Improved particle swarm optimization algorithm-based wind-solar complementary combined optimization scheduling method

Abstract

The application claims a wind-solar complementary combined optimization scheduling method based on an improved particle swarm algorithm, which takes maximizing the wind-solar combined complementary rate and minimizing the abandoned wind and light quantity as the optimization target, takes the energy balance constraint as the basic power constraint in the system, considers the operation economy of the thermal power and the minimum abandoned wind and light quantity as the target, and adds a thermal power unit model to establish the combined optimization scheduling. The method includes using the complementary optimal matching region, the complementary optimal wind-solar capacity ratio and other basic data, and combining the improved particle swarm algorithm to conduct the scheduling research on the combined power generation system considering the wind-solar complementarity. In the application, the wind-solar complementary rate is maximized as the optimization target, which has strong practical significance. Meanwhile, the optimization method considers minimizing the abandoned wind and light quantity on the basis of maximizing the complementary rate, which not only effectively reduces the resource waste, but also realizes the efficient utilization of energy.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of energy planning, and mainly relates to a wind-solar complementary joint optimization scheduling method based on an improved particle swarm algorithm. BACKGROUND

[0002] At present, the global countries are facing a series of development difficulties such as energy shortage and increasingly serious environmental pollution, and the utilization of renewable resources such as light energy and wind energy gradually becomes a research hotspot of people. Among them, solar energy and wind energy are considered to have broad development prospects due to their advantages such as inexhaustible and endless. However, as an intermittent energy, solar energy and wind energy have instability, and the geographical distribution of energy also presents dispersion, so it is of great significance to establish a wind-solar complementary energy optimization method to improve the utilization rate of wind-solar resources.

[0003] Wind-solar complementation refers to the full utilization of resources and uninterrupted energy supply by comprehensively utilizing the complementarity of wind energy resources and solar energy resources in time and region. The wind-solar resources can have certain complementarity because the climate in China presents monsoon characteristics. Related researches have found that in the long term, the wind energy resources are relatively abundant in spring and winter, but the solar energy resources are relatively short at the same time; in summer and autumn, the solar energy resources are relatively abundant, and the wind energy resources are relatively short. If from the short term, in a day, the solar energy resources are relatively abundant in the daytime, which can make up for the shortage of wind energy resources at this time; and the wind energy resources are relatively abundant at night, which can make up for the shortage of solar energy resources. Therefore, the superior climate conditions in China can bring natural wind-solar complementarity, which provides a basis for related wind-solar resource complementary research.

[0004] The basic particle swarm algorithm principle is to simulate the intelligent response of a biological population as a template. Taking bird predation as an example, assume that a group of birds are placed in a designated area and allowed to search for food randomly. Beforehand, all birds do not know the specific location of the food, but they know the distance between themselves and the food. Therefore, in order to find the food, these birds must first search the area closest to the food. In the particle swarm algorithm, each individual is called a particle, and the algorithm relies on information exchange between particles to achieve common evolution within the population. During the solution process, each particle will fly in space at a certain speed, and the process of each particle flying is called the search process of the individual. During the search process of the individual, each particle will fly in space at a certain speed, and the flying speed of the particle can be dynamically adjusted by the individual historical optimum and the population historical optimum. Therefore, the particle itself has two attributes: speed attribute and position attribute. The speed is used to describe the speed of movement, and the position is used to describe the direction of movement. The solution obtained by each particle after its own optimal search is called the individual extreme value, and the optimal individual extreme value in the entire population is the current global optimal solution. Through continuous iteration calculation, the speed and position attributes are updated, and the optimal solution that meets the termination condition is obtained.

[0005] Therefore, the wind and light complementary joint optimization scheduling method based on the improved particle swarm algorithm solves the wind and light resource complementary scheduling target by using the improved particle swarm intelligent optimization algorithm, so as to ensure the speed and accuracy of the solution. This method can make the complementary optimization performance of wind energy and solar energy reach the best, thereby improving the utilization rate of energy.

