Optimal configuration method for wind, solar and geothermal integrated energy system considering supply and demand randomness

Abstract

The present invention discloses a method for optimizing the configuration of a wind, solar, and geothermal energy integrated energy system considering the randomness of supply and demand. The system is composed of a number of units, and the method includes two layers of optimization: the first layer uses a parameter selection algorithm to generate unit parameters; the second layer uses an approximate dynamic programming method to solve the random programming results of the wind, solar, and geothermal energy integrated energy system considering the randomness of supply and demand under given parameters; the first layer parameter selection algorithm generates new unit capacity parameters according to the second layer random programming results, and the second layer random programming method solves again according to the new unit capacity parameters until the capacity parameters of each unit are traversed or the capacity parameters of each unit converge, and the optimized expected configuration parameters of the wind, solar, and geothermal energy integrated energy system are obtained. The present invention uses an approximate dynamic programming method to solve the random programming problem of the wind, solar, and geothermal energy integrated energy system to shorten the design cycle of the wind, solar, and geothermal energy integrated energy system and improve the adaptability of the system to supply and demand fluctuations during operation.

Description

Technical Field

[0001] The present invention relates to an integrated energy system composed of wind energy, solar energy and geothermal energy and related technical fields, and in particular to an optimization configuration method for an integrated energy system of wind, solar and geothermal energy taking into account the randomness of supply and demand. Background Art

[0002] Using wind, solar and geothermal energy to form an efficient integrated energy system can meet the electricity, heating and cooling needs of a building or region while reducing carbon emissions.

[0003] In the integrated energy system composed of wind, solar and geothermal energy, due to the volatility and randomness of renewable energy and loads such as cooling, heating and electricity, it is impossible to find a set of parameters that minimize the total cost of the system over its entire life cycle under all source and load scenarios when planning and designing various types of equipment in the system. Therefore, in the early planning stage of the system, in order to obtain highly adaptable system design parameters, it is necessary to consider the impact of supply and demand randomness on system operation, which involves the stochastic planning problem of the integrated energy system. Due to the large number of parameters to be designed, system control and state variables, solving the stochastic planning energy management of the integrated energy system of wind, solar and geothermal energy will greatly increase the computational complexity of the design stage. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and provide an optimization configuration method for a wind, solar, and geothermal integrated energy system taking into account the randomness of supply and demand. The approximate dynamic programming method is used to solve the random programming problem of the wind, solar, and geothermal integrated energy system, so as to shorten the design cycle of the wind, solar, and geothermal integrated energy system and improve the system's adaptability to supply and demand fluctuations during operation.

[0005] To achieve the above object, the technical solution of the present invention is:

[0006] A method for optimizing the configuration of a wind, solar and geothermal integrated energy system considering the randomness of supply and demand, wherein the system is composed of a plurality of units and the method includes two levels of optimization:

[0007] The first layer uses parameter selection algorithm to generate unit parameters;

[0008] The second layer adopts the approximate dynamic programming method to solve the stochastic programming results of the wind, solar and geothermal integrated energy system considering the randomness of supply and demand under given parameters;

[0009] The first-level parameter selection algorithm generates new unit capacity parameters according to the second-level random programming results. The second-level random programming method solves again according to the new unit capacity parameters until the capacity parameters of each unit are traversed or the capacity parameters of each unit converge, and the optimized expected configuration parameters of the wind, solar and geothermal integrated energy system are obtained.

[0010] Furthermore, the second layer adopts an approximate dynamic programming method including:

[0011] S1: Establish a nonlinear efficiency model for wind, solar and geothermal integrated energy systems;

[0012] S2: Generate random source-load scenarios for wind, solar and geothermal integrated energy systems;

[0013] S3: Constructing the stochastic programming problem of energy management for wind, solar and geothermal integrated energy systems;

[0014] S4: Approximate dynamic programming is used to iteratively solve the approximate value function in random scenarios.

[0015] Furthermore, the wind, solar and geothermal integrated energy system includes the following units: wind turbine, photovoltaic, ground source heat pump, battery, hot water storage tank or cold water storage tank.

[0016] Furthermore, in step S1, the nonlinear operating characteristics of battery energy storage and ground source heat pump are considered.

