A sensor optimization configuration method based on energy balance

Abstract

This invention relates to a sensor optimization configuration method based on energy balance, belonging to the field of networked system measurement. The method includes the following steps: S101: Establishing an energy network model of the networked system, the network has b edges; S102: Obtaining the edge-cut correlation matrix Q based on the energy network model; S103: Performing elementary transformations on Q column-wise. If m sensors (m < b) need to be deployed, selecting b-m column vectors to form matrix C, such that the number of non-zero rows in matrix C is maximized and the columns are full rank. The edges corresponding to the remaining m column vectors are the required sensor deployment locations. Using this method, maximum measurement redundancy can be obtained under the condition of the same number of sensors.

Description

Technical Field

[0001] This invention belongs to the field of networked system measurement, and specifically relates to a sensor optimization configuration method based on energy balance. Background Technology

[0002] Given a fixed number of sensors, maximizing measurement redundancy in a networked system through sensor configuration is a crucial and fundamental issue in networked system monitoring. Gui Weihua et al. proposed an equivalent space fault detection method based on optimal sensor configuration, selecting measurement points with the goal of maximizing fault information to minimize the overall system measurement cost. Chinese invention patent CN110727909B discloses a sensor configuration redundancy determination method and system based on energy balance, providing a method for quantifying measurement redundancy for different sensor configurations in a networked system. However, it does not address how to deploy sensors to maximize measurement redundancy when the number of sensors is fixed. Summary of the Invention

[0003] (a) Technical problems to be solved

[0004] The technical problem to be solved by this invention is how to provide a sensor optimization configuration method based on energy balance, so as to solve the problem of how to deploy sensors to maximize measurement redundancy when the number of sensors is fixed.

[0005] (II) Technical Solution

[0006] To address the aforementioned technical problems, this invention proposes a sensor optimization configuration method based on energy balance, which includes the following steps:

[0007] S101: Use graph theory to establish an energy network model of a networked system. Model a networked system as a directed graph, where the nodes of the graph correspond to the devices that transform and distribute energy in the networked system, the edges correspond to the various pipes that transmit energy in the physical network, and the direction of the edges corresponds to the direction of energy flow.

[0008] S102: Obtain the edge cut incidence matrix based on the network model. Let the number of nodes in the network be . The total number of source nodes and sink nodes is The source node is a node that only outputs energy and matter, and the sink node is a node that only inputs energy and matter. The number of edges in the network is... ;

[0009] S103: Matrix Perform elementary transformations column-wise, then select from them. column vectors form a matrix To maximize the number of non-zero rows and ensure full rank, the remaining rows... The edges corresponding to each column vector are the edges where sensors need to be deployed.

[0010] Furthermore, in S101, the networked system is a physical system.

[0011] Furthermore, in S101, the networked system is a power system.

[0012] Furthermore, the conduit in S101 is a cable.

[0013] Furthermore, in S101, the networked system is a gas system.

[0014] Furthermore, the pipe in S101 is a liquid pipe.

[0015] Furthermore, in S103, a matrix is ​​constructed. At that time, the number of non-zero rows should be maximized, reaching a maximum of 100%. .

[0016] Furthermore, in step 102, record This refers to the number of ordinary nodes in the network. An ordinary node is a node that both inputs and outputs energy and matter. The node-related edge cut matrix of the network is:

[0017]

[0018] in,

[0019] .

[0020] Furthermore, in step S103, at least the following configuration is required: One sensor, capable of detecting the rest The edge where no sensors are deployed is estimated.

[0021] Furthermore, step S103 also includes: The steps for selecting the location of each sensor configuration are as follows: For the matrix... Perform elementary row operations, from Select from column vectors There are 1 linearly independent column vectors, and the edges corresponding to the remaining columns are the edges where sensors need to be configured.

[0022] (III) Beneficial Effects

[0023] This invention proposes a sensor optimization configuration method based on energy balance. The method includes the following steps: S101: Establishing an energy network model of a networked system, wherein the network has... Edge cut; S102: Obtain the edge cut incidence matrix based on the energy network model. S103: will Perform elementary transformations column-wise, if deployment is required. One sensor ( Select one of them The matrix is ​​composed of column vectors. , making the matrix If the number of non-zero rows is the largest and the column has full rank, then the remaining rows are... The edges corresponding to each column vector represent the locations where sensors need to be deployed. The method of this invention can achieve maximum measurement redundancy with the same number of sensors. Attached Figure Description

[0024] Figure 1 This is a network model according to an embodiment of the present invention;

[0025] Figure 2 This is a sensor configuration method for maximizing measurement redundancy;

[0026] Figure 3 Deployment method 1, which requires the minimum number of sensors to meet the estimation requirements;

[0027] Figure 4 Deployment method two with the minimum number of sensors required to meet the estimation requirements;

[0028] Figure 5 This is a flowchart of the method of the present invention. Detailed Implementation

[0029] To make the objectives, contents, and advantages of the present invention clearer, the specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples.