[0006] CN114204549A, a method for joint optimization and operation of a wind-solar-storage cluster considering energy storage sharing, includes the following steps: establishing a wind-solar-storage cluster including a plurality of new energy power stations and an energy storage sharing aggregator; the resource transaction behavior of the cluster with the external system is carried out through the energy storage sharing aggregator; considering the complementary characteristics of the new energy power stations, establishing a resource scheduling strategy within the wind-solar-storage cluster; based on the resource scheduling strategy within the wind-solar-storage cluster, establishing a revenue model for a single new energy power station within the wind-solar-storage cluster; using a scenario-based stochastic programming method to model the uncertainty of market electricity prices and new energy power station output; establishing a target function considering the revenue model of a single new energy power station, market electricity prices, and new energy power station output, and taking the maximum overall revenue of the wind-solar-storage cluster as the target; solving the target function by using an improved particle swarm algorithm, and outputting the optimal wind-solar-storage cluster joint operation scheme.

[0007] The application discloses a wind-solar complementary joint optimization scheduling method based on an improved particle swarm algorithm. In the scheduling research, the intelligent optimization algorithm selected is the particle swarm algorithm, and the model is solved through the algorithm, which has the advantages of fast search speed, but has the defect of being prone to falling into a local optimal solution. SUMMARY

[0008] The application aims to solve the problems of the prior art. The application discloses a wind-solar complementary joint optimization scheduling method based on an improved particle swarm algorithm.

[0009] The application discloses a wind-solar complementary joint optimization scheduling method based on an improved particle swarm algorithm.

[0010] First, the output model of the wind driven generator and the photovoltaic cell panel is determined according to the wind-solar complementary distributed cooperative operation mode, so as to obtain the output power of the wind driven generator and the photovoltaic cell panel. The application discloses a wind-solar complementary joint optimization scheduling method based on an improved particle swarm algorithm.

[0011] Further, the wind driven generator output model specifically comprises:

[0012] The wind driven generator contains a cut-in wind speed and a cut-out wind speed, and can only generate available power output when running between the cut-in wind speed and the cut-out wind speed, and takes the rated wind speed as a boundary.

[0013]

[0014] In the formula, p r represents the rated power of the wind driven generator, v r is the rated wind speed of the wind driven generator, v in , v outrespectively represent the cut-in wind speed and the cut-out wind speed, v is the real-time wind speed of the selected research area;

[0015] The photovoltaic power generation output model specifically comprises:

[0016] The photovoltaic power generation utilizes the photovoltaic effect to convert light energy into electric energy, and the output power is closely related to the solar radiation intensity and the environmental temperature parameter, so the photovoltaic cell panel output power model formula is:

[0017] p pv =ηSI[1-0.005(T0+25)]

[0018] In the formula, η is the photoelectric conversion efficiency, and 16.6% is commonly used; S is the area of the solar photovoltaic cell panel, the area can directly reflect the photovoltaic rated power and installation, I is the solar radiation intensity; T0 is the ambient temperature.

[0019] Further, the maximum wind-light joint complementary rate and the minimum abandoned wind and light amount are taken as the target, and specifically comprising:

[0020] The complementary rate is used to express the optimal target in scheduling, and the economic efficiency of thermal power operation and the minimum abandoned wind and light amount are considered as the target, and the overall target description formula is as follows:

[0021]

[0022]

[0023] min f1=L W +L S

[0024] In the formula, λ is the complementary rate, C is the operation cost of the complementary joint power generation system, u i is the start-stop state of the thermal power unit i at the t period, [t, T] is the total period of the daily scheduling of the thermal power unit, a, b, and c are the coal combustion cost coefficients of the thermal power unit, and the investment and maintenance costs are not considered, so only the coal combustion cost of the thermal power unit is included, f1 is the abandoned amount, L W and L S are the abandoned wind amount and the abandoned light amount respectively;

[0025] The target is all converted into the minimum target formula as follows:

[0026]

[0027]

[0028] min f1=L W +L S

[0029]

[0030]

[0031]

[0032] wherein, δ B is the standard deviation of the output power of the wind-solar hybrid complement, δ A is the standard deviation of the output power of the energy with large capacity proportion, P(t) is the output power value of the corresponding device at each time, P maxW is the average output power of the corresponding device, P maxS (t) and P W (t) represent the output power of the wind turbine and the photovoltaic device at time t respectively, P S (t) and P W (t) represent the actual scheduling values of the wind power and the photovoltaic power respectively.

