Micro-grid multi-element collaborative configuration method based on random fractal search algorithm

Abstract

The invention discloses a micro-grid multi-element collaborative configuration method based on a random fractal search algorithm, and the method comprises the steps: building a comprehensive energy system model which comprises a photovoltaic power supply, a wind turbine, an energy storage system and a diesel generator, and defining the operation parameters and constraint conditions of all assemblies; energy leveling cost, power supply loss probability and energy consumption of a virtual load are comprehensively considered, maximization of power supply and minimization of energy cost in the hybrid micro-grid system are taken as final objectives, and a multi-objective optimization function is constructed; a random fractal search algorithm is adopted to carry out optimal configuration on all components of the system, and the optimal scale and operation parameters are determined through iterative calculation and self-adaptive adjustment; after optimization, the result is analyzed, the performance of the system under different operation conditions is verified, the design of the hybrid energy system which operates independently can be optimized, the electric power cost is reduced, and meanwhile the stability and reliability of power supply are improved.

Description

Technical Field

[0001] The present invention relates to the field of new energy system design, and in particular to a microgrid multi-factor collaborative configuration method based on a random fractal search algorithm, which is particularly suitable for optimal design and resource configuration in an independently operated hybrid energy system. Background Art

[0002] With the continuous growth of global energy demand and the rapid development of renewable energy technology, independently operated hybrid energy systems are increasingly widely used in remote and off-grid areas. In these areas, diesel generators have become the main source of power supply. Although their working process is stable and controllable, the power generation cost is high and there are environmental pollution problems. At the same time, the ubiquity of wind and solar energy resources makes wind power generation and photovoltaic power generation typical renewable energy generation methods with broad application prospects. However, wind power generation and photovoltaic power generation are sensitive to meteorological factors and their reliability is relatively low. Therefore, configuring energy storage systems to achieve time-shifted energy is an effective means to improve reliability, but the high cost of energy storage significantly affects the economic feasibility of microgrids. In addition, limited by the instability and randomness of wind, solar, and diesel energy storage systems, battery energy storage systems may reduce their lifespan and increase operation and maintenance costs due to the influx of excess electricity. Traditional optimization design methods, such as particle swarm optimization (PSO) algorithms, face the disadvantages of high computational complexity and slow convergence speed when solving system scale optimization problems, and it is difficult to achieve ideal results in multi-objective optimization. As an emerging optimization algorithm, the random fractal search algorithm has the advantages of high computational efficiency and strong global search capabilities, and performs particularly well in single-objective optimization and simple multi-objective problems. However, since the design and optimization of microgrid systems involve multiple interdependent objectives, and these objectives often conflict with each other, existing studies have shown that the random fractal search algorithm may not be able to fully consider the trade-offs between various objectives when dealing with highly nonlinear and multi-conflicting objectives, resulting in insufficient robustness of the final solution. Therefore, applying the random fractal search algorithm to the optimization design of microgrid systems not only has important research value and practical significance, but also provides new ideas for solving the technical development problems of existing models. Summary of the invention

[0003] The purpose of the present invention is to propose a microgrid multi-factor collaborative configuration method based on a random fractal search algorithm, which can optimize the design of independently operated hybrid energy systems, reduce electricity costs, and improve the stability and reliability of power supply.

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

[0005] A microgrid multi-element collaborative configuration method based on a random fractal search algorithm, the microgrid multi-element collaborative configuration method comprising the following steps:

[0006] Step 1: Establish a comprehensive energy system model including photovoltaic power sources, wind turbines, energy storage systems and diesel generators, and define the operating parameters and constraints of each component;

[0007] Step 2, comprehensively considering the levelized cost of energy, the probability of power supply loss and the energy consumption of virtual loads, with maximizing power supply and minimizing energy cost in the hybrid microgrid system as the ultimate goal, constructing a multi-objective optimization function;

[0008] Step 3: Use the random fractal search algorithm to optimize the configuration of each component of the system, and determine the optimal scale and operating parameters through iterative calculation and adaptive adjustment; after optimization, analyze the results to verify the performance of the system under different operating conditions.

