Optical oil storage comprehensive power supply system optimization design method considering full-period multiple constraints

Abstract

The invention discloses a full-period multi-constraint-considered optimal design method for an optical storage and oil storage integrated power supply system, and relates to the technical field of design and optimization of a new energy micro-grid system. Firstly, load, climate, economic and technical parameters are input; secondly, establishing an optimization model which takes photovoltaic, energy storage and diesel engine capacity as decision variables, takes minimization of full life cycle leveling cost LCOE as a target function, covers initial investment, operation maintenance, fuel and equipment replacement cost, and takes power supply reliability and equipment operation limitation as constraint conditions; and finally, solving the model by adopting an intelligent optimization algorithm, and outputting optimal configuration. Through full-life-cycle cost analysis and high-precision operation simulation, the design objective of optimal long-term economy of the system is achieved on the premise that power supply reliability is strictly guaranteed, the problems that an existing empirical design method is rough and poor in economy, reliability cannot be quantitatively guaranteed and the like are solved, and the method is particularly suitable for accurate design of a micro-grid in an off-grid area.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of new energy micro-grid system design and optimization, in particular to an optimal design method for a photovoltaic-battery-diesel integrated power supply system considering full-cycle multi-constraints. BACKGROUND

[0002] The photovoltaic-battery-diesel integrated power supply system is energy complementary, and is an effective solution to the power supply problem of unpowered areas, islands and important facilities. However, the existing design method has significant limitations: firstly, the current design mostly relies on the experience of engineers, and adopts a trial-and-error mode of "estimation-verification", which is tedious and difficult to obtain a global optimal solution, and is prone to lead to conservative (high cost) or aggressive (insufficient reliability) system configuration; secondly, most of the existing optimization methods take single initial investment cost as the target, ignoring the long-term costs such as operation and maintenance, fuel consumption and equipment replacement, and cannot truly reflect the economy of the system throughout its life cycle; thirdly, the design process generally lacks high-precision modeling of the specific climate conditions (such as solar irradiance and ambient temperature) of the project site, and does not fully consider key operating constraints such as the minimum load rate of diesel generators and the charging and discharging limits of batteries, resulting in a large deviation between the design scheme and the actual operation effect; fourthly, the evaluation of power supply reliability is often too rough, and the quantitative reliability index (such as load power supply shortage rate LPSP) is not integrated into the optimization model as a core constraint, and the power supply quality cannot be guaranteed from the design source.

[0003] Therefore, there is an urgent need in the art for an automated, refined and multi-objective collaborative optimization design method to scientifically solve the optimal configuration problem of the photovoltaic-battery-diesel integrated power supply system under complex multi-constraint conditions. SUMMARY

[0004] Therefore, the present application proposes an optimal design method for a photovoltaic-battery-diesel integrated power supply system considering full-cycle multi-constraints, which is suitable for the capacity configuration optimization design of a photovoltaic-battery-diesel integrated power supply system in an off-grid scenario, and provides a scientific, accurate and efficient optimal design method for a photovoltaic-battery-diesel integrated power supply system. The method establishes a full-life-cycle cost model, integrates high-resolution climate data and equipment operating constraints, and uses an intelligent optimization algorithm for automated global search, and finally obtains an optimal system configuration scheme with the lowest full-life-cycle cost under the condition of meeting strict power supply reliability requirements.

[0005] To achieve the above purpose, the technical solution adopted by the present application is as follows:

[0006] The optimal design method for a photovoltaic-battery-diesel integrated power supply system considering full-cycle multi-constraints comprises the following steps:

[0007] Step S1, input parameter preparation: obtain and input basic parameters including load demand data, regional climate data, device technical and economic parameters, and power supply reliability indicators; the device technical and economic parameters include photovoltaic module efficiency, operation and maintenance cost, and battery life model parameters;

[0008] Step S2, optimization model establishment: taking the rated power of the photovoltaic array, the rated capacity of the battery, and the rated power of the diesel generator as decision variables, combining the input parameters of step S1, taking the minimum system full life cycle levelized cost LCOE as the objective function, and setting multiple constraint conditions including power supply reliability constraints, energy storage system operation constraints, and diesel generator operation constraints;

