Multi-source data state estimation method and device based on model compensation and Kalman filtering

Abstract

The invention provides a multi-source data state estimation method and device based on model compensation and Kalman filtering, and the method comprises the steps: obtaining multi-source measurement data in a power system, and generating pseudo measurement data between adjacent measurement data through employing an interpolation method; filtering the obtained multi-source measurement data and pseudo measurement data by using an unscented Kalman filtering algorithm to obtain a preliminary state estimation value; establishing an actual power grid mathematical model and a simplified model, and calculating a difference value of the two models as a compensation amount; and superposing the compensation quantity to the initial state estimation value to obtain a final state estimation value. According to the method, the sampling step length is shortened, discretization errors caused by differentiation are reduced, and the filtering result is consistent with the model as much as possible by using the unscented Kalman filtering algorithm. And finally, through an error compensation model, the compensation correction of the simplified model is superposed to a preliminary estimation result of the Kalman filtering algorithm, so that the model precision is improved, and the influence of process noise is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of power system state estimation, and in particular to a multi-source data state estimation method and device based on model compensation and Kalman filtering. Background Art

[0002] The dynamic state estimation of power system was first performed based on mean square error filtering, which can achieve good estimation results when process noise and measurement noise obey Gaussian distribution. However, in actual engineering, due to the influence of complex situations such as communication interference, equipment aging, and telemetry anomalies, the measurement data may have problems such as non-Gaussian noise, data missing, or even bad data. Under the above interference, the estimation effect of traditional Kalman filter (KF) will be greatly reduced, and even cause wrong estimation.

[0003] On the other hand, the intermittent and volatile nature of new energy and the rapid and discrete nature of power electronics also make the system operating conditions unpredictable, further making the uncertainty problem of the system more obvious. When there are parameter errors or bad measurement data, the estimation effect will be greatly reduced. Summary of the invention

[0004] In view of this, the present invention provides a multi-source data state estimation method and device based on model compensation and Kalman filtering to solve at least one of the above-mentioned problems.

[0005] In order to achieve the above object, the present invention adopts the following scheme:

[0006] According to a first aspect of the present invention, there is provided a multi-source data state estimation method based on model compensation and Kalman filtering, the method comprising: acquiring multi-source measurement data in an electric power system, and generating pseudo-measurement data between adjacent measurement data using an interpolation method; filtering the acquired multi-source measurement data and pseudo-measurement data using an unscented Kalman filtering algorithm to obtain a preliminary state estimation value; establishing a mathematical model of an actual power grid and a simplified model, and calculating the difference between the two models as a compensation amount; and superimposing the compensation amount on the preliminary state estimation value to obtain a final state estimation value.

[0007] As an embodiment of the present invention, the multi-source measurement data in the above method comes from at least one of PMS3.0, distribution automation system, electricity consumption information collection system, and marketing SG186 system.

[0008] As an embodiment of the present invention, the method of generating pseudo measurement data between adjacent measurement data using interpolation includes: using linear interpolation to estimate the null value at the i-th moment based on the measurement value at the k-1 moment and the measurement value at the k moment, where i is between k-1 and k.

[0009] As an embodiment of the present invention, the above method uses an unscented Kalman filter algorithm for filtering to obtain a preliminary state estimate, including: using an unscented transformation to obtain a Sigma sampling point set based on the state variable estimate and covariance matrix at time k; calculating the mean and covariance matrix of the one-step prediction value of the Sigma sampling point set; generating a new Sigma point set based on the mean of the one-step prediction value, substituting it into the measurement equation to obtain a new predicted measurement value, and further obtaining an error covariance matrix and a cross-covariance matrix; using the error covariance matrix and the cross-covariance matrix to calculate the Kalman gain at time k+1, and obtaining the state estimate and covariance matrix of the system at the next moment; repeating the above steps until the number of cycles is greater than the interpolation factor and the data sampling is completed to obtain a preliminary state estimate.

[0010] As an embodiment of the present invention, the simplified model in the above method is a model obtained by equivalently replacing each micro-source part of the actual power grid mathematical model.

[0011] According to a second aspect of the present invention, there is provided a multi-source data state estimation device based on model compensation and Kalman filtering, the device comprising: a data acquisition unit, used to acquire multi-source measurement data in a power system; an interpolation unit, used to generate pseudo measurement data between adjacent measurement data using an interpolation method; a filtering unit, used to filter the acquired multi-source measurement data and pseudo measurement data using an unscented Kalman filtering algorithm to obtain a preliminary state estimation value; a model compensation unit, used to establish a mathematical model of an actual power grid and a simplified model, and calculate the difference between the two models as a compensation amount; and a state estimation unit, used to superimpose the compensation amount on the preliminary state estimation value to obtain a final state estimation value.

