A distribution network adaptive dynamic state estimation method, device, equipment and medium

Abstract

The present invention relates to a distribution network adaptive dynamic state estimation method, device, equipment and medium, wherein the method comprises the following steps: establishing a distribution network dynamic state estimation model, including a state equation and a measurement equation, wherein the measurement equation is a linear measurement equation for power system state estimation based on PMU measurement; setting a measurement noise covariance matrix based on the statistical characteristics of the phasor error of the PMU device, and adaptively estimating the measurement noise according to the latest measurement before each state estimation; using the Kalman filtering method to perform distribution network dynamic state estimation, and in the Kalman filtering process, considering the time-varying posterior statistics of the process noise, and adaptively adjusting the process noise covariance matrix through the forgetting memory method. Compared with the prior art, the present invention has the advantages of being able to accurately estimate the degree of state change and being able to track the dynamic changes of the system.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network state estimation, and in particular to a distribution network adaptive dynamic state estimation method, device, equipment and medium. Background Art

[0002] State estimation is a method that can use the redundancy of real-time measurement systems to improve data accuracy, automatically eliminate erroneous information caused by random interference, and estimate or predict the operating status of the system. State estimation in the distribution network can quickly and accurately obtain the real-time state and dynamic power flow of the entire distribution network system when the measurement data is incomplete. The emergence and development of synchronized phasor measurement units (PMUs) provide a new way to obtain real-time measurement data and dynamic power flow in distribution networks. The data obtained by PMUs have exact time stamps, short data update cycles, and can directly measure the voltage phasors of nodes and the current phasors of branches. Based on these characteristics, PMUs can bring more and more accurate measurement data to distribution network state estimation. If PMUs are configured on a large scale in the distribution network, only the voltage and current phasor measurements are sufficient to estimate the system state, which improves the calculation efficiency and accuracy. Therefore, PMUs are widely used in distribution network state estimation.

[0003] The setting of the system process noise covariance matrix Q is the key to whether the dynamic state estimation can track the dynamic changes of the system. At present, when the system is in steady state, since the state changes little, the setting of the system process noise covariance matrix Q can be directly set to a constant based on prior knowledge. However, when the system changes, this method cannot accurately estimate the degree of change of the state. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a distribution network adaptive dynamic state estimation method, device, equipment and medium. Based on PMU measurement, the measurement noise is adaptively estimated according to the latest measurement before each state estimation, and the system process noise is reflected according to the average value and variance of the difference between the measured quantity and the predicted quantity, and the adaptive adjustment of the process noise covariance matrix is ​​set to realize dynamic state estimation.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] According to a first aspect of the present invention, a method for adaptive dynamic state estimation of a distribution network is provided, the method comprising the following steps:

[0007] Establishing a distribution network dynamic state estimation model, including a state equation and a measurement equation, wherein the measurement equation is a linear measurement equation for power system state estimation based on PMU measurement;

[0008] Based on the statistical characteristics of the phasor error of the PMU device, the measurement noise covariance matrix is ​​set to adaptively estimate the measurement noise based on the latest measurement before each state estimation;

[0009] The Kalman filtering method is used to estimate the dynamic state of the distribution network. In the Kalman filtering process, the time-varying posterior statistics of the process noise are considered, and the covariance matrix of the process noise is adaptively adjusted through the forgetting memory method.

[0010] As a preferred technical solution, the process of establishing the measurement equation is:

[0011] When the PMU measurement satisfies the system observability, the node voltage phasor is used as the state quantity, and the linear measurement equation for power system state estimation based on PMU measurement is obtained:

[0012] ,

[0013] In the formula, and They are PMU node voltage phasor measurement and branch current phasor measurement respectively; E for n dimensional identity matrix, Y is the node-branch admittance matrix, and the Jacobian matrix of the measurement equation is expressed as ; and are the measurement noise of voltage phasor and current phasor respectively.

