A multi-objective optimal configuration method and system for metering points in a DC distribution network

Abstract

The present invention discloses a multi-objective optimal configuration method and system for metering points in a DC distribution network, including: collecting system parameters of the DC distribution network and establishing a state estimation model of the DC distribution network; calculating the node voltage estimation error rate according to the state estimation model and the corresponding measured values, taking maximizing the node voltage estimation error rate as the first evaluation index, and taking minimizing the first evaluation index as the first optimization objective; taking the number of configured DC watt-hour meters as the second evaluation index, taking minimizing the second evaluation index as the second optimization objective, and establishing a multi-objective optimal configuration model according to the first optimization objective and the second optimization objective; jointly solving the multi-objective optimal configuration model according to the genetic algorithm and the quadratic constraint quadratic programming algorithm to obtain the optimal configuration of the metering points in the DC distribution network; which can improve the metering resolution of the DC distribution network while reducing the configuration cost of the metering points in the DC distribution network.

Description

Technical Field

[0001] The present invention relates to the field of metering point configuration in DC distribution networks, and particularly to a multi-objective optimal configuration method and system for metering points in DC distribution networks. Background Art

[0002] Driven by a series of policies such as "dual carbon" and the construction of a new power system, large-scale distributed new energy sources are connected to the distribution network. Common distributed new energy sources such as photovoltaic, wind power, and energy storage are usually direct current or are rectified into direct current and are connected to the AC distribution network through inverters. Moreover, in recent years, with the continuous development of power electronics technology, the overall structure on the load side has also changed greatly. When connecting to the AC distribution system, loads such as electric vehicles, computers, and mobile phones need to be equipped with AC / DC devices for power supply, and household variable-frequency devices such as air conditioners, refrigerators, and washing machines and energy storage systems also need to go through AC / DC / AC devices to achieve frequency conversion and ensure power supply quality and reliability. Against this background, forms such as DC distribution networks and AC / DC hybrid distribution networks have emerged. In the operation of DC distribution networks, DC metering is one of the most important guarantee links, and accurate and reliable DC metering is particularly crucial for the safe and stable operation of DC distribution networks. Similar to AC distribution networks, the operation control of DC distribution networks depends on the results of state estimation, and the accuracy, economy, and reliability of state estimation depend on the configuration of each measurement node.

[0003] However, the current setting of metering points in DC distribution networks is mainly based on relevant technical standards and specifications and combined with relevant engineering experience, and it is difficult to simultaneously meet the requirements in terms of observability, metering resolution, and configuration economy, and the adaptability to the needs of various application scenarios is poor. Therefore, it is of great significance to develop an optimal configuration method for metering points in DC distribution networks. Summary of the Invention

[0004] The purpose of the present invention is to address the above-mentioned deficiencies of the prior art and propose a multi-objective optimal configuration method and system for metering points in DC distribution networks, which can improve the metering resolution of DC distribution networks while reducing the configuration cost of metering points in DC distribution networks.

[0005] In the first aspect, the present invention provides a multi-objective optimal configuration method for metering points in DC distribution networks, including:

[0006] Collect system parameters of the DC distribution network, and establish a state estimation model of the DC distribution network according to the system parameters;

[0007] According to the state estimation model and the corresponding measured values, calculate the node voltage estimation error rate, use maximizing the node voltage estimation error rate as the first evaluation index, and use minimizing the first evaluation index as the first optimization goal; wherein, the system parameters include: measured values of node active power, line active power, and line current.

[0008] Take the number of DC energy meters configured as the second evaluation index, minimize the second evaluation index as the second optimization goal, and establish a multi-objective optimization configuration model according to the first optimization goal and the second optimization goal;

[0009] Jointly solve the multi-objective optimization configuration model according to the genetic algorithm and the quadratic constraint quadratic programming algorithm to obtain the optimal configuration of the metering points of the DC distribution network, so as to configure the DC ammeters of the DC distribution network according to the optimal configuration of the metering points.

[0010] The present invention estimates the node voltage state of the DC distribution network, obtains the estimation error rate of the DC distribution network according to the state estimation and the actual voltage measurement value, takes the estimation error rate as the first evaluation index for state evaluation, and configures the metering table according to the first evaluation index. Compared with the prior art, it is not necessary to rely on the engineering experience of the staff to configure the metering table, so as to avoid subjective and empirical metering table configuration, and further improve the configuration accuracy and feasibility of the metering table; moreover, take the number of DC energy meters configured as the second evaluation index, establish a joint multi-objective model to maximize the estimation error rate and minimize the number of DC energy meters configured, and jointly solve with the genetic algorithm and the quadratic constraint quadratic programming, which can solve the overall optimal solution of the model, so as to reduce the cost of configuring the metering table while improving the dosage resolution, and therefore also has higher applicability.

