand at each time period of the day, the maximum power generation of each power generation unit, the minimum power generation of each power generation unit, and the number of plug-in vehicles that can be connected to the power grid, and use the electricity demand at each time period and the number of plug-in vehicles that can be connected to the power grid as the input parameters of the model;

[0007] S2: Taking the lowest cost and lowest emission generated by the power generation unit and plug-in vehicles as the optimization objectives, the optimal scheduling model of the power generation unit is constructed by constraining the power generation of the power generation unit and the charging and discharging power of the plug-in vehicles;

[0008] S3: The peak-shaving and valley-filling algorithm is used to distribute the charging and discharging of plug-in vehicles. The charging and discharging behaviors of plug-in vehicles are distributed to different time periods according to the power demand during peak and valley periods.

[0009] S4: Optimizing the power generation of the power generation unit using a multi-objective genetic algorithm;

[0010] S5: Through optimization and solution, the power generation of the power generation unit at each moment and the number of plug-in vehicles connected to the power grid are obtained.

[0011] In the above-mentioned method for optimizing the dispatch of the power generation unit of a plug-in vehicle in combination with the dynamic economy and emission, in step S2, the power generation unit optimization dispatch model has two objective functions, one is to minimize the cost, and the other is to minimize the emission:

[0012] ;

[0013] in, represents the emissions from the power generation unit, represents the cost of the power generation unit, Represents the cost of charging and discharging a plug-in vehicle;

[0014] The emissions from the power generation unit are:

[0015] ;

[0016] In the formula, is the total power generation time of the power generation unit, Both are power generation units The emission coefficient is, N is the number of power generation units, It is a power generation unit At the moment of power generation;

[0017] The cost of the power generation unit is:

[0018] ;

[0019] In the formula, It is a power generation unit The cost factor, and It is a power generation unit The valve point effect coefficient, For power generation unit The minimum power generation capacity, Indicates power generation unit power generation efficiency;

[0020] The cost of charging and discharging a plug-in vehicle is:

[0021] ;

[0022] In the formula, is the number of plug-in vehicles participating in the grid, It is Battery cost per kilowatt-hour of power consumption for a plug-in car battery, It is The battery capacity of a plug-in car, is the labor cost of replacing the battery, Indicates The discharge capacity of a plug-in car, DOD is the depth of discharge, It is the battery life at a set depth of discharge.

[0023] In the above-mentioned method for optimizing the dispatch of dynamic economy and emission combined with the plug-in vehicle power generation unit, in step S2, the constraints include:

[0024] Power generation constraints:

[0025] ;

[0026] In the formula, For power generation unit The minimum power generation capacity, For power generation unit Maximum power generation capacity;

[0027] Constraints on plug-in car battery capacity:

[0028] ;

[0029] In the formula, Indicates the minimum battery charge of a plug-in vehicle. Indicates the maximum battery capacity of a plug-in vehicle. For the plug-in cars at the time Battery capacity;

[0030] Constraints on charging and discharging of plug-in vehicles:

[0031] ;

[0032] In the formula, It is plug-in cars at the time The charge capacity, It is plug-in cars at the time The discharge capacity, and They are The maximum charge and discharge capacity of a plug-in vehicle;

[0033] Power generation constraints after the power generation unit is combined with the plug-in vehicle:

[0034] ;

[0035] In the formula, and Plug-in vehicles at the time The discharge power and charging power of It's time The power load demand, represents circuit loss;

[0036] Determine the circuit loss as follows:

[0037] ;

[0038] In the formula, For connecting busbar and busbar No. The admittance of the line, is the number of lines in the system, , The busbar and busbar The voltage, For bus and busbar The phase angle difference between

[0039] The voltage constraint of the power generation unit is determined as follows:

[0040] ;

[0041] In the formula, Indicates that it is located on the busbar The voltage, and Busbar The minimum and maximum voltage limits, is the total number of buses in the system;

[0042] The power flow constraint of the line is determined as follows:

[0043] ;

[0044] In the formula, For the The power flow of the line, For the Maximum power flow limit for each line;

[0045] The ramp rate constraint of the power generation unit is determined as follows:

[0046] ;

[0047] In the formula, and Power generation unit The upper and lower limits of the climbing slope, For power generation unit The power generation at the previous moment, For power generation unit The amount of electricity generated at the current moment.