[0009] Step 1 further comprises:

[0010] Step 11, establish a mathematical model of the photovoltaic power source, where the power output P of the photovoltaic power source PV (t) is calculated by the following formula:

[0011]

[0012] Where P Rated is the rated power of the photovoltaic cell, G(t) is the solar radiation intensity at time t, G Ref is the radiation intensity under standard test conditions, T amb (t) is the ambient temperature, T Ref is the reference temperature, K T is the temperature coefficient;

[0013] Step 12: Establish a wind turbine model. The power output P of the wind turbine wind (t) is calculated according to the following formula:

[0014]

[0015] Where V(t) is the wind speed, V cut is the cut-in wind speed, V rated is the rated wind speed, V cutout To cut out the wind speed;

[0016] Step 13: Establish an energy storage system model and calculate the battery capacity according to the following formula:

[0017]

[0018] In the formula, El is the load; AD is the number of self-sufficient days, DOD is the depth of discharge, η inv is the inverter efficiency, η bis the battery efficiency; the battery state of charge SOC(t) at each time point t is an indicator to measure the state of the battery energy storage system; the operation of the battery energy storage system is affected by its maximum state of charge SOC max and minimum state of charge SOC min The battery charging power is represented by E ch (t) represents the discharge power, E disch (t) represents; wherein, the battery charging and the energy change in the battery during charging are as follows:

[0019] E ch (t) = η ch ·(P renewable (t)-(P load (t) / η inv ))·Δt (4)

[0020] P renewable =P PV (t)+P wind (t) (5)

[0021] SOC(t)=Eb(t) (6)

[0022] Eb(t)=Eb(t-1)·(1-σ)+E ch (t) (7)

[0023] Where η ch is the battery charging efficiency; Eb(t) is the battery energy at time t; Eb(t-1) is the battery energy at time t-1; σ is the battery self-discharge rate, that is, the energy loss ratio of the battery when it is not in use; P load (t) is the hourly load demand;

[0024] The battery discharge and the energy changes in the battery during discharge are shown below:

[0025] E disch (t)=(P load (t) / η inv )-(P renewable (t))·Δt·η disch (8)

[0026] Eb(t)=Eb(t-1)·(1-σ)-E disch (t) (9)

[0027] Where η disch is the battery discharge efficiency;

[0028] Step 14: Establish a diesel generator model. The fuel consumption F of the diesel generator consumption (t) and total efficiency η overall They are:

[0029] F consumption (t)=0.246·P diesel (t)+0.08415·P diesel rated (10)

[0030] η overall =η brake thermal +η generator (11)

[0031] Where P diesel (t) is the actual output power of the diesel generator at time t, P diesel rated is the rated power of the diesel generator; η brake thermal is the braking thermal efficiency; η generator Generator efficiency, where the coefficients 0.246 and 0.08415 represent approximate coefficients of fuel consumption.

[0032] Step 15, establish an inverter model. The inverter converts the DC power generated by the hybrid renewable energy system into AC power to supply the load. Its efficiency is:

[0033]

[0034] P=P out / P n (13)

[0035]

[0036]

[0037] Where, P is the standardized output power; P0 is the inverter no-load power loss; P n is the rated power of the inverter; P out is the actual output power of the inverter; K is the efficiency correction factor; η 10 and η 100 are the inverter efficiencies at 10% and 100% rated power, respectively.

[0038] Step 2 further includes:

[0039] Step 21, construct a multi-objective optimization function (fitness function of random fractal search algorithm):

[0040] Fitness Function:F(x)=Min(LCOE+LPSP+Dummy excess ) (16)

[0041] The levelized cost of energy (LCOE) is calculated by the following formula:

[0042]

[0043] Where NPC stands for net present value and CRF is the capital recovery factor, which depends on the actual interest rate i r and system life cycle R;P load Represents the total load demand within 8760 hours; the calculation formulas for NPC and CRF are:

[0044]

[0045] NPC=TAC / CRF(19)

[0046]

[0047] Where N pv 、N wind 、N DG 、N battery , and N inverter Respectively represent the number of components in the system; PV cost 、Wind cost , DG cost , Batt cost and Inv cost Represent the costs of photovoltaic, wind turbine, diesel generator, battery and inverter respectively; the cost of each component is calculated by the following formula:

