A method for optimizing the ratio of wind, solar, diesel and storage capacity based on stochastic programming

Abstract

A method for optimizing the ratio of wind, solar, diesel and storage capacity based on random programming includes the following steps: Step 1: Establishing a wind, solar, diesel and storage output model; Step 2: Establishing a technical and economic model; Step 3: Establishing wind, solar, diesel and storage system constraints; Step 4: Establishing a model with the objective functions of minimum comprehensive cost and minimum load power shortage rate; Step 5: Using a genetic algorithm based on random simulation to solve the model. In the established random programming model, the foundation constraint model is used to fully consider the uncertainty of wind and solar output. Compared with the deterministic solution method, the genetic algorithm based on random simulation technology is used in the solution method, and the elite operator and elimination operator are introduced to expand the search space and improve the accuracy of the solution result.

Description

Technical Field

[0001] The present invention relates to the field of new energy capacity technology, and in particular to a method for optimizing the ratio of wind, solar, diesel and storage capacity based on random programming. Background Art

[0002] Distributed generation is an emerging technology that uses clean energy to generate electricity. It brings a new prospect for energy application and leads the direction of future electric energy development. Therefore, it has attracted more and more attention from scholars. Compared with traditional centralized power generation such as thermal power and hydropower, distributed generation has the characteristics of small scale, low investment, short construction period, clean and environmentally friendly, reliable power supply and diverse power generation methods. Although the advantages of distributed generation technology are outstanding, with the increase in the penetration rate of distributed generation technology, the problems it brings are also constantly emerging. When distributed power sources are connected to the large power grid or powered separately, the number and capacity of distributed power sources must meet strict requirements to ensure the reliable operation of the power grid; therefore, the optimal configuration of microgrid capacity is an important part of the design of microgrids. It refers to the determination of the capacity of distributed power sources and energy storage devices in the planned microgrid through optimization methods based on natural resources such as wind and light in the planned area and the characteristics of loads, so that the entire microgrid can improve its economy and environmental protection under the premise of ensuring power supply reliability. Scientific and reasonable configuration schemes are of great significance to reducing the construction cost of microgrids, ensuring power supply reliability, improving the level of renewable energy consumption and reducing pollution emissions. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a method for optimizing the capacity ratio of wind, solar, diesel and storage based on random programming, so as to solve the capacity ratio problem of independent power grids during planning and design. The premise of the economic operation of independent power grids is to ensure the load demand, and secondly consider the investment cost and operation cost, so as to reasonably arrange the output of each distributed power source. Based on this, the present invention models the economic operation of independent power grids, constrains the system reliability with the load power shortage rate, considers the uncertainty of the output of new energy, and establishes a multi-objective random programming model to solve the economic operation of independent power grids, aiming to provide new ideas and methods for the early planning of independent power grids.

[0004] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0005] A method for optimizing the ratio of wind, solar, diesel and storage capacity based on stochastic programming, the method comprising the following steps:

[0006] Step 1. Establish a wind, solar, diesel and storage output model;

[0007] Step 2: Establish a technical and economic model;

[0008] Step 3: Establish constraints on the wind, solar, and diesel storage systems;

[0009] Step 4, establish a model with the objective function of minimum comprehensive cost and minimum load power shortage rate;

[0010] Step 5. Use genetic algorithm based on random simulation to solve the model.

[0011] In the above Step 1, the wind, solar, diesel and storage output model is established as follows:

[0012] Step 1.1, the wind power output model is:

[0013] Wind power generation has strong randomness because its output mainly depends on natural wind speed, and wind speed is affected by natural weather and has uncertainty. After obtaining wind speed data for a certain area for one year or longer, it is analyzed that the annual average wind speed in the wind speed area exceeds 4.5m / s and is relatively stable. Therefore, the Rayleigh distribution function is selected from many wind speed simulation methods to describe wind power output, because the Rayleigh distribution function has high requirements for the average wind speed. The Rayleigh distribution function can describe the distribution of wind speed with appropriate accuracy. Wind power output is suitable for the Rayleigh distribution function, as follows:

[0014]

[0015] Where: υ is the measured wind speed, θ are the shape parameters and scale quantities of the wind speed curve. The above formula can be used to calculate the expected value and variance of the wind speed, thereby determining the output power distribution of the wind farm. The specific calculation is as follows:

[0016]

[0017]

[0018] Where: t is the natural wind speed at the moment, υ in , rated and out are the cut-in, rated and cut-out wind speeds of the wind farm respectively;

[0019] Step 1.2, the photovoltaic output model is:

[0020] Photovoltaic power generation has similar characteristics to wind power generation, which mainly depends on the intensity of solar radiation. The beta distribution function can be regarded as a probability distribution of probability. Since the light intensity is random, the beta distribution function gives the probability of all probabilities, which can well describe the randomness of photovoltaic output. The photovoltaic power generation output distribution function is as follows:

[0021]

[0022] Where: θ is the solar radiation intensity, α and β are the shape parameters of the Beta distribution function respectively. The above formula can be used to obtain the expected value u and variance δ of the solar photovoltaic radiation intensity, and the values ​​of the parameters α and β can be further estimated. The specific calculation is as follows:

[0023]

[0024]

[0025] According to the above formula, the solar photovoltaic radiation intensity distribution function can be determined. Combined with the photoelectric conversion formula, the output power of PV at time t can be calculated:

[0026]

[0027]

[0028] Where: η PV and S PV are the solar radiation area and radiation efficiency, θ t represents the solar radiation intensity at time t;

[0029] Step 1.3, diesel generator output model is:

[0030] The power change of diesel generators is flexible and can vary between 0 and rated power. The fuel consumption is a curve that slopes upward with no-load fuel consumption as the starting point. When wind and solar power generation cannot meet the load requirements or there is a short-term load increase, the diesel generator needs to generate power to meet the load requirements. The energy consumption characteristics can be used to represent the power of the diesel generator:

[0031] V hy (t) = cP sc (t)+dP ce (t)

[0032] Where: V hy (t) represents the fuel consumption of the diesel generator, in L, c, d represent the fuel curve coefficient, in L / kW·h, P sc (t) represents the actual power of the diesel generator, P ce (t) represents the rated output power of the generator.

[0033] In the above Step 2, the technical economic model established is:

[0034] Step 2.1, load power failure rate is:

[0035] The load power shortage rate indicates the probability that the system power generation cannot meet the load demand within a certain period of time. It can be used as an indicator to measure the reliability of the power grid. When the load power shortage rate is 0, it means that the power generated by the independent power grid system can meet the load demand within the time period T. When the load power shortage rate is 1, the independent power grid system cannot meet the load demand, resulting in power outages or load shedding, affecting the safety and reliability of the system. The load power shortage rate within the evaluation period T can be expressed as the ratio of the load power shortage amount to the total load demand within the period. The independent power grid system load power shortage rate expression method is:

[0036]

[0037] Where: P FH (t) is the system load demand at time t, P PV (t)η1 is the average power emitted by the photovoltaic module at time t, P PW (t)η2 is the average power generated by wind power, P CY (t) represents the average power generated by the diesel generator, P CN (t) represents the average power generated by the energy storage, η1 is the efficiency of inverter C1, and η2 is the efficiency of inverter C2;

[0038] Load power shortage Q QD (t) is a positive value, and the load power failure rate Q is defined QDL for:

[0039]

[0040] Where: t0 is the initial time; n is the time series; Q QDL The value range of is [0-1];

[0041] Step 2.2, initial investment cost and operating cost are:

[0042]

[0043]

[0044]

[0045] In the formula: i represents four different types of power sources, namely wind power, photovoltaic power, energy storage battery, and diesel generator, N = 4, C gd represents the fixed investment cost of wind power, photovoltaic power, energy storage, and diesel generators, C csi It is expressed as the initial investment cost of different types of power sources. For energy storage batteries, it is mainly related to capacity and power, while for wind power and photovoltaic power, it is mainly related to rated power. yx It is expressed as the operation and management costs of different types of power sources. Wind power and photovoltaic power are mainly related to the actual operating power, Pit represents the actual operating power of each power supply at time t, C hs (l,r) represents the capital recovery factor, r represents the interest rate, and l represents the service life of the power supply;

[0046] C csj =f(P EDj )=K j P EDj

[0047] C INIbat =f(P BD ,E BD )=K bp P BD +K ep E BD

[0048] Where: j represents wind power, photovoltaic power, energy storage battery and diesel generator respectively, C csj Expressed as the initial investment cost of different types of power supplies, P EDj Expressed as the rated power of different power supplies, kW; K j Expressed as the power cost coefficient of different power sources, yuan / kW; P BD Expressed as the rated power of the energy storage battery, kW; E BD Expressed as the rated capacity of the energy storage battery, kW·h; K bp Expressed as the power cost coefficient of energy storage battery, RMB / kW; K ep Expressed as the energy cost coefficient of the energy storage battery, yuan / kW·h;

[0049] C yxj (P j (t)) = k yxxj P j (t)

[0050] Where: j represents wind power, photovoltaic power, energy storage battery and diesel generator, C yxj Represents the operation and management costs of different power types, P j (t) represents the actual operating power of different power supplies, kW; k yxxj Represents the operation and management cost coefficient of different power sources, RMB / kW·h.