[0048] In the above-mentioned method for optimizing the dispatch of dynamic economy and emission combined with the plug-in vehicle power generation unit, the specific steps of step S3 are as follows:

[0049] S31: Initialize parameters; divide a day into peak hours, normal hours, and off-peak hours according to the 24-hour power load demand, and determine the available plug-in vehicles, peak hours, and off-peak hours power demand and restriction conditions;

[0050] S32: Allocate remaining resources; according to the power demand, give priority to allowing plug-in vehicles to discharge during peak hours and feed the power back to the power grid. When the plug-in vehicles discharge to the power grid during peak hours, the battery state of charge must be kept within the set range. During off-peak hours, plug-in vehicles will be given priority to be charged. The number of plug-in vehicles is allocated to peak and off-peak periods, and the initial allocation is an average allocation;

[0051] S33: Calculate the remaining resources; after allocating the charging and discharging resources of the plug-in vehicles during the peak period and the off-peak period, calculate the remaining unallocated plug-in vehicles. If some plug-in vehicles cannot be fully discharged during the peak period or are not fully charged during the off-peak period, these remaining resources are allocated to other time periods through a uniform allocation strategy to ensure the balance of power load in each time period;

[0052] S34: Evenly distribute the remaining resources; through the process of peak shaving and valley filling, the remaining plug-in vehicles are evenly distributed to the peak and valley periods as needed, with the periods with the least demand being filled first, until the resources are exhausted or all demand is met.

[0053] In the above-mentioned method for optimizing the dispatch of dynamic economy and emission combined with the plug-in vehicle power generation unit, the specific steps of step S4 are as follows:

[0054] S41: defining an optimization objective function, the objective function includes two aspects: the minimum cost and the minimum emission generated by the power generation unit and the plug-in vehicle, the power generation of the power generation unit and the number of plug-in vehicles connected are used as optimization variables;

[0055] S42: Population initialization: Initialize various parameters of the genetic algorithm, including the population size , maximum number of iterations , crossover probability and mutation probability, and initialize the first generation population, each individual in the population represents a power generation allocation scheme and plug-in vehicle access strategy;

[0056] S43: Fitness evaluation: Perform fitness evaluation on the population, calculate the two objective function values ​​of each individual, use non-dominated sorting to divide the population into different fronts, and calculate the crowding distance of each individual to measure the diversity of individuals in the same front;

[0057] S44: Parent selection: Use tournament selection to select parents based on non-dominated sorting and crowding distance to form a mating pool;

[0058] S45: Generate offspring: Perform crossover and mutation operations on the parent generation in the mating pool to generate new offspring;

[0059] S46: Constraint check: perform constraint check on the generated children to ensure that the power generation allocation scheme and plug-in vehicle access strategy of each child conform to the actual physical constraints;

[0060] S47: Population merging: merge the current population with the newly generated offspring, calculate the objective function value again, and perform non-dominated sorting and crowding distance calculations;

[0061] S48: Select a new population: Select the top population from the merged population based on the non-dominated sorting result and crowding distance. Individuals ensure the diversity of the next generation population and the approach to the Pareto frontier;

[0062] S49: Iterative update: Repeat steps S43 to S48 until the preset maximum number of iterations is reached. , gradually optimize the non-dominated solutions of the population;

[0063] S410: Output Pareto optimal solution: Output the final Pareto optimal solution set, that is, the individual ranked 1 in the non-dominated sorting, as the final solution, representing the optimal power generation allocation plan of the power generation unit and the optimal access strategy of the plug-in vehicle.

[0064] In the above-mentioned method for optimizing the dispatch of dynamic economy and emission of plug-in vehicle power generation units, in step S45, crossover uses simulated binary crossover or two-point crossover to generate diversified offspring, and mutation uses a combination of polynomial mutation and non-uniform mutation to ensure population diversity and gradually approach the optimal solution; a probability selection method is used to select the type of crossover and mutation, and the probability Determined by the following formula:

[0065] ;

[0066] In the formula Indicates the current iteration number, Indicates the maximum number of iterations, randomly generating a random number between 0 and 1 ,if < , then simulated binary crossover and non-uniform mutation are used; if , a combination of two-point crossover and polynomial mutation is used.

[0067] In the above-mentioned method for optimizing the dispatch of dynamic economy and emission of plug-in electric vehicle power generation units, in step S45, the simulated binary crossover is a crossover method for generating diversified offspring, by adjusting the crossover coefficient Control the crossover effect, the formula is as follows:

[0068] ;

[0069] in, Determined by the following formula:

[0070] ;

[0071] In the formula, is the distribution index that controls the crossover strength, is a randomly generated probability value, , is the parent individual used for crossover operation, , They are the two offspring after the crossover operation.

[0072] In the above-mentioned method for optimizing the dispatch of dynamic economy and emission combined with the plug-in vehicle power generation unit, in step S45, two-point crossover is used to generate diversified offspring from the parent generation, and new offspring are generated by selecting two crossover points and exchanging intermediate gene fragments, and the formula is as follows:

[0073] ;

[0074] ;

[0075] In the formula, Represents the chromosome genes, is the parent individual used for crossover operation, , Respectively Middle The value of genes; the chromosome of the parent individual contains D genes; and are two randomly generated cross position indices, where ; , are the two offspring after the crossover operation, , Respectively , Middle The value of a gene.