[0014] As a preferred technical solution, the statistical characteristics of the phasor error of the PMU device are used to set the measurement noise covariance matrix, and the measurement noise is adaptively estimated according to the latest measurement before each state estimation, specifically:

[0015] Assumptions is an element measured by PMU, and its amplitude and phase measurement errors are and , according to the Euclidean distance, The measurement error It is expressed as:

[0016] ,

[0017] In the formula, express The true value of , Respectively The amplitude and phase of

[0018] Assume that the amplitude error measured by the PMU is and phase error The mean is and , the standard deviation is and The phasor error is The mean of is expressed as:

[0019] ,

[0020] In the formula, ,right Perform Taylor expansion and ignore higher-order terms to obtain , then the phasor error The variance of is expressed as:

[0021] ,

[0022] Assuming that the amplitude and phase measured by the PMU are independently distributed with a zero-mean Gaussian distribution, and are all equal to 0, and and Independent of each other, phasor error The variance of simplifies to:

[0023] ,

[0024] Use phasor amplitude measurement To replace the unknown phasor amplitude true value , in the known and In the case of statistical properties, The value of is approximately estimated by mathematical statistics methods;

[0025] The measurement noise covariance matrix is ​​set by the diagonal matrix of the measurement error variance R , the measurement noise is adaptively estimated based on the latest measurement before each state estimation:

[0026] ,

[0027] In the formula, express R The diagonal elements of The first i A measurement.

[0028] As a preferred technical solution, the method for adaptively adjusting the process noise covariance matrix is:

[0029] The system process noise is approximated as the difference between the state estimate and the predicted value, and an unbiased estimate of the process noise variance is calculated;

[0030] The weight factor of adaptive variation of process noise is given by forgetting memory method;

[0031] The unbiased estimate of the process noise variance is adjusted by the weight factor, and the adaptive adjustment equation of the process noise covariance matrix is ​​obtained.

[0032] As a preferred technical solution, the system process noise is approximately:

[0033] ,

[0034] Among them, ω k and v k are the system process noise and measurement noise respectively, x k is the true value of the state at time k, is the estimated value of the state at time k, is the estimation error at time k, f() is the state transfer equation, which reflects the dynamic changes of the distribution network operation state. is the state transfer matrix after the linearization of the state transfer equation, K k is the Kalman filter gain, r k is the innovation value, that is, the difference between the measured value and the measured predicted value.

[0035] As a preferred technical solution, the unbiased estimate of the process noise variance is:

[0036] ,

[0037] Among them, Q k is the process noise covariance matrix at time k, N is the number of samples, is the average value of the system process noise in N samplings, and P is the estimation error covariance matrix.

[0038] As a preferred technical solution, the adaptive adjustment equation of the process noise covariance matrix is ​​expressed as:

[0039] ,

[0040] in,

[0041] ,

[0042] Q k is the process noise covariance matrix at time k, d k is the weight factor, b is the forgetting factor, and b k is the kth power of the forgetting factor.

[0043] As a preferred technical solution, the state equation is expressed as: ;

[0044] The measurement equation is expressed as: ;

[0045] Among them, ω k and v k are the system process noise and measurement noise, ω k and v k The covariances of are Q and R respectively.

[0046] As a preferred technical solution, the Kalman filtering process is divided into two parts: prediction update part and new information update part, wherein the prediction update part calculates the state variables at the next moment according to the current system state and the system state equation, and solves the prediction error covariance matrix; the new information update part is when the measurement information at the next moment is received, the state value is updated considering the statistical characteristics of the noise, and the estimation error covariance matrix is ​​calculated; the calculation equation is expressed as:

[0047] Prediction equation: ;

[0048] Innovation correction equation: ;

[0049] Among them, Q k is the process noise covariance matrix at time k, F k for k The state transfer matrix at time , is the estimated value of the state at time k, is the state prediction value at time k, P k for k The estimated error covariance matrix at time , for k The forecast error covariance matrix at time t, H k for k The Jacobian matrix of the measurement equation at the moment, R k for k The measurement noise covariance matrix at time z k for k The measured value at the moment, K k for k Kalman filter gain at time t.