[0011] Further, the jointly solving the multi-objective optimization configuration model according to the genetic algorithm and the quadratic constraint quadratic programming algorithm to obtain the optimal configuration of the metering points of the DC distribution network includes:

[0012] Obtain the first metering point configuration result of the line according to the genetic algorithm, and in each iteration process of the genetic algorithm, use the quadratic constraint quadratic programming algorithm to solve and obtain the node voltage state estimation value;

[0013] Calculate the first evaluation index according to the node voltage state estimation value and the corresponding measurement value to obtain the first evaluation index result, and count the number of DC energy meters configured to obtain the second evaluation index result; wherein, take the first metering point configuration result corresponding to the optimal first evaluation index result and the second evaluation index result as the second metering point configuration result;

[0014] When the preset population evolution algebra is reached or the preset convergence condition is satisfied, take the second metering point configuration result and the corresponding first evaluation index result and second evaluation index result as the optimal configuration of the metering points of the DC distribution network.

[0015] In the present invention, the metering point configuration result is obtained through a genetic algorithm. During the iteration process, a quadratic constraint quadratic programming is used to solve the first evaluation index, so that the optimal first evaluation index representing the accuracy can be obtained for each iteration. After the iteration ends, the first evaluation indexes corresponding to all metering point configuration results are obtained, and thus the optimal metering point configuration result is used as the optimal configuration of the metering points in the DC distribution network. By jointly solving the genetic algorithm and the quadratic constraint quadratic programming, the accuracy of multi-objective configuration can be improved, the subjective judgment relying too much on the engineering experience of the staff can be avoided, the metering resolution of the DC distribution network can be improved, and the configuration cost of the metering points in the DC distribution network can be reduced.

[0016] Further, obtaining the first metering point configuration result of the line according to the genetic algorithm includes:

[0017] Define the binary configuration variables of the line metering points, and initialize the first metering point configuration result of the population size according to the binary configuration variables; wherein, when the binary configuration variable takes 0, it means that no DC watt-hour meter is configured on the line, and when the binary configuration variable takes 1, it means that a DC watt-hour meter is configured on the line.

[0018] Further, solving to obtain the node voltage state estimation value by using the quadratic constraint quadratic programming algorithm includes:

[0019] Introduce intermediate variables to linearize the first objective function of the state estimation model of the DC distribution network to obtain the second objective function without absolute value; wherein, the intermediate variables satisfy the relational expression established by the measured value and the measured equation value containing the state variables, and are restricted by the upper and lower bounds of the relational expression.

[0020] In the present invention, by introducing intermediate variables to process the objective function, a linearized objective function can be obtained, which is convenient for quickly solving the multi-objective optimization configuration model, thereby improving the efficiency of the metering table configuration for the DC distribution network.

[0021] Further, establishing the state estimation model of the DC distribution network according to the system parameters includes:

[0022] Taking the node voltage as the state variable, taking the node active power, the line active power and the line current as the measured values, taking the weighted sum of the absolute values of the differences between the measured values and the measured equation values containing the state variables as the first objective function, and taking the line active power measurement equation, the line current measurement equation and the node active power measurement equation as the constraint conditions, establish the state estimation model of the DC distribution network.

[0023] Further, calculating the node voltage estimation error rate according to the state estimation model and the corresponding measured values, and taking maximizing the node voltage estimation error rate as the first evaluation index includes:

[0024] Take the ratio of the absolute value of the difference between the node voltage state estimation value and the corresponding measured value of the state estimation model to the node voltage state estimation value as the node voltage estimation error rate, and take maximizing the node voltage estimation error rate as the first evaluation index.

[0025] Preferably, the first evaluation index can be expressed as:

[0026]

[0027] where Ω is the set of nodes, and are the voltage measured value of node i and the node voltage state estimation value respectively.

[0028] Preferably, the second objective function can be expressed as:

[0029] The second objective function can be expressed as:

[0030]

[0031] where N is the dimension of the measured values of the node active power, line active power and line current; ω k is the weight coefficient of the kth measured value corresponding to the state estimation; are the kth measured value and the state estimation value respectively; G k (x) represents the measurement equation containing state variables corresponding to the kth measured value. The measurement equation containing state variables includes: line active power measurement equation, line current measurement equation and node active power measurement equation, and the corresponding constraint conditions are constraint condition c1, constraint condition c2 and constraint condition c3 respectively; and are the line active power of line ij, the line current measurement equation value of line ij and the node active power measurement equation value of node i respectively; and are the node voltage state estimation values corresponding to node i and node j respectively; x is the state variable vector, α k is the first binary configuration variable after linearization processing, M is a preset positive number, z k is the intermediate variable corresponding to the kth measured value, the constraint condition c4 is the upper and lower bound constraint of the intermediate variable, g ij is the conductance parameter of line ij; Ω is the set of nodes.

[0032] Furthermore, the collection of the system parameters of the DC distribution network includes: collecting the node-line connection relationship of the DC distribution network, the conductance parameters of the lines, and the measured values of the node active power, line active power and line current.