[0017] Furthermore, in step S2, the historical operating data of wind turbines, photovoltaics, air temperature, soil temperature, and cooling, heating, and electricity loads are used to generate random working curves of wind power, photovoltaics, geothermal energy, and cooling, heating, and electricity loads in the wind-solar-geothermal integrated energy system using the Latin supersampling method. The time span of the curve is year, month, and day, and the sampling period is minute and hour.

[0018] Furthermore, in step S3, the stochastic programming problem of energy management of the wind, solar and geothermal integrated energy system is expressed in Bellman recursive form.

[0019] Furthermore, in step S4, when approximate dynamic programming is used to solve the stochastic planning energy management of the wind, solar and geothermal integrated energy system, the charge state of the battery and the heat storage or cold storage tank is used as the state variable, a three-dimensional value function is constructed, and then the source-load random working curve generated by S2 is iterated in turn to obtain the global optimal approximate solution of the stochastic planning energy management of the wind, solar and geothermal integrated energy system.

[0020] Furthermore, the three-dimensional value function is implemented in the form of a three-dimensional data table, or is approximated as the sum of a value function with the battery state of charge as a state variable and a value function with the heat storage or cold storage tank state of charge as a state variable.

[0021] Furthermore, in step S4, when approximate dynamic programming is used to solve the stochastic planning energy management of the wind, solar and geothermal integrated energy system, when the hot water storage tank and the cold water storage tank exist at the same time, a three-dimensional value function is constructed to form a four-dimensional data table, or the value function is approximated to the sum of a value function with the battery state of charge as a state variable, a value function with the hot water storage tank state of charge as a state variable, and a value function with the cold water storage tank state of charge as a state variable, and then the value function is used to iteratively solve.

[0022] Furthermore, the parameter selection algorithm is a grid generation method or a heuristic algorithm.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] The present invention adopts an approximate dynamic programming method to solve the stochastic programming problem of a wind, solar, and geothermal integrated energy system, so as to shorten the design cycle of the wind, solar, and geothermal integrated energy system and improve the adaptability of the system to supply and demand fluctuations during operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a schematic diagram of the composition of the wind, solar and geothermal integrated energy system;

[0026] Figure 2 A flow chart of a method for optimizing configuration of a wind-solar-geothermal integrated energy system taking into account the randomness of supply and demand provided by an embodiment of the present invention;

[0027] Figure 3 This is a flow chart of the second-level solution of the energy management problem of a wind, solar and geothermal integrated energy system taking into account the randomness of supply and demand using approximate dynamic programming. DETAILED DESCRIPTION

[0028] Example:

[0029] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0030] See also Figure 1 As shown, the wind, solar and geothermal integrated energy system includes but is not limited to the following units: wind turbines, photovoltaics, ground source heat pumps, batteries, hot water storage tanks or cold water storage tanks.

[0031] See also Figure 2As shown, the method for optimizing the configuration of a wind, solar, and geothermal energy integrated energy system considering the randomness of supply and demand provided in this embodiment includes two levels of optimization: the first level uses a parameter selection algorithm such as a grid partitioning method or a heuristic algorithm to generate unit parameters such as battery capacity, heat storage or cold storage tank capacity, wind turbine power capacity, photovoltaic power capacity, and ground source heat pump unit power capacity; the second level uses an approximate dynamic programming method to quickly solve the random planning results of the wind, solar, and geothermal energy integrated energy system considering the randomness of supply and demand under given parameters; the first level parameter selection algorithm generates new unit capacity parameters based on the second level random planning results; the second level random programming method solves again based on the new unit capacity parameters until the capacity parameters of each unit are traversed or the capacity parameters of each unit converge, and the optimized expected configuration parameters of the wind, solar, and geothermal energy integrated energy system are obtained.

[0032] It can be seen that this method uses the approximate dynamic programming method to solve the stochastic programming problem of the wind, solar and geothermal integrated energy system, so as to shorten the design cycle of the wind, solar and geothermal integrated energy system and improve the system's adaptability to supply and demand fluctuations during operation.

[0033] In one specific embodiment, see Figure 3 As shown, the second layer of the wind-solar-geothermal integrated energy system optimization configuration method considering the randomness of supply and demand adopts approximate dynamic programming to solve the energy management problem of the wind-solar-geothermal integrated energy system considering the randomness of supply and demand, including the following steps:

[0034] S1: Establish a nonlinear efficiency model for wind, solar and geothermal integrated energy systems;

[0035] S2: Generate random source-load scenarios for the wind, solar, and geothermal integrated energy system, with a time span including but not limited to years, months, and days;

[0036] S3: Constructing the stochastic programming problem of energy management for wind, solar and geothermal integrated energy systems;

[0037] S4: Approximate dynamic programming is used to iteratively solve the approximate value function in random scenarios.