[0050] According to a second aspect of the present invention, there is provided a distribution network adaptive dynamic state estimation device for implementing the above method, the device comprising:

[0051] State estimation model building module: building a distribution network dynamic state estimation model, including a state equation and a measurement equation, wherein the measurement equation is a linear measurement equation for power system state estimation based on PMU measurement;

[0052] Measurement noise adaptive estimation module: Based on the statistical characteristics of the phasor error of the PMU device, the measurement noise covariance matrix is ​​set, and the measurement noise is adaptively estimated according to the latest measurement before each state estimation;

[0053] Dynamic state estimation module: The Kalman filtering method is used to estimate the dynamic state of the distribution network. During the Kalman filtering process, the time-varying posterior statistics of the process noise are considered, and the process noise covariance matrix is ​​adaptively adjusted through the forgetting memory method.

[0054] According to a third aspect of the present invention, there is provided an electronic device, comprising a memory and a processor, wherein a computer program is stored in the memory, and the method described above is implemented when the processor executes the program.

[0055] According to a fourth aspect of the present invention, there is provided a computer-readable storage medium having a computer program stored thereon, wherein the program implements the method described when executed by a processor.

[0056] Compared with the prior art, the present invention sets the measurement noise covariance matrix according to PMU measurement, has higher estimation accuracy, and helps to improve the accuracy of subsequent adaptive adjustment of process noise covariance; on the basis of adaptively setting the measurement noise covariance matrix, the present invention reflects the system process noise according to the average value and variance of the difference between the measured quantity and the predicted quantity, and realizes adaptive adjustment of the process noise covariance matrix considering the time-varying a posteriori statistics of the process noise, thereby realizing dynamic state estimation, and can well track the dynamic changes of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 is a flow chart of the method of the present invention;

[0058] Figure 2 The true value and state estimation value of the voltage phasor under the dynamic state estimation of process noise without adaptive setting in an embodiment, wherein (2a) and (2b) are the comparisons of the true value and state estimation value of the voltage amplitude of two nodes, respectively, and (2c) and (2d) are the true value and state estimation value of the voltage phase angle of two nodes, respectively;

[0059] Figure 3 The true value and state estimation value of the voltage phasor under the dynamic state estimation of the process noise in an adaptive setting in an embodiment, wherein (3a) and (3b) are respectively the comparison between the true value and state estimation value of the voltage amplitude of two nodes, and (3c) and (3d) are respectively the true value and state estimation value of the voltage phase angle of two nodes;

[0060] Figure 4: is the amplitude and phase error of each node with or without adaptive update of Q in an embodiment, wherein (4a) and (4b) are the amplitude and phase error of the unobservable node, respectively, and (4c) and (4d) are the amplitude and phase error of the observable node, respectively;

[0061] Figure 5 is a dynamic state estimation value of an unobservable node in an embodiment, wherein (5a) and (5b) are comparisons of the true value and the state estimation value of the voltage amplitude of the two nodes, and (5c) and (5d) are comparisons of the true value and the state estimation value of the voltage phase angle of the two nodes, respectively;

[0062] Figure 6 1 is an amplitude and phase error curve under different forgetting factors b in an embodiment, wherein (6a), (6b), (6c), and (6d) represent the amplitude and phase error curves when b=0.995, b=0.95, b=0.98, and b=0.65, respectively;

[0063] Figure 7 is the system topology diagram in Example 2;

[0064] Figure 8 are the true value and state estimated value of the system power flow calculation in Example 2, wherein (8a) and (8b) are the comparisons of the true value and state estimated value of the voltage amplitude of the two nodes, respectively, and (8c) and (8d) are the true value and state estimated value of the voltage phase angle of the two nodes, respectively;

[0065] Fig. 9 is the state estimation error curve in Example 2, wherein (9a) is the voltage amplitude state estimation error curve, and (9b) is the voltage phase angle state estimation error curve;

[0066] Fig.10 is the state estimation error of a single node in Example 2, where (10a) is the voltage amplitude state estimation error and (10b) is the voltage phase angle state estimation error. DETAILED DESCRIPTION

[0067] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0068] Unless otherwise defined, the technical terms or scientific terms involved in this application should be understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a", "the" and the like involved in this application do not indicate a quantitative limitation, and may represent the singular or plural. The terms "include", "comprise", "have" and any of their variations involved in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The "multiple" involved in this application refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there may be three relationships, for example, "A and / or B" can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects before and after are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects.