[0051] Step 2.3: Penalty costs for abandoning new energy

[0052] When the load power shortage is negative, there will be a surplus of electricity. This excess electricity may be wasted. In order to prevent energy waste caused by over-configuration of system capacity during capacity planning, energy waste penalty costs must be introduced:

[0053]

[0054] Where: w , v The penalty costs for abandoning wind and solar power in period t respectively; Indicates the abandoned wind power of the wind farm and the abandoned light power of the photovoltaic power station at time t;

[0055] Step 2.4. Revenue from subsidies for new energy power generation

[0056] Traditional thermal power generation has high safety and stability. Compared with it, renewable energy power generation does not have much advantage because it has the defects of randomness and instability. However, it has the characteristics of being highly clean and pollution-free. In addition, the country supports the development of renewable energy, so the country will give certain power generation subsidies to renewable energy power generation.

[0057] The subsidy cost can be expressed as:

[0058]

[0059] Where: C bt is the power generation subsidy cost, P bt is the subsidy electricity price yuan / kW·h; P bt (t) represents the power generated by renewable energy, kW;

[0060] Step 2.5: Pollution control cost model

[0061] Diesel generators will generate pollution during the power generation process. The pollutants emitted include carbon dioxide, nitrogen oxides, sulfur dioxide, etc. The pollution control costs are:

[0062]

[0063] Where: C zw represents the pollution control cost, k represents the number of different pollutant types; m is the type of pollutant discharged, α k Indicates the treatment cost coefficient of different pollutants, yuan / kg; β k Indicates the emission coefficient of different pollutants, kg / kW·h.

[0064] In the above Step 3, the constraint model of the wind-solar-diesel-storage system is established as follows:

[0065] Step 3.1. Wind and solar power output constraints

[0066]

[0067] Where: To contribute to wind power, is the photovoltaic output, S w represents the actual installed capacity of wind power, S vIndicates the actual installed capacity of photovoltaic power;

[0068] Step 3.2, load balancing constraints

[0069]

[0070] Where: Indicates the output power of the diesel generator. To contribute to wind power, For photovoltaic power generation, To discharge the energy storage, is the load sequence, charging energy storage;

[0071] Step 3.3: Wind and solar power curtailment constraints

[0072]

[0073] Where: is the total amount of wind power abandoned, is the total amount of abandoned solar power, α is the wind abandonment ratio, β is the abandoned solar power ratio, is the installed capacity of wind power at time t, is the installed capacity of photovoltaic power at time t;

[0074] Step 3.4: Electric energy storage constraints

[0075] (1) When there is excess power generation, the battery energy storage system is charged. If the charging power and the remaining power are within the battery's constraints, the battery is charged at the maximum power. When the charging power or the remaining power exceeds the limit, the battery energy storage system is charged at the maximum allowable power, and the rest is consumed by the standby load to avoid wasting energy;

[0076] (2) If the power generation capacity is insufficient, the power shortage will be provided by the battery energy storage system first. If the battery energy storage system alone can meet the load demand, the energy storage battery will discharge at maximum power to meet the load demand. If the discharge power or the remaining power exceeds the limit, the battery energy storage system will discharge at the maximum allowable power, and the rest will be supplemented by the diesel generator within its rated power range;

[0077]

[0078]

[0079]

[0080] Where: SOC d,t is the current capacity of the electric energy storage, E max Maximum installed capacity of electric energy storage, is the energy storage power constraint, The discharge rate of the energy storage, Energy storage charging rate, The state of discharge of electrical energy storage, Energy storage charging status, The charging capacity state of the energy storage at the previous moment, The discharge capacity state of the energy storage at the previous moment, SOC d,t-1 is the capacity of the energy storage at the previous moment, It is the state of energy storage release. It is the energy storage charging state. It is a 0-1 variable to be optimized.

[0081] In the above Step 4, the stochastic planning model of wind, solar, diesel and energy storage is established as follows:

[0082] Since the wind-solar hybrid system is random and unstable, the economic operation of the system is improved under the premise of ensuring system safety. The load power shortage rate indicates the probability that the system power generation cannot meet the load demand within a certain period of time. The load power shortage rate within the evaluation period T can be expressed as the ratio of the load power shortage to the total load demand during this period, which is used as an evaluation indicator to measure system safety.

[0083] When planning an independent microgrid, the capacity of each wind, solar, diesel and storage unit is not determined in advance, so the capacity of each unit is often regarded as a random variable and usually has multidimensionality, diversity and multi-criteria. For different economic and technical requirements, the solution method is also different. The method adopted in this paper is chance constrained planning, which is mainly aimed at the situation where the constraints contain random variables and the decision must be made before the random variables are observed to be realized. Considering that the decision may not meet the constraints when unfavorable situations occur, a principle is adopted: that is, the decision is allowed to not meet the constraints to a certain extent, but the decision should make the probability of the constraints being established not less than a certain confidence level β. Based on this, this paper selects the capacity of each wind, solar, diesel and storage unit as the decision variable for multi-objective optimization, and the objective function as the fitness function. Under certain constraints, the optimal wind, solar and storage combination is obtained to minimize the objective function value. F1 is selected as the lowest system load power shortage rate, and F2 is selected as the lowest system comprehensive cost:

[0084]

[0085] minF2=C zong +F qy +C jh +C bt -C zw

[0086]

[0087] Where: β represents the confidence level of the chance constraint given in advance.