[0033] In a second aspect, the present invention provides a multi-objective optimal configuration system for metering points in a DC distribution network, including:

[0034] A parameter unit for collecting system parameters of the DC distribution network and establishing a state estimation model of the DC distribution network according to the system parameters;

[0035] A first optimization objective unit for calculating the node voltage estimation error rate according to the state estimation model and corresponding measurement values, taking maximizing the node voltage estimation error rate as a first evaluation index, and taking minimizing the first evaluation index as a first optimization objective; wherein, the system parameters include: measurement values of node active power, line active power, and line current;

[0036] A second optimization objective unit for taking the number of DC watt-hour meters configured as a second evaluation index and taking minimizing the second evaluation index as a second optimization objective;

[0037] A multi-objective optimal configuration model unit for establishing a multi-objective optimal configuration model according to the first optimization objective and the second optimization objective;

[0038] A calculation unit for jointly solving the multi-objective optimal configuration model according to a genetic algorithm and a quadratic constraint quadratic programming algorithm to obtain the optimal configuration of metering points in the DC distribution network, so as to configure DC ammeters in the DC distribution network according to the optimal configuration of metering points. Description of the Drawings

[0039] Figure 1 is a schematic flow chart of a multi-objective optimal configuration method for metering points in a DC distribution network provided by an embodiment of the present invention;

[0040] Figure 2 is a schematic diagram of the topology and branch table settings of a multi-objective optimal configuration simulation model provided by an embodiment of the present invention;

[0041] Figure 3 is a schematic diagram of the solution result of a multi-objective optimal configuration model of a multi-objective optimal configuration simulation model provided by an embodiment of the present invention;

[0042] Figure 4 is a schematic structural diagram of a multi-objective optimal configuration system for metering points in a DC distribution network provided by an embodiment of the present invention. Detailed Embodiments

[0043] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0044] See Figure 1 , which is a schematic flow chart of a multi-objective optimal configuration method for metering points in a DC distribution network provided by an embodiment of the present invention, including steps S11 to S14, specifically:

[0045] Step S11: Collect the system parameters of the DC distribution network, and establish a state estimation model of the DC distribution network according to the system parameters.

[0046] Among them, the collection of the system parameters of the DC distribution network includes: collecting the node-line connection relationship of the DC distribution network, the conductance parameters of the lines, and the measured values of the node active power, line active power, and line current. Specifically, collect the network topology structure parameters, key equipment parameters, and the measured values of nodes and lines of the DC distribution network; among them, the measured values include: the injected node active power and voltage measured values of each power generation and consumption node, and the line active power and line current transmitted by the lines.

[0047] Collect the network topology structure parameters, key equipment parameters, measured values of node quantities, and measured values of lines of the DC distribution network; among them, the network topology structure parameters of the DC distribution network include: the node-line connection relationship of the DC distribution network and the impedance value of the DC line; the key equipment parameters of the DC distribution network include: the rated capacity, conversion efficiency, and loss parameters of the AC / DC converter and DC / DC converter; the measured values of the nodes of the DC distribution network include: the injected active power and actual voltage measured values of each power generation and consumption node; the measured values of the lines of the DC distribution network include: the line active power and current transmitted by each line.

[0048] According to the system parameters, establishing a state estimation model of the DC distribution network includes:

[0049] Taking the node voltage as the state variable, taking the node active power, line active power, and line current as the measured values, and taking the weighted sum of the absolute values of the differences between the measured values and the measured equation values containing the state variables as the first objective function, and taking the line active power measurement equation, line current measurement equation, and node active power measurement equation as the constraint conditions, establish a state estimation model of the DC distribution network.

[0050] By means of metering devices, i.e., meters, configured on the nodes and lines of a DC distribution network, redundant data such as the active power, current, and voltage of the nodes and lines can be collected. Based on the collected redundant data, state estimation of the DC distribution network can be carried out, and different degrees of redundant data will produce state estimation results with different accuracies. The present invention calculates the observation accuracy through the state estimation of the DC distribution network. Specifically, taking the node voltage as the state variable and the actual voltage, node active power, branch active power, and branch current as the measured values, a state estimation model of the DC distribution network is established.

[0051] Preferably, the first objective function of the state estimation model is to minimize the sum of the absolute values of the differences between the estimated node voltage values and the measured values; among them, the measured values used include: node active power, line active power, and line current.

[0052] Preferably, the state estimation model can be expressed as:

[0053]

[0054] where N is the dimension of the measured values of the node active power, line active power, and line current; ω k is the weight coefficient corresponding to the kth measured value for state estimation; are respectively the kth measured value and the state estimation G corresponding to the measured value k (x) represents the measurement equation containing the state variable corresponding to the kth measured value, and the measurement equation containing the state variable includes: line active power measurement equation, line current measurement equation, and node active power measurement equation; x is the state variable vector, and Ω is the set of nodes.

[0055] Preferably, the line active power measurement equation, line current measurement equation, and node active power measurement equation can be respectively expressed as:

[0056]

[0057]

[0058]

[0059] where, and are respectively the line active power of line ij, the estimated line current of line ij, and the node active power of node i; and are respectively the state variables corresponding to nodes i and j; g ij is the admittance of line ij; Ω is the set of nodes.