[0076] In the above-mentioned method for optimizing the dispatch of dynamic economy and emission of plug-in electric vehicle power generation units, in step S45, the polynomial mutation increases the population diversity by introducing a small mutation in the solution space, and the mutation effect is determined by the mutation amplitude. Determine, the formula is as follows:

[0077] ;

[0078] in, Determined by the following formula:

[0079] ;

[0080] In the formula, Indicates the first The value of each gene; and Respectively represent The minimum and maximum values ​​allowed for each gene are used to limit the value range of the gene; is the variation range, which is used to control the variation range of genes. The value of is calculated based on the random number and the distribution index; is a randomly generated probability value in the range of [0, 1], which is used to determine the direction and amplitude of mutation; is the multinomial distribution exponent, controlling the strength and range of variation.

[0081] In the above-mentioned method for optimizing the dispatch of dynamic economy and emission of plug-in vehicle power generation units, in step S45, the non-uniform variation dynamically adjusts the variation amplitude according to the number of iterations. , in order to enhance convergence and gradually approach the optimal solution, the formula is as follows:

[0082] ;

[0083] In the formula, Indicates the first The value of each gene; and Respectively represent The minimum and maximum values ​​allowed for each gene; Indicates the current iteration number, Indicates the maximum number of iterations; is a randomly generated probability value, ranging from [0, 1]; is a random number in [0,1]; is the non-uniformity coefficient, which is used to control the variation amplitude to gradually decrease with the number of iterations.

[0084] The beneficial effects of the present invention are:

[0085] 1. The present invention can effectively balance the power load, reduce the demand for power generation units, and significantly improve the safety and stability of power grid operation by coordinating the charging and discharging behavior of plug-in vehicles with power generation units, discharging plug-in vehicles during peak periods and charging plug-in vehicles during off-peak periods. At the same time, the participation of plug-in vehicles greatly reduces peak loads and reduces the demand for power grid expansion, which has good economic and environmental benefits.

[0086] 2. Through the peak-shaving and valley-filling algorithm, the present invention can intelligently allocate the charging and discharging resources of plug-in vehicles, give priority to meeting the power demand during peak periods, and optimize the charging plan of plug-in vehicles during valley periods when the power is sufficient, ensuring that the battery is in the best charge state, thereby maximizing the synergy between plug-in vehicles and the power grid.

[0087] 3. The present invention has made a number of improvements in the multi-objective genetic algorithm used. First, by introducing a crossover strategy combining simulated binary crossover (SBX) and two-point crossover (Two-Point Crossover), the population can better explore the solution space while maintaining the diversity of solutions, avoiding falling into the local optimum. Secondly, by combining the dual mutation mechanism of polynomial mutation (Polynomial Mutation) and non-uniform mutation (non-uniform Mutation), it is ensured that the solution space is fully explored in the early iterations, and the solution can be fine-tuned in the later stage, thereby accelerating the convergence to the global optimal solution; these improvements greatly improve the convergence speed of the algorithm and its ability to solve complex multi-objective problems.

[0088] 4. The present invention adopts a multi-objective genetic algorithm for optimization, which can simultaneously consider power generation costs and emissions, optimize the power generation of power generation units and the access strategy of plug-in vehicles, and provide a set of Pareto optimal solutions that balance economic and environmental benefits through a non-dominated sorting method, ensuring that the scheduling scheme can be flexibly adapted under different operating conditions.

[0089] 5. The plug-in vehicle charging and discharging scheduling model of the present invention fully considers the charge state of the plug-in vehicle battery and the number of plug-in vehicles connected. While meeting the load demand of the power grid, it also maximizes the life of the plug-in vehicle battery, reduces maintenance costs, and improves system reliability.

[0090] 6. The scheduling method of the present invention is simple and easy to use, can effectively handle complex multi-objective optimization problems, and adapt to various power demand scenarios in actual power grid operation; the crossover and mutation operation of the genetic algorithm is adopted, so that the population can fully explore in the solution space, effectively avoid falling into the local optimal solution, and enhance the robustness of the method. BRIEF DESCRIPTION OF THE DRAWINGS

[0091] Figure 1 It is the overall flow chart of the present invention.

[0092] Figure 2 Flowchart for peak load shaving and valley filling for plug-in vehicles.

[0093] Figure 3 This is the flow chart of the multi-objective genetic algorithm.

[0094] Figure 4 This is the algorithm result diagram of the present invention. DETAILED DESCRIPTION

[0095] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0096] like Figure 1 As shown, a dynamic economic and emission optimization scheduling method combining a plug-in vehicle power generation unit includes the following steps:

[0097] S1: Collect the electricity demand at each time period of the day, the maximum power generation Pmax of each power generation unit, the minimum power generation Pmin of each power generation unit, and the number of plug-in vehicles that can be connected to the power grid, and use the electricity demand at each time period and the number of plug-in vehicles that can be connected to the power grid as the input parameters of the model.