[0088] In the above Step 5, when solving the optimal capacity configuration of energy storage, the following steps are adopted:

[0089] Step 5.1. Input data such as wind speed and light intensity, and obtain the wind and solar output value through the probability model;

[0090] Step 5.2, set the system parameters and encode the output of each unit using real number encoding;

[0091] Step 5.3, based on active power balance, capacity constraints of each unit, wind and solar curtailment constraints, and power exchange constraints, obtain the capacity combination range of each wind, solar and storage unit;

[0092] Step 5.4, determine the optimal wind, solar and storage capacity combination within the time period t;

[0093] Step 5.5, within the time period t, an initial population P0 of size N will be randomly generated, during which all individuals will be quickly non-dominated sorted, and genetic operations will be performed through selection, crossover and mutation. Thus, a child population Q0 of size N will be generated, and then elite selection and retention will be performed, P0 and Q0 will be merged into a population of 2N, and then R0 will be non-dominated sorted to obtain a non-dominated individual set, so as to select and generate the parent population P1, and determine whether the population is still changing. If it is still changing, continue the algorithm, otherwise the algorithm ends and outputs the optimal population;

[0094] Step 5.6, according to the ideal point method, min||F(x)-F * The feasible solution of || is regarded as a satisfactory solution, and in this paper, the modulus is defined as:

[0095]

[0096] Among them, F1 * , They are the minimum values ​​of the objective functions F1 and F2 in this paper respectively. The minimum F1 is defined as the optimal solution with the highest reliability, and the minimum F2 is defined as the optimal solution with the highest economic efficiency.

[0097] The present invention provides a method for optimizing the ratio of wind, solar, diesel and storage capacity based on random programming. Compared with the prior art, the present invention has the following technical effects:

[0098] In the established stochastic programming model, the foundation constraint model is used to fully consider the uncertainty of wind and solar power output. Compared with the deterministic solution method, it is more in line with reality. In the solution method, a genetic algorithm based on random simulation technology is adopted, and elite operators and elimination operators are introduced to expand the search space and improve the accuracy of the solution results. BRIEF DESCRIPTION OF THE DRAWINGS

[0099] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:

[0100] Figure 1 is the normalized wind speed curve;

[0101] Figure 2 is the normalized light intensity curve;

[0102] Figure 3 It is the result of system capacity ratio; DETAILED DESCRIPTION

[0103] In order to make the purpose, technical solutions and advantages of the present invention more clear, the following content will systematically and completely describe the specific technical solutions of the present invention in conjunction with the accompanying drawings provided according to the present invention.

[0104] A method for optimizing the ratio of wind, solar, diesel and storage capacity based on stochastic programming, the method comprising the following steps:

[0105] Step 1. Establish a wind, solar, diesel and storage output model;

[0106] Step 2: Establish a technical and economic model;

[0107] Step 3: Establish constraints on the wind, solar, and diesel storage systems;

[0108] Step 4, establish a model with the objective function of minimum comprehensive cost and minimum load power shortage rate;

[0109] Step 5. Use genetic algorithm based on random simulation to solve the model.

[0110] In the above Step 1, the wind-solar-diesel-storage output model is established as follows:

[0111] Step 1.1, the wind power output model is:

[0112] Wind power generation has strong randomness because its output mainly depends on natural wind speed, which is affected by natural weather and has uncertainty. After obtaining wind speed data for a year or longer in a certain area, the Rayleigh distribution function can describe the distribution of wind speed with appropriate accuracy. Wind power output is suitable for the Rayleigh distribution function, as follows:

[0113]

[0114] Where: υ is the measured wind speed, θ are the shape parameters and scale quantities of the wind speed curve. The above formula can be used to calculate the expected value and variance of the wind speed, thereby determining the output power distribution of the wind farm. The specific calculation is as follows:

[0115]

[0116]

[0117] Where: t is the natural wind speed at the moment, υ in , rated and out are the cut-in, rated and cut-out wind speeds of the wind farm respectively;

[0118] Step 1.2, the photovoltaic output model is:

[0119] Photovoltaic power generation has similar characteristics to wind power generation. It mainly depends on the intensity of solar radiation. The beta distribution function can be regarded as a probability distribution of probability. Since the light intensity is random, the beta distribution function gives the probability of all probabilities, which can well describe the randomness of photovoltaic output. The photovoltaic power generation output distribution function is as follows:

[0120]