[0038] In a specific embodiment, the wind, solar and geothermal integrated energy system in S1 takes into account the nonlinear working characteristics of equipment such as battery energy storage and ground source heat pumps.

[0039] In a specific embodiment, the S2 uses historical operation data of wind turbines, photovoltaics, air temperature, soil temperature, cold, heat and electricity loads, etc., and can use Latin supersampling and other methods to generate random working curves of wind power, photovoltaics, geothermal, cold, heat and electricity loads of the wind\solar\geothermal integrated energy system. The time span of the curve can be years, months, days, etc., and the sampling period can be minutes, hours, etc.;

[0040] In a specific embodiment, the stochastic programming problem of energy management of the wind, solar and geothermal integrated energy system in S3 can be expressed in Bellman recursive form.

[0041] In a specific embodiment, when S4 uses approximate dynamic programming to solve the stochastic planning energy management of the wind, solar and geothermal integrated energy system, the charge state of the battery and the heat storage or cold storage water tank is used as the state variable to construct a three-dimensional value function. The value function can be implemented in the form of a three-dimensional data table, or it can be approximated as the sum of the value function with the battery charge state as the state variable and the value function with the heat storage or cold storage water tank charge state as the state variable. The value function is then iterated on the source-load random working curve generated by S2 in turn to obtain the global optimal approximate solution of the stochastic planning energy management of the wind, solar and geothermal integrated energy system.

[0042] In a specific embodiment, when S4 uses approximate dynamic programming to solve the stochastic planning energy management of the wind, solar and geothermal integrated energy system, when the hot water storage tank and the cold water storage tank exist at the same time, a three-dimensional value function can be constructed to form a four-dimensional data table, or the value function can be approximated to the sum of a value function with the battery state of charge as a state variable, a value function with the hot water storage tank state of charge as a state variable, and a value function with the cold water storage tank state of charge as a state variable, and then the value function is used to iteratively solve.

[0043] Below, the planning and design of the wind-solar integrated energy system is used as an example to further illustrate this method:

[0044] The operation model of the wind-solar integrated energy system can be expressed as:

[0045] min J (1-1)

[0046] subject to:

[0047] J=J pv +J wt +J b +J ct +J ht +J ph,c +J ph,h +J o (1-2)

[0048]

[0049]

[0050]

[0051]

[0052]

[0053]

[0054]

[0055] J o (p e ∑P g (k)Δt (1-10)

[0056] P.S pv (k)R s (t)×A pv ×η pv (1-11)

[0057] P.S wt (k)N wt ×P wt,base (v) (1-12)

[0058] E b (k)(1-τ b )E b (k-1)+x b P.S b,ch (k)Δtη b,ch +(1-x b )P b,dis (k)Δt / η b,dis (1-13)

[0059] E ct (k)(1-τ ct )E ct (k-1)+x ct P.S ct,ch (k)Δtη ct,ch +(1-x ct )P ct,dis (k)Δt / η ct,dis (1-14)

[0060] E ht (k)(1-τ ht )E ht (k-1)+x ht P.S ht,ch (k)Δtη ht,ch +(1-x ht )P ht,dis (k)Δt / η ht,dis (1-15)

[0061] Q hp,c (k)C cop,c (k)P hp,c (k) (1-16)

[0062] Q hp,h (k)=C cop,h (k)P hp,h (k) (1-17)

[0063] P ph,h (k)+P ph,c (k)+x b P b,ch (k)+(1-x b )P b,ch (k)=P pv (k)+P wt (k)+P g (k)+P d,e (k) (1-18)

[0064] Q hp,c (k)+x ct P ct,dis (k)+(1-x ct )P ct,dis (k)=P d,c (k) (1-19)

[0065] Q hp,h (k)+x ht P ht,dis (k)+(1-x ht )P ht,dis (k)=P d,h (k) (1-20)

[0066] 0≤P b,dis ≤P b,max (1-21)

[0067] 0≤P b,ch ≤P b,max (1-22)