[0069] Example 1

[0070] This embodiment provides a method for adaptively estimating the dynamic state of a distribution network. Figure 1 As shown, the method comprises the following steps:

[0071] S1, establish a dynamic state estimation model for the distribution network, including state equations and measurement equations.

[0072] Traditional static state estimation completely relies on the measurement set at a certain moment to obtain a snapshot of the system's operating state at that time section, while ignoring the process of system state changes in continuous time. The idea of ​​dynamic state estimation is to track system state changes by recursively updating the estimate, using the state information of the previous moment and the current measurement to better estimate the current state and predict the state at the next moment. At present, the main method for dynamic state estimation of distribution networks is the Kalman filtering method. The Kalman filter combines the system's state equation and measurement equation into a state space model to describe the model of the random linear system. According to the statistical characteristics of noise, it follows the linear unbiased minimum mean square error criterion to filter out random interference in the measurement information, thereby obtaining the best estimate of the state variable. It is a process of linear filtering composed of measurement equations and state equations. The measurement equations and state equations are expressed as:

[0073] The state equation is expressed as: ;

[0074] The measurement equation is expressed as: ;

[0075] Among them, ω k and v k are the system process noise and measurement noise, which are independent of each other and have a mean of zero. k and v k The covariances of are Q and R respectively.

[0076] Since PMU can bring more and more accurate measurement data to the distribution network state estimation, if PMU is deployed on a large scale in the distribution network, only the voltage and current phasor measurements are sufficient to estimate the system state, which improves the calculation efficiency and accuracy. In the p-type equivalent branch ij simplified from the power network branch, the current phasor can be expressed as a voltage phasor according to Ohm's law and Kirchhoff's law:

[0077] ;

[0078] ;

[0079] In the formula, , , and Branch i - j The node voltage phasors and branch current phasors of , , and Branch i - j The admittance is calculated.

[0080] When the PMU measurement satisfies the system observability, the node voltage phasor is used as the state quantity, and the linear measurement equation for power system state estimation based on PMU measurement can be obtained:

[0081] ;

[0082] In the formula, and They are PMU node voltage phasor measurement and branch current phasor measurement respectively; for n dimensional identity matrix, is the node-branch admittance matrix, and the Jacobian matrix of the measurement equation is expressed as ; and are the measurement noise of voltage phasor and current phasor respectively.

[0083] S2, based on the statistical characteristics of the phasor error of the PMU device, sets the measurement noise covariance matrix, and adaptively estimates the measurement noise according to the latest measurement before each state estimation.

[0084] The amplitude and phase of PMU measurements are usually assumed to follow independent zero-mean Gaussian distributions. is an element measured by PMU, and its amplitude and phase measurement errors are and , according to the Euclidean distance, The measurement error It is expressed as:

[0085] ,

[0086] In the formula, express The true value of , Respectively The amplitude and phase of

[0087] Assume that the amplitude error measured by the PMU is and phase error The mean is and , the standard deviation is and The phasor error is The mean of is expressed as:

[0088] ,

[0089] In the formula, ,right Perform Taylor expansion and ignore higher-order terms to obtain , then the phasor error The variance of is expressed as:

[0090] ,

[0091] Assuming that the amplitude and phase measured by the PMU are independently distributed with a zero-mean Gaussian distribution, and are all equal to 0, and and Independent of each other, phasor error The variance of simplifies to:

[0092] ,

[0093] Since the PMU amplitude error is generally a few thousandths of the true amplitude value, after identifying and processing the PMU bad data, the phasor amplitude measurement value can be used To replace the unknown phasor amplitude true value , in the known and In the case of statistical properties, The value of is approximately estimated by mathematical statistics methods.

[0094] The measurement noise covariance matrix is ​​set by the diagonal matrix of the measurement error variance R , the measurement noise is adaptively estimated based on the latest measurement before each state estimation:

[0095] ,

[0096] In the formula, express R The diagonal elements of The first i A measurement.