[0009] Step S3, optimization solution: generate a candidate configuration scheme by using an intelligent optimization algorithm, perform hourly annual operation simulation on each candidate scheme, judge whether it meets the multiple constraint conditions, further calculate the corresponding objective function value, and finally iterate and optimize the candidate scheme according to the objective function value until the convergence condition is met;

[0010] Step S4, result output: output the system configuration scheme that meets the multiple constraint conditions and makes the objective function optimal: namely, the corresponding photovoltaic array rated power, battery rated capacity, and diesel generator rated power.

[0011] Further, the regional climate data in step S1 includes hourly horizontal total irradiance and ambient temperature data under a typical meteorological year.

[0012] Further, the minimum system full life cycle levelized cost LCOE includes initial investment, operation and maintenance, fuel cost, equipment replacement cost, and residual value.

[0013] Further, the power supply reliability constraint in step S2 is that the load power shortage rate LPSP needs to be less than or equal to the set maximum allowed value LPSP of the power supply reliability indicator. max The LPSP value is obtained by hourly simulation calculation for 8760 hours in a year.

[0014] Further, the energy storage system operation constraint in step S2 includes a battery life model based on discharge depth; the battery life model is used to dynamically estimate the battery life and determine the replacement cost according to the discharge depth in the operation process.

[0015] The energy storage system operation constraint includes a battery state of charge SOC constraint, a charge and discharge power constraint, and a SOC dynamic update model constraint based on energy conservation.

[0016] Further, the diesel generator operation constraint in step S2 is that its output power cannot be lower than the minimum operation power ratio k gen.minto avoid its operation in the low efficiency zone.

[0017] Further, the intelligent optimization algorithm in the step S3 is a genetic algorithm, a particle swarm optimization algorithm or a differential evolution algorithm.

[0018] Further, the annual operation simulation in the step S3 follows a preset energy management strategy, and the priority order is: preferentially using photovoltaic power generation, supplementing by discharging the storage battery when the photovoltaic power generation is insufficient, and starting the diesel generator when the storage battery is insufficient.

[0019] Compared with the prior art, the application has the beneficial effects as follows:

[0020] 1. Global optimality: the intelligent optimization algorithm is used for global search, which overcomes the defects of experience design and local optimization method, and can obtain the economic optimal configuration in the true sense.

[0021] 2. Whole cycle economy: LCOE is used as the target function, and the whole process cost from construction, operation to scrap is considered, so that the user is provided with a long-term economic optimal scheme.

[0022] 3. Operation reliability: LPSP is used as a hard constraint, and multiple constraint conditions of the energy storage system operation constraint and the diesel generator operation constraint are combined, and the design scheme is verified through accurate hourly simulation, so that the strict reliability requirement of the design scheme can be met.

[0023] 4. Actual adaptability: through integration of high-resolution climate data and fine equipment operation model, the design scheme can accurately adapt to the specific resource conditions and equipment operation characteristics of the project site, and has strong practicability. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 It is a schematic diagram of the light-storage-oil comprehensive power supply system in the embodiment of the application.

[0025] Figure 2 It is a general flowchart of the light-storage-oil comprehensive power supply system optimization design method considering the whole cycle and multiple constraints in the embodiment of the application.

[0026] Figure 3 It is an optimization solution flowchart of the genetic algorithm used in the embodiment of the application. DETAILED DESCRIPTION

[0027] The content of the application will be further described below in combination with the drawings and specific embodiments.

[0028] The light-storage-oil comprehensive power supply system optimization design method considering the whole cycle and multiple constraints, as shown in Figure 2 , includes the following processes:

[0029] S1: Input parameter preparation

[0030] Collect and input all the basic parameters required for optimization calculation:

[0031] Load demand data: typical annual hourly (8760 hours) load power sequence P of the target power supply area load (t), unit: kW.