[0098] The power demand and power generation limit corresponding to each power generation unit are different. For example, when there are 6 power generation units, the power generation limit is shown in Table 1:

[0099]

[0100] S2: Taking the lowest cost and lowest emission generated by the power generation unit and plug-in vehicles as the optimization goal, the optimal scheduling model of the power generation unit is constructed by constraining the power generation of the power generation unit and the charging and discharging amount of the plug-in vehicle.

[0101] In step S2, the objective functions of the power generation unit optimization scheduling model are two: one is to minimize the cost, and the other is to minimize the emission:

[0102] ;

[0103] in, represents the emissions from the power generation unit, represents the cost of the power generation unit, Represents the cost of charging and discharging a plug-in vehicle.

[0104] The emissions from the power generation unit are:

[0105] ;

[0106] In the formula, is the total power generation time of the power generation unit, Both are power generation units The emission coefficient depends on the type of power generation unit. Table 2 shows the emission coefficient of a certain type of power generation unit. N is the number of power generation units. It is a power generation unit At the moment of power generation.

[0107]

[0108] The cost of the power generation unit is:

[0109] ;

[0110] In the formula, It is a power generation unit The cost coefficient depends on the type of power generation unit. Table 3 shows the cost coefficient of a certain type of power generation unit. and It is a power generation unit The valve point effect coefficient, For power generation unit The minimum power generation capacity, Indicates power generation unit power generation efficiency.

[0111]

[0112] The cost objective function of plug-in vehicle charging and discharging is:

[0113] ;

[0114] In the formula, is the number of plug-in vehicles participating in the grid, It is Battery cost per kilowatt-hour of power consumption for a plug-in car battery, It is The battery capacity of a plug-in car, is the labor cost of replacing the battery, Indicates The discharge capacity of a plug-in car, DOD is the depth of discharge, It is the battery life at a set depth of discharge.

[0115] The constraints include:

[0116] Power generation constraints:

[0117] ;

[0118] In the formula, For power generation unit The minimum power generation capacity, For power generation unit Maximum power generation.

[0119] Constraints on plug-in car battery capacity:

[0120] ;

[0121] In the formula, Indicates the minimum battery charge of a plug-in vehicle. Indicates the maximum battery capacity of a plug-in vehicle. For the plug-in cars at the time This means that the battery charge level must always be maintained between these two limits to ensure safe and efficient operation of the battery.

[0122] Constraints on charging and discharging of plug-in vehicles:

[0123] ;

[0124] In the formula, It is plug-in cars at the time The charge capacity, It is plug-in cars at the time The discharge capacity, and They are The maximum charge and discharge capacity of a plug-in vehicle. This ensures that the charging and discharging process does not exceed the battery's capabilities, protecting the battery life.

[0125] Power generation constraints after the power generation unit is combined with the plug-in vehicle:

[0126] ;

[0127] In the formula, and Plug-in vehicles at the time The discharge power and charging power of It's time The power load demand, Represents circuit loss. This ensures that the system can meet all power demands during operation.

[0128] Determine the circuit loss as follows:

[0129] ;

[0130] In the formula, For connecting busbar and busbar No. The admittance of the line, is the number of lines in the system, , The busbar and busbar The voltage, For bus and busbar The formula calculates the power lost during power transmission through line parameters and voltage difference.

[0131] The voltage constraint of the power generation unit is determined as follows:

[0132] ;

[0133] In the formula, Indicates that it is located on the busbar The voltage, and Busbar The minimum and maximum voltage limits, is the total number of buses in the system. This ensures that the voltage remains within acceptable ranges, thus maintaining system stability and safety.

[0134] The power flow constraint of the line is determined as follows:

[0135] ;

[0136] In the formula, For the The power flow of the line, For the This constraint ensures that the power flow of a line does not exceed its carrying capacity, thus avoiding overload conditions.

[0137] The ramp rate constraint of the power generation unit is determined as follows:

[0138] ;

[0139] In the formula, and Power generation unit The upper and lower limits of the climbing slope, For power generation unit The power generation at the previous moment, For power generation unit The power generation at the current moment. This constraint is used to control the output change of the generator in different time periods, ensure that the response speed of the generator is within an acceptable range, and avoid system fluctuations.

[0140] S3: The peak-shaving and valley-filling algorithm is used to distribute the charging and discharging of plug-in vehicles. The charging and discharging behaviors of plug-in vehicles are allocated to different time periods according to the power demand during peak and valley periods.