[0097] S3, the Kalman filtering method is used to estimate the dynamic state of the distribution network. In the Kalman filtering process, the time-varying posterior statistics of the process noise are considered, and the process noise covariance matrix is ​​adaptively adjusted through the forgetting memory method.

[0098] The Kalman filter process is divided into two parts: prediction update and new information update. The prediction update part calculates the state variables at the next moment according to the current system state and the system state equation, and solves the prediction error covariance matrix. The new information update part is to update the state value considering the statistical characteristics of noise when receiving the measurement information at the next moment, and calculate the estimation error covariance matrix. The calculation equation is expressed as:

[0099] Prediction equation: ;

[0100] Innovation correction equation: ;

[0101] Among them, Q k is the process noise covariance matrix at time k, F k for k The state transfer matrix at time , is the estimated value of the state at time k, is the state prediction value at time k, P k for k The estimated error covariance matrix at time , for k The forecast error covariance matrix at time t, H k for k The Jacobian matrix of the measurement equation at the moment, R k for k The measurement noise covariance matrix at time z k fork The measured value at the moment, K k for k Kalman filter gain at time t.

[0102] The prediction of the system state usually adopts the Holt two-parameter exponential smoothing method. In order to avoid ambiguity in the purpose of this application, this embodiment will not elaborate on this process.

[0103] The setting of the system process noise covariance matrix Q is the key to whether the dynamic state estimation can track the dynamic changes of the system very well. At present, when the system is in steady state, since the state changes little, the setting of the system process noise covariance matrix Q can be directly set to a constant based on prior knowledge. However, when the system changes, this method cannot accurately predict the degree of change of the state. The system process noise can be reflected according to the average value and variance of the difference between the measured quantity and the predicted quantity, thereby setting the process noise covariance matrix Q. According to the Q matrix setting method, the dynamic state estimation under short circuit fault is performed, and the system meets the critical observable condition. The results in one embodiment are as follows: Figure 2 shown.

[0104] Depend on Figure 2 It can be seen that the dynamic state can also track the dynamic changes of the system well. The dynamic state estimation is compared with the static state estimation when the system does not meet the observability, and the difference between the two methods is analyzed. The results are shown in Table 1, where DSE represents static state estimation, SSE represents dynamic state estimation, TAEm represents amplitude error, and TAEa represents phase error.

[0105] Table 1 Comparison between dynamic state estimation and static state estimation

[0106]

[0107] For unobservable systems, the accuracy of the amplitude and phase angle of DSE in the PMU observable area is higher than that of SSE. However, for the unobservable area, the state estimation of DSE is very inaccurate. This is because the system process noise ω of the unobservable nodes k It is impossible to estimate accurately.

[0108] Therefore, this embodiment proposes an adaptive dynamic state estimation considering the time-varying posterior statistics of process noise, and the method of adaptively adjusting the process noise covariance matrix is:

[0109] S31, approximates the system process noise as the difference between the state estimate and the predicted value:

[0110] ,

[0111] Among them, ω k and v k are the system process noise and measurement noise respectively, x kis the true value of the state at time k, is the estimated value of the state at time k, is the estimation error at time k, f() is the state transfer equation, which reflects the dynamic changes of the distribution network operation state. is the state transfer matrix after the linearization of the state transfer equation, K k is the Kalman filter gain, r k is the innovation value, that is, the difference between the measured value and the measured predicted value.

[0112] And calculate the unbiased estimate of the process noise variance:

[0113] ,

[0114] Among them, Q k is the process noise covariance matrix at time k, N is the number of samples, is the average value of the system process noise in N samplings, and P is the estimation error covariance matrix.

[0115] S32, considering that the most recent samples obviously have a more important influence, the weight factor of the process noise adaptively changes through the forgetting memory method:

[0116] ,

[0117] Among them, d k is the weight factor, b is the forgetting factor, and b k is the k-th power of the forgetting factor.

[0118] S33, by adjusting the unbiased estimate of the process noise variance through the weight factor, the adaptive adjustment equation of the process noise covariance matrix is ​​obtained:

[0119] ,

[0120] Among them, Q k is the process noise covariance matrix at time k.