[0032] Regional climate data: typical meteorological year (TMY) data of the project location, including hourly horizontal total irradiance G(t) (unit: W / m 2 ) and ambient temperature T a (t) (unit: ℃).

[0033] Economic parameters: including photovoltaic component unit power cost (Yuan / kW), battery unit capacity cost (Yuan / kWh), diesel generator unit power cost (Yuan / kW), system balancing component (inverter, controller, cable, etc.) cost (Yuan), system annual maintenance cost (unit: Yuan / year), discount rate d, project life cycle N (years).

[0034] Technical parameters: including photovoltaic component efficiency , inverter efficiency , battery charging efficiency , battery discharging efficiency , maximum battery charging power , maximum battery discharging power , upper and lower limits of battery state of charge , , maximum allowable battery discharge depth (This parameter is used to limit the lower limit of SOC operation to protect the battery, and is also a key input for evaluating battery life), diesel generator idle or idling fuel consumption coefficient (unit: L / kWh), diesel generator marginal fuel consumption coefficient (unit: L / kWh), minimum operating power ratio of diesel generator (percentage of rated power) .

[0035] Design constraint index: maximum allowable load power shortage rate .

[0036] S2: Optimization model establishment

[0037] Decision variable definition: define the system configuration parameters to be optimized as the decision vector , where: Pnominal is the nominal power of the photovoltaic array (kW), Pnominal is the nominal power of the photovoltaic array (kW), Pnominal is the nominal power of the photovoltaic array (kW).

[0038] Objective function construction: the objective function is to minimize the system's levelized cost of energy (LCOE) over the whole life cycle. This objective function comprehensively covers the initial investment cost, operation and maintenance cost, fuel cost, equipment replacement cost and residual value of the system.

[0039]

[0040] wherein:

[0041] Cinitial represents the initial investment cost, .

[0042] Cannual represents the annual maintenance cost of the system, which can be calculated according to the historical values of previous years.

[0043] Cfuel represents the total fuel cost present value, , Cfuel,y represents the average price of fuel in the yth year, which is calculated according to the historical values of previous years, T represents the length of each time step, Cfuel represents the average fuel consumption per hour (unit: L / h), in the system operation simulation, for any given candidate configuration scheme, the diesel generator's rated power is , wherein, Pnominal is the rated power of the diesel generator in the candidate scheme (i.e. one of the decision variables), Pactual represents the actual output power of the diesel generator (KW) at simulation time t.

[0044] Cbattery represents the battery replacement cost present value, N represents the number of times of battery replacement, bat and the replacement year t k are calculated by the battery life model based on the depth of discharge (DoD).

[0045] Cresidual represents the residual value present value, SALV represents the nominal residual value of the entire system at the end of the project.

[0046] Ptotal,y represents the total load power supply in the yth year, assuming that the load power remains stable in each year during the life cycle of the project, .

[0047] Constraint setting:

[0048] 1. System power supply reliability constraint:

[0049]

[0050] where, denotes the length of each time step, denotes the load power at time t (unit: kW), denotes the actual output power of the photovoltaic array at time t (unit: kW), denotes the output power of the diesel generator at time t (unit: kW), denotes the discharge power of the battery at time t (unit: kW), denotes the charging power of the battery at time t (unit: kW).

[0051] 2. Battery operation constraint:

[0052] Power constraint: and

[0053] SOC dynamic constraint: , denotes the state of charge of the battery at time t.

[0054] SOC range constraint:

[0055] Depth of discharge (DoD) constraint:

[0056] Life model and replacement logic constraint: The end-of-life condition of the battery is determined by its cumulative cycle aging degree. Usually, a life model based on depth of discharge (DoD) is used for estimation.

[0057]

[0058] Reference cycle number at a specific depth of discharge (DoD) provided by the battery manufacturer. When the replacement logic is triggered.

[0059] wherein the calculation of the life model and replacement logic constraint is real-time calculation, when the value of SOC(1) is assigned to SOC(t) at the current time t, and the replacement logic calculation starts from the current time t; the calculation of the SOC dynamic constraint, the SOC range constraint, the depth of discharge (DoD) constraint, and the life model and replacement logic constraint is restarted.

