CN-121984102-A - Capacity configuration method and system for photovoltaic-photo-thermal hybrid system

CN121984102ACN 121984102 ACN121984102 ACN 121984102ACN-121984102-A

Abstract

The application discloses a capacity configuration method and system of a photovoltaic-photo-thermal hybrid system, and belongs to the technical field of new energy power generation and power system optimization. The method solves the technical problems that the high-proportion renewable energy base is difficult to consume surplus photovoltaic power, the utilization rate of a photo-thermal system is low, and the economical efficiency and the reliability of a single optimization target are difficult to consider in a local consumption scene. Firstly, constructing a multi-objective capacity configuration optimization model which aims at minimizing the electrical cost of the leveling degree and minimizing the load loss probability and takes the photovoltaic capacity, the heat storage duration and the like as decision variables. And then, adopting a multi-objective Bayesian optimization algorithm based on a Gaussian process to carry out efficient solution on the model. And finally, outputting a group of pareto optimal capacity configuration schemes reflecting the optimal trade-off of the cost and the reliability, and providing flexible selection for planning decisions. The application can obviously improve the economical efficiency, the reliability and the calculation efficiency of the system.

Inventors

DING YING
TANG AIPING
LIN JIE
WEI SIQI
WANG BANGYAN
WU BO
WANG XIULI
MA QIYUE
ZHANG LIKAI
LIU LI
CHENG LIU
QIU JUNHUA

Assignees

西安交通大学
珠海电力设计院有限公司

Dates

Publication Date: 20260505
Application Date: 20260408

Claims (10)