[0121] Where: θ is the solar radiation intensity, α and β are the shape parameters of the Beta distribution function respectively. The above formula can be used to obtain the expected value u and variance δ of the solar photovoltaic radiation intensity, and the values ​​of the parameters α and β can be further estimated. The specific calculation is as follows:

[0122]

[0123]

[0124] According to the above formula, the solar photovoltaic radiation intensity distribution function can be determined. Combined with the photoelectric conversion formula, the output power of PV at time t can be calculated:

[0125]

[0126]

[0127] Where: η PV and S PV are the solar radiation area and radiation efficiency, θ t represents the solar radiation intensity at time t;

[0128] Step 1.3, diesel generator output model is:

[0129] The power change of diesel generators is flexible and can vary between 0 and rated power. When wind and solar power generation cannot meet the load requirements or there is a short-term load increase, the diesel generator needs to generate power to meet the load requirements. The energy consumption characteristics can be used to represent the power of the diesel generator:

[0130] V hy (t) = cP sc (t)+dP ce (t)

[0131] Where: V hy (t) represents the fuel consumption of the diesel generator, the unit is L, c, d represent the fuel curve coefficient, the unit is L / kW·h.

[0132] In the above Step 2, the technical economic model established is:

[0133] Step 2.1, load power failure rate is:

[0134] The load power shortage rate indicates the probability that the system power generation cannot meet the load demand within a certain period of time. The load power shortage rate within the evaluation period T can be expressed as the ratio of the load power shortage amount to the total load demand during the period. The load power shortage rate of the independent power grid system is expressed as follows:

[0135]

[0136] Where: P FH (t) is the system load demand at time t, P PV (t)η1 is the average power emitted by the photovoltaic module at time t, P PW (t)η2 is the average power generated by wind power, P CY (t) represents the average power generated by the diesel generator, P CN (t) represents the average power generated by the energy storage, η1 is the efficiency of inverter C1, and η2 is the efficiency of inverter C2;

[0137] Load power shortage Q QD (t) is a positive value, and the load power failure rate Q is defined QDL for:

[0138]

[0139] Where: t0 is the initial time; n is the time series; Q QDL The value range of is [0-1]. When it is 0, it means that the power generated by the independent power grid system can meet the load demand in the time period T. When it is 1, the independent power grid system cannot meet the load demand.

[0140] Step 2.2, initial investment cost and operating cost are:

[0141]

[0142]

[0143]

[0144] In the formula: i represents four different types of power sources, namely wind power, photovoltaic power, energy storage battery, and diesel generator, N = 4, C gd represents the fixed investment cost of wind power, photovoltaic power, energy storage, and diesel generators, C csi It is expressed as the initial investment cost of different types of power sources. For energy storage batteries, it is mainly related to capacity and power, while for wind power and photovoltaic power, it is mainly related to rated power. yx It is expressed as the operation and management costs of different types of power sources. Wind power and photovoltaic power are mainly related to the actual operating power, P it represents the actual operating power of each power supply at time t, C hs (l,r) represents the capital recovery factor, r represents the interest rate, and l represents the service life of the power supply;

[0145] C csj =f(P EDj )=K j P EDj

[0146] C INIbat =f(P BD ,E BD )=K bp P BD +K ep E BD

[0147] Where: j represents wind power, photovoltaic power, energy storage battery and diesel generator respectively, C csj Expressed as the initial investment cost of different types of power supplies, P EDj Expressed as the rated power of different power supplies, kW; K j Expressed as the power cost coefficient of different power sources, yuan / kW; P BD Expressed as the rated power of the energy storage battery, kW; E BD Expressed as the rated capacity of the energy storage battery, kW·h; K bp Expressed as the power cost coefficient of energy storage battery, RMB / kW; K ep Expressed as the energy cost coefficient of the energy storage battery, yuan / kW·h;

[0148] C yxj (P j (t)) = k yxxj P j (t)

[0149] Where: j represents wind power, photovoltaic power, energy storage battery and diesel generator, C yxj Represents the operation and management costs of different power types, P j (t) represents the actual operating power of different power supplies, kW; k yxxjRepresents the operation and management cost coefficient of different power sources, RMB / kW·h.

[0150] Step 2.3: Penalty costs for abandoning new energy

[0151]

[0152] Where: w , v The penalty costs for abandoning wind and solar power in period t respectively; Indicates the abandoned wind power of the wind farm and the abandoned light power of the photovoltaic power station at time t;

[0153] Step 2.4. Revenue from subsidies for new energy power generation

[0154] Traditional thermal power generation has high safety and stability. Compared with it, renewable energy power generation does not have much advantage because it has the defects of randomness and instability. However, it has the characteristics of being highly clean and pollution-free. In addition, the country supports the development of renewable energy, so the country will give certain power generation subsidies to renewable energy power generation.