[0048] C component =C cost +O&M cost +R cost (twenty one)

[0049] In the formula, C component , C cost 、O&M cost , R cost They are the total cost, purchase cost, operation and maintenance cost, and replacement cost of each component;

[0050] The power supply loss probability LPSP is:

[0051]

[0052] LPS(t)=P load (t)-((P renewable (t))+SOC(t-1)-SOC min )·η inv +P diesel (t) (23)

[0053] Dummy Load excess for:

[0054]

[0055] Where P dump (t) represents the total energy consumed in the virtual load at time t;

[0056] Step 22, define the decision variables and their ranges:

[0057] X=[N pv N wt N Battery N AD N diesel ] (25)

[0058]

[0059]

[0060]

[0061]

[0062]

[0063] In the formula, and are the minimum values ​​of the photovoltaic unit, wind turbine, battery, diesel generator and independent operation days, respectively; and They represent the maximum values ​​of photovoltaic units, wind turbines, batteries, diesel generators and independent operation days respectively.

[0064] Step 3 further includes:

[0065] Step 31, generate an initial population, each individual represents a configuration scheme of the hybrid energy system; set the configuration and operation parameters of each component of the system; perform fitness function evaluation on all configuration schemes, calculate various performance indicators including levelized cost of energy LCOE, net present value cost NPC, power outage probability LPSP, so as to judge the preliminary feasibility of each configuration scheme, and sort all individuals according to the fitness function value;

[0066] Step 32, based on the initial evaluation, start the optimization algorithm and iteratively update the configuration parameters of each component of the system; in each iteration, perform fitness evaluation based on the updated system design and compare it with the termination condition; if the termination condition is not met, continue the optimization process until the termination condition is met, and output the final optimization result, including the optimal configuration plan of the system and the optimal values ​​of relevant economic and performance indicators.

[0067] Furthermore, in step 32, the process of iteratively updating the configuration parameters of each component of the system includes the following steps:

[0068] Step 321, each particle diffuses around its current position so that each screened solution generates other solutions; the global optimal solution is found by using Gaussian random walk. The configuration parameters of the solution (such as the capacity of the photovoltaic system, the charging and discharging strategy of the battery, etc.) will be updated in each iteration;

[0069] G walk1 =Gaussian(P best ,σ)+(r1·P best -r2·P i ) (31)

[0070] G walk2 =Gaussian(P i ,σ) (32)

[0071]

[0072] In the formula, G walk1 , G walk2 is a solution generated based on Gaussian random walk, used to explore the global optimal solution; the standard deviation used in the diffusion process depends on the number of iterations iter and the distance between the current position of the solution and the optimal solution; P best is the optimal solution in the current population; P i is the seed solution; r1 and r2 are random numbers in the range [0,1];

[0073] Step 322, use formula (34) to calculate a probability value for each solution in the set, and sort all solutions according to their fitness values. The fitness function is the objective function of step 21 above; use formula (35) to perform the first update, adjust the configuration parameters of the system components, and when the condition r is met <P Ai Update all points when

[0074]

[0075] P i ′(j)=P r (j)-r·(P t (j)-P i (j)) (35)

[0076] Among them, P Ai is the probability value of the ith solution; P r (j), P t (j) are two randomly selected candidate solutions; P i (j) is a randomly selected solution from the previous update process; Pi ′(j) is the new position of the solution, r is a random number in the range [0, 1]; the purpose of formula (35) is to change the position of the bad candidate solution while keeping the best solution unchanged so that the best solution can be passed to the next iteration;

[0077] Step 323, recalculate the fitness function value, and use formula (34) to sort the fitness function values ​​of all points for the third time;

[0078] Step 324, use formula (36) and (37) to make a second update for all points, and the random fractal search algorithm randomly switches between the two. If the fitness function value of the new solution is better than the old value, the new solution replaces the old solution:

[0079] P i "(j) = P i ′(j)-r·(P i ′(j)-P best ); r′≤0.5 (36)

[0080] P i "(j) = P i ′(j)-r·(P i ′(j)-P r ′(j)); r′>0.5 (37)

[0081] Among them, P i ′(j) and P r ′(j) represents a random solution randomly selected from the first update process; P i ″(j) represents the solution after the second update;

[0082] Step 323, repeating steps 321 and 322 until the maximum number of iterations is reached. When the termination condition is reached, the solution with the best fitness value is finally output, and its corresponding configuration scheme performs best in each objective function (LCOE, NPC, LPSP), that is, the hybrid energy system configuration with the lowest cost, the best economy, and the highest power supply reliability.