[0060] 3. Diesel generator output power constraint: or , Typical value range is .

[0061] 4. Photovoltaic output model: where:

[0062] , represents the working temperature of the photovoltaic panel at time t, represents the temperature under standard test conditions (usually 25℃) , represents the power temperature coefficient, taking values in the range [-0.20% / ℃, -0.50% / ℃], NOCT represents the nominal panel working temperature (usually 45℃).

[0063] S3: Optimization solution

[0064] Intelligent optimization algorithms (such as genetic algorithms) are used to solve the above model:

[0065] a) Algorithm initialization: set algorithm parameters (population size, iteration number, crossover mutation probability, etc.).

[0066] b) Fitness evaluation: for each candidate solution X i =

, ,

[0067] Load power calculation: from the input load data;

[0068] Photovoltaic output calculation: calculated according to the photovoltaic model in S2 based on the input meteorological data;

[0069] Battery charging power and discharging power : first complete SOC state check; charging and discharging power decision (execute preset energy management strategy), preferentially use photovoltaic power generation, then dispatch the battery, and finally start and stop the diesel generator. When , photovoltaic surplus, diesel generator does not work, the remaining power is used to charge the battery (must meet the battery operation constraint); when , photovoltaic is insufficient, supplemented by battery discharging, start diesel generator when battery power is insufficient;

[0070] Diesel generator output power : start-stop judgment: when , start the diesel generator to charge the load and the battery (subject to the diesel generator output power constraint);

[0071] Energy shortage calculation: after the above power distribution, if the load cannot be satisfied, there is a power shortage;

[0072] Update the battery SOC and the battery replacement judgment;

[0073] After the simulation is completed, it is judged whether the reliability index LPSP meets the hard constraint , in addition, whether it meets the battery operation constraint, if it meets, calculate the objective function value f(X i ) under this configuration; if not, record the objective function value f(X i ) under this configuration as 0;

[0074] c) Iterative optimization: the algorithm selects, crosses and mutates according to the objective function value, generates a new generation of population, and repeats step b) until the convergence condition is met.

[0075] d) Result output: output the global optimal solution and its corresponding performance index.

[0076] The global optimal solution is the scheme with the minimum objective function value under the premise that the objective function value is not 0;

[0077] As shown in the following, a remote area microgrid project is taken as an example to illustrate the present application in detail. It should be emphasized that the present embodiment is only used for fully disclosing and understanding the present application, and does not limit the protection scope of the present application. Figure 1

[0078] S1 input parameters:

[0079] Load: peak power 80kW, annual total electricity consumption about 288MWh.

[0080] Location: an island in the East China Sea (north latitude 30.0°, east longitude 123.0°), using the TMY data of this place (using NASA-SSE database).

[0081] Economic parameters: =3000 yuan / kW, =1600 yuan / kWh, =1800 yuan / kW, =400000 yuan, =10000 yuan / year, d=6%, N=25 years, ,​ = … =8.0 liter / L, SALV is 5% of initial investment.

[0082] Technical parameters: =18%, =98%, = =96%, =200 kW, =200 kW, , , SOC(1)=0.5, , =0.08 L / kWh, =0.24 L / kWh, =0.35.

[0083] Design constraints: =0.5%.

[0084] S2 & S3 optimization process and results: as shown in Figure 3 , the population-based intelligent algorithm is used for optimization solution, and the specific parameter settings are as follows:

[0085] Chromosome coding: real number coding is used, and the chromosome structure

, ,

[0086] A candidate configuration

, ,

[0087] That is, in the chromosome, pv bit binary represents the rated power of photovoltaic array, bat bit binary represents the rated capacity of battery, and gen bit binary represents the rated power of diesel generator, and the binary codes of the three are concatenated; after generating a new population each time, it is necessary to judge whether the three decision variables corresponding to each individual meet the search range, if not, randomly change the binary code to meet the search range setting; ; ; ;

[0088] The population size is set to 100, and the maximum iteration number is 500.

