Three-tower one-machine photo-thermal mirror field cooperative control method based on game theory model

Abstract

The invention relates to the field of photo-thermal mirror field control, and discloses a three-tower one-machine photo-thermal mirror field cooperative control method based on a game theory model, and the method comprises the following steps: calculating the power contribution of each heliostat to each heat absorber at a certain moment; according to the obtained power contribution of each heliostat to each heat absorber at a certain moment, a game theory model is constructed by taking a total power value of a maximized heliostat field as a game target and taking a combination scheme that each heliostat is distributed to which heat absorber as a strategy space; the strategy space is updated through iterative optimization until the convergence condition is met, that is, all heat absorbers cannot improve effectiveness through unilateral strategy change, and the final strategy space is obtained; and if falling into local optimum, applying random disturbance to the equilibrium state, and restarting the dynamic process until convergence. According to the method disclosed by the invention, the game theory model is introduced into the cooperative control field of the multi-tower photo-thermal system for the first time, and the overall energy efficiency optimization of the system is realized by calculating the dynamic game behavior between the heat absorption towers.

Description

Technical Field

[0001] The present invention relates to the field of photothermal mirror field control, and in particular to a three-tower-one-machine photothermal mirror field collaborative control method based on a game theory model. Background Art

[0002] Tower-type solar thermal power generation is one of the new clean energy storage methods. It not only has the advantage of low power generation cost, but also can achieve 24-hour continuous power generation compared to photovoltaic power generation. Solar thermal power generation is also used in winter heating, oil extraction and other fields. With the expansion of solar thermal installed capacity, the scale of mirror fields will gradually increase in the future, and large-capacity units will inevitably require the configuration of large-scale mirror fields. Affected by resource conditions and the relevant technical performance of heliostats, when the mirror field exceeds a certain scale, the optical performance of the mirror field will be greatly reduced due to the increase in the radius of the mirror field. There is an upper limit optimal value for the area of ​​a single mirror field. Therefore, multiple towers and one machine, especially three towers and one machine solutions, have gradually come into people's view. "Three towers and one machine" architecture (the heliostats M in the three mirror fields collect heat, and the three absorbers T share a set of heat storage and power generation system G, such as Figure 1 (As shown) Through mirror field sharing and large-scale steam turbine units, the concentration efficiency and thermoelectric conversion efficiency can be significantly improved, and the unit cost can be reduced.

[0003] There are two types of three-tower-one-machine. One is that the three mirror fields do not intersect and the heliostats are not shared. This form of control is relatively simple and each mirror field is controlled separately. The other is that the three mirror fields do intersect (such as Figure 1 As shown in Figure 1, this type of heliostat shares common mirrors, significantly reducing costs and making it highly competitive. The coordinated control strategy for the three-tower, one-unit solar array is particularly important for current research into its commercial application. However, developing a scientific and rational coordinated control strategy for the three mirror fields (specifically assigning each heliostat to a specific receiver for tracking the sun) remains a challenge. Currently, this strategy relies solely on manual experience or proximity allocation, resulting in varying spot shapes and energy levels for each heliostat at different times. This control strategy results in low field efficiency, significantly impacting the improvement of photovoltaic conversion efficiency. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides a collaborative control method for a three-tower-one-machine solar thermal mirror field based on a game theory model, so as to achieve the purpose of scientifically and rationally formulating a collaborative control strategy for the three mirror fields and optimizing the overall energy efficiency of the system.

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

[0006] A method for collaborative control of a three-tower and one-unit solar thermal mirror field based on a game theory model includes the following steps:

[0007] Step 1: Based on the Monte Carlo ray tracing method, calculate the power contribution of each heliostat to each receiver at a certain moment;

[0008] Step 2: Based on the power contribution of each heliostat to each receiver at a certain moment, a game theory model is constructed with maximizing the total power value of the mirror field as the game goal and the combination of which receiver each heliostat is assigned to as the strategy space.

[0009] Step three: Update the strategy space through iterative optimization until the convergence condition is met, that is, all heat absorbers cannot improve their utility by unilaterally changing their strategies, and obtain the final strategy space; if it falls into a local optimum, apply random perturbations to the equilibrium state and restart the dynamic process until convergence.

[0010] In the above scheme, in step 2, in the constructed game theory model, the game subjects are three heat absorbers, denoted as heat absorber j, j = 1, 2, 3.