[0083] Compared with the prior art, the present invention has the following beneficial effects:

[0084] First, the microgrid multi-factor collaborative configuration method based on the random fractal search algorithm of the present invention constructs a system load and energy generation model and uses the random fractal search algorithm to optimize the configuration of each component of the system to determine its optimal scale.

[0085] Second, the multi-factor coordinated configuration method of microgrid based on random fractal search algorithm of the present invention comprehensively considers the levelized cost of energy (LCOE), power supply loss probability (LPSP) and energy consumption of virtual load to form a multi-objective optimization function. Through the application of random fractal search algorithm, the optimal resource allocation under different operating conditions is achieved, which effectively reduces the overall cost of the system and improves the stability and reliability of power supply;

[0086] Third, the microgrid multi-factor collaborative configuration method based on the random fractal search algorithm of the present invention optimizes the design of an independent hybrid microgrid system consisting of photovoltaic power sources, wind turbines, energy storage systems and diesel generators, aiming to reduce electricity costs and ensure stable and reliable power supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 It is a flow chart of the microgrid multi-element collaborative configuration method based on random fractal search algorithm of the present invention;

[0088] Figure 2 It is a schematic diagram of the diffusion process of the random fractal search algorithm of the present invention;

[0089] Figure 3 It is a schematic diagram of the updating process of the random fractal search algorithm of the present invention. DETAILED DESCRIPTION

[0090] The embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings.

[0091] See also Figure 1 The present invention discloses a microgrid multi-element collaborative configuration method based on a random fractal search algorithm, comprising the following steps:

[0092] S1, system model and objective function design: Establish a comprehensive energy system model including photovoltaic power sources, wind turbines, energy storage systems and diesel generators, and define the operating parameters and constraints of each component. Consider the levelized cost of energy (LCOE), power supply loss probability (LPSP) and the energy consumption of virtual loads, and construct a multi-objective optimization function.

[0093] S2, Randomized fractal search algorithm optimization and result analysis: The randomized fractal search algorithm is used to optimize the configuration of each component of the system, and the optimal scale and operating parameters are determined through iterative calculation and adaptive adjustment. After optimization, the results are analyzed to verify the performance of the system under different operating conditions, ensure that the energy load requirements are met at the lowest cost, and improve the stability and reliability of the system.

[0094] Specifically, the S1 step specifically includes:

[0095] Step S1-1, constructing a system model, as follows:

[0096] 1) Establish a mathematical model of photovoltaic power supply, where the power output P of the photovoltaic power supply PV (t) can be calculated by the following formula:

[0097]

[0098] Where P Rated is the rated power of the photovoltaic cell, G(t) is the solar radiation intensity at time t, G Ref is the radiation intensity under standard test conditions, T amb (t) is the ambient temperature, T Ref is the reference temperature, K T is the temperature coefficient;

[0099] 2) Establish a wind turbine model, the power output P of the wind turbine wind (t) is calculated according to the following formula:

[0100]

[0101] Where V(t) is the wind speed, V cut is the cut-in wind speed, V rated is the rated wind speed, V cutout To cut out the wind speed;

[0102] 3) Establish the energy storage system model and calculate the battery capacity according to the following formula:

[0103]

[0104] In the formula, El is the load; AD is the number of self-sufficient days, DOD is the depth of discharge, η inv is the inverter efficiency, η b is the battery efficiency; the battery state of charge SOC(t) at each time point t is an indicator to measure the state of the battery energy storage system; the operation of the battery energy storage system is affected by its maximum state of charge SOC max and minimum state of charge SOC min The battery charging power is represented by E ch (t) represents the discharge power, E disch (t) represents; wherein, the battery charging and the energy change in the battery during charging are as follows:

[0105] The battery charging and the energy changes in the battery during charging are shown below:

[0106] E ch (t) = η ch ·(P renewable (t)-(P load (t) / ηinv ))·Δt (4)

[0107] P renewable =P PV (t)+P wind (t) (5)

[0108] SOC(t)=Eb(t) (6)

[0109] Eb(t)=Eb(t-1)·(1-σ)+E ch (t) (7)

[0110] Where η ch is the battery charging efficiency; Eb(t) is the battery energy at time t; Eb(t-1) is the battery energy at time t-1; σ is the battery self-discharge rate, that is, the energy loss ratio of the battery when not in use; Formula (7) reflects the energy change in the battery during charging.

[0111] The battery discharge and the energy changes in the battery during discharge are shown below:

[0112] E disch (t)=(P load (t) / η inv )-(P renewable (t))·Δt·η disch (8)

[0113] Eb(t)=Eb(t-1)·(1-σ)-E disch (t) (9)

[0114] η disch is the battery discharge efficiency, and formula (9) reflects the change of energy in the battery during discharge.

[0115] 4) Establish a diesel generator model, the fuel consumption F of the diesel generator consumption (t) and total efficiency η overall They are:

[0116] F consumption (t)=0.246·P diesel (t)+0.08415·P diesel rated (10)

[0117] η overall =η brake thermal +η generator (11)

[0118] Where P diesel (t) is the actual output power of the diesel generator at time t, P diesel rated is the rated power of the diesel generator; η brake thermalis the braking thermal efficiency; η generator Generator efficiency, where the coefficients 0.246 and 0.08415 represent approximate coefficients of fuel consumption.

[0119] 5) Establish an inverter model. The inverter converts the DC power generated by the hybrid renewable energy system (HRES) into AC power to supply the load. Its efficiency is:

[0120]

[0121] P=P out / P n (13)

[0122]

[0123]

[0124] Where, P is the standardized output power; P0 is the inverter no-load power loss; P n is the rated power of the inverter; P out is the actual output power of the inverter; K is the efficiency correction factor; η 10 and η 100 are the inverter efficiencies at 10% and 100% rated power, respectively.

[0125] Step S1-2, construct a multi-objective optimization function:

[0126] 1) The main goal is to maximize the power supply while minimizing the energy cost in the hybrid microgrid system. The optimization problem aims to simultaneously minimize the levelized cost of energy (LCOE), power supply loss probability (LPSP), and dummy load (Edummy).

[0127] F(x)=Min(LCOE+LPSP+Dummy excess ) (16)

[0128] Specifically, the levelized cost of energy (LCOE) can be calculated as follows:

[0129]

[0130] Where NPC stands for net present value and CRF is the capital recovery factor, which depends on the real interest rate (i r =13%) and system life cycle (R = 24 years); P load Represents the total load demand within 8760 hours. NPC and CRF are:

[0131]

[0132] NPC=TAC / CRF (19)

[0133]

[0134] Where N pv 、N wind 、N DG 、N battery , and N inverter Respectively represent the number of components in the system; PV cost 、Wind cost , DG cost , Batt cost and Inv cost Represents the cost of photovoltaic, wind turbine, diesel generator, battery and inverter respectively. The cost of each component can be calculated by the following formula:

[0135] C component =C cost +O&M cost +R cost (twenty one)

[0136] In the formula, C component , C cost 、O&M cost , R cost They are the total cost, purchase cost, operation and maintenance cost, and replacement cost of each component.

[0137] Specifically, the loss of power supply probability (LPSP) is:

[0138]

[0139] LPS(t)=P load (t)-((P renewable (t))+SOC(t-1)-SOC min )·η inv +P diesel (t) (23)

[0140] In addition, when conducting system reliability analysis, P load (t)>P generation (t).

[0141] Specifically, the dummy load excess )for:

[0142]

[0143] P dump (t) represents the total energy consumed in the virtual load at time t.

[0144] 2) The decision variables and their ranges are defined as follows:

[0145] X=[N pv N wt N Battery N AD N diesel ] (25)

[0146]

[0147]

[0148]

[0149]

[0150]

[0151] and are the minimum values ​​of the photovoltaic unit, wind turbine, battery, diesel generator and independent operation days, respectively; and They represent the maximum values ​​of photovoltaic units, wind turbines, batteries, diesel generators and independent operation days respectively.