1. The capacity configuration method of the photovoltaic-photo-thermal hybrid system is characterized by comprising the following steps of: building a photovoltaic-photo-thermal hybrid system capable of performing annual operation simulation on an hour-by-hour scale, wherein the photovoltaic-photo-thermal hybrid system comprises a photovoltaic power station, a photo-thermal power station with a fused salt thermal energy storage system, a storage battery and an electric heater; A local consumption cooperative operation strategy is prepared based on the photovoltaic-photo-thermal hybrid system, and under the local consumption cooperative operation strategy, the leveling degree electricity cost and the load loss probability are calculated according to the light rejection amount and the load loss amount; The method comprises the steps of taking the minimized leveling degree electric cost and the minimized load loss probability as dual optimization targets, determining the rated capacity of a photovoltaic power station, the capacity of a storage battery and the thermal energy storage time of a fused salt heat energy storage system as decision variables, setting the value constraint range of each decision variable, and constructing a multi-target capacity configuration optimization model of the photovoltaic-photo-thermal hybrid system; Generating initial sample points by Latin hypercube sampling, constructing a proxy model of a multi-target capacity configuration optimization model by Gaussian process regression, iterating an optimization acquisition function to adaptively select new sample points and update the proxy model until the iteration termination condition is met, and outputting a capacity configuration scheme set with pareto optimal characteristics.
2. The capacity configuration method of the photovoltaic-photo-thermal hybrid system according to claim 1, wherein the photo-thermal power station with the molten salt thermal energy storage system adopts a solar tower power generation system, the solar tower power generation system comprises heliostat fields which are arranged in a radial staggered mode, an endothermic tower is arranged in the center of each heliostat field, a receiver is arranged at the top of each heliostat field, and the electric heater is connected between the photovoltaic power station and the molten salt thermal energy storage system and used for converting photovoltaic surplus electric energy into heat energy and storing the heat energy in the molten salt thermal energy storage system.
3. The method of capacity allocation of a photovoltaic-photo-thermal hybrid system according to claim 2, wherein during the heat transfer between the receiver and the electric heater, the equations of mass and energy balance of molten salt in the hot tank and the cold tank of the receiver are as follows: Wherein, the 、 The molten salt is the mass of molten salt in the hot tank and the cold tank; 、 loading and unloading mass flows, respectively; the molten salt mass flow is heated by an electric heater; 、、 The temperatures of the hot tank, the receiver outlet and the electric heater outlet are respectively; Is the specific heat capacity of constant pressure.
4. The method for configuring capacity of a photovoltaic-photothermal hybrid system according to claim 1, wherein the solar tower power generation model comprises a heliostat field and a receiver, and incident solar energy is reflected by the heliostat field consisting of thousands of heliostats and converged to the receiver at the top of the heat absorption tower; The photovoltaic power generation model comprises a photovoltaic power station, and the power generation power of the photovoltaic power station The calculation is as follows: Wherein, the The number of the photovoltaic modules; An effective area for a single component; is the total irradiance of the horizontal plane; Is the efficiency of the photovoltaic module; Is inverter efficiency; Is a derating coefficient; the energy balance of the storage battery considers the self-discharge rate Efficiency of charging And discharge efficiency The energy balance is: Wherein, the 、 Respectively discharging power and charging power; For the time step size of the time step, Is the energy of the storage battery at the time t+1, Is the energy of the storage battery at the time t, The self-discharge efficiency of the storage battery is achieved; 、 Respectively charging efficiency and discharging efficiency, and t is time.
5. The photovoltaic-photothermal hybrid system capacity configuration method of claim 1, wherein the local absorption co-operation strategy comprises: The method comprises the steps of firstly utilizing the output of a photovoltaic power station to meet local load, discharging and supplementing the photovoltaic power station and a storage battery when the output of the photovoltaic power station is insufficient, charging the storage battery when the output of the photovoltaic power station is excessive, and converting electric energy into heat energy through the electric heater and storing the heat energy in the fused salt heat energy storage system when the storage battery is still excessive after the storage battery is charged.
6. The method for configuring capacity of a photovoltaic-photo-thermal hybrid system according to claim 1, wherein calculating the leveling electrical cost and the load shedding probability according to the amount of light shedding and the amount of load shedding under the local consumption cooperative operation strategy comprises: S0, inputting meteorological conditions and local load data, initializing state variables and statistics of a system, and setting simulation starting time T=1; s1, for each simulation time T, executing the following steps: S11, energy supply and load matching: (a) The photovoltaic power station gives out power to be preferentially supplied to a local load, so that the residual load power and the photovoltaic surplus power at the moment are obtained; (b) If the residual load power is larger than zero, the available heat energy in the fused salt heat energy storage system is preferentially utilized to drive the photo-thermal power station to generate electricity so as to supplement the load; (c) If the residual load is still left after the completion of the step (b), recording the power of the residual load as the load losing quantity at the current moment; And S12, processing the surplus energy: (d) If the photovoltaic surplus power exists, the photovoltaic surplus power is preferentially used for charging the storage battery; (e) After the storage battery is charged, if photovoltaic surplus power still exists and the fused salt heat energy storage system is not full, converting the part of electric energy into heat energy through an electric heater and storing the heat energy into the fused salt heat energy storage system; (f) If the photovoltaic surplus power still exists after the step (e) is finished, recording the photovoltaic surplus power as the light discarding quantity at the current moment; s13, updating state variables and statistics of the system, and entering the next simulation time T+1; And S2, after all the moments in the preset simulation period are traversed, calculating the load loss probability and the leveling degree electricity cost based on the accumulated load loss quantity and the total system cost.
7. The photovoltaic-photo-thermal hybrid system capacity configuration method according to claim 1, wherein the minimizing the electrical cost of the leveling degree and the minimizing the probability of load shedding are in a double optimization objective: Probability of no load The expression is: Wherein, the Is the first An hour load demand; 、、 the output power of the photovoltaic, the photo-thermal unit and the storage battery are respectively; Leveling degree electric cost The expression is: Wherein, the Is the installation cost, consisting of direct and indirect capital costs; is the annual cost of the product, Is the reset cost of the device; 、、 The generated energy of the first year of the photo-thermal unit, the generated energy of light Fu Di for one year and the generated energy of the first year of the storage battery are respectively, 、、 The annual attenuation rate of the photo-thermal unit, the annual attenuation rate of the photovoltaic unit and the annual attenuation rate of the storage battery are respectively; Is the discount rate; And The life expectancy of the photovoltaic-photo-thermal hybrid model containing heat energy storage and the storage battery are respectively, n is the first year number and m is the second year number.
8. The method for configuring the capacity of the photovoltaic-photo-thermal hybrid system according to claim 1, wherein the system operation physical constraints of the multi-objective capacity configuration optimization model of the photovoltaic-photo-thermal hybrid system comprise photovoltaic output constraints, storage battery operation constraints and photo-thermal and heat storage system operation constraints; The storage battery operation constraint comprises an energy balance constraint, a capacity limit constraint and a power limit constraint; The operation constraint of the photo-thermal and heat storage system comprises energy collection constraint, heat energy balance constraint, temperature limit constraint and heat storage capacity constraint.
9. The method for configuring capacity of a photovoltaic-photo-thermal hybrid system according to claim 1, wherein generating initial sample points by using latin hypercube sampling, constructing a proxy model of a multi-objective capacity configuration optimization model by gaussian process regression, iterating an optimization acquisition function to adaptively select new sample points and update the proxy model until an iteration termination condition is satisfied, and outputting a capacity configuration scheme set with pareto optimal characteristics, comprising: Generating an initial sample point set in the value range of the decision variable, obtaining an objective function value through simulation to form an initial training data set, and establishing a Gaussian process proxy model for each objective function based on the data set; performing iterative optimization, and in each iteration, performing the following substeps: selecting a next candidate sample point to be evaluated by optimizing an acquisition function based on a current Gaussian process agent model; Inputting the selected candidate sample points into the multi-objective capacity configuration optimization model for simulation to obtain the actual objective function value; adding the new sample points and the objective function values thereof into a training data set, and updating a Gaussian process agent model; and repeating iterative optimization until a preset termination condition is met, and outputting the finally obtained non-dominant solution set as a capacity configuration scheme set with pareto optimal characteristics.
10. A photovoltaic-photothermal hybrid system capacity configuration system, comprising: The photovoltaic-photo-thermal hybrid system building module is used for building a photovoltaic-photo-thermal hybrid system capable of performing annual operation simulation on an hour-by-hour scale, and the photovoltaic-photo-thermal hybrid system comprises a photovoltaic power station, a photo-thermal power station with a fused salt thermal energy storage system, a storage battery and an electric heater; The operation strategy making module is used for making a local consumption cooperative operation strategy based on the photovoltaic-photo-thermal hybrid system, and calculating the leveling degree electric cost and the load loss probability according to the light rejection amount and the load loss amount under the local consumption cooperative operation strategy; the optimization model construction module is used for determining the rated capacity of the photovoltaic power station, the capacity of the storage battery and the thermal energy storage time of the fused salt heat energy storage system as decision variables by taking the minimized leveling degree electric cost and the minimized load loss probability as dual optimization targets, setting the value constraint range of each decision variable and constructing a multi-target capacity configuration optimization model of the photovoltaic-photo-thermal hybrid system; and the agent model iteration module is used for generating initial sample points by Latin hypercube sampling, constructing an agent model of the multi-target capacity configuration optimization model by Gaussian process regression, iterating and optimizing an acquisition function to self-adaptively select new sample points and update the agent model until the iteration termination condition is met, and outputting a capacity configuration scheme set with pareto optimal characteristics.