[0060] By solving the above DC distribution network state estimation model, the state estimation of the voltage at each node can be obtained. Comparing it with the actual voltage measurement value, and defining the maximum estimation error rate of the node voltage as the first evaluation index of the observation accuracy.

[0061] Step S12: According to the state estimation model and the corresponding measurement values, calculate the node voltage estimation error rate, take the maximum node voltage estimation error rate as the first evaluation index, and take minimizing the first evaluation index as the first optimization goal; wherein, the system parameters include: the measured values of node active power, line active power, and line current.

[0062] Among them, according to the state estimation model and the corresponding measurement values, calculating the node voltage estimation error rate, and taking the maximum node voltage estimation error rate as the first evaluation index includes: taking the ratio of the absolute value of the difference between the node voltage state estimation value of the state estimation model and the corresponding measurement value to the node voltage state estimation value as the node voltage estimation error rate, and taking maximizing the node voltage estimation error rate as the first evaluation index.

[0063] Preferably, the first evaluation index can be expressed as:

[0064]

[0065] wherein, Ω is the set of nodes, and are the voltage measurement value of node i and the node voltage state estimation value respectively.

[0066] Step S13: Take the number of DC watt-hour meters configured as the second evaluation index, take minimizing the second evaluation index as the second optimization goal, and establish a multi-objective optimization configuration model according to the first optimization goal and the second optimization goal.

[0067] It should be noted that in the DC distribution network, the data measurement of nodes and lines all uses DC watt-hour meters. Therefore, the number of DC watt-hour meters configured is used as the second evaluation index of the economy of its metering equipment. The second evaluation index can be expressed as:

[0068] λ2 = N m ,

[0069] wherein, N m is the number of DC watt-hour meters configured.

[0070] The first evaluation index for characterizing the observation accuracy and the second evaluation index for characterizing the cost respectively correspond to the first optimization goal and the second optimization goal of the multi-objective optimization configuration model.

[0071] Preferably, the multi-objective optimization configuration model is a multi-objective model that minimizes the first optimization objective and the second optimization objective, and the multi-objective optimization configuration model can be expressed as:

[0072]

[0073] s.t.s ij ∈ {0, 1},

[0074] where s ij is the second binary configuration variable without linearization processing, Ω is the set of nodes, and are the voltage measurement value and the state estimate of the voltage at node i respectively, f1 and λ1 are the first optimization objective and the first evaluation index respectively, and f2 and λ2 are the second optimization objective and the second evaluation index respectively.

[0075] Specifically, when the second binary configuration variable takes 0, it means that a DC ammeter is not configured on this line, and when the second binary configuration variable takes 1, it means that a DC ammeter is configured on this line.

[0076] Exemplarily, the second binary configuration variable can be expressed as:

[0077]

[0078] where the measurement values include: node active power, line active power, and line current.

[0079] Step S14: Jointly solve the multi-objective optimization configuration model according to the genetic algorithm and the quadratic constraint quadratic programming algorithm to obtain the optimal configuration of the metering points of the DC distribution network, so as to configure the DC ammeters of the DC distribution network according to the optimal configuration of the metering points.

[0080] Jointly solving the multi-objective optimization configuration model according to the genetic algorithm and the quadratic constraint quadratic programming algorithm to obtain the optimal configuration of the metering points of the DC distribution network includes: obtaining the first metering point configuration result of the line according to the genetic algorithm, and in each iteration process of the genetic algorithm, solving to obtain the node voltage state estimation value by using the quadratic constraint quadratic programming algorithm according to the first metering point configuration result; calculating the first evaluation index according to the node voltage state estimation value and the corresponding measurement value to obtain the first evaluation index result, and counting the number of DC energy meter configurations to obtain the second evaluation index result; where the first metering point configuration result corresponding to the optimal first evaluation index result and the second evaluation index result is used as the second metering point configuration result; when the preset population evolution algebra is reached or the preset convergence condition is satisfied, the second metering point configuration result and the corresponding first evaluation index result and second evaluation index result are used as the optimal configuration of the metering points of the DC distribution network.

[0081] The present invention obtains the metering point configuration result through a genetic algorithm. During the iteration process, a quadratic constraint quadratic programming is used to solve the first evaluation index, so that the optimal first evaluation index representing the accuracy can be obtained for each iteration. After the iteration ends, the first evaluation indexes corresponding to all metering point configuration results are obtained, and thus the optimal metering point configuration result is used as the optimal configuration of the metering points in the DC distribution network. By jointly solving the genetic algorithm and the quadratic constraint quadratic programming, the accuracy of multi-objective configuration can be improved, the subjective judgment relying too much on the engineering experience of the staff can be avoided, the metering resolution of the DC distribution network can be improved, and the configuration cost of the metering points in the DC distribution network can be reduced.