[0068] 0≤P ct,ch ≤P ct,max (1-23)

[0069] 0≤P ct,dis ≤P ct,max (1-24)

[0070] 0≤P ht,ch ≤P ht,max (1-25)

[0071] 0≤P ht,dis ≤P ht,max (1-26)

[0072] 0<P ph,c <P ph,cmax (1-27)

[0073] 0<P ph,h <P ph,hmax (1-28)

[0074] In formula (1), J, J pv , J wt , J b , J ct , J ht and J ph They are the total cost of the wind-solar-geothermal integrated energy system during its life cycle, photovoltaic cost, wind turbine cost, battery cost, cold water storage tank cost, hot water storage tank cost and ground source heat pump cost, J o is the operating cost of the wind, solar and geothermal integrated energy system during its life cycle, p pv , p wt , p b , p ct , p ht and p hp are the unit prices of photovoltaic, wind turbine, battery, cold water storage tank, hot water storage tank and ground source heat pump, respectively. pv,om , p wt,om , p b,om , p ht,om , p ct,om and p hp,om are the unit maintenance prices of photovoltaic, wind turbine, battery, cold water storage tank, hot water storage tank and ground source heat pump, respectively. g is the electricity price, r is the annual profit, n is the expected life cycle of the system, R s is the radiant light intensity, A pv is the photovoltaic area, η pv is the photovoltaic efficiency, N wt is the number of fans, P wt,base is the fan output power, v is the fan speed, P b,max , P ct,max , P ht,max , P ph,cmax and P ph,hmax are the maximum power of the battery, cold water storage tank, hot water storage tank and ground source heat pump for cooling and heating, P b,dis , P b,ch , P ct,dis , P ct,ch , P ht,dis and P ht,ch They are the charging and discharging power, cold storage and release power, and heat storage and release power of the battery, cold water storage tank, and hot water storage tank, respectively. b , x ct and x ht They are the charging and discharging power states of the battery, cold water tank, and hot water tank, respectively. hp,c and Q hp,tare the cooling output power and heating output power of the cooling ground source heat pump and the heating ground source heat pump respectively, C cop,c and C cop,h are cooling energy efficiency and heating energy efficiency, P hp,c and P hp,h are the cooling power and heating power of the ground source heat pump, τ b , τ ct and τ ht are the self-consumption coefficients of the battery, cold water tank and hot water tank, respectively, E b , E ct and E ht are the existing electricity, cold and heat energy of the battery, cold water tank and hot water tank respectively, Δt is the sampling time, η b,ch and η b,dis is the charging and discharging efficiency of the battery, η ct,ch and η ct,dis is the cold power input and output efficiency of the cold storage tank, η ht,ch and η ht,dis is the thermal power input and output efficiency, P d,e , P d,c and P d,h They are the required electricity, cooling and heating load power respectively.

[0075] In formula (1), the efficiency or energy efficiency of various types of equipment can be selected as a constant or a nonlinear function according to needs.

[0076] After establishing the wind, solar and ground source scenarios and the cooling, heating and electricity load scenarios, for any set of design parameters A pv , N wt , P b,max , P ct,max , P ht,max , P ph,cmax and P ph,hmax The feasible solution of , and any random source-load scenario, establish the following stochastic planning problem:

[0077] J o (k-1,E b ,E ct ,E ht )=min(c(E b ,E ct ,E ht )+EJ o (k,E b ,E ct ,E ht )) (2)

[0078] In formula (2), c is the cost function when the system state transitions.

[0079] If the value function is approximated as:

[0080]

[0081] In formula (3), J o_y , y = {b, ct, ht} is the approximate independent value function of various energy storage devices, then the approximate dynamic programming method can be used to quickly iterate and solve the expected system operating cost under the generated source-load scenario set. Combined with formula (1-2), the total cost of the system during its life cycle under the selected design parameters can be obtained.

[0082] If the grid partitioning method is adopted, after the feasible set of system design parameters is grid partitioned, it is only necessary to traverse all parameter combinations on the grid, repeatedly use approximate dynamic programming to solve the operating expectations under the parameter combination, and then obtain the total cost of the parameter combination in the life cycle. Finally, all grid parameters are compared to obtain the optimal solution of formula (1), which greatly reduces the difficulty of solving, shortens the design cycle of the wind, solar and geothermal integrated energy system, and improves the system's adaptability to supply and demand fluctuations during operation.