[0121] According to the process noise Q adaptive adjustment equation obtained above, adaptive dynamic state estimation considering the time-varying a posteriori statistics of the process noise is performed, and the PMU configuration satisfies the critical observability of the system. In one embodiment, the system has a circuit breaker fault at the 3rd second and recloses at the 10th second. Figure 3 The dynamic state estimation curves of two nodes in the system are given. The Q adaptive adjustment method is compared with the method of setting Q according to the average and variance of the difference between the measured and predicted quantities. Table 2 gives the node voltage estimation errors of the two Q setting methods.

[0122] Table 2 Node voltage estimation errors of two Q setting methods

[0123] QNo Adaptive Q Adaptive TAE 0.0011 0.0015 TAE 0.0963 0.1548

[0124] from Figure 3 It can be seen that the adaptive update of the Q matrix can well track the power flow changes under the circuit breaker fault. As can be seen from Table 2, when the critical value is observable, the non-adaptive setting of the Q matrix has a significant effect on the process error ω k The estimate of is better because when the critical value is observable, the process error of all nodes can be accurately estimated based only on the difference between the current measured value and the predicted value. However, the adaptive update of the Q matrix takes into account the process error at the previous moment, which leads to an increase in the error.

[0125] The dynamic state estimation is performed when the system PMU does not meet the observability for comparison. The operating state of the system is the same as when it is critically observable. Figure 4 The amplitude and phase angle errors of each node with and without adaptive update of Q are given. Figure 4 It can be seen that the state estimation without Q matrix adaptation will produce a large estimation error for unobservable nodes, while the Q matrix adaptation can estimate the process error of unobservable nodes according to the K matrix, thereby estimating the state of unobservable nodes, such as Figure 5 As shown. Since the adaptive adjustment of the Q matrix is ​​a combination of the previous process errors, when the system mutates, the previous process errors are inaccurate, and the forgetting factor can be adjusted as needed. The forgetting factor b adjustment simulation verification is performed on the critical observable system, and the results are shown as follows Figure 6 And as shown in Table 3.

[0126] Table 3 Amplitude and phase angle errors under different forgetting factors b

[0127] b 0.995 0.95 0.8 0.65 TAE 0.0016 0.0015 0.0013 0.0012 TAE 0.1577 0.1488 0.1347 0.1447

[0128] Depend on Figure 6 It can be seen from Table 3 that 3s and 10s are the two mutation moments of circuit breaking and reclosing respectively. When the forgetting factor b is larger, the less previous information is included, the better the mutation situation is, but the larger the error is at the steady state; on the contrary, when the forgetting factor b is smaller, the more previous information is included, the worse the mutation situation is, but the smaller the error is at the steady state, which can be intuitively seen from the errors at 3s and 10s in the above four figures.

[0129] Example 2

[0130] In order to explore the effect of the proposed method in actual cases, a state estimation simulation under hybrid measurement is carried out on a distribution network in a certain area of ​​Shanghai. Figure 7 As shown. Figure 7It can be seen that the distribution network has 51 nodes. PMUs are configured for nodes 2, 4, 5, 9, 12, 18, 20, 21, 24, 30, 31, 34, 37, 38, and 41, and the remaining nodes are measured by the SCADA system. The errors of the measuring devices are shown in the table. Figure 8 The true value of the system power flow calculation within 15s and the estimated value of the mixed state estimation are given. Fig. 9 The state estimation error curve is given. Table 4 gives the voltage amplitude and phase angle error of a single node in each state estimation. Fig.10 The average error of each state estimation of a single node is given.

[0131] Table 4 Node voltage estimation error considering pseudo measurement

[0132] Amplitude error / pu <![CDATA[5.014×10 -4 ]]> Phase angle error / ° 0.0678

[0133] Depend on Figure 8 and Fig. 9 It can be seen that the state estimation value of the proposed method fluctuates around the true value, the amplitude error of the node voltage fluctuates within 2×10-3p.u., and the phase angle error of the node voltage fluctuates within 0.3°, which can effectively track the power flow of the system. It can be seen from Table 4 that the average error of each state estimation is very small, and the amplitude error is only 0.5%. The proposed method has high accuracy. Fig.10 It can be seen that except for the relatively large average voltage amplitude of node 1, the errors of the other nodes are similar, indicating that this method is effective in estimating the state of each node with high accuracy.

