Distributed economic dispatch method and device based on dynamic weight

Abstract

The application discloses a distributed economic dispatching method and device based on dynamic weights, and the method comprises the following steps: acquiring a power system node cost, and initializing a node cost and a node price difference state variable; based on the initialized node cost and the node price difference state variable, a communication weighting diagram weight is corrected in real time, and an economic dispatching is iteratively calculated; based on the result of the iterative calculation, a distributed economic dispatching optimal output is calculated and executed. The application can improve the operation efficiency of the distributed economic dispatching under a sparse communication topology, and improve the economy and safety of the power system operation.

Description

Technical Field

[0001] This invention relates to the field of power system economic dispatch technology, and in particular to a distributed economic dispatch method and apparatus based on dynamic weights. Background Technology

[0002] In the energy internet, distributed resources are playing an increasingly important role, and the operation and optimization of massive distributed resources face enormous computing and communication pressures. Distributed algorithms can provide algorithmic support for edge computing, achieving operational optimization through communication between adjacent nodes, while mitigating threats such as single points of failure, high computing pressure, and communication congestion. Distributed resources can achieve automatic optimized operation through distributed economic scheduling, integrating distributed resource clusters into a whole. This not only satisfies power balance in the most economical way but also coordinates and optimizes with the active distribution network in a timely manner.

[0003] Considering the cost and time required for channel construction, the communication topology among massive distributed resources will tend to be sparse while maintaining connectivity, resulting in low overall connectivity. Simultaneously, some distributed resources may have relatively dense communication topologies. This coexistence of dense and sparse communication topologies will lead to inconsistent convergence speeds in distributed consensus algorithms, with slow convergence regions hindering the consensus achievement rate, ultimately impacting the operational efficiency of distributed economic scheduling. Summary of the Invention

[0004] The present invention aims to at least partially solve one of the technical problems in the related art.

[0005] To address this, a distributed economic dispatching method and apparatus based on dynamic weights are proposed. This invention dynamically adjusts the weights of the communication weighting graph according to the communication topology, thereby accelerating the convergence speed of slow convergence links, speeding up the convergence of the distributed economic dispatching algorithm, and improving the economy and security of power system operation.

[0006] To achieve the above objectives, a first aspect of the present invention proposes a distributed economic scheduling method based on dynamic weights, comprising:

[0007] Obtain the node cost of the power system and initialize the node cost and node price difference state variables; based on the initialized node cost and node price difference state variables, adjust the weight of the communication weighting graph in real time and perform iterative calculation on economic dispatch; based on the results of the iterative calculation, calculate and execute the optimal output of distributed economic dispatch.

[0008] In addition, the distributed economic scheduling method based on dynamic weights according to the above embodiments of the present invention may also have the following additional technical features:

[0009] Furthermore, in one embodiment of the present invention, the step of obtaining the power system node cost and initializing the node cost and node price difference state variables includes: obtaining and converting the key cost parameter information of each node; setting the initial estimated value of the virtual power deficit and the initial estimated value of the price sensitivity of each node according to the key cost parameter information; and setting the initial node price, the initial line price difference and the initial node price difference of each node.

[0010] Furthermore, in one embodiment of the present invention, the step of real-time correction of the communication weighting graph weights based on the initial node cost and node price difference state variables, and iterative calculation of economic scheduling, includes: generating communication weighting graph weights based on the latest node price difference; exchanging node state variables based on the communication weighting graph weights; calculating node prices and node price difference state variables based on the node state variables; and determining whether convergence has occurred based on the node prices and deciding whether to continue iterating or terminate the calculation.

[0011] Furthermore, in one embodiment of the present invention, the step of calculating and executing the optimal output of distributed economic scheduling based on the result of the iterative calculation includes: calculating the optimal output of each node according to the node price obtained after the calculation is terminated; and making each node actually output according to the optimal output of the node.

[0012] The distributed economic dispatch method based on dynamic weights in this invention can dynamically adjust the weights of the communication weighting graph according to the communication topology, accelerate the convergence speed of slow convergence links, speed up the convergence of the distributed economic dispatch algorithm, and improve the economy and security of power system operation.

[0013] To achieve the above objectives, a second aspect of the present invention provides a distributed economic scheduling device based on dynamic weights, comprising:

[0014] The node initialization module is used to obtain the node cost of the power system and initialize the node cost and node price difference state variables; the correction calculation module is used to correct the weight of the communication weighting graph in real time based on the initialized node cost and node price difference state variables, and perform iterative calculation on economic dispatch; the result output module is used to calculate and execute the optimal output of distributed economic dispatch based on the results of the iterative calculation.