[0030] This invention addresses the problems of the prior art by providing a method for optimizing sensor configuration in a networked energy transmission system. This method maximizes measurement redundancy given a given number of sensors.

[0031] This invention discloses a sensor optimization configuration method based on energy balance, which includes the following steps: S101: Establishing an energy network model of a networked system, wherein the network has Edge cut; S102: Obtain the edge cut incidence matrix based on the energy network model. S103: will Perform elementary transformations column-wise, if deployment is required. One sensor ( Select one of them The matrix is ​​composed of column vectors. , making the matrix If the number of non-zero rows is the largest and the column has full rank, then the remaining rows are... The edges corresponding to each column vector represent the locations where sensors need to be deployed. The method of this invention can achieve maximum measurement redundancy with the same number of sensors.

[0032] To address the aforementioned technical problems, this invention discloses a sensor optimization configuration method based on energy balance, comprising the following steps:

[0033] S101: Utilize graph theory to establish an energy network model for a networked system. A networked system is modeled as a directed graph, where the nodes correspond to devices that transform and distribute energy within the networked system, the edges correspond to various pipes in the physical network that transmit energy, such as cables and liquid pipes, and the direction of the edges corresponds to the direction of energy flow. The networked system is a physical system.

[0034] S102: Obtain the edge cut incidence matrix based on the network model. Let the number of nodes in the network be . The total number of source nodes and sink nodes is The source node is a node that only outputs energy and matter, and the sink node is a node that only inputs energy and matter. The number of edges in the network is... The number of sensors is ,remember This refers to the number of ordinary nodes in the network. An ordinary node is a node that both inputs and outputs energy and matter. The node-related edge cut matrix of the network is

[0035]

[0036] in,

[0037]

[0038] S103: Matrix Perform elementary transformations column-wise, then select from them. column vectors form a matrix To maximize the number of non-zero rows and ensure full rank, the remaining rows... The edges corresponding to each column vector are the edges where sensors need to be deployed.

[0039] In addition, the energy balance-based sensor optimization configuration method according to the above embodiments of the present invention may also have the following additional technical features:

[0040] In some examples, in S101, a networked system refers to a type of networked system that has energy transmission and conversion capabilities, such as a power system or a gas system.

[0041] In some examples, in S101, establishing the energy network model of the networked system involves abstracting the pipelines that transmit energy in the networked system as edges of the network model, and abstracting the devices that process and transform energy as nodes of the network model.

[0042] In some examples, S103, according to the definition of information redundancy, , where g' is The number of non-zero rows in the matrix should be minimized. The number of non-zero rows is large, and the columns must be of full rank. Therefore, when constructing the matrix... At that time, the number of non-zero rows g' should be maximized, which can reach a maximum of 100%. .

[0043] In some examples, S103 will transform the matrix Perform elementary transformations column-wise, then select from them. column vectors form a matrix To maximize the number of non-zero rows and ensure full rank, the remaining rows... The edges corresponding to each column vector are the edges where sensors need to be deployed.

[0044] Example 1:

[0045] like Figure 1 The network shown is configured such that, assuming there are 6 sensors that need to be deployed across the network's 8 edges, according to the maximum measurement redundancy deployment strategy, we have...

[0046] 1) The edge cut incidence matrix is ​​obtained as follows

[0047]

[0048] 2) For the matrix Perform elementary column transformations to obtain

[0049]

[0050] 3) If column is selected and Then the matrix If the maximum number of rows with full rank and not all values ​​being zero is found, then the number of edges used to deploy the sensors is... The redundancy at this point is 2; the redundancy of other deployment schemes does not exceed 2. The sensor configuration method is as follows: Figure 2 As shown.

[0051]

[0052] Another embodiment of the present invention, such as Figure 1 The network model shown has the following network parameters. This requires determining the number and deployment locations of sensors so that the values ​​of edges in the network that are not directly measured can be estimated with a minimum number of sensors. Its matrix... Elementary row operations yield the following expression.

[0053]

[0054] Minimum number of sensors

[0055] If selected The edge deployment, then and The two data matrices are respectively

[0056]

[0057] At this point, the sensors should be deployed on the four edges e4, e6, e7, and e8, as shown in the following figure. Figure 3 As shown.