[0152] Specifically, step S2 specifically includes:

[0153] Step S2-1, generate the initial population, each individual represents a configuration scheme of the hybrid energy system. Then carry out detailed system design, covering the configuration and operation parameter setting of components such as photovoltaic power sources, wind turbines, and energy storage systems. Subsequently, the fitness function is evaluated for all configuration schemes, and their performance indicators (such as levelized energy cost LCOE, net present value cost NPC, power outage probability LPSP, etc.) are calculated to determine their initial feasibility.

[0154] Step S2-2, based on the initial evaluation, the optimization algorithm is started to iteratively update the size and configuration of the system components. In each iteration, the fitness evaluation is performed based on the updated system design and compared with the termination condition. If the termination condition is not met, the optimization process continues; once the termination condition is met, the final optimization result is output, including the optimal configuration of the system and the optimal values ​​of relevant economic and performance indicators.

[0155] Specifically, the optimization algorithm of step S2-2 is a random fractal search algorithm, which mainly includes two core processes: diffusion process and update process to achieve efficient global optimization. The following is a description of these two processes:

[0156] 1) Diffusion process: During the diffusion process, each particle diffuses around its current position, which is considered as the convergence point. Figure 2 As shown. Each screened solution generates other solutions, thereby increasing the chance of finding the global optimal solution. The random fractal search algorithm searches for the global optimal solution by using Gaussian random walks, which are implemented by formulas (31) and (32), and randomly exchanges these solutions. This process focuses on the best particles generated by the diffusion process, thereby increasing the probability of finding the global optimal solution.

[0157] G walk1 =Gaussian(P best ,σ)+(r1·P best -r2·P i ) (31)

[0158] G walk2 =Gaussian(P i ,σ) (32)

[0159]

[0160] 2) Update process: The update process of the random fractal search algorithm is divided into two stages. Figure 3 shown.

[0161] First update process: It is carried out immediately after the diffusion process. In this stage, all solutions are sorted according to their fitness values, which are determined by the objective function. Then, a probability value is calculated for each solution in the set using formula (34):

[0162]

[0163] P i ′(j)=P r (j)-r·(P t (j)-P i (j)) (35)

[0164] Among them, P i ′(j) is the new position of the solution, and r is a random number in the range [0, 1].

[0165] Second update process: Same as the first update process, use formula (34) to sort the solutions and select the solution when the condition r<P is satisfied. Ai All points are updated at the same time. Subsequently, further updates are performed using formulas (36) and (37), and the random fractal search algorithm randomly exchanges these solutions. If the fitness function value of the new solution is better than the old value, the old solution is replaced by the new solution.

[0166] P i ″(j)=Pi ′(j)-r·(P i ′(j)-P best ); r′≤0.5 (36)

[0167] P i ″(j)=P i ′(j)-r·(P i ′(j)-P r ′(j)); r′>0.5 (37)

[0168] Among them, P i ′(j) and P r ′(j) represents a point randomly selected from the first update process.

[0169] The random fractal search algorithm repeats these steps throughout the iterative process until the maximum number of iterations is reached, thus ensuring the accuracy and effectiveness of the optimization process.

[0170] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0171] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0172] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0173] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0174] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0175] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A microgrid multi-factor collaborative configuration method based on random fractal search algorithm, characterized in that: The microgrid multi-element collaborative configuration method comprises the following steps: Step 1: Establish a comprehensive energy system model including photovoltaic power sources, wind turbines, energy storage systems and diesel generators, and define the operating parameters and constraints of each component; Step 2, comprehensively considering the levelized cost of energy, the probability of power supply loss and the energy consumption of virtual loads, with maximizing power supply and minimizing energy cost in the hybrid microgrid system as the ultimate goal, constructing a multi-objective optimization function; Step 3: Use the random fractal search algorithm to optimize the configuration of each component of the system, and determine the optimal scale and operating parameters through iterative calculation and adaptive adjustment; after optimization, analyze the results to verify the performance of the system under different operating conditions.