[0141] like Figure 2 As shown, the specific steps of step S3 are:

[0142] S31: Initialize parameters; divide a day into peak hours, normal hours, and off-peak hours according to the 24-hour power load demand, and determine the available plug-in vehicles, peak hours, and off-peak hours power demand and restriction conditions;

[0143] S32: Allocate remaining resources; according to the power demand, give priority to discharging plug-in vehicles during peak hours and feed the power back to the grid, thereby reducing the pressure on the power generation unit; when discharging to the grid during peak hours, plug-in vehicles must keep the battery charge state within the set range to avoid over-discharge; during off-peak hours, plug-in vehicles will be given priority to be charged to prepare for the discharge demand during subsequent peak hours; allocate the number of plug-in vehicles to peak and off-peak periods, with an initial equal allocation;

[0144] S33: Calculate the remaining resources; after allocating the charging and discharging resources of the plug-in vehicles during the peak period and the off-peak period, calculate the remaining unallocated plug-in vehicles. If some plug-in vehicles cannot be fully discharged during the peak period or are not fully charged during the off-peak period, these remaining resources are reasonably allocated to other time periods through a uniform allocation strategy to ensure the balance of power load in each time period;

[0145] S34: Evenly distribute the remaining resources; through the process of peak shaving and valley filling, the remaining plug-in vehicles are evenly distributed to the peak and valley periods as needed, with the periods with the least demand being filled first, until the resources are exhausted or all demand is met.

[0146] For discharge behavior during peak periods, electric vehicles are given priority to discharge when the power demand is high; during the valley period, electric vehicles are ensured to be fully charged to avoid insufficient power during the next peak period. This peak-shaving and valley-filling strategy can effectively alleviate the load pressure of the power grid during peak periods, fill the power gap during valley periods, and balance the power supply and demand of the entire power grid.

[0147] S4: Optimize the power generation of the power generation unit using a multi-objective genetic algorithm.

[0148] like Figure 3 As shown, the specific steps of step S4 are:

[0149] S41: Define an optimization objective function, which includes two objectives: the minimum cost and the minimum emission generated by the power generation unit and the plug-in vehicle. The power generation of the power generation unit and the number of plug-in vehicles connected are used as optimization variables.

[0150] S42: Population initialization: Initialize various parameters of the genetic algorithm, including the population size , maximum number of iterations , crossover probability and mutation probability, and initialize the first generation population. Each individual in the population represents a power generation allocation scheme and plug-in vehicle access strategy.

[0151] S43: Fitness evaluation: Perform fitness evaluation on the population, calculate the two objective function values ​​of each individual, use non-dominated sorting to divide the population into different fronts, and at the same time, calculate the crowding distance of each individual to measure the diversity of individuals in the same front.

[0152] S44: Parent selection: Use tournament selection to select parents based on non-dominated sorting and crowding distance to form a mating pool.

[0153] S45: Generate offspring: Perform crossover and mutation operations on the parent generation in the mating pool to generate new offspring.

[0154] Crossover uses simulated binary crossover or two-point crossover to produce diverse offspring. Mutation uses a combination of polynomial mutation and non-uniform mutation to ensure population diversity and gradually approach the optimal solution. Probabilistic selection is used to select the type of crossover and mutation. Determined by the following formula:

[0155] ;

[0156] In the formula Indicates the current iteration number, Indicates the maximum number of iterations, randomly generating a random number between 0 and 1 ,if < , then simulated binary crossover and non-uniform mutation are used; if , a combination of two-point crossover and polynomial mutation is used.

[0157] Simulated binary crossover is a crossover method used to generate diverse offspring by adjusting the crossover coefficient Control the crossover effect, the formula is as follows:

[0158] ;

[0159] in, Determined by the following formula:

[0160] ;

[0161] In the formula, is the distribution index that controls the crossover strength, is a randomly generated probability value, , is the parent individual used for crossover operation, , They are the two offspring after the crossover operation.

[0162] Two-point crossover is used to generate diverse offspring from the parent generation. New offspring are generated by selecting two crossover points and exchanging the intermediate gene fragments. The formula is as follows:

[0163] ;

[0164] ;

[0165] In the formula, Represents the chromosome genes, is the parent individual used for crossover operation, , Respectively Middle The value of genes; the chromosome of the parent individual contains D genes; and are two randomly generated cross position indices, where ; , are the two offspring after the crossover operation, , Respectively , Middle The value of a gene.

[0166] Polynomial mutation increases population diversity by introducing small mutations in the solution space. The effect of mutation is determined by the magnitude of the mutation. Determine, the formula is as follows:

[0167] ;

[0168] in, Determined by the following formula:

[0169] ;

[0170] In the formula, Indicates the first The value of each gene; and Respectively represent The minimum and maximum values ​​allowed for each gene are used to limit the value range of the gene; is the variation range, which is used to control the variation range of genes. The value of is calculated based on the random number and the distribution index; is a randomly generated probability value in the range of [0, 1], which is used to determine the direction and amplitude of mutation; is the multinomial distribution exponent, controlling the strength and range of variation.

[0171] Non-uniform mutation dynamically adjusts the mutation amplitude according to the number of iterations , in order to enhance convergence and gradually approach the optimal solution, the formula is as follows:

[0172] ;

[0173] In the formula, Indicates the first The value of each gene; and Respectively represent The minimum and maximum values ​​allowed for each gene; Indicates the current iteration number, Indicates the maximum number of iterations; is a randomly generated probability value, ranging from [0, 1]; is a random number in [0,1]; is the non-uniformity coefficient, which is used to control the variation amplitude to gradually decrease with the number of iterations.