[0033] Further, the energy balance constraint equation in the minimization target formula is:

[0034] P S (t) + P H (t) = P LD (t)

[0035] The energy balance constraint is the basic power constraint in the system, and the formula needs to satisfy that the sum of the real-time scheduling value P W (t) of the wind power, the real-time scheduling value P S (t) of the photovoltaic power and the real-time scheduling value P H (t) of the traditional thermal power as a backup is equal to the load demand P LD (t).

[0036] Further, the device constraint can be divided into two parts, i.e. the output constraint of each energy device and the ramping constraint of the thermal power unit, and the output constraint formula of each energy is:

[0037]

[0038] wherein, P H,min and P H,max represent the minimum output and the maximum output of the thermal power unit as a backup; P W,max and P S,max are the maximum outputs of the wind turbine and the photovoltaic device.

[0039] The ramping constraint formula of the thermal power unit is:

[0040] -r d Δt≤P H (t+1)-P H (t)≤r uΔt

[0041] In the formula, r d r u These are the rate limits for load reduction and loading of thermal power units during the dispatch period, respectively.

[0042] Furthermore, the improved particle swarm optimization algorithm specifically includes:

[0043] The algorithm flow is constructed using a mathematical model. Let the target space be an N-dimensional search space, and the number of particles be M. For particle i at time t, its position attributes... and speed attribute V i t as follows:

[0044]

[0045]

[0046] in, and Let v be the upper and lower limits of particle i in N-dimensional space. min,id and v max,id Let be the minimum and maximum velocities of particle i in N-dimensional space;

[0047] The optimal position of an individual during the search process Represented as:

[0048]

[0049] Global optimal position Represented as:

[0050]

[0051] At time t+1, the particle's velocity and position attributes are:

[0052]

[0053]

[0054] Wherein, c1 and c2 are collectively referred to as acceleration constants, which are individual learning factors and social learning factors, respectively; r1 and r2 are random numbers and are uniformly distributed on [0,1]. The improved particle swarm algorithm adds a weighted inertia factor ω to the dual attributes of the traditional particle swarm. The computational speed and accuracy of the particle swarm are controlled by continuously changing the value of the inertia factor during the search process.

[0055] Furthermore, the velocity and position attributes of the improved particle swarm optimization algorithm can be represented as:

[0056]

[0057]

[0058] When the weight inertia factor ω is 1, it is a traditional particle swarm dual attribute calculation formula;

[0059] ω in the iteration process, ω participates in the iteration t The calculation formula is:

[0060]

[0061] In the formula, ω t is the inertia weight factor under the tth iteration that can participate in the iteration change, ω max represents the maximum value of the weight factor, which is 0.9, ω min represents the minimum value of the weight factor, which is 0.4, N max is the maximum number of iterations, and the adjustment of the dual attributes in the improved particle swarm is realized according to the change of the inertia factor.

[0062] Further, the wind and light complementary joint scheduling optimization method using the improved particle swarm algorithm specifically comprises:

[0063] (1) Input device parameters: input wind turbine parameters, photovoltaic device parameters, thermal power unit parameters, and meteorological data;

[0064] (2) Input algorithm parameter initialization: input the weight inertia factor, the acceleration constant, and the number of iterations, and determine the local optimum, the global optimum, and the speed information according to the output of the wind turbine and the photovoltaic device;

[0065] (3) Calculate the fitness: calculate the fitness value considering economy, complementarity, and abandonment according to the target formula;

[0066] (4) Update the local optimum: compare the fitness with the initial local optimum, if the fitness is better, update the local optimum; if the initial local optimum is more suitable, no change occurs;

[0067] (5) Update the global optimum: compare the updated local optimum with the initial global optimum to determine whether to update;

[0068] (6) Update the algorithm dual attributes: update the particle position and particle speed dual attributes according to the dual attribute formula;

[0069] (7) Check whether the convergence condition has been met: when the convergence condition is met, stop the algorithm iteration, and output the optimal solution of economy, complementarity, and abandonment, as well as the output of each device. If the convergence condition is not met, continue the iteration according to the formula to update the weight factor until the iteration is stopped.