2. The microgrid multi-factor collaborative configuration method based on random fractal search algorithm according to claim 1 is characterized in that: Step 1 further comprises: Step 11, establish a mathematical model of the photovoltaic power source, where the power output P of the photovoltaic power source PV (t) is calculated by the following formula: Where P Rated is the rated power of the photovoltaic cell, G(t) is the solar radiation intensity at time t, G Ref is the radiation intensity under standard test conditions, T amb (t) is the ambient temperature, T Ref is the reference temperature, K T is the temperature coefficient; Step 12: Establish a wind turbine model. The power output P of the wind turbine wind (t) is calculated according to the following formula: Where V(t) is the wind speed, V cut is the cut-in wind speed, V rated is the rated wind speed, V cutout To cut out the wind speed; Step 13: Establish an energy storage system model and calculate the battery capacity according to the following formula: In the formula, El is the load; AD is the number of self-sufficient days, DOD is the depth of discharge, η inv is the inverter efficiency, η b is the battery efficiency; the battery state of charge SOC(t) at each time point t is an indicator to measure the state of the battery energy storage system; the operation of the battery energy storage system is affected by its maximum state of charge SOC max and minimum state of charge SOC min The battery charging power is represented by E ch (t) represents the discharge power, E disch (t) represents; wherein, the battery charging and the energy change in the battery during charging are as follows: From ch (t)=η ch ·(P renewable (t)-(P load (t) / η inv ))·Δt (4) P renewable =P PV (t)+P wind (t) (5) SOC(t)=Eb(t)(6) Eb(t)=Eb(t-1)·(1-σ)+E ch (t) (7) Where η ch is the battery charging efficiency; Eb(t) is the battery energy at time t; Eb(t-1) is the battery energy at time t-1; σ is the battery self-discharge rate, that is, the energy loss ratio of the battery when it is not in use; P load (t) is the hourly load demand; The battery discharge and the energy changes in the battery during discharge are shown below: From disch (t)=(P load (t) / η inv )-(P renewable (t))·Δt·η disch (8) Eb(t)=Eb(t-1)·(1-σ)-E disch (t) (9) Where η disch is the battery discharge efficiency; Step 14: Establish a diesel generator model. The fuel consumption F of the diesel generator consumption (t) and total efficiency η overall They are: F consumption (t)=0.246·P diesel (t)+0.08415·P dieselrated (10) η overall =η brakethermal +η generator (11) In the formula, P diesel (t) is the actual output power of the diesel generator at time t, P dieselrated is the rated power of the diesel generator; η brakethermal is the braking thermal efficiency; η generator Generator efficiency, where the coefficients 0.246 and 0.08415 represent the approximate coefficients of fuel consumption; Step 15, establish an inverter model. The inverter converts the DC power generated by the hybrid renewable energy system into AC power to supply the load. Its efficiency is: P=P out / P n (13) Where, P is the standardized output power; P0 is the inverter no-load power loss; P n is the rated power of the inverter; P out is the actual output power of the inverter; K is the efficiency correction factor; η 10 and η 100 are the inverter efficiencies at 10% and 100% rated power, respectively.

3. The microgrid multi-factor collaborative configuration method based on random fractal search algorithm according to claim 1 is characterized in that: Step 2 further includes: Step 21, construct a multi-objective optimization function: Fitness Function:F(x)=Min(LCOE+LPSP+Dummy excess ) (16) The levelized cost of energy (LCOE) is calculated by the following formula: Where NPC stands for net present value and CRF is the capital recovery factor, which depends on the actual interest rate i r and system life cycle R;P load Represents the total load demand within 8760 hours; the calculation formulas for NPC and CRF are: NPC=TAC / CRF(19)TAC=N pv ×PV cost ×N wind ×Wind cost +N DG ×DG cost (20)+N battery ×Batt cost +N inverter ×Inv cost Where N pv 、N wind 、N DG 、N battery , and N inverter Respectively represent the number of components in the system; PV cost 、Wind cost , DG cost , Batt cost and Inv cost Represent the costs of photovoltaic, wind turbine, diesel generator, battery and inverter respectively; the cost of each component is calculated by the following formula: C component =C cost +O&M cost +R cost (21) In the formula, C component , C cost 、O&M cost , R cost They are the total cost, purchase cost, operation and maintenance cost, and replacement cost of each component; The power supply loss probability LPSP is: LPS(t)=P load (t)-((P renewable (t))+SOC(t-1)-SOC min )·η inv +P diesel (t) (23) Dummy Load excess for: Where P dump (t) represents the total energy consumed in the virtual load at time t; Step 22, define the decision variables and their ranges: X=[N pv N wt N Battery N AD N diesel ](25) In the formula, and are the minimum values ​​of the photovoltaic unit, wind turbine, battery, diesel generator and independent operation days, respectively; and They represent the maximum values ​​of photovoltaic units, wind turbines, batteries, diesel generators and independent operation days respectively.