[0155] The subsidy cost can be expressed as:

[0156]

[0157] Where: C bt is the power generation subsidy cost, P bt is the subsidy electricity price yuan / kW·h; P bt (t) represents the power generated by renewable energy, kW;

[0158] Step 2.5: Pollution control cost model

[0159] Diesel generators will generate pollution during the power generation process. The pollutants emitted include carbon dioxide, nitrogen oxides, sulfur dioxide, etc. The pollution control costs are:

[0160]

[0161] Where: C zw represents the pollution control cost, k represents the number of different pollutant types; m is the type of pollutant discharged, α k Indicates the treatment cost coefficient of different pollutants, yuan / kg; β k Indicates the emission coefficient of different pollutants, kg / kW·h.

[0162] In the above Step 3, the constraint model of the wind-solar-diesel-storage system is established as follows:

[0163] Step 3.1. Wind and solar power output constraints

[0164]

[0165] Where: To contribute to wind power, is the photovoltaic output, S w represents the actual installed capacity of wind power, S v Indicates the actual installed capacity of photovoltaic power;

[0166] Step 3.2, load balancing constraints

[0167]

[0168] Where: Indicates the output power of the diesel generator. To contribute to wind power, For photovoltaic power generation, To discharge the energy storage, is the load sequence, Charging energy storage;

[0169] Step 3.3: Wind and solar power curtailment constraints

[0170]

[0171] Where: is the total amount of wind power abandoned, is the total amount of abandoned solar power, α is the wind abandonment ratio, β is the abandoned solar power ratio, is the installed capacity of wind power at time t, is the installed capacity of photovoltaic power at time t;

[0172] Step 3.4: Electric energy storage constraints

[0173]

[0174]

[0175]

[0176] Where: SOC d,t is the current capacity of the electric energy storage, E max Maximum installed capacity of electric energy storage, is the energy storage power constraint, The discharge rate of the energy storage, Energy storage charging rate, The state of discharge of electrical energy storage, Energy storage charging status, The charging capacity state of the energy storage at the previous moment, The discharge capacity state of the energy storage at the previous moment, SOC d,t-1 is the capacity of the energy storage at the previous moment, It is the state of energy storage release. It is the energy storage charging state. It is a 0-1 variable to be optimized.

[0177] In the above Step 4, the stochastic planning model of wind, solar, diesel and energy storage is established as follows:

[0178] Since the wind-solar hybrid system is random and unstable, the economic operation of the system is improved under the premise of ensuring system safety. The load power shortage rate indicates the probability that the system power generation cannot meet the load demand within a certain period of time. The load power shortage rate within the evaluation period T can be expressed as the ratio of the load power shortage to the total load demand during this period, which is used as an evaluation indicator to measure system safety.

[0179] Select F1 as the system with the lowest load power shortage rate, and select F2 as the system with the lowest comprehensive cost:

[0180]

[0181] minF2=C zong +F qy +C jh +C bt -C zw

[0182]

[0183] Where: β represents the confidence level of the chance constraint given in advance.

[0184] In the above Step 5, when solving the optimal capacity configuration of energy storage, the following steps are adopted:

[0185] Step 5.1. Input data such as wind speed and light intensity, and obtain the wind and solar power output value through the probability model;

[0186] Step 5.2, set the system parameters and encode the output of each unit using real number encoding;

[0187] Step 5.3, based on active power balance, capacity constraints of each unit, wind and solar curtailment constraints, and power exchange constraints, obtain the capacity combination range of each wind, solar and storage unit;

[0188] Step 5.4, determine the optimal wind, solar and storage capacity combination within the time period t;

[0189] Step 5.5, within the time period t, an initial population P0 of size N will be randomly generated, during which all individuals will be quickly non-dominated sorted, and genetic operations will be performed through selection, crossover and mutation. Thus, a child population Q0 of size N will be generated, and then elite selection and retention will be performed, P0 and Q0 will be merged into a population of 2N, and then R0 will be non-dominated sorted to obtain a non-dominated individual set, so as to select and generate the parent population P1, and determine whether the population is still changing. If it is still changing, continue the algorithm, otherwise the algorithm ends and outputs the optimal population;

[0190] Step 5.6, according to the ideal point method, min||F(x)-F * ||A feasible solution is a satisfactory solution, and in this paper, the modulus is defined as:

[0191]

[0192] Among them, F1 * , They are the minimum values ​​of the objective functions F1 and F2 in this paper respectively. The minimum F1 is defined as the optimal solution with the highest reliability, and the minimum F2 is defined as the optimal solution with the highest economic efficiency.