[0174] S46: Constraint check: perform constraint check on the generated children to ensure that the power generation allocation scheme and plug-in vehicle access strategy of each child conform to the actual physical constraints;

[0175] S47: Population merging: merge the current population with the newly generated offspring, calculate the objective function value again, and perform non-dominated sorting and crowding distance calculations;

[0176] S48: Select a new population: Select the top population from the merged population based on the non-dominated sorting result and crowding distance. Individuals ensure the diversity of the next generation population and the approach to the Pareto frontier;

[0177] S49: Iterative update: Repeat steps S43 to S48 until the preset maximum number of iterations is reached. , gradually optimize the non-dominated solutions of the population;

[0178] S410: Output Pareto optimal solution: Output the final Pareto optimal solution set, that is, the individual ranked 1 in the non-dominated sorting, as the final solution, representing the optimal power generation allocation plan of the power generation unit and the optimal access strategy of the plug-in vehicle.

[0179] The advantages of the improved algorithm in the present invention compared with the common multi-objective genetic algorithm (NSGA-II) are:

[0180] 1. Adaptive crossover and mutation strategy: The improved algorithm adjusts the crossover and mutation methods according to the number of iterations, which enhances adaptability and search efficiency, while NSGA-II usually uses fixed crossover and mutation rates, which may not be able to adapt to the needs of different stages.

[0181] 2. Dynamic constraint checking: Constraint checking is performed in each generation to ensure that each individual meets the physical constraints and enhance the feasibility of the solution. However, NSGA-II has a relatively simple constraint handling in this regard, which may lead to solutions that do not meet the constraints.

[0182] 3. Selection mechanism based on crowding distance: By combining crowding distance and fitness for selection, the improved algorithm can better maintain population diversity and avoid early convergence, while NSGA-II mainly relies on non-dominated sorting and crowding distance, which may lead to insufficient diversity in some cases.

[0183] 4. Flexibility of multi-objective optimization: The improved algorithm not only optimizes power generation costs and emissions, but can also adjust targets based on actual conditions. This flexibility makes it more widely used in complex power systems, while NSGA-II is relatively fixed in target setting.

[0184] 5. Improved offspring generation: Combining multiple crossover and mutation methods, the improved algorithm can generate more diverse offspring, which helps to explore the solution space more comprehensively, while the mutation and crossover operations of NSGA-II are relatively simple.

[0185] Before comparing and explaining the effects of the algorithm of the present invention, other algorithms are briefly described: MHBA (hybrid bat algorithm), which searches for solutions by imitating the behavior of bats and combines other strategies to enhance local search capabilities. It is suitable for fine-tuning, but has high computational complexity. MOEA / D (multi-objective optimization algorithm based on decomposition), decomposes the multi-objective problem into several single-objective sub-problems and optimizes each sub-problem separately. This method converges quickly, but is sensitive to the way the problem is decomposed.

[0186] Compared with MHBA and MOEA / D, the proposed algorithm Muti-NSGA2 enhances the diversity and local convergence of solutions by combining methods such as simulated binary crossover, two-point crossover, polynomial mutation and non-uniform mutation. Figure 4 As shown, Figure 4 The horizontal axis is the sum of the costs of the power generation unit and the plug-in vehicle, and the vertical axis is the emissions of the power generation unit, which is Figure 4 It can be seen that the algorithm of the present invention performs well in search capability and solution diversity. It not only covers a wider range in global search, but also fine-tunes the solution through various mutation and crossover strategies, so that it can quickly approach the optimal solution while maintaining diversity. This multivariate operation has significant advantages in dynamic environments or complex optimization problems.

[0187] When using the algorithm of this invention for optimization, attention should be paid to the population size. , maximum number of iterations , crossover probability and mutation probability Parameter selection, different parameters will result in different search results, such as Figure 1 The parameter results are shown in Table 4:

[0188]

[0189] The parameter settings depend on the scale of the problem, as well as the parameter settings of the generator set and the tram.

[0190] S5: Through optimization and solution, the power generation of the power generation unit at each moment and the number of plug-in vehicles connected to the power grid are obtained.