[0070] The advantages and beneficial effects of the present application are as follows:

[0071] The present application proposes a wind-solar complementary joint optimization scheduling method based on an improved particle swarm algorithm, constructs a wind-solar and photovoltaic output model and a joint optimization scheduling model considering the complementary characteristics of wind-solar based on the complementary rate, and to some extent, not only improves the utilization rate of the two renewable energy sources of wind and solar, but also ensures the stability of the joint complementary output and the safety of the standby output. By using the improved particle swarm algorithm with a weight inertia factor to solve the scheduling model, the abandoned quantity reduction value considering the complementary requirement in system scheduling and the coal-fired cost saving amount of thermal power units are obtained, thereby the importance of considering complementarity to the joint complementary system scheduling is clarified, and this method can also provide a model reference for the scheduling of the complementary system in practical engineering.

[0072] The traditional wind-solar complementarity mainly refers to the time complementarity between wind energy and solar energy, and few studies consider maximizing the wind-solar complementary rate considering the factors such as power transmission scheduling cost. As described in the steps of claim 3, the present application constructs a joint optimization scheduling model considering the complementary characteristics of wind-solar based on the complementary rate, uses the complementary rate to express the optimal complementary target in scheduling, and considers the economic efficiency of thermal power operation and the minimum abandoned wind and solar as the target. To some extent, not only the utilization rate of the two renewable energy sources of wind and solar is improved, but also the stability of the joint complementary output and the safety of the standby output are ensured.

[0073] The current scheduling research has many problems such as high dimension, large amount of calculation, and calculation deviation, so it is particularly important to select a suitable model solving algorithm. The current mainstream intelligent optimization algorithms include particle swarm algorithm and genetic algorithm, which have fast search speed but are easy to fall into local optimal solution. As shown in the steps of claim 7, the improved particle swarm algorithm of the present application is based on the traditional particle swarm algorithm, and a weight inertia factor is added in the original self-optimization and global optimization processes, thereby avoiding the problem that the traditional particle swarm algorithm is easy to fall into local optimal solution. BRIEF DESCRIPTION OF DRAWINGS

[0074] Figure 1 is a schematic diagram of a wind-solar complementary distribution collaborative operation mode provided by the present application.

[0075] Figure 2 is a flowchart of the improved particle swarm algorithm provided by the present application. DETAILED DESCRIPTION

[0076] The technical solutions in the embodiments of the present application will be described in detail below with reference to the drawings in the embodiments of the present application. The described embodiments are only a part of the embodiments of the present application.

[0077] The technical solution of the present application to solve the above technical problems is:

[0078] The method of the application firstly determines the output model of wind power and photovoltaic according to a wind-solar complementary distribution collaborative operation mode, and a schematic diagram of the wind-solar complementary distribution collaborative operation mode is as shown in Figure 1 ; then the wind-solar complementary is jointly scheduled and optimized based on an improved particle swarm algorithm, and a flowchart of the improved particle swarm algorithm is as shown in Figure 2 .

[0079] An optimal complementary target is established in the wind-solar scheduling, and a fluctuation theory index is used for description. Meanwhile, the economic efficiency of thermal power operation and the minimum amount of abandoned wind and light are considered as targets, and the overall target description formula is as follows:

[0080]

[0081]

[0082] min f1=L W +L S

[0083] In the formula, λ is a complementary rate, C is an operation cost of a complementary combined power generation system. The investment and maintenance costs are not considered, and therefore only the coal combustion cost of a thermal power unit is included. f1 is an abandoned amount, L W and L S are an abandoned wind amount and an abandoned light amount respectively.

[0084] Further, all the targets are converted into a minimum target formula as follows:

[0085]

[0086]

[0087] min f1=L W +L S

[0088]

[0089]

[0090]

[0091] In the formula, δ B is a standard deviation of output power after wind-solar joint complementation, δ A is a standard deviation of output power of an energy with a large capacity proportion, P(t) is an output power value of a corresponding device at each time, and P is an average output power of a corresponding device. P maxW (t) and P maxS (t) respectively represent output powers of a wind power generator and a photovoltaic device at t time, and P W(t) represents the actual scheduling value of wind power and light power respectively. S (t) represents the actual scheduling value of wind power and light power respectively.

[0092] Further, the energy balance constraint equation is:

[0093] P W (t) represents the actual scheduling value of wind power and light power respectively. S (t) represents the actual scheduling value of wind power and light power respectively. H (t) represents the actual scheduling value of wind power and light power respectively. LD (t) represents the actual scheduling value of wind power and light power respectively.