[0015] The distributed economic dispatching device based on dynamic weights in this invention can dynamically adjust the weights of the communication weighting graph according to the communication topology, thereby accelerating the convergence speed of the slow convergence stage, speeding up the convergence of the distributed economic dispatching algorithm, and improving the economy and security of power system operation.

[0016] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0017] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0018] Figure 1 A flowchart of a distributed economic scheduling method based on dynamic weights according to an embodiment of the present invention;

[0019] Figure 2 This is an architecture diagram of a distributed economic scheduling method based on dynamic weights according to an embodiment of the present invention;

[0020] Figure 3 This is a schematic diagram of a distributed economic scheduling device based on dynamic weights according to an embodiment of the present invention.

[0021] Figure 4 This is a schematic diagram of another distributed economic scheduling device based on dynamic weights according to an embodiment of the present invention.

[0022] Figure 5 This is a schematic diagram of another distributed economic scheduling device based on dynamic weights according to an embodiment of the present invention;

[0023] Figure 6 This is a schematic diagram of another distributed economic scheduling device based on dynamic weights according to an embodiment of the present invention. Detailed Implementation

[0024] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0025] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0026] The following description, with reference to the accompanying drawings, describes a distributed economic scheduling method and apparatus based on dynamic weights according to embodiments of the present invention.

[0027] Figure 1 This is a flowchart of a distributed economic scheduling method based on dynamic weights according to an embodiment of the present invention.

[0028] like Figure 1 As shown, the method includes, but is not limited to, the following steps:

[0029] S1, obtain the node cost of the power system and initialize the node cost and node price difference state variables.

[0030] Specifically, in this embodiment of the invention, obtaining the node cost of the power system and initializing the node cost state variables and price difference state variables includes three steps: (1-1) obtaining and converting the key cost parameter information of each node; (1-2) setting the initial estimated value of the virtual power deficit and the initial estimated value of the price sensitivity of each node according to the unit cost information; and (1-3) setting the initial node price, the initial line price difference, and the initial node price difference for each node, such as... Figure 2 As shown.

[0031] (1-1) Obtain and convert the key cost parameter information of each node.

[0032] In this embodiment of the invention, the economic scheduling model is as follows:

[0033] Objective function:

[0034]

[0035] Constraints:

[0036]

[0037]

[0038]

[0039] Where N is the number of nodes, and in this specific embodiment, each node corresponds to one generating unit. Equations (1) and (2) indicate that the objective is to minimize the power generation cost of each generating unit, λ i (p i Let p be the marginal cost function of the unit output, N be the number of units, and p be the marginal cost function of the unit output. i To provide power to the generator unit, α i β i γ i All of these are parameters in the unit output cost function, γ i For the hypothetical minimum cost of the unit, β i α is the marginal growth rate of marginal cost with respect to unit output. i The output of the hypothetical unit under zero marginal cost obtained by linear extrapolation; Equation (4) represents the power balance constraint, d i Let be the load at node i.

[0040] In this embodiment of the invention, the key cost parameter information refers to α i β i γ i d i .

[0041] (1-2) Set the initial estimate of virtual power deficit and initial estimate of price sensitivity for each node based on the unit cost information.

[0042] Initialize virtual power deficit estimate s i The formula is described as follows:

[0043]

[0044] Where, d i Let α be the load at unit i, i.e., the initial output of unit i. i The output of the hypothetical unit under zero marginal cost obtained by linear extrapolation. This represents the estimated virtual power deficit after the k-th iteration.

[0045] Initialize the initial estimate of price sensitivity b i The formula is described as follows:

[0046]

[0047] in, This represents the estimated price sensitivity value after the k-th iteration.

[0048] (1-3) Set the initial node price, initial line price difference, and initial node price difference for each node.

[0049] Initial node price λ i The formula is described as follows:

[0050]

[0051] in Let λ represent the node price in the k-th iteration. max In this embodiment, λ is the pre-set maximum price. max It can be set to 10 times the price under normal circumstances.

[0052] Initial line price difference The formula is described as follows:

[0053]

[0054] in, Let c represent the price difference between the nodes at both ends of the line in the k-th iteration. c can be any pre-given positive real number.

[0055] Initial node price spread The formula is described as follows:

[0056]

[0057] in, This represents the maximum price difference between adjacent lines of node i in the k-th iteration.

[0058] S2, based on the initial node cost and node price difference state variables, corrects the weights of the communication weighting graph in real time and performs iterative calculations on economic scheduling.