[0193] According to the wind and solar output model established above, the improved NSGA2 is used to solve the model. The algorithm sets the population size W = 50, the maximum evolutionary generation gen = 100, the number of random simulations of chance constraints is 300 times, and the confidence level is 0.9. Figure 1 , Figure 2 The following are the normalized curves of wind speed and light intensity. Table 1 and Table 2 are the technical parameters of the system. The optimization results are shown in Figure 3 As shown in the figure, the economic optimal solution, comprehensive optimal solution and safety optimal solution are as follows Figure 3 It is marked in the middle that the load power shortage rate of the economic optimal solution is about 2.7%, and the lowest investment and operating cost is 56,000 yuan at this time. The load power shortage rate of the optimal solution for safety is 0.1%, which can ensure the safe and reliable operation of the system, but at the expense of economy. For the wind, solar, diesel and storage system, the load power shortage rate is generally required to be below 0.5% to ensure the safety of the system. In the comprehensive optimal solution, the load power shortage rate is 0.5%, and the investment and operating cost is about 113,000 yuan. Compared with the economic optimal solution, the cost has doubled, but the safety has increased by 5 times; compared with the safety optimal solution, the safety has decreased by 2 times, but the economy has increased by 3 times, and it can operate safely and stably within the safety and reliability requirements; in summary, the comprehensive optimal solution can be selected as the optimal ratio scheme of wind, solar, diesel and storage, among which the capacity ratio of wind, solar and storage is 1:0.8:2.05.

[0194] Table 1 Technical parameters of wind, solar and diesel storage

[0195]

[0196] Table 2 Pollution control cost parameters

[0197]

Claims

1. A method for optimizing the ratio of wind, solar, diesel and storage capacity based on stochastic programming, characterized in that: The steps of the method include: Step 1. Establish a wind, solar, diesel and storage output model; Step 2: Establish a technical and economic model; Step 3: Establish constraints on the wind, solar, and diesel storage systems; The established wind-solar-diesel-storage system constraint model is as follows: Step 3.

1. Wind and solar power output constraints ； ； Where: To contribute to wind power, For photovoltaic power generation, represents the actual installed capacity of wind power, Indicates the actual installed capacity of photovoltaic power; Step 3.2, load balancing constraints ； Where: Indicates the output power of the diesel generator. To contribute to wind power, For photovoltaic power generation, To discharge the energy storage, is the load sequence, charging energy storage; Step 3.3: Wind and solar power curtailment constraints ； Where: is the total amount of wind power abandoned, is the total amount of abandoned light, is the wind abandonment ratio, For the ratio of abandoned light, is the installed capacity of wind power at time t, is the installed capacity of photovoltaic power at time t; Step 3.4: Electric energy storage constraints ； ； ； ； ； Where: is the current capacity of the electric energy storage, Maximum installed capacity of electric energy storage, is the energy storage power constraint, The discharge rate of the energy storage, Energy storage charging rate, The state of discharge of electrical energy storage, Energy storage charging status, The charging capacity state of the energy storage at the previous moment, The discharge capacity state of the energy storage at the previous moment, is the capacity of the energy storage at the previous moment, It is the state of energy storage release. It is the energy storage charging state. , It is a 0-1 variable to be optimized; Step 4, establish a model with the objective function of minimum comprehensive cost and minimum load power shortage rate; The established stochastic planning model of wind, solar, diesel and energy storage is as follows: Since the wind-solar hybrid system is random and unstable, the economic operation of the system is improved under the premise of ensuring system safety. The load power shortage rate indicates the probability that the system power generation cannot meet the load demand within a certain period of time. The load power shortage rate within the evaluation period T can be expressed as the ratio of the load power shortage to the total load demand during this period, which is used as an evaluation indicator to measure system safety. Select To minimize the power shortage rate of system load, select To minimize the overall system cost: ； ； Where: represents the confidence level of the chance constraint given in advance; Step 5: Use the genetic algorithm based on random simulation to solve the model. When solving the optimal capacity configuration of energy storage, the following steps are adopted: Step 5.

1. Input wind speed and light data, and obtain wind and solar output values ​​through probability model; Step 5.2, set the system parameters and encode the output of each unit using real number encoding; Step 5.3, based on active power balance, capacity constraints of each unit, wind and solar curtailment constraints, and power exchange constraints, obtain the capacity combination range of each wind, solar and storage unit; Step 5.4, in the time period Determine the optimal wind, solar and storage capacity combination within this period; Step 5.5, in the time period A random number of size The initial population During this period, all individuals are quickly non-dominated sorted, and genetic operations are performed through selection, crossover and mutation; thus, the number of individuals is The offspring population , followed by elite selection and retention, and Merge to a population of 2 population, and then Perform non-dominated sorting to obtain a non-dominated individual set, thereby selecting the parent population , determine whether the population is still changing, if it is still changing, continue the algorithm, otherwise the algorithm ends and outputs the optimal population; Step5.6, according to the ideal point method, min|| ||A feasible solution is a satisfactory solution, and in this paper, the modulus is defined as: ； in , They are the objective functions in this paper , The minimum value of The minimum is the most reliable optimal solution. The minimum is the economically optimal solution.