[0058] If select the edge To deploy sensors, and The two data matrices are shown in the following formulas, and the sensor configuration is as follows: Figure 4 As shown.

[0059]

[0060] Example 2:

[0061] This invention discloses a sensor optimization configuration method based on energy balance, comprising the following steps:

[0062] S101: Establish an energy network model for the networked system, setting its network parameters as follows: Represents the number of network nodes. This represents the number of source nodes and sink nodes. Represents the number of edges in the network;

[0063] S102: Obtain the edge cut incidence matrix based on the network model. ;

[0064] S103: Matrix Perform elementary transformations column-wise, then select from them. The matrix is ​​composed of column vectors. To maximize the number of non-zero rows and ensure full rank, the remaining rows... The edges corresponding to each column vector are the edges where sensors need to be deployed.

[0065] Furthermore, in S101, a networked system refers to a type of networked system such as a power system or a gas system that has energy transmission and conversion capabilities.

[0066] Furthermore, in step S102, the edge cut correlation matrix is ​​obtained based on the network model. .

[0067] Furthermore, in S103, the matrix Perform elementary transformations column-wise, then select from them. column vectors form a matrix To maximize the number of non-zero rows and ensure full rank, the remaining rows... The edges corresponding to each column vector are the edges where sensors need to be deployed.

[0068] Furthermore, at least the following configuration is required: One sensor can monitor the rest Estimating the edges where no sensors are deployed. The steps for selecting the location of each sensor configuration are as follows: For the matrix... Perform elementary row operations; from Select from column vectors There are 1 linearly independent column vectors, and the edges corresponding to the remaining columns are the edges where sensors need to be configured.

[0069] This invention discloses a sensor optimization configuration method based on energy balance, which includes the following steps: S101: Establishing an energy network model of a networked system, wherein the network has Edge cut; S102: Obtain the edge cut incidence matrix based on the energy network model. S103: will Perform elementary transformations column-wise, if deployment is required. One sensor ( Select one of them The matrix is ​​composed of column vectors. , making the matrix If the number of non-zero rows is the largest and the column has full rank, then the remaining rows are... The edges corresponding to each column vector represent the locations where sensors need to be deployed. The method of this invention can achieve maximum measurement redundancy with the same number of sensors.

[0070] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A sensor optimization configuration method based on energy balance, characterized in that, The method includes the following steps: S101: Use graph theory to establish an energy network model of a networked system. Model a networked system as a directed graph, where the nodes of the graph correspond to the devices that transform and distribute energy in the networked system, the edges correspond to the various pipes that transmit energy in the physical network, and the direction of the edges corresponds to the direction of energy flow. S102: Obtain the edge cut incidence matrix based on the network model. Let the number of nodes in the network be . The total number of source nodes and sink nodes is The source node is a node that only outputs energy and matter, and the sink node is a node that only inputs energy and matter. The number of edges in the network is... ; S103: Matrix Perform elementary transformations column-wise, then select from them. column vectors form a matrix To maximize the number of non-zero rows and ensure full rank, the remaining rows... The edges corresponding to each column vector are the edges where sensors need to be deployed; in, In step S103, at least the following configuration is required: One sensor, capable of detecting the rest Estimate the edges where no sensors are deployed; In step S102, record This refers to the number of ordinary nodes in the network. An ordinary node is a node that both inputs and outputs energy and matter. The node-related edge cut matrix of the network is: in, 。 2. The sensor optimization configuration method based on energy balance as described in claim 1, characterized in that, In S101, the networked system is a physical system.

3. The sensor optimization configuration method based on energy balance as described in claim 1, characterized in that, In S101, the networked system is the power system.

4. The sensor optimization configuration method based on energy balance as described in claim 3, characterized in that, The conduit in S101 is a cable.

5. The sensor optimization configuration method based on energy balance as described in claim 1, characterized in that, In S101, the networked system is a gas system.

6. The sensor optimization configuration method based on energy balance as described in claim 5, characterized in that, The pipe in S101 is a liquid pipe.

7. The sensor optimization configuration method based on energy balance as described in claim 1, characterized in that, In S103, a matrix is ​​constructed. At that time, the number of non-zero rows should be maximized, reaching a maximum of 100%. .

8. The sensor optimization configuration method based on energy balance as described in claim 1, characterized in that, Step S103 further includes: The steps for selecting the location of each sensor configuration are as follows: For the matrix... Perform elementary row operations, from Select from column vectors There are 1 linearly independent column vectors, and the edges corresponding to the remaining columns are the edges where sensors need to be configured.

Images