[0089] With the above configuration, the genetic optimization algorithm is run. In the simulation process, the life model and replacement logic constraints are calculated according to the hourly charging and discharging data and the SOC data, and then the replacement cost is calculated . Finally, the optimal configuration is obtained as:

[0090] Photovoltaic capacity: = 380 kW

[0091] Battery capacity: = 2200 kWh

[0092] Diesel generator power: = 120 kW

[0093] The verification simulation of the scheme is carried out, the full life cycle normalized cost (LCOE) is 1.18 yuan / kWh, and the annual load power supply shortage rate (LPSP) is 0.42%, which fully meets the design requirements. Compared with the initial experience scheme (400 kW, 2500 kWh, 150 kW), the total cost is reduced by about 12% under the premise of ensuring higher reliability.

[0094] The above examples prove that the method of the present application can efficiently and reliably design a photovoltaic storage oil comprehensive power supply system with good technical and economic performance.

[0095] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any change or replacement that can be easily thought of by those skilled in the art within the technical range disclosed by the present application should be covered within the protection scope of the present application.

Claims

1. A method for optimizing design of an integrated power supply system of optical storage oil considering full-cycle multi-constraints, characterized in that, The method comprises the following steps: Step S1, input parameter preparation: obtaining and inputting basic parameters including load demand data, regional climate data, equipment technical and economic parameters, and power supply reliability indicators; the equipment technical and economic parameters include photovoltaic module efficiency, operation and maintenance cost, and battery life model parameters; Step S2, optimization model establishment: taking the rated power of the photovoltaic array, the rated capacity of the battery, and the rated power of the diesel generator as decision variables, combining the input parameters of step S1, taking the minimization of the system full life cycle levelized cost LCOE as the objective function, and setting multiple constraint conditions including power supply reliability constraints, energy storage system operation constraints, and diesel generator operation constraints; Step S3, optimization solution: generating a candidate configuration scheme by using an intelligent optimization algorithm, performing hourly annual operation simulation on each candidate scheme, judging whether the multiple constraint conditions are met, further calculating the corresponding objective function value, and finally iteratively optimizing the candidate scheme according to the objective function value until the convergence condition is met; Step S4, result output: outputting the system configuration scheme that meets the multiple constraint conditions and optimizes the objective function: namely, the corresponding rated power of the photovoltaic array, the rated capacity of the battery, and the rated power of the diesel generator. 2.The method of claim 1, wherein, The regional climate data in step S1 includes hourly horizontal total irradiance and ambient temperature data under a typical meteorological year. 3.The method of claim 1, wherein, The minimization of the system full life cycle levelized cost LCOE includes initial investment, operation and maintenance, fuel cost, equipment replacement cost, and residual value. 4.The method of claim 1, wherein, The power supply reliability constraint in step S2 is that the load power shortage rate LPSP needs to be less than or equal to a set maximum allowable value of the power supply reliability index LPSP max ; the LPSP value is obtained by hourly simulation calculation for 8760 hours in a year.

5. The method for optimal design of integrated power supply system of optical storage oil considering full-cycle multi-constraints according to claim 1, characterized in that, The energy storage system operation constraints in step S2 include a battery life model based on discharge depth; the battery life model is used to dynamically estimate the battery life and determine the replacement cost according to the discharge depth during operation. The energy storage system operation constraints include battery state of charge SOC constraints, charge and discharge power constraints, and SOC dynamic update model constraints based on energy conservation. 6.The method of claim 1, wherein, The diesel generator operation constraint in the step S2 is that its output power cannot be lower than the minimum operation power proportion k of its rated power gen.min to avoid its operation in the low efficiency region. 7.The method of claim 1, wherein, The intelligent optimization algorithm in step S3 is a genetic algorithm, a particle swarm optimization algorithm, or a differential evolution algorithm. 8.The method of claim 1, wherein, The annual operation simulation in step S3 follows a predetermined energy management strategy, and the priority order is: preferentially using photovoltaic power generation, supplementing by battery discharge when photovoltaic power generation is insufficient, and starting the diesel generator when the battery power is insufficient.

Images