2. The method for optimizing the ratio of wind, solar, diesel and storage capacity based on stochastic programming according to claim 1 is characterized in that: In the above Step 1, the wind-solar-diesel-storage output model is established as follows: Step 1.1, the wind power output model is: ； Where: is the measured wind speed, , are the shape parameters and scale quantity of the wind speed curve respectively. The above formula can be used to calculate the expected value and variance of the wind speed, thereby determining the output power distribution of the wind farm. The specific calculation is as follows: ； Where: is the natural wind speed at the moment, , and are the cut-in, rated and cut-out wind speeds of the wind farm respectively; Step 1.2, the photovoltaic output model is: The photovoltaic power generation output distribution function is as follows: ； Where: is the solar radiation intensity, and are the shape parameters of the Beta distribution function respectively. The above formula can be used to obtain the expected value of solar photovoltaic radiation intensity. and variance , the parameters can be further estimated and The specific calculation is as follows: ； According to the above formula, the solar photovoltaic radiation intensity distribution function can be determined. Combined with the photoelectric conversion formula, the PV at the time can be calculated. Output power: ； ； Where: and are the solar radiation area and radiation efficiency, Indicates time Solar radiation intensity; Step 1.3, diesel generator output model is: ； Where: Expressed as the fuel consumption of the diesel generator, unit L, , They represent the fuel curve coefficient respectively, with the unit of L / kW·h.

3. The method for optimizing the ratio of wind, solar, diesel and storage capacity based on stochastic programming according to claim 2 is characterized in that: In the aforementioned Step 2, the technical economic model established is: Step 2.1, load power failure rate is: The load power shortage rate indicates the probability that the system power generation cannot meet the load demand within a certain period of time. The load power shortage rate within the evaluation period T can be expressed as the ratio of the load power shortage amount to the total load demand during the period. The load power shortage rate of the independent power grid system is expressed as follows: ； Where: is the system load demand at time t, is the average power emitted by the photovoltaic module at time t, is the average power generated by wind power, Indicates the average power generated by the diesel generator. represents the average power generated by the energy storage, For inverter efficiency, For inverter efficiency; Load power shortage is a positive value, defining the load power failure rate for: ； Where: is the initial moment; n is the time series; The value range is [0-1]. When it is 0, it means that in the time period The power generated by the independent power grid system can meet the load demand. When the independent power grid system cannot meet the load demand; Step 2.2, initial investment cost and operating cost are: ； ； ； Where: Represented as four different types of power sources, namely wind power, photovoltaic power, energy storage battery, and diesel generator. =4, represents the fixed investment cost of wind power, photovoltaic, energy storage, and diesel generators, It is expressed as the initial investment cost of different types of power sources. For energy storage batteries, it is related to capacity and power, while for wind power and photovoltaic power, it is related to rated power. It is expressed as the operation and management costs of different types of power sources. Wind power and photovoltaic power are related to the actual operating power. Indicates that each power supply is The actual operating power at the moment, Expressed as the capital recovery coefficient, Expressed as interest rate, Expressed as the service life of the power supply; ； ； Where: Represent wind power, photovoltaic power, energy storage battery and diesel generator respectively. Expressed as the initial investment cost of different types of power supplies, Expressed as the rated power of different power supplies, kW; Expressed as the power cost coefficient of different power sources, yuan / kW; Expressed as the rated power of the energy storage battery, kW; Expressed as the rated capacity of the energy storage battery, kW·h; Expressed as the energy storage battery power cost coefficient, yuan / kW; Expressed as the energy cost coefficient of the energy storage battery, yuan / kW·h; ； Where: They represent wind power, photovoltaic power, energy storage battery and diesel generator respectively. Represents the operating and management costs of different power types. Represents the actual operating power of different power supplies, kW; Represents the operation and management cost coefficient of different power sources, RMB / kW·h; Step 2.3: Penalty costs for abandoning new energy ； Where: , The penalty costs for abandoning wind and solar power in period t respectively; , Indicates the abandoned wind power of the wind farm and the abandoned light power of the photovoltaic power station at time t; Step 2.

4. Revenue from subsidies for new energy power generation The subsidy cost can be expressed as: ； Where: To subsidize the cost of electricity generation, The subsidized electricity price is RMB / kW·h; Expressed as renewable energy power generation, kW; Step 2.5: Pollution control cost model Diesel generators will generate pollution during the power generation process. The pollutants emitted include carbon dioxide, nitrogen oxides, and sulfur dioxide. The pollution control costs are: ； Where: represents the pollution control costs, Indicates different pollutant type numbers; is the type of pollutants emitted, Indicates the treatment cost coefficient of different pollutants, yuan / kg; Indicates the emission coefficient of different pollutants, kg / kW·h.

Images