[0082] Among them, obtaining the first metering point configuration result of the line according to the genetic algorithm includes: defining the binary configuration variable of the line metering point, and initializing the first metering point configuration result of the population size according to the binary configuration variable; wherein, when the binary configuration variable takes 0, it means that no DC energy meter is configured on this line, and when the binary configuration variable takes 1, it means that a DC energy meter is configured on this line.

[0083] Using the quadratic constraint quadratic programming algorithm to solve and obtain the node voltage state estimation value includes: introducing an intermediate variable to linearize the first objective function of the state estimation model of the DC distribution network to obtain a second objective function without absolute value; wherein, the intermediate variable satisfies the relational expression established by the measured value and the measured equation value containing the state variable, and is restricted by the upper and lower bounds of the relational expression.

[0084] The present invention adopts introducing an intermediate variable to process the objective function, and can obtain a linearized objective function, which is convenient for quickly solving the multi-objective optimization configuration model, thereby improving the efficiency of the metering table configuration for the DC distribution network.

[0085] Exemplarily, according to the genetic algorithm, the optimal configuration of the metering points in the upper layer of the multi-objective optimization configuration model is solved, that is, the configuration situation of the metering tables for each line is obtained according to the genetic algorithm. Specifically, when initializing the genetic algorithm, the binary configuration variable of the line is defined, and a line metering point configuration scheme with a set population size is initialized and generated; for each line metering point configuration scheme, the quadratic constraint quadratic programming algorithm is used to solve the first optimization objective. It should be noted that the essence of the DC distribution network observation accuracy evaluation model is a quadratic constraint quadratic programming model with absolute value terms, and the absolute value data items of the first objective function of the first optimization objective need to be linearized to obtain the second objective function.

[0086] Preferably, the second objective function can be expressed as:

[0087]

[0088] where N is the dimension of the measured values of the active power of the node, the active power of the line, and the line current; ω k is the weight coefficient of the k-th measured value corresponding to the state estimation; are the k-th measured value and the state estimation value respectively; G k (x) represents the measurement equation containing state variables corresponding to the k-th measured value. The measurement equation containing state variables includes: the active power measurement equation of the line, the current measurement equation of the line, and the active power measurement equation of the node. The corresponding constraint conditions are constraint condition c1, constraint condition c2, and constraint condition c3 respectively; and are the active power of line ij, the measured value of the current measurement equation of line ij, and the measured value of the active power measurement equation of node i respectively; and are the node voltage state estimation values corresponding to nodes i and j respectively; x is the state variable vector, α k is the first binary configuration variable after linearization processing, M is a preset positive number, z k is the intermediate variable corresponding to the k-th measured value, and constraint condition c4 is the upper and lower bound constraint of the intermediate variable, g ij is the conductance parameter of line ij; Ω is the node set.

[0089] Similarly, when the first binary configuration variable takes 0, it means that a DC ammeter is not configured on this line; when the first binary configuration variable takes 1, it means that a DC ammeter is configured on this line.

[0090] After the above linearization processing, the quadratic constraint quadratic programming algorithm can be used to quickly solve the state estimation of the voltage of each node, and then calculate the observation accuracy of the DC distribution network. For the metering point configuration of each line, count the number of DC watt-hour meters configured to obtain the number of DC watt-hour meters configured for the second optimization target; through the calculation of crossover, mutation, and selection operators, retain the scheme with the optimal objective value in each generation of population; when the set population evolution algebra is reached or the convergence condition is satisfied, output the optimal configuration of the metering points of the DC distribution network. Among them, the optimal configuration of the metering points includes the results of the first optimization target and the results of the second optimization target.

[0091] Exemplarily, see Figure 2, which is a schematic diagram of the topology and branch table settings of the multi-objective optimization configuration simulation model provided by an embodiment of the present invention. The figure includes: 6 distributed photovoltaics, 2 electric vehicle charging loads, 1 energy storage device, and several conventional DC loads. Considering the influence of on-site operating condition changes on the metering accuracy of DC energy meters, a random error that follows a normal distribution with a mean of 0 and a standard deviation of 5% of the measured value is superimposed on each measured value. In addition to configuring metering points at each node, a total of 32 lines need to be decided whether to configure DC energy meters. According to the above steps, a multi-objective optimization configuration model for metering points in a DC distribution network is established. See Figure 3 , which is a schematic diagram of the solution results of the multi-objective optimization configuration model of the multi-objective optimization configuration simulation model provided by an embodiment of the present invention. The results show that when the number of configured line metering points increases, the estimated error rate of node iodine salt maximization decreases accordingly. It can be seen that the method of the present invention can provide a multi-objective optimization configuration scheme for metering points in a DC distribution network, effectively taking into account the configuration cost and metering accuracy.

[0092] See Figure 4 , which is a schematic diagram of the structure of the multi-objective optimization configuration system for metering points in a DC distribution network provided by an embodiment of the present invention, including: a parameter unit 41, a first optimization objective unit 42, a second optimization objective unit 43, a multi-objective optimization configuration model unit 44, and a calculation unit 45.