[0011] In the above scheme, in step 2, in the constructed game theory model, the game goal is to maximize the total power value of the mirror field:

[0012]

[0013] Among them, S j is the strategy space, which here refers to the set of heliostats in receiver j, S j =(x 1j ,x 2j ,..x ij ,...,x nj ), where x ij Indicates whether heliostat i is assigned to receiver j for sun tracking, x ij = 0 means that heliostat i will not be assigned to receiver j for sun tracking, x ij =1 means that heliostat i is assigned to receiver j for sun tracking and satisfies:

[0014] x ij =0 or 1 and

[0015] In the above scheme, in step 2, in the constructed game theory model, the utility function of heat sink j is:

[0016]

[0017] in, represents the value gain of heat sink j, represents the over-capacity penalty, which prevents over-fitting from affecting the optimal solution during the iteration process. λ is the penalty coefficient, which is called over-capacity penalty in game theory. P j is the upper limit of the power of the heat sink j, wij is the power contribution of the i-th heliostat on the receiver j, i = 1, 2, ..., n, and n is the total number of heliostats in the field.

[0018] In the above scheme, in step 2, in the game theory model constructed, the Nash equilibrium goal is that in the equilibrium state, all heat sinks cannot improve their utility by unilaterally changing their strategies:

[0019] For any Both

[0020] in, Refers to the strategy space under the Nash equilibrium state, S j ′ refers to the strategy space under other states, is the utility function of heat sink j in the Nash equilibrium state, U j (S j ′ ) refers to the utility function of heat sink j in other states.

[0021] In the above scheme, in step 3, initialization is performed before iterative optimization, and the heliostats are randomly assigned to each receiver. That is, each heliostat selects a combination of receivers to which it is assigned, forming the initial strategy space S1, S2, and S3, in which whether each heliostat is assigned to the receiver satisfies the quantity constraint And x ij =0 or 1, x ij Indicates whether heliostat i is assigned to receiver j for sun tracking, x ij = 0 means that heliostat i will not be assigned to receiver j for sun tracking, x ij =1 means that heliostat i is assigned to receiver j for sun tracking.

[0022] In the above scheme, in step 3, during the iterative optimization process, each heat sink adjusts its own strategy space S under the fixed strategy of other heat sinks. j , is chosen to maximize the utility function U of heat sink j j (S j ), that is, solving a sub-problem, strategy space S j The update ratio depends on its relative fitness:

[0023]

[0024] in, is the average utility corresponding to the three heat sinks, represents the power of heat absorber j at time t+1, represents the power of heat absorber j at time t;

[0025] Maximize U j (S j ) actually solves the following model:

[0026]

[0027] in, Refers to the strategy space under the Nash equilibrium state.

[0028] In the above scheme, in step 3, if the system falls into a local optimum, random perturbations are applied to the equilibrium state. By randomly exchanging some heliostats or randomly increasing or decreasing the number of heliostats of a certain receiver under the constraints, the dynamic process is restarted until convergence.

[0029] Through the above technical solution, the present invention provides a three-tower-one-machine solar thermal mirror field coordinated control method based on a game theory model, which has the following beneficial effects:

[0030] 1. This invention introduces game theory models (especially evolutionary game algorithms) into the field of coordinated control of multi-tower solar thermal systems for the first time, and optimizes the overall energy efficiency of the system by calculating the dynamic game behavior between the heat-absorbing towers.

[0031] 2. Compared with traditional linear programming and dynamic programming methods, this method is based on a game theory model, which can avoid the curse of dimensionality and quickly solve large-scale (tens of thousands of heliostats) multi-heat absorption tower (three heat absorption tower) scenarios. It also ensures global optimality based on the concept of perturbation in game theory. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for describing the embodiments or the prior art.

[0033] Figure 1 This is a schematic diagram of the three-tower-one-unit solar thermal mirror field;

[0034] Figure 2 This is a flow chart of a collaborative control method for a three-tower-one-machine solar thermal mirror field based on a game theory model disclosed in an embodiment of the present invention.

[0035] In the figure, T, absorber; G, heat storage and power generation system; M, heliostat. DETAILED DESCRIPTION

[0036] The technical solutions in the embodiments of the present invention will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present invention.

[0037] The present invention provides a three-tower-one-machine photothermal mirror field collaborative control method based on a game theory model. Figure 2 As shown, the following steps are included:

[0038] Step 1: Based on the Monte Carlo ray tracing method, calculate the power contribution of each heliostat M to each receiver T at a certain moment; record w ij is the power contribution of the i-th heliostat (i=1,2,...,n) on the receiver j (j=1,2,3). Let P1, P2, and P3 be the maximum power on the receiver 1, receiver 2, and receiver 3, respectively. If this power limit is exceeded, the receiver will be damaged.