[0094] The energy balance constraint is the basic power constraint in the system, and the formula needs to satisfy the sum of the real-time scheduling value P W (t) of wind power, the real-time scheduling value P S (t) of photovoltaic and the real-time scheduling value P H (t) of traditional thermal power as backup equal to the load demand P LD (t).

[0095] Further, in the wind-solar complementary model set in this paper, the equipment constraint can be divided into two parts, namely the output constraint of each energy equipment and the ramping constraint of thermal power units. The formula of each energy output constraint is:

[0096]

[0097] In the formula, P H,min and P H,max represent the minimum output and maximum output of thermal power units as backup; P W,max and P S,max are the maximum output of wind turbine and photovoltaic equipment.

[0098] Further, the formula of thermal power unit ramping constraint is:

[0099] -r d Δt≤P H (t+1)-P H (t)≤r u Δt

[0100] In the formula, r d and r u are the rate limit values of load shedding and loading of thermal power units within the scheduling period.

[0101] Further, the algorithm flow is constructed by mathematical model, the target space is N-dimensional search space, and the number of particles is M. For particle i at time t, the position attribute and the velocity attribute V i t are as follows:

[0102]

[0103]

[0104] wherein, and are the upper and lower limit values of particle i in N-dimensional space, v min,id and v max,id are the minimum and maximum velocities of particle i in N-dimensional space.

[0105] Further, the optimal position of the individual in the search process is expressed as:

[0106]

[0107] The global optimal position is expressed as:

[0108]

[0109] And at t+1, the velocity and position attributes of the particle are:

[0110]

[0111]

[0112] wherein, c1, c2 are collectively referred to as acceleration constants, and are an individual learning factor and a social learning factor, respectively; r1, r2 are random numbers, and are uniformly distributed on [0, 1]. The improved particle swarm algorithm adds a weight inertia factor ω to the double attributes of the traditional particle swarm, and controls the calculation speed and accuracy of the particle swarm by constantly changing the value of the inertia factor in the search process.

[0113] Further, the velocity and position attributes of the improved particle swarm algorithm can be expressed as:

[0114]

[0115]

[0116] When the weight inertia factor ω is 1, it is the double attribute calculation formula of the traditional particle swarm. In the solving process, we always want to improve the global convergence ability in the early stage of calculation, and quickly approach the global optimal solution; and want to have strong local convergence ability in the later stage to ensure the accuracy of the solution.

[0117] Further, ω in the iteration process, ω t The calculation formula is:

[0118]

[0119] In the formula, ω tIn order to participate in the inertia weight factor in the t-th iteration of the iterative change, ω max This represents the maximum value of the weighting factor, typically taken as 0.9, ω. min This represents the minimum value of the weighting factor, typically taken as 0.4, N. max This represents the maximum number of iterations. The modulation of dual attributes in particle swarm optimization is achieved based on changes in the inertia factor.

[0120] Furthermore, the method for optimizing joint scheduling of wind-solar hybrid systems using an improved particle swarm optimization algorithm specifically includes:

[0121] (1) Input equipment parameters: Input wind turbine parameters, photovoltaic equipment parameters, thermal power unit parameters, meteorological data, etc.

[0122] (2) Input algorithm parameter initialization: Input weight inertia factor, acceleration constant, iteration number, and determine local optimum, global optimum and speed information based on the output of wind turbine and photovoltaic equipment.

[0123] (3) Calculate fitness: Calculate the fitness value considering economy, complementarity and abandonment based on the target formula.

[0124] (4) Update local optimum: Compare the fitness with the initial local optimum. If the fitness is better, update it to the local optimum; if the initial local optimum is more suitable, do not change it.

[0125] (5) Update the global optimum: Compare the updated local optimum with the initial global optimum to determine whether to update.

[0126] (6) Update the dual attributes of the algorithm: Update the particle position and particle velocity dual attributes according to the dual attribute formula.

[0127] (7) Check if the convergence condition has been met: When the convergence condition is met, stop the algorithm iteration, output the optimal solution in terms of economy, complementarity, and waste, as well as the output status of each device. If the convergence condition is not met, update the weight factors according to the formula and continue iterating until it stops.