[0093] After the parameter unit 41 obtains the system parameters, it transmits the system parameters to the first optimization objective unit 42 and the second optimization objective unit 43; after the first optimization objective unit 42 and the second optimization objective unit 43 respectively receive the system parameters, they respectively establish a first optimization objective representing the observation accuracy and a second optimization objective representing the configuration cost. The first optimization objective unit 42 and the second optimization objective unit 43 then respectively transmit the first optimization objective and the second optimization objective to the multi-objective optimization configuration model unit 44; after the multi-objective optimization configuration model unit 44 receives the first optimization objective and the second optimization objective, it establishes a combined multi-objective optimization configuration model and transmits the multi-objective optimization configuration model to the calculation unit 45; after the calculation unit 45 receives the multi-objective optimization configuration model, it performs a combined solution according to the genetic algorithm and quadratic constraint quadratic programming to obtain the optimal configuration of the metering points of the DC distribution network, so as to configure the DC ammeters of the DC distribution network according to the optimal configuration of the metering points.

[0094] The parameter unit 41 is used to collect the system parameters of the DC distribution network and establish a state estimation model of the DC distribution network according to the system parameters.

[0095] Among them, the collection of the system parameters of the DC distribution network includes: collecting the node-line connection relationship of the DC distribution network, the conductance parameters of the lines, and the measured values of node active power, line active power, and line current.

[0096] Based on the system parameters, a state estimation model of the DC distribution network is established, including: taking the node voltage as the state variable, taking the node active power, line active power, and line current as the measured values, and taking the weighted sum of the absolute values of the differences between the measured values and the values of the measurement equations containing the state variables as the first objective function, and taking the line active power measurement equation, line current measurement equation, and node active power measurement equation as the constraint conditions to establish the state estimation model of the DC distribution network.

[0097] The first optimization objective unit 42 is used to calculate the node voltage estimation error rate according to the state estimation model and the corresponding measured values, take maximizing the node voltage estimation error rate as the first evaluation index, and take minimizing the first evaluation index as the first optimization objective; wherein, the system parameters include: the measured values of node active power, line active power, and line current.

[0098] Among them, calculating the node voltage estimation error rate according to the state estimation model and the corresponding measured values, and taking maximizing the node voltage estimation error rate as the first evaluation index includes: taking the ratio of the absolute value of the difference between the node voltage state estimation value of the state estimation model and the corresponding measured value to the node voltage state estimation value as the node voltage estimation error rate, and taking maximizing the node voltage estimation error rate as the first evaluation index.

[0099] Preferably, the first evaluation index can be expressed as:

[0100]

[0101] Among them, Ω is the set of nodes, and are the voltage measured value of node i and the node voltage state estimation value respectively.

[0102] The second optimization objective unit 43 is used to take the number of DC ammeters configured as the second evaluation index, and take minimizing the second evaluation index as the second optimization objective.

[0103] The multi-objective optimization configuration model unit 44 is used to establish a multi-objective optimization configuration model according to the first optimization objective and the second optimization objective.

[0104] The calculation unit 45 is used to jointly solve the multi-objective optimization configuration model according to the genetic algorithm and the quadratic constraint quadratic programming algorithm to obtain the optimal configuration of the metering points of the DC distribution network, so as to configure the DC ammeters of the DC distribution network according to the optimal configuration of the metering points.

[0105] The multi-objective optimal configuration model is jointly solved according to the genetic algorithm and the quadratic constrained quadratic programming algorithm to obtain the optimal configuration of the metering points of the DC distribution network, including: obtaining the first metering point configuration result of the line according to the genetic algorithm, and in each iteration process of the genetic algorithm according to the first metering point configuration result, using the quadratic constrained quadratic programming algorithm to solve and obtain the node voltage state estimation value; calculating the first evaluation index according to the node voltage state estimation value and the corresponding measured value to obtain the first evaluation index result, and counting the number of DC watt-hour meter configurations to obtain the second evaluation index result; wherein, the first metering point configuration result corresponding to the optimal of the first evaluation index result and the second evaluation index result is used as the second metering point configuration result; when the preset population evolution algebra is reached or the preset convergence condition is satisfied, the second metering point configuration result and the corresponding first evaluation index result and second evaluation index result are used as the optimal configuration of the metering points of the DC distribution network.

[0106] Among them, obtaining the first metering point configuration result of the line according to the genetic algorithm includes: defining the binary configuration variable of the line metering point, and initializing the first metering point configuration result of the population size according to the binary configuration variable; wherein, when the binary configuration variable takes 0, it means that no DC watt-hour meter is configured on this line, and when the binary configuration variable takes 1, it means that a DC watt-hour meter is configured on this line.

[0107] Using the quadratic constrained quadratic programming algorithm to solve and obtain the node voltage state estimation value includes: introducing an intermediate variable to linearize the first objective function of the state estimation model of the DC distribution network to obtain the second objective function without absolute value; wherein, the intermediate variable satisfies the relational expression established by the measured value and its corresponding measured equation value containing the state variable, and is restricted by the upper and lower bounds of the relational expression.