[0039] Step 2: Based on the power contribution of each heliostat M to each receiver T at a certain moment, a game theory model is constructed with maximizing the total power value of the mirror field as the game goal and the combination of which receiver T each heliostat chooses to be assigned to as the strategy space.

[0040] Game theory models include:

[0041] (1) Game subjects: three heat absorbers, denoted as heat absorber j, j = 1, 2, 3.

[0042] (2) Game goal: maximize the total power value of the mirror field:

[0043]

[0044] Among them, S j is the strategy space, which here refers to the set of heliostats in receiver j, S j =(x 1j ,x 2j ,..x ij ,...,x nj ), where x ij Indicates whether heliostat i is assigned to receiver j for sun tracking, x ij = 0 means that heliostat i will not be assigned to receiver j for sun tracking, x ij =1 means that heliostat i is assigned to receiver j for sun tracking and satisfies:

[0045] x ij =0 or 1 and

[0046] (3) Strategy space: Each heliostat chooses which combination of receivers to assign to it, strategy S j =(x 1j ,x 2j ,..x ij ,...,x nj );

[0047] (4) Utility function (taking heat sink j as an example):

[0048]

[0049] in, represents the value gain of heat sink j, represents the over-capacity penalty, which prevents over-fitting from affecting the optimal solution during the iteration process. λ is the penalty coefficient, which is called over-capacity penalty in game theory. P j is the upper limit of the power of the heat sink j, w ij is the power contribution of the i-th heliostat on the receiver j, i = 1, 2, ..., n, and n is the total number of heliostats in the field.

[0050] (5) Nash equilibrium goal: In the equilibrium state, all heat sinks cannot improve their utility by unilaterally changing their strategies:

[0051] For any Both

[0052] in, Refers to the strategy space under the Nash equilibrium state, S j ′ refers to the strategy space under other states, is the utility function of heat sink j in the Nash equilibrium state, U j (S j ′ ) refers to the utility function of heat sink j in other states.

[0053] Step three: Update the strategy space through iterative optimization until the convergence condition is met, that is, all heat absorbers cannot improve their utility by unilaterally changing their strategies, and obtain the final strategy space; if it falls into a local optimum, apply random perturbations to the equilibrium state and restart the dynamic process until convergence.

[0054] The specific steps include:

[0055] (1) Initialization: Randomly assign heliostats to each receiver, that is, each heliostat selects a combination of receivers to which it is assigned, forming an initial strategy space S1, S2, S3, where each heliostat is assigned to a receiver to satisfy the quantity constraint. And x ij =0 or 1, x ij Indicates whether heliostat i is assigned to receiver j for sun tracking, x ij = 0 means that heliostat i will not be assigned to receiver j for sun tracking, x ij =1 means that heliostat i is assigned to receiver j for sun tracking.

[0056] (2) Iterative optimization: each heat sink adjusts its own strategy space S while fixing the strategies of other heat sinks. j , is chosen to maximize the utility function U of heat sink j j (Sj ), that is, solving a sub-problem, strategy space S j The update ratio depends on its relative fitness:

[0057]

[0058] in, is the average utility corresponding to the three heat sinks, represents the power of heat absorber j at time t+1, represents the power of heat absorber j at time t;

[0059] Maximize U j (S j ) actually solves the following model:

[0060]

[0061] in, Refers to the strategy space under the Nash equilibrium state.

[0062] (3) Convergence condition: All heat sinks cannot improve their utility (reach Nash equilibrium) by unilaterally changing their strategies, that is, for any Both

[0063] in, Refers to the strategy space under the Nash equilibrium state, S j ′ refers to the strategy space under other states, is the utility function of heat sink j in the Nash equilibrium state, U j (S j ′ ) refers to the utility function of heat sink j in other states.

[0064] (4) Perturbation and global optimization: If the system falls into a local optimum, random perturbations are applied to the equilibrium state. The dynamic process is restarted until convergence by randomly exchanging some heliostats (for example, randomly selecting one heliostat from two receivers for exchange) or randomly increasing or decreasing the number of heliostats in a certain receiver under the constraints.

[0065] Perform iterative calculations according to steps 2 and 3 to obtain the final strategy space. At this time, the total power of the three receivers is maximized and the status of each heliostat is obtained, that is, which receiver each heliostat should be assigned to for tracking the sun.