[0128] This invention utilizes an improved particle swarm optimization algorithm, using the obtained capacity allocation as the basis for capacity setting, to establish a regional joint complementary power generation optimization scheduling model considering wind-solar complementarity. The scheduling model is solved with the objectives of maximizing wind-solar complementarity, minimizing thermal power unit costs, and minimizing wind and solar curtailment. Considering wind-solar complementarity not only stabilizes the output of thermal power reserves during scheduling but also reduces thermal power costs and wind and solar curtailment, while simultaneously improving the utilization rate of wind and solar energy.

[0129] The systems, apparatuses, modules, or units illustrated in the above examples can be specifically implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0130] Computer readable media includes permanent and non-permanent, removable and non-removable media, which can be implemented by any method or technology to store information. The information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device. According to the definition herein, computer readable media does not include transitory media, such as modulated data signals and carriers.

[0131] It should also be noted that the terms "comprising", "including", or any other variant thereof are intended to cover non-exclusive inclusion, so that processes, methods, articles or devices including a series of elements not only include those elements, but also include other elements not explicitly listed or inherent to such processes, methods, articles or devices. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article or device including the element.

[0132] The above examples are to be understood only as illustrative of the application and not a limitation of the scope of protection of the application. After reading the specification, those skilled in the art can make various changes or modifications to the application, and these equivalent changes and modifications also fall within the scope defined by the claims of the application.

Claims

1. A wind-solar complementary joint optimization scheduling method based on an improved particle swarm algorithm, characterized in that, It comprises the following steps: First, the output model of the wind turbine and photovoltaic is determined according to the wind-solar complementary distribution cooperative operation mode to obtain the output power of the wind power and photovoltaic panel; the maximum wind-solar complementary rate and the minimum abandoned wind and light amount are taken as the target; the thermal power unit is added as a backup, the complementary optimal matching area and the complementary optimal wind-solar capacity ratio are used as the basic data, and the improved particle swarm algorithm is combined to optimize and dispatch the wind-solar complementary combined power generation system; The wind turbine output model specifically comprises: The wind turbine comprises a cut-in wind speed and a cut-out wind speed, and can only generate a usable power output when running between the cut-in wind speed and the cut-out wind speed, and takes the rated wind speed as a boundary, so the real-time output model formula of the wind turbine is: where p r represents the rated power of the wind turbine, v r is the rated wind speed of the wind turbine, v in , v out represent the cut-in and cut-out wind speeds, respectively, and v is the real-time wind speed of the selected study area. The photovoltaic power generation output model specifically comprises: The photovoltaic power generation converts light energy into electric energy by using the photovoltaic effect, and the output power is closely related to the solar radiation intensity and the environmental temperature parameter, so the photovoltaic panel output power model formula is: p pv = ηsI [1 - 0.005(T0+25)] In the formula, η is the photoelectric conversion efficiency, which is commonly used 16.6%; S is the area of the solar photovoltaic panel, which can directly reflect the photovoltaic rated power and installation; I is the solar radiation intensity; T0 is the ambient temperature; The maximum wind-solar complementary rate and the minimum abandoned wind and light amount are taken as the target, specifically comprising: The complementary rate is used to express the complementary optimal target in the dispatching, and the economic efficiency of the thermal power operation and the minimum abandoned wind and light amount are considered as the target, and the overall target description formula is as follows: min f1 = L W +L S where λ is the complementary rate, C is the operating cost of the complementary combined power generation system, u i is the on-off state of the thermal power unit i in the t period, [t, T] is the total period of the thermal power unit daily scheduling, a, b, c are the coal cost coefficients of the thermal power unit, the investment and maintenance costs are not taken into account, so only the coal cost of the thermal power unit is included, f1 is the abandoned quantity, L W and L S are the abandoned wind and light quantities, respectively; All the targets are converted into the minimum target formula as follows: min f1 = L W +L S In the formula, δ B is the standard deviation of the output power after wind-solar combined complementation, δ A is the standard deviation of the output power of the energy with large capacity proportion, P(t) is the output power value of the corresponding device at each time, P maxW is the average output power of the corresponding device, P maxS (t) and P W (t) represent the output power of the wind turbine and the photovoltaic device at time t, respectively, and P S (t) and P S (t) represent the actual scheduling values of wind power and photovoltaic power, respectively.