[0108] Preferably, the second objective function can be expressed as:

[0109]

[0110] Among them, N is the dimension of the measured values of the node active power, line active power and line current; ω k is the weight coefficient corresponding to the state estimation of the kth measured value; are respectively the kth measured value and the state estimation value; G k (x) represents the measured equation containing the state variable corresponding to the kth measured value, and the measured equation containing the state variable includes: the line active power measured equation, the line current measured equation and the node active power measured equation, and the corresponding constraint conditions are constraint condition c1, constraint condition c2 and constraint condition c3 respectively; and They are respectively the active power of line ij, the measured equation value of the line current of line ij, and the measured equation value of the active power of node i; and the estimated values of the node voltages corresponding to nodes i and j respectively; x is the state variable vector, and α k is the first binary configuration variable after linearization processing, M is a preset positive number, and z k is the intermediate variable corresponding to the k-th measured value. The constraint condition c4 is the upper and lower bound constraint of the intermediate variable, and g ij is the conductance parameter of line ij; Ω is the set of nodes.

[0111] The present invention estimates the state of the node voltages in the DC distribution network, obtains the estimated error rate of the DC distribution network based on the state estimation and the actual voltage measurement values, uses the estimated error rate as the first evaluation index for state evaluation, and configures the metering devices according to the first evaluation index. Compared with the prior art, it is not necessary to rely on the engineering experience of the staff to configure the metering devices, so as to avoid subjective and empirical configuration of the metering devices, and further improve the configuration accuracy and feasibility of the metering devices; moreover, taking the number of configured DC energy meters as the second evaluation index, a joint multi-objective model is established to maximize the estimated error rate and minimize the number of configured DC energy meters, and the genetic algorithm and quadratic constraint quadratic programming are used for joint solution, which can solve the overall optimal solution of the model, so as to improve the dosage resolution while reducing the cost of configuring the number of metering devices, and thus has higher applicability.

[0112] Those skilled in the art should understand that the embodiments of the present application may also provide a computer program product. Therefore, the present application may be implemented in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application may be implemented in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0113] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0114] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to work in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction device that implements the functions specified in one or more processes and / or one or more blocks of the process Figure 1 one process or a plurality of processes and / or blocks Figure 1 of the function specified in one or more blocks.

[0115] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, such that a series of operational steps are performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one or more processes and / or one or more blocks of the process Figure 1 one process or a plurality of processes and / or blocks Figure 1 of the function specified in one or more blocks.

[0116] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the technical principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. A multi-objective optimal configuration method for metering points in a DC distribution network, characterized in that, Including: Collecting system parameters of a DC distribution network, and establishing a state estimation model of the DC distribution network according to the system parameters; Calculating a node voltage estimation error rate according to the state estimation model and corresponding measured values, taking maximizing the node voltage estimation error rate as a first evaluation index, and taking minimizing the first evaluation index as a first optimization objective; wherein, the system parameters include: measured values of node active power, line active power, and line current; Taking the number of configured DC energy meters as a second evaluation index, taking minimizing the second evaluation index as a second optimization objective, and establishing a multi-objective optimization configuration model according to the first optimization objective and the second optimization objective; Jointly solving the multi-objective optimization configuration model according to a genetic algorithm and a quadratic constraint quadratic programming algorithm to obtain an optimal configuration of metering points of the DC distribution network, so as to configure DC ammeters of the DC distribution network according to the optimal configuration of metering points; Wherein, the jointly solving the multi-objective optimization configuration model according to a genetic algorithm and a quadratic constraint quadratic programming algorithm to obtain an optimal configuration of metering points of the DC distribution network includes: Obtaining a first metering point configuration result of a line according to a genetic algorithm, and solving to obtain a node voltage state estimation value by using a quadratic constraint quadratic programming algorithm in each iteration process of the genetic algorithm according to the first metering point configuration result; Calculating the first evaluation index according to the node voltage state estimation value and corresponding measured values to obtain a first evaluation index result, and counting the number of configured DC energy meters to obtain a second evaluation index result; wherein, taking the first metering point configuration result corresponding to the optimal first evaluation index result and second evaluation index result as a second metering point configuration result; When a preset population evolution algebra is reached or a preset convergence condition is satisfied, taking the second metering point configuration result and corresponding first evaluation index result and second evaluation index result as the optimal configuration of metering points of the DC distribution network.

2. The multi-objective optimal configuration method for metering points in a DC distribution network according to claim 1, wherein, The obtaining a first metering point configuration result of a line according to a genetic algorithm includes: Defining binary configuration variables of line metering points, and initializing a first metering point configuration result of a population size according to the binary configuration variables; wherein, when the binary configuration variable takes 0, it means that no DC energy meter is configured on this line, and when the binary configuration variable takes 1, it means that a DC energy meter is configured on this line.