[0083] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable ordinary technicians in the field to understand the content of the present invention and implement it accordingly, and they cannot be used to limit the protection scope of the present invention. Any equivalent changes or modifications made based on the essence of the content of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for optimizing the configuration of a wind, solar and geothermal integrated energy system considering the randomness of supply and demand, wherein the system comprises a plurality of units, characterized in that: The method involves two levels of optimization: The first layer uses a parameter selection algorithm to generate unit parameters; the parameter selection algorithm is a grid generation method or a heuristic algorithm; The second layer uses the approximate dynamic programming method to solve the stochastic programming results of the wind-solar-geothermal integrated energy system considering the randomness of supply and demand under given parameters; The first-layer parameter selection algorithm generates new unit capacity parameters based on the second-layer random programming results; The second-level random programming method is solved again according to the new unit capacity parameters until the capacity parameters of each unit are traversed or converged, and the optimal expected configuration parameters of the wind, solar and geothermal integrated energy system are obtained; The second layer adopts an approximate dynamic programming method including: S1: Establish a nonlinear efficiency model for wind, solar and geothermal integrated energy systems; S2: Generate random source-load scenarios for wind, solar and geothermal integrated energy systems; S3: Constructing the stochastic programming problem of energy management for wind, solar and geothermal integrated energy systems; S4: Approximate dynamic programming is used to iteratively solve the approximate value function in random scenarios.

2. The method for optimizing the configuration of a wind, solar and geothermal integrated energy system considering the randomness of supply and demand as claimed in claim 1, characterized in that: The wind, solar and geothermal integrated energy system includes the following units: wind turbines, photovoltaics, ground source heat pumps, batteries, hot water storage tanks or cold water storage tanks.

3. The method for optimizing the configuration of a wind, solar and geothermal integrated energy system considering the randomness of supply and demand as claimed in claim 2, characterized in that: In step S1, the nonlinear operating characteristics of battery energy storage and ground source heat pump are considered.

4. The method for optimizing the configuration of a wind, solar and geothermal integrated energy system considering the randomness of supply and demand as claimed in claim 2, characterized in that: In step S2, the historical operating data of wind turbines, photovoltaics, air temperature, soil temperature, and cooling, heating and electricity loads are used to generate random working curves of wind power, photovoltaics, geothermal energy, cooling, heating and electricity loads of the wind-solar-geothermal integrated energy system using the Latin supersampling method. The time span of the curve is year, month, and day, and the sampling period is minute and hour.

5. The method for optimizing the configuration of a wind, solar and geothermal integrated energy system considering the randomness of supply and demand as claimed in claim 1 or 2, characterized in that: In step S3, the stochastic programming problem of energy management of the wind, solar and geothermal integrated energy system is expressed in Bellman recursive form.

6. The method for optimizing the configuration of a wind, solar and geothermal integrated energy system considering the randomness of supply and demand as claimed in claim 4, characterized in that: In step S4, when approximate dynamic programming is used to solve the stochastic planning energy management of the wind, solar and geothermal integrated energy system, the charge state of the battery and the heat storage or cold storage tank is used as the state variable, and a three-dimensional value function is constructed. Then, the source-load random working curve generated in step S2 is iterated in turn to obtain the global optimal approximate solution of the stochastic planning energy management of the wind, solar and geothermal integrated energy system.

7. The method for optimizing the configuration of a wind, solar and geothermal integrated energy system considering the randomness of supply and demand as claimed in claim 6, characterized in that: The three-dimensional value function is implemented in the form of a three-dimensional data table, or is approximated as the sum of a value function with the battery charge state as a state variable and a value function with the heat storage or cold storage water tank charge state as a state variable.

8. The method for optimizing the configuration of a wind, solar and geothermal integrated energy system considering the randomness of supply and demand as claimed in claim 2, characterized in that: In step S4, when approximate dynamic programming is used to solve the stochastic planning energy management of the wind, solar and geothermal integrated energy system, when the hot water storage tank and the cold water storage tank exist at the same time, a three-dimensional value function is constructed to form a four-dimensional data table, or the value function is approximated to the sum of a value function with the battery state of charge as a state variable, a value function with the hot water storage tank state of charge as a state variable, and a value function with the cold water storage tank state of charge as a state variable, and then the value function is used to iteratively solve.

Images