[0059] Specifically, the real-time correction of communication weighted graph weights and iterative calculation of economic scheduling includes four steps: (2-1) generating communication weighted graph weights based on the latest node price difference; (2-2) exchanging node state variables based on the communication weighted graph weights; (2-3) calculating node prices and node price difference state variables based on the exchanged node state variables; and (2-4) determining whether convergence has occurred based on node prices and deciding whether to continue iterating or terminate the calculation. Figure 2 As shown.

[0060] (2-1) Generate communication weights based on the latest node price difference.

[0061] In this embodiment of the invention, the communication model is defined as follows: Nodes connected by transmission lines are considered to be able to communicate, and are referred to as mutually adjacent. The set of nodes adjacent to node i is defined as D(i). Node i can send information to all adjacent nodes, and this information path can be described as the edges of the directed graph G. A node in the directed graph can send information to itself, meaning that nodes necessarily have self-loops. The directed graph G can be described by the adjacency matrix A, which is defined as follows:

[0062] A = [a ij ] N×N (10)

[0063]

[0064] Assigning weights to the edges of a directed graph creates a weighted directed graph, which can be described by a matrix Q and has the following characteristics:

[0065] Q = [q] ij ] N×N (12)

[0066]

[0067] Where, q ij To assign weights to the graph, in this embodiment of the invention, q is dynamically adjusted in each iteration. ij Achieve better performance than fixed weight q ij Faster convergence. Weighted graph weights. The correction is made by node j, and the correction method is as follows:

[0068]

[0069]

[0070] in, This represents the weighted graph weights used in the k-th iteration. This refers to the latest node price difference observed at node j. For node j, the node price difference observed is the one calculated in round k-1. The node price differences observed at other neighboring nodes can only be received in round k-1. Therefore, node j can only obtain the node price differences calculated in round k-2 of other nodes.

[0071] (2-2) Exchange the state variables of each node according to the weight of the communication weight graph.

[0072] In the k-th iteration, node j will change its own state variable and Multiply by the weights of the adjacent edges in the communication weighting graph. Then the product is sent along the direction of the edges in the directed graph. For example, in a directed graph, the edge weight from node j to node i is... Then node j sends to node i and

[0073] remove and In addition, node j also sends the node price calculated by node j in the (k-1)th round along the edges of the directed graph. Node spread

[0074] (2-3) Calculate the node price and node price difference state variables based on the node state variables obtained from the exchange.

[0075] Node i receives the product sent by all its neighboring nodes j and Then, simply add them together and update the corresponding state variables. The formula is described as follows:

[0076]

[0077]

[0078] Node i calculates its price based on this. The formula is described as follows:

[0079]

[0080] Node i further calculates the line price difference The formula is described as follows:

[0081]

[0082] Recalculate the node price difference The formula is described as follows:

[0083]

[0084] (2-4) Determine whether convergence has been achieved based on the node price and decide whether to continue iterating or terminate the calculation.

[0085] Determine whether the marginal cost meets the iteration termination condition. The iteration termination condition is that the iteration process terminates when the marginal cost error is less than the error tolerance δ. The formula is described as follows:

[0086]

[0087] If the convergence condition is not met, increment the iteration number k by 1, jump to step (2-1) to generate the latest communication weighting graph weights based on the latest node price difference, and then continue to execute (2-2), (2-3), and (2-4) in sequence until convergence.

[0088] S3, based on the results of iterative calculations, calculates and executes the optimal output of distributed economic scheduling.

[0089] Specifically, in this embodiment of the invention, calculating and executing the optimal output of distributed economic scheduling includes two steps: (3-1) calculating the optimal output of each node based on the node price obtained after the calculation is terminated; and (3-2) each node performs its actual output according to its optimal output, such as... Figure 2 As shown.

[0090] (3-1) Calculate the optimal output of each node based on the node price obtained after the calculation is terminated.

[0091] After terminating the iteration process, node i is determined based on the node price at the time of termination. Calculate the optimal power generation output:

[0092]

[0093] (3-2) Each node will generate power The power output is adjusted by issuing orders to the generating units to achieve economic dispatch.

[0094] According to the distributed economic dispatch method based on dynamic weights of the present invention, the node cost state variables and price difference state variables are initialized; the weights of the communication weighting graph are corrected in real time, and the economic dispatch is calculated iteratively; the optimal output of the distributed economic dispatch is calculated and executed, which can accelerate the convergence of the distributed economic dispatch algorithm and improve the economy and security of power system operation.

[0095] To achieve the above embodiments, such as Figure 3As shown, this embodiment also provides a distributed economic scheduling device 10 based on dynamic weights. The device 10 includes a node initialization module 100, a correction calculation module 200, and a result output module 300.

[0096] The node initialization module 100 is used to obtain the node cost of the power system and initialize the node cost and node price difference state variables;

[0097] The correction calculation module 200 is used to correct the weights of the communication weighting graph in real time based on the initial node cost and node price difference state variables, and to perform iterative calculations on economic scheduling.