2. The wind-solar complementary joint optimization scheduling method based on the improved particle swarm algorithm according to claim 1, characterized in that, The energy balance constraint equation in the minimum target formula is: P W (t)+P S (t)+P H (t)=P LD (t) The energy balance constraint is the base power constraint in the system, the equation needs to satisfy the wind power real-time scheduling value P W (t), the photovoltaic real-time scheduling value P S (t) and the traditional thermal power real-time scheduling value P H (t) as a backup. The sum of the above three values is equal to the load demand P LD (t).

3. The wind-solar complementary joint optimization scheduling method based on the improved particle swarm algorithm according to claim 2, characterized in that, The equipment constraint can be divided into two parts, namely the output constraint of each energy equipment and the climbing constraint of the thermal power unit, and the energy output constraint formula is as follows: where P H,min and P H,max represent the minimum and maximum power output of the backup thermal power unit, respectively; P W,max and P S,max are the maximum power outputs of the wind turbine and photovoltaic device, respectively. The climbing constraint formula of the thermal power unit is as follows: - r d Δt < P H (t + 1) - P H (t) < r u Δt In the formula, r d , r u are the rate limits for load shedding and loading of thermal power units within the dispatch period, respectively.

4. The wind-solar complementary joint optimization scheduling method based on the improved particle swarm algorithm according to claim 3, characterized in that, The improved particle swarm algorithm specifically comprises: With mathematical model to construct algorithm flow, set target space as N-dimensional search space, particle number as M, for particle i at t time, its position attribute and velocity attribute V i t As follows: wherein, and are upper and lower limit values for particle i in N-dimensional space, v min,id and v max,id are minimum and maximum velocities for particle i in N-dimensional space; Optimal position of individuals in search process is represented as: Global optimum position is represented as: At t+1 time, the speed and position attributes of the particle are as follows: In the formula, c1 and c2 are collectively called acceleration constants, and are respectively an individual learning factor and a social learning factor; r1 and r2 are random numbers, and are uniformly distributed in [0, 1]; the improved particle swarm algorithm adds a weight inertia factor ω in the double attributes of the traditional particle swarm, and the value of the inertia factor is changed in the search process to control the calculation speed and accuracy of the particle swarm.

5. The wind-solar complementary joint optimization scheduling method based on the improved particle swarm algorithm according to claim 4, characterized in that, The speed and position attributes of the improved particle swarm algorithm can be expressed as follows: When the weight inertia factor ω is 1, it is the double attribute calculation formula of the traditional particle swarm; ω in the iteration process, ω participating in the iteration t The calculation formula is: In the formula, ω t is the inertia weight factor in the tth iteration of participating in the iterative change, ω max represents the maximum value of the weight factor, which is 0.9, ω min represents the minimum value of the weight factor, which is 0.4, N max is the maximum number of iterations, and the improvement of the dual attribute in the particle swarm is realized according to the change of the inertia factor.

6. The wind-solar complementary joint optimization scheduling method based on the improved particle swarm algorithm according to claim 5, characterized in that, Specifically comprising: (1) inputting equipment parameters: inputting wind turbine parameters, photovoltaic equipment parameters, thermal power unit parameters, and meteorological data; (2) inputting algorithm parameter initialization: inputting the weight inertia factor, the acceleration constant, and the iteration number, and determining the local optimum, the global optimum, and the speed information according to the output of the wind turbine and the photovoltaic equipment; (3) calculating the fitness: calculating the fitness value considering the economy, the complementarity, and the abandoned amount according to the target formula; (4) updating the local optimum: comparing the fitness with the initial local optimum, if the fitness is better, the local optimum is updated; if the initial local optimum is more suitable, no change occurs. (5) Update global optimum: compare the updated local optimum with the initial global optimum to determine whether to update; (6) Update algorithm double attribute: update the double attribute of particle position and particle velocity according to the double attribute formula; (7) Check whether the convergence condition has been met: when the convergence condition is met, stop the algorithm iteration, output the optimal solution of economy, complementarity and waste, and the output of each device; if the convergence condition is not met, continue the iteration until it is stopped after updating the weight factor according to the formula.

Images