3. The multi-objective optimal configuration method for metering points in a DC distribution network according to claim 1, characterized in that, The solving to obtain a node voltage state estimation value by using a quadratic constraint quadratic programming algorithm includes: Introducing intermediate variables to linearly process a first objective function of the state estimation model of the DC distribution network to obtain a second objective function without absolute value; wherein, the intermediate variables satisfy a relational expression established by a measured value and a measured equation value containing state variables, and are restricted by upper and lower bounds of the relational expression.

4. The multi-objective optimal configuration method for metering points in a DC distribution network according to claim 1, characterized in that The establishing a state estimation model of the DC distribution network according to the system parameters includes: Taking the node voltage as the state variable, the node active power, line active power and line current as the measurement values, and the weighted sum of the absolute values of the differences between the measurement values and the corresponding measurement equation values containing the state variables as the first objective function, and taking the line active power measurement equation, line current measurement equation and node active power measurement equation as the constraint conditions, a state estimation model of the DC distribution network is established.

5. The multi-objective optimal configuration method for metering points in a DC distribution network according to claim 1, characterized in that, Calculating the node voltage estimation error rate according to the state estimation model and the corresponding measurement values, and taking maximizing the node voltage estimation error rate as the first evaluation index, including: Taking the ratio of the absolute value of the difference between the node voltage state estimation value of the state estimation model and the corresponding measurement value to the node voltage state estimation value as the node voltage estimation error rate, and taking maximizing the node voltage estimation error rate as the first evaluation index.

6. The multi-objective optimal configuration method for metering points in a DC distribution network according to claim 5, wherein, The first evaluation index can be expressed as: where Ω is the set of nodes, and are the measured voltage value of node i and the estimated value of node voltage state respectively.

7. The multi-objective optimal configuration method for metering points in a DC distribution network according to claim 3, characterized in that The second objective function can be expressed as: where N is the dimension of the measured values of the active power of the node, the active power of the line, and the line current; ω k is the weight coefficient of the k-th measured value corresponding to the state estimation; are respectively the k-th measured value and the state estimation value; G k (x) represents the measurement equation with state variables corresponding to the k-th measured value. The measurement equation with state variables includes: the active power measurement equation of the line, the line current measurement equation, and the active power measurement equation of the node. The corresponding constraint conditions are constraint condition c1, constraint condition c2, and constraint condition c3 respectively; and are respectively the active power of line ij, the value of the line current measurement equation of line ij, and the value of the active power measurement equation of node i; and are respectively the state estimation values of the node voltages corresponding to nodes i and j; x is the state variable vector, α k is the first binary configuration variable after linearization processing, M is a preset positive number, z k is the intermediate variable corresponding to the k-th measured value, and the constraint condition c4 is the upper and lower bound constraint of the intermediate variable, g ij is the conductance parameter of line ij; Ω is the node set.

8. The multi-objective optimal configuration method for metering points in a DC distribution network according to claim 1, wherein Collecting the system parameters of the DC distribution network, including: collecting the node-line connection relationship of the DC distribution network, the conductance parameters of the lines, and the measurement values of the node active power, line active power and line current.

9. A multi-objective optimal configuration system for metering points in a DC distribution network, characterized in that, Including: A parameter unit for collecting the system parameters of the DC distribution network and establishing a state estimation model of the DC distribution network according to the system parameters; A first optimization objective unit for calculating the node voltage estimation error rate according to the state estimation model and the corresponding measurement values, taking maximizing the node voltage estimation error rate as the first evaluation index, and taking minimizing the first evaluation index as the first optimization objective; wherein, the system parameters include: the measurement values of the node active power, line active power and line current; A second optimization objective unit for taking the number of configured DC wattmeters as the second evaluation index and taking minimizing the second evaluation index as the second optimization objective; A multi-objective optimization configuration model unit for establishing a multi-objective optimization configuration model according to the first optimization objective and the second optimization objective; A calculation unit for jointly solving the multi-objective optimization configuration model according to the genetic algorithm and the quadratic constraint quadratic programming algorithm to obtain the optimal configuration of the metering points of the DC distribution network, so as to configure the DC ammeters of the DC distribution network according to the optimal configuration of the metering points; Among them, jointly solving the multi-objective optimization configuration model according to the genetic algorithm and the quadratic constraint quadratic programming algorithm to obtain the optimal configuration of the metering points of the DC distribution network, including: Obtaining the first metering point configuration result of the line according to the genetic algorithm, and in each iteration process of the genetic algorithm, solving to obtain the node voltage state estimation value by using the quadratic constraint quadratic programming algorithm; Calculating the first evaluation index according to the node voltage state estimation value and the corresponding measurement values to obtain the first evaluation index result, and counting the number of configured DC wattmeters to obtain the second evaluation index result; wherein, taking the first metering point configuration result corresponding to the optimal first evaluation index result and the second evaluation index result as the second metering point configuration result; When the preset population evolution algebra is reached or the preset convergence condition is satisfied, the second metering point configuration result, the corresponding first evaluation index result, and the second evaluation index result are used as the optimal configuration of the metering points of the DC distribution network.

Images