Claims

1. A method for optimizing the dispatch of dynamic economic and emission-based power generation units for plug-in electric vehicles, characterized in that: The following steps are involved: S1: Collect the electricity demand at each time period of the day, the maximum power generation of each power generation unit, the minimum power generation of each power generation unit, and the number of plug-in vehicles that can be connected to the power grid, and use the electricity demand at each time period and the number of plug-in vehicles that can be connected to the power grid as the input parameters of the model; S2: Taking the lowest cost and lowest emission generated by the power generation unit and plug-in vehicles as the optimization objectives, the optimal scheduling model of the power generation unit is constructed by constraining the power generation of the power generation unit and the charging and discharging power of the plug-in vehicles; S3: The peak-shaving and valley-filling algorithm is used to distribute the charging and discharging of plug-in vehicles. The charging and discharging behaviors of plug-in vehicles are distributed to different time periods according to the power demand during peak and valley periods. S4: Optimizing the power generation of the power generation unit using a multi-objective genetic algorithm; S5: Obtain the power generation of the power generation unit at each moment and the number of plug-in vehicles connected to the power grid through optimization solution; In step S2, the objective functions of the power generation unit optimization scheduling model are two: one is to minimize the cost, and the other is to minimize the emission: ； in, represents the emissions from the power generation unit, represents the cost of the power generation unit, Represents the cost of charging and discharging a plug-in vehicle; The emissions from the power generation unit are: ； In the formula, is the total power generation time of the power generation unit, Both are power generation units The emission coefficient is, N is the number of power generation units, It is a power generation unit At the moment of power generation; The cost of the power generation unit is: ； In the formula, It is a power generation unit The cost factor, and It is a power generation unit The valve point effect coefficient, For power generation unit The minimum power generation capacity, Indicates power generation unit power generation efficiency; The cost of charging and discharging a plug-in vehicle is: ； In the formula, is the number of plug-in vehicles participating in the grid, It is Battery cost per kilowatt-hour of power consumption for a plug-in car battery, It is The battery capacity of a plug-in car, is the labor cost of replacing the battery, Indicates The discharge capacity of a plug-in car, DOD is the depth of discharge, It is the battery life at a set depth of discharge.

2. The method for dynamic economic and emission optimization scheduling in combination with a plug-in vehicle power generation unit according to claim 1, characterized in that: In step S2, the constraints include: Power generation constraints: ； In the formula, For power generation unit The minimum power generation capacity, For power generation unit Maximum power generation capacity; Constraints on plug-in vehicle battery capacity: ； In the formula, Indicates the minimum battery charge of a plug-in vehicle. Indicates the maximum battery capacity of a plug-in vehicle. For the Plug-in cars at the time Battery capacity; Constraints on charging and discharging of plug-in vehicles: ； In the formula, It is Plug-in cars at the time The charge capacity, It is Plug-in cars at the time The discharge capacity, and They are The maximum charge and discharge capacity of a plug-in vehicle; Power generation constraints after the power generation unit is combined with the plug-in vehicle: ； In the formula, and Plug-in vehicles at the time The discharge power and charging power of It's time The power load demand, represents circuit loss; Determine the circuit loss as follows: ； In the formula, For connecting busbar and busbar No. The admittance of the line, is the number of lines in the system, , The busbar and busbar The voltage, For bus and busbar The phase angle difference between The voltage constraint of the power generation unit is determined as follows: ； In the formula, Indicates that it is located on the busbar The voltage, and Busbar The minimum and maximum voltage limits, is the total number of buses in the system; The power flow constraint of the line is determined as follows: ； In the formula, For the The power flow of the line, For the Maximum power flow limit for each line; The ramp rate constraint of the power generation unit is determined as follows: ； In the formula, and Power generation unit The upper and lower limits of the climbing slope, For power generation unit The power generation at the previous moment, For power generation unit The amount of electricity generated at the current moment.

3. The method for dynamic economic and emission optimization scheduling in combination with a plug-in vehicle power generation unit according to claim 2, characterized in that: The specific steps of step S3 are: S31: Initialize parameters; divide a day into peak hours, normal hours, and off-peak hours according to the 24-hour power load demand, and determine the available plug-in vehicles, peak hours, and off-peak hours power demand and restriction conditions; S32: Allocate remaining resources; according to the power demand, give priority to allowing plug-in vehicles to discharge during peak hours and feed the power back to the power grid. When the plug-in vehicles discharge to the power grid during peak hours, the battery state of charge must be kept within the set range. During off-peak hours, plug-in vehicles will be given priority to be charged. The number of plug-in vehicles is allocated to peak and off-peak periods, and the initial allocation is an average allocation; S33: Calculate the remaining resources; after allocating the charging and discharging resources of the plug-in vehicles during the peak period and the off-peak period, calculate the remaining unallocated plug-in vehicles. If some plug-in vehicles cannot be fully discharged during the peak period or are not fully charged during the off-peak period, these remaining resources are allocated to other time periods through a uniform allocation strategy to ensure the balance of power load in each time period; S34: Evenly distribute the remaining resources; through the process of peak shaving and valley filling, the remaining plug-in vehicles are evenly distributed to the peak and valley periods as needed, with the periods with the least demand being filled first, until the resources are exhausted or all demand is met.