[0066] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A three-tower-one-unit solar thermal mirror field collaborative control method based on a game theory model, characterized in that: The steps include: Step 1: Based on the Monte Carlo ray tracing method, calculate the power contribution of each heliostat to each receiver at a certain moment; Step 2: Based on the power contribution of each heliostat to each receiver at a certain moment, a game theory model is constructed with maximizing the total power value of the mirror field as the game goal and the combination of which receiver each heliostat is assigned to as the strategy space. Step three: Update the strategy space through iterative optimization until the convergence condition is met, that is, all heat absorbers cannot improve their utility by unilaterally changing their strategies, and obtain the final strategy space; if it falls into a local optimum, apply random perturbations to the equilibrium state and restart the dynamic process until convergence.

2. The method for collaborative control of a three-tower-one-unit solar thermal mirror field based on a game theory model according to claim 1 is characterized in that: In step 2, in the constructed game theory model, the game subjects are three heat absorbers, denoted as heat absorber j, j = 1, 2, 3.

3. The method for collaborative control of a three-tower-one-unit solar thermal mirror field based on a game theory model according to claim 1 is characterized in that: In step 2, in the constructed game theory model, the game goal is to maximize the total power value of the mirror field: Among them, S j is the strategy space, which here refers to the set of heliostats in receiver j, S j =(x 1j ,x 2j ,..x ij ,...,x nj ), where x ij Indicates whether heliostat i is assigned to receiver j for sun tracking, x ij = 0 means that heliostat i will not be assigned to receiver j for sun tracking, x ij =1 means that heliostat i is assigned to receiver j for sun tracking and satisfies: x ij =0 or 1 and 4. The method for collaborative control of a three-tower-one-unit solar thermal mirror field based on a game theory model according to claim 1 is characterized in that: In step 2, in the constructed game theory model, the utility function of heat sink j is: in, represents the value gain of heat sink j, represents the over-capacity penalty, which prevents over-fitting from affecting the optimal solution during the iteration process. λ is the penalty coefficient, which is called over-capacity penalty in game theory. P j is the upper limit of the power of the heat sink j, w ij is the power contribution of the i-th heliostat on the receiver j, i = 1, 2, ..., n, and n is the total number of heliostats in the field.

5. The method for collaborative control of a three-tower-one-unit solar thermal mirror field based on a game theory model according to claim 1 is characterized in that: In step 2, in the game theory model constructed, the Nash equilibrium goal is that in the equilibrium state, all heat sinks cannot improve their utility by unilaterally changing their strategies: For any Both in, Refers to the strategy space under the Nash equilibrium state, S j ′ refers to the strategy space under other states, is the utility function of heat sink j in the Nash equilibrium state, U j (S j ′ ) refers to the utility function of heat sink j in other states.

6. The method for collaborative control of a three-tower-one-unit solar thermal mirror field based on a game theory model according to claim 1 is characterized in that: In step 3, initialization is performed before iterative optimization, and the heliostats are randomly assigned to each receiver. That is, each heliostat selects a combination of receivers to which it is assigned, forming the initial strategy space S1, S2, and S3, in which whether each heliostat is assigned to the receiver satisfies the quantity constraint And x ij =0 or 1, x ij Indicates whether heliostat i is assigned to receiver j for sun tracking, x ij = 0 means that heliostat i will not be assigned to receiver j for sun tracking, x ij =1 means that heliostat i is assigned to receiver j for sun tracking.

7. The method for collaborative control of a three-tower-one-unit solar thermal mirror field based on a game theory model according to claim 1 is characterized in that: In step 3, during the iterative optimization process, each heat sink adjusts its own strategy space S while fixing the strategies of other heat sinks. j , is chosen to maximize the utility function U of heat sink j j (S j ), that is, solving a sub-problem, strategy space S j The update ratio depends on its relative fitness: in, is the average utility corresponding to the three heat sinks, represents the power of heat absorber j at time t+1, represents the power of heat absorber j at time t; Maximize U j (S j ) actually solves the following model: in, Refers to the strategy space under the Nash equilibrium state.

8. The method for collaborative control of a three-tower-one-unit solar thermal mirror field based on a game theory model according to claim 1 is characterized in that: In step 3, if the system falls into a local optimum, random perturbations are applied to the equilibrium state. The dynamic process is restarted until convergence by randomly exchanging some heliostats or randomly increasing or decreasing the number of heliostats of a certain receiver under the constraints.

Images