[0134] Example 3

[0135] The above is an introduction to a method embodiment. The following is a further explanation of the solution of the present invention through an apparatus embodiment.

[0136] A distribution network adaptive dynamic state estimation device, used to implement the method described in the above embodiment 1, comprising:

[0137] State estimation model building module: building a distribution network dynamic state estimation model, including a state equation and a measurement equation, wherein the measurement equation is a linear measurement equation for power system state estimation based on PMU measurement;

[0138] Measurement noise adaptive estimation module: Based on the statistical characteristics of the phasor error of the PMU device, the measurement noise covariance matrix is ​​set, and the measurement noise is adaptively estimated according to the latest measurement before each state estimation;

[0139] Dynamic state estimation module: The Kalman filtering method is used to estimate the dynamic state of the distribution network. During the Kalman filtering process, the time-varying posterior statistics of the process noise are considered, and the process noise covariance matrix is ​​adaptively adjusted through the forgetting memory method.

[0140] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the described module can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0141] Example 4

[0142] The electronic device of the present invention includes a central processing unit (CPU), which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) or loaded from a storage unit into a random access memory (RAM). In the RAM, various programs and data required for device operation can also be stored. The CPU, ROM and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0143] Multiple components in the device are connected to the I / O interface, including: input units, such as keyboards, mice, etc.; output units, such as various types of displays, speakers, etc.; storage units, such as disks, optical disks, etc.; and communication units, such as network cards, modems, wireless communication transceivers, etc. The communication unit allows the device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunication networks.

[0144] The processing unit performs the various methods and processes described above, such as methods S1 to S3. For example, in some embodiments, methods S1 to S3 may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on the device via a ROM and / or a communication unit. When the computer program is loaded into the RAM and executed by the CPU, one or more steps of methods S1 to S3 described above may be performed. Alternatively, in other embodiments, the CPU may be configured to execute methods S1 to S3 in any other appropriate manner (e.g., by means of firmware).

[0145] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0146] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code can be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0147] In the context of the present invention, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0148] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for adaptive dynamic state estimation of a distribution network, characterized in that: The method comprises the following steps: Establishing a distribution network dynamic state estimation model, including a state equation and a measurement equation, wherein the measurement equation is a linear measurement equation for power system state estimation based on PMU measurement; Based on the statistical characteristics of the phasor error of the PMU device, the measurement noise covariance matrix is ​​set to adaptively estimate the measurement noise based on the latest measurement before each state estimation; The Kalman filter method is used to estimate the dynamic state of the distribution network. In the Kalman filter process, the time-varying posterior statistics of the process noise are considered, and the covariance matrix of the process noise is adaptively adjusted through the forgetting memory method. When the PMU measurement satisfies the system observability, the node voltage phasor is used as the state quantity, and the linear measurement equation for power system state estimation based on PMU measurement is obtained: , In the formula, and They are PMU node voltage phasor measurement and branch current phasor measurement respectively; E for n dimensional identity matrix, Y is the node-branch admittance matrix, and the Jacobian matrix of the measurement equation is expressed as ; and are the measurement noise of voltage phasor and current phasor respectively; Based on the statistical characteristics of the phasor error of the PMU device, the measurement noise covariance matrix is ​​set, and the measurement noise is adaptively estimated according to the latest measurement before each state estimation, specifically: Assumptions is an element measured by PMU, and its amplitude and phase measurement errors are and , according to the Euclidean distance, The measurement error It is expressed as: ， In the formula, express The true value of , Respectively The amplitude and phase of Assume that the amplitude error measured by the PMU is and phase error The mean is and , the standard deviation is and The phasor error is The mean of is expressed as: ， In the formula, ,right Perform Taylor expansion and ignore the higher-order terms to obtain , then the phasor error The variance of is expressed as: ， Assuming that the amplitude and phase measured by the PMU are independently distributed with a zero-mean Gaussian distribution, and are all equal to 0, and and Independent of each other, phasor error The variance of simplifies to: ， Use phasor amplitude measurement To replace the unknown phasor amplitude true value , in the known and In the case of statistical properties, The value of is approximately estimated by mathematical statistics methods; The measurement noise covariance matrix is ​​set by the diagonal matrix of the measurement error variance R , the measurement noise is adaptively estimated based on the latest measurement before each state estimation: ， In the formula, express R The diagonal elements of The first i A measurement.