[0098] The result output module 300 is used to calculate and execute the optimal output of distributed economic scheduling based on the results of iterative calculation.

[0099] Furthermore, such as Figure 4 As shown, in this embodiment of the invention, the node initialization module 100 includes:

[0100] The parameter acquisition module 101 is used to acquire and convert key cost parameter information for each node.

[0101] Node setting module 102 is used to set the initial estimate of virtual power deficit and the initial estimate of price sensitivity for each node based on key cost parameter information; and,

[0102] The price setting module 103 is used to set the initial node price, initial line price difference, and initial node price difference for each node.

[0103] Furthermore, such as Figure 5 As shown, in this embodiment of the invention, the correction calculation module 200 includes:

[0104] The weight generation module 201 is used to generate communication weighting graph weights based on the latest node price difference.

[0105] Node transformation module 202 is used to obtain node state variables based on the weight exchange of the communication weighted graph;

[0106] Price calculation module 203 is used to calculate node price and node price difference state variable based on node state variable;

[0107] The calculation and judgment module 204 is used to determine whether convergence has been achieved based on the node price and decide whether to continue iterating or terminate the calculation.

[0108] Furthermore, such as Figure 6 As shown, in this embodiment of the invention, the result output module 300 includes:

[0109] Optimal output module 301 is used to calculate the optimal output of each node based on the node price obtained after the calculation is terminated; and

[0110] The actual output module 302 is used to calculate the actual output of each node based on the node's optimal output.

[0111] The distributed economic dispatching device based on dynamic weights in this invention can improve the operational efficiency of distributed economic dispatching under sparse communication topology, and enhance the economy and security of power system operation.

[0112] It should be noted that the foregoing explanation of the embodiment of the distributed economic scheduling method based on dynamic weights also applies to the distributed economic scheduling device based on dynamic weights in this embodiment, and will not be repeated here.

[0113] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0114] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A dynamic weight-based distributed economic dispatch method, characterized in that, Includes the following steps: Obtain the node cost of the power system and initialize the node cost and node price difference state variables; Based on the initial node cost and node price difference state variables, the weights of the communication weighting graph are corrected in real time, and the economic scheduling is iteratively calculated. Based on the results of the iterative calculation, calculate and execute the optimal output of the distributed economic scheduling; The process of obtaining the power system node cost and initializing the node cost and node price difference state variables includes: Obtain and convert key cost parameter information for each node; Based on the aforementioned key cost parameter information, set the initial estimate of the virtual power deficit and the initial estimate of the price sensitivity for each node; and, Set the initial node price, initial line price difference, and initial node price difference for each node; The process of adjusting the communication weighting graph weights in real time based on the initial node cost and node price difference state variables, and iteratively calculating the economic scheduling, includes: Generate communication weights based on the latest node price differences; The node state variables are obtained by exchanging the weights in the communication weighted graph. Calculate the node price and the node price difference state variable based on the node state variables; Based on the node price, determine whether convergence has occurred and decide whether to continue iterating or terminate the calculation; The calculation and execution of optimal output for distributed economic scheduling based on the results of the iterative calculation includes: Calculate the optimal output of each node based on the node price obtained after the calculation is terminated; and Each node will output its actual power based on its optimal output.

2. A dynamic weight based distributed economic dispatch apparatus, characterized in that, include: The node initialization module is used to obtain the node cost of the power system and initialize the node cost and node price difference state variables; The correction calculation module is used to correct the weights of the communication weighting graph in real time based on the initial node cost and node price difference state variables, and to perform iterative calculations on economic scheduling. The result output module is used to calculate and execute the optimal output of distributed economic scheduling based on the results of the iterative calculation. The node initialization module includes: The parameter acquisition module is used to acquire and convert key cost parameter information for each node; The node setting module is used to set the initial estimate of the virtual power deficit and the initial estimate of the price sensitivity for each node based on the key cost parameter information; and, The price setting module is used to set the initial node price, initial line price difference, and initial node price difference for each node; the correction calculation module includes: The weight generation module is used to generate communication weight graph weights based on the latest node price difference; The node transformation module is used to obtain node state variables based on the weight exchange of the communication weighting graph; The price calculation module is used to calculate the node price and the node price difference state variable based on the node state variable; the calculation judgment module is used to determine whether convergence has occurred based on the node price and decide whether to continue iteration or terminate the calculation; the result output module includes: The optimal output module is used to calculate the optimal output of each node based on the node price obtained after the calculation is terminated; and the actual output module is used to make each node output actual output based on the optimal output of the node.

Images