Description

Capacity configuration method and system for photovoltaic-photo-thermal hybrid system Technical Field The application belongs to the technical field of new energy power generation and power system optimization, and particularly relates to a capacity configuration method of a photovoltaic-photo-thermal hybrid system. Background The photovoltaic power generation has the advantages of short construction period, quick cost reduction and the like, but the output characteristic of the photovoltaic power generation is highly dependent on solar irradiance, has obvious volatility and non-schedulability, and is difficult to independently bear long-term continuous power supply tasks. In contrast, solar thermal power generation (Concentrated solar power, CSP) configured with the molten salt thermal energy storage system can convert intermittent solar energy into schedulable electric energy through a heat collection-heat storage-power generation link, smooth renewable energy output fluctuation to a certain extent, and improve capacity factor and power supply reliability of the system. Therefore, the photovoltaic and the photo-thermal system are integrated, the complementary advantages of low photovoltaic cost and photo-thermal schedulability are fully exerted, and the method is an important development direction for improving the comprehensive utilization efficiency of solar energy. Existing researches show that under different technical routes and typical working conditions, the integrated PV-CSP system can remarkably improve the operation characteristics of the system. For example, by changing the key parameters of photovoltaic capacity, CSP capacity, solar power, thermal energy storage scale and the like, the leveling degree electric cost and the capacity factor can be considered in a certain range, and under the condition of the given base load requirement, the photo-thermal unit and the heat storage system are reasonably configured, so that the effective capacity and the economy of the system are improved. On the other hand, the storage battery and the electric heater are introduced into the integrated system, and an electric-thermal cooperative energy storage structure is adopted, so that the photovoltaic power station can be charged with electricity preferentially when the output of the photovoltaic power station is surplus, and after the storage battery is close to saturation, the residual electric energy is converted into heat energy through the electric heater and stored into the thermal energy storage tank, so that the electricity discarding level is reduced to a certain extent, and the flexible scheduling capability of the system is enhanced. However, the following disadvantages still exist in the capacity configuration and operation optimization of the photovoltaic-photo-thermal hybrid system in the prior art: (1) In the setting of load scenes and operation modes, most of work focuses on the analysis of constant base load or simplified load curves, the operation characteristics of typical renewable energy bases facing actual power requirements and constraint conditions in a local consumption scene are not fully considered, the influence of actual load fluctuation and power consumption behaviors on system configuration is difficult to accurately reflect, the capacity configuration scheme is easily mismatched with the actual power consumption requirements, the problems of insufficient power supply reliability caused by small installed capacity, investment redundancy caused by large installed capacity, high local power rejection rate and the like are easily caused. (2) In terms of capacity configuration methods, the prior researches mostly adopt parameter scanning or traditional heuristic algorithms (such as genetic algorithms and the like) to jointly optimize photovoltaic, photo-thermal and energy storage units. The method generally needs larger population scale and iteration times to obtain a more ideal solution set, has higher calculation cost, is difficult to extract sufficient information from limited simulation times in time when facing a complex energy flow model based on sequential simulation, and is sensitive to calculation resources in an optimization process. (3) In the aspect of the depiction of the multi-objective trade-off relationship, a plurality of indexes such as economy, reliability and the like are considered in part of the research, but the multi-objective problem is often converted into a single-objective problem by means of weight summation, layering optimization and the like, the influence of subjective weight setting is easy to be received, the pareto trade-off relationship between key indexes such as load loss probability, leveling degree electricity cost and the like is difficult to be systematically and clearly represented, and the proper configuration scheme is not easy to be selected by planners according to different decision preferences. Disclosure of Invention Aiming at the tec