4. The microgrid multi-factor collaborative configuration method based on random fractal search algorithm according to claim 1 is characterized in that: Step 3 further includes: Step 31, generate an initial population, each individual represents a configuration scheme of the hybrid energy system; set the configuration and operation parameters of each component of the system; perform fitness function evaluation on all configuration schemes, calculate various performance indicators including levelized cost of energy LCOE, net present value cost NPC, power outage probability LPSP, so as to judge the preliminary feasibility of each configuration scheme, and sort all individuals according to the fitness function value; Step 32, based on the initial evaluation, start the optimization algorithm and iteratively update the configuration parameters of each component of the system; in each iteration, perform fitness evaluation based on the updated system design and compare it with the termination condition; if the termination condition is not met, continue the optimization process until the termination condition is met, and output the final optimization result, including the optimal configuration plan of the system and the optimal values ​​of relevant economic and performance indicators.

5. The microgrid multi-factor collaborative configuration method based on random fractal search algorithm according to claim 1 is characterized in that: In step 32, the process of iteratively updating the configuration parameters of each component of the system includes the following steps: In step 321, each particle diffuses around its current position so that each screened solution generates other solutions; the global optimal solution is found by using Gaussian random walk; the configuration parameters of the solution are updated in each iteration: G walk1 =Gaussian(P best ,σ)+(r1·P best -r2·P i )(31)G walk2 =Gaussian(P i ,σ)(32) In the formula, G walk1 , G walk2 is a solution generated based on Gaussian random walk, used to explore the global optimal solution; the standard deviation used in the diffusion process depends on the number of iterations iter and the distance between the current position of the solution and the optimal solution; P best is the optimal solution in the current population; P i is the seed solution; r1 and r2 are random numbers in the range [0,1]; Step 322, using formula (34) to calculate a probability value for each solution in the set, all solutions are sorted according to their fitness values, and the fitness function is the objective function of the above step 21; using formula (35) to perform the first update, adjust the configuration parameters of the system components, and when the condition r<P is satisfied Ai Update all points when: P i ′(j)=P r (j)-r·(P t (j)-P i (j))(35) Among them, P Ai is the probability value of the ith solution; P r (j), P t (j) are two randomly selected candidate solutions; P i (j) is a randomly selected solution from the previous update process; P i ′(j) is the new position of the solution, r is a random number in the range [0, 1]; the purpose of formula (35) is to change the position of the bad candidate solution while keeping the best solution unchanged so that the best solution can be passed to the next iteration; Step 323, recalculate the fitness function value, and use formula (34) to sort the fitness function values ​​of all points for the third time; Step 324, use formula (36) and (37) to make a second update for all points, and the random fractal search algorithm randomly switches between the two; if the fitness function value of the new solution is better than the old value, the new solution replaces the old solution: P i "(j)=P i ′(j)-r·(P i ′(j)-P best );r′≤0.5 (36) P i "(j)=P i ′(j)-r·(P i ′(j)-P r ′(j));r′>0.5 (37) Among them, P i ′(j) and P r ′(j) represents a random solution randomly selected from the first update process; P i "(j) represents the solution after the second update; Step 323, continuously repeating steps 321 and 322 until the maximum number of iterations is reached; when the termination condition is reached, the solution with the optimal fitness value is finally output, and its corresponding configuration scheme performs best in each objective function, that is, the hybrid energy system configuration with the lowest cost, the best economy, and the highest power supply reliability.

Images