2. A distribution network adaptive dynamic state estimation method according to claim 1, characterized in that: The method of adaptively adjusting the process noise covariance matrix is: The system process noise is approximated as the difference between the state estimate and the predicted value, and an unbiased estimate of the process noise variance is calculated; The weight factor of adaptive variation of process noise is given by forgetting memory method; The unbiased estimate of the process noise variance is adjusted by the weight factor, and the adaptive adjustment equation of the process noise covariance matrix is ​​obtained.

3. A method for adaptive dynamic state estimation of a distribution network according to claim 2, characterized in that: The system process noise is approximately: ， Among them, ω k and v k are the system process noise and measurement noise respectively, x k is the true value of the state at time k, is the estimated value of the state at time k, is the estimation error at time k, f() is the state transfer equation, which reflects the dynamic changes of the distribution network operation state. is the state transfer matrix after the linearization of the state transfer equation, K k is the Kalman filter gain, r k is the innovation value, that is, the difference between the measured value and the measured predicted value.

4. A method for adaptive dynamic state estimation of a distribution network according to claim 3, characterized in that: The unbiased estimate of the process noise variance is: ， Among them, Q k is the process noise covariance matrix at time k, N is the number of samples, is the average value of the system process noise in N samplings, and P is the estimation error covariance matrix.

5. A method for adaptive dynamic state estimation of a distribution network according to claim 3, characterized in that: The adaptive adjustment equation of the process noise covariance matrix is ​​expressed as: ， in, ， Q k is the process noise covariance matrix at time k, d k is the weight factor, b is the forgetting factor, and b k is the kth power of the forgetting factor.

6. A method for adaptive dynamic state estimation of a distribution network according to claim 1, characterized in that: The state equation is expressed as: ; The measurement equation is expressed as: ; Among them, ω k and v k are the system process noise and measurement noise, ω k and v k The covariances of are Q and R respectively.

7. A method for adaptive dynamic state estimation of a distribution network according to claim 1, characterized in that: The Kalman filter process is divided into two parts: prediction update part and new information update part. The prediction update part calculates the state variables at the next moment according to the current system state and the system state equation, and solves the prediction error covariance matrix; the new information update part is to update the state value considering the noise statistical characteristics when receiving the measurement information at the next moment, and calculate the estimation error covariance matrix; the calculation equation is expressed as: Prediction equation: ; Innovation correction equation: ; Among them, Q k is the process noise covariance matrix at time k, F k for k The state transfer matrix at time , is the estimated value of the state at time k, is the state prediction value at time k, P k for k The estimated error covariance matrix at time , for k The forecast error covariance matrix at time t, H k for k The Jacobian matrix of the measurement equation at the moment, R k for k The measurement noise covariance matrix at time z k for k The measured value at the moment, K k for k Kalman filter gain at time t.

8. A distribution network adaptive dynamic state estimation device, characterized in that: For implementing the method according to any one of claims 1 to 7, the device comprises: State estimation model building module: building a distribution network dynamic state estimation model, including a state equation and a measurement equation, wherein the measurement equation is a linear measurement equation for power system state estimation based on PMU measurement; Measurement noise adaptive estimation module: Based on the statistical characteristics of the phasor error of the PMU device, the measurement noise covariance matrix is ​​set, and the measurement noise is adaptively estimated according to the latest measurement before each state estimation; Dynamic state estimation module: The Kalman filtering method is used to estimate the dynamic state of the distribution network. During the Kalman filtering process, the time-varying posterior statistics of the process noise are considered, and the process noise covariance matrix is ​​adaptively adjusted through the forgetting memory method.

9. An electronic device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images