4. The method for dynamic economic and emission optimization scheduling in combination with a plug-in vehicle power generation unit according to claim 2, characterized in that: The specific steps of step S4 are: S41: defining an optimization objective function, the objective function includes two aspects: the minimum cost and the minimum emission generated by the power generation unit and the plug-in vehicle, the power generation of the power generation unit and the number of plug-in vehicles connected are used as optimization variables; S42: Population initialization: Initialize various parameters of the genetic algorithm, including the population size , maximum number of iterations , crossover probability and mutation probability, and initialize the first generation population, each individual in the population represents a power generation allocation scheme and plug-in vehicle access strategy; S43: Fitness evaluation: Perform fitness evaluation on the population, calculate the two objective function values ​​of each individual, use non-dominated sorting to divide the population into different fronts, and calculate the crowding distance of each individual to measure the diversity of individuals in the same front; S44: Parent selection: Use tournament selection to select parents based on non-dominated sorting and crowding distance to form a mating pool; S45: Generate offspring: Perform crossover and mutation operations on the parent generation in the mating pool to generate new offspring; S46: Constraint check: perform constraint check on the generated children to ensure that the power generation allocation scheme and plug-in vehicle access strategy of each child conform to the actual physical constraints; S47: Population merging: merge the current population with the newly generated offspring, calculate the objective function value again, and perform non-dominated sorting and crowding distance calculations; S48: Select a new population: Select the top population from the merged population based on the non-dominated sorting result and crowding distance. Individuals ensure the diversity of the next generation population and the approach to the Pareto frontier; S49: Iterative update: Repeat steps S43 to S48 until the preset maximum number of iterations is reached. , gradually optimize the non-dominated solutions of the population; S410: Output Pareto optimal solution: Output the final Pareto optimal solution set, that is, the individuals ranked 1 in the non-dominated sorting, as the final solution, representing the optimal power generation allocation plan of the power generation unit and the optimal access strategy of the plug-in vehicle.

5. The method for dynamic economic and emission optimization scheduling in combination with a plug-in vehicle power generation unit according to claim 4, characterized in that: In step S45, crossover uses simulated binary crossover or two-point crossover to generate diversified offspring, and mutation uses a combination of polynomial mutation and non-uniform mutation to ensure population diversity and gradually approach the optimal solution; a probabilistic selection method is used to select the type of crossover and mutation, and the probability Determined by the following formula: ； In the formula Indicates the current iteration number, Indicates the maximum number of iterations, randomly generating a random number between 0 and 1 ,if < , then simulated binary crossover and non-uniform mutation are used; if , a combination of two-point crossover and polynomial mutation is used.

6. The method for dynamic economic and emission optimization scheduling in combination with a plug-in vehicle power generation unit according to claim 4, characterized in that: In step S45, the simulated binary crossover is a crossover method for generating diversified offspring by adjusting the crossover coefficient. Control the crossover effect, the formula is as follows: ； in, Determined by the following formula: ； In the formula, is the distribution index that controls the crossover strength, is a randomly generated probability value, , is the parent individual used for crossover operation, , They are the two offspring after the crossover operation.

7. The method for dynamic economic and emission optimization scheduling in combination with a plug-in vehicle power generation unit according to claim 4, characterized in that: In step S45, two-point crossover is used to generate diversified offspring from the parent generation, and new offspring are generated by selecting two crossover points and exchanging the intermediate gene fragments. The formula is as follows: ； ； In the formula, Represents the chromosome genes, is the parent individual used for crossover operation, , Respectively Middle The value of genes; the chromosome of the parent individual contains D genes; and are two randomly generated cross position indices, where ; , are the two offspring after the crossover operation, , Respectively , Middle The value of a gene.

8. The method for dynamic economic and emission optimization scheduling in combination with a plug-in vehicle power generation unit according to claim 4, characterized in that: In step S45, polynomial mutation increases population diversity by introducing small mutations in the solution space. The mutation effect is determined by the mutation amplitude. Determine, the formula is as follows: ； in, Determined by the following formula: ； In the formula, Indicates the first The value of each gene; and Respectively represent The minimum and maximum values ​​allowed for each gene are used to limit the value range of the gene; is the variation range, which is used to control the variation range of genes. The value of is calculated based on the random number and the distribution index; is a randomly generated probability value in the range of [0, 1], which is used to determine the direction and amplitude of mutation; is the multinomial distribution exponent, controlling the strength and range of variation.

9. The method for dynamic economic and emission optimization scheduling in combination with a plug-in vehicle power generation unit according to claim 4, characterized in that: In step S45, the non-uniform mutation dynamically adjusts the mutation amplitude according to the number of iterations. , in order to enhance convergence and gradually approach the optimal solution, the formula is as follows: ； In the formula, Indicates the first The value of each gene; and Respectively represent The minimum and maximum values ​​allowed for each gene; Indicates the current iteration number, Indicates the maximum number of iterations; is a randomly generated probability value, ranging from [0, 1]; is a random number in [0,1]; is the non-uniformity coefficient, which is used to control the variation amplitude to gradually decrease with the number of iterations.

Images