[0012] As an embodiment of the present invention, the multi-source measurement data comes from at least one of PMS3.0, distribution automation system, electricity consumption information collection system, and marketing SG186 system.

[0013] As an embodiment of the present invention, the interpolation unit is specifically used to estimate the null value at the i-th moment according to the measured value at the k-1 moment and the measured value at the k moment by using a linear interpolation method, wherein i is between k-1 and k.

[0014] As an embodiment of the present invention, the above-mentioned filtering unit uses an unscented Kalman filtering algorithm to perform filtering, and obtaining a preliminary state estimate includes: using an unscented transformation to obtain a Sigma sampling point set based on the state variable estimate and covariance matrix at time k; calculating the mean and covariance matrix of the one-step prediction value of the Sigma sampling point set; generating a new Sigma point set based on the mean of the one-step prediction value, substituting it into the measurement equation to obtain a new predicted measurement value, and further obtaining an error covariance matrix and a cross-covariance matrix; using the error covariance matrix and the cross-covariance matrix to calculate the Kalman gain at time k+1, and obtaining the state estimate and covariance matrix of the system at the next moment; repeating the above steps until the number of cycles is greater than the interpolation factor and the data sampling is completed to obtain a preliminary state estimate.

[0015] As an embodiment of the present invention, the simplified model is a model obtained by equivalently replacing each micro-source part of the actual power grid mathematical model.

[0016] According to a third aspect of the present invention, there is provided an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0017] 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 computer program implements the steps of the above method when executed by a processor.

[0018] The multi-source data state estimation method and device based on model compensation and Kalman filtering proposed in the present invention first collects multi-source measurement data, and adds pseudo-measurements between adjacent measurement data through an interpolation algorithm, thereby shortening the sampling step and reducing the discretization error caused by differentiation. Then, by using an unscented Kalman filter algorithm, the noise is filtered out so that the filtering result is as consistent as possible with the model. Finally, through the error compensation model, the compensation correction of the simplified model is superimposed on the preliminary estimation result of the Kalman filter algorithm, thereby improving the model accuracy and reducing the impact of process noise. Therefore, the present invention greatly improves the state estimation accuracy under conditions of complex noise, bad data, and load mutations, so that the bus voltage remains stable near the reference value. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0020] Figure 1 It is a flow chart of a multi-source data state estimation method based on model compensation and Kalman filtering provided in an embodiment of the present application;

[0021] Figure 2 It is a structural schematic diagram of a simplified model provided by an embodiment of the present invention;

[0022] Figure 3 It is a flowchart of a multi-source data state estimation method provided in an embodiment of the present application;

[0023] Figure 4 Schematic diagram of control effect under Gaussian white noise provided by an embodiment of the present invention;

[0024] Figure 5 Schematic diagram of control effect under Gaussian biased noise provided by an embodiment of the present invention;

[0025] Figure 6 It is a flowchart of a multi-source data state estimation method based on model compensation and Kalman filtering provided by another embodiment of the present application;

[0026] Figure 7 It is a structural schematic diagram of a multi-source data state estimation device based on model compensation and Kalman filtering provided in an embodiment of the present application;

[0027] Figure 8 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0028] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0029] The information collected in the technical solution of this application is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant data comply with the relevant laws, regulations and standards of relevant countries and regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0030] The acquisition, transmission, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0031] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0032] like Figure 1 The figure is a flow chart of a multi-source data state estimation method based on model compensation and Kalman filtering provided by an embodiment of the present application, the method comprising the following steps:

[0033] Step S101: acquiring multi-source measurement data in a power system, and generating pseudo measurement data between adjacent measurement data using an interpolation method.

[0034] Preferably, the multi-source measurement data acquired in this step comes from at least one of PMS3.0, distribution automation system, power consumption information collection system, and marketing SG186 system.

[0035] PMS3.0 system: collects comprehensive real-time data on the power system, including information on power generation, transmission, transformation and distribution.

[0036] Distribution automation system: mainly collects operating data of the distribution network, such as line current, voltage, switch status, etc.

[0037] Electricity consumption information collection system: collects users' electricity consumption data, such as power consumption, power factor, etc.

[0038] Marketing SG186 system: collects marketing data related to users' electricity consumption, such as electricity prices and electricity charges.

[0039] The type of data obtained depends on the specific application scenario and requirements, and may include, for example, voltage amplitude, current amplitude, phase angle, active power, reactive power, etc.

[0040] Due to the limitation of sensor sampling frequency, it can generally only reach 50-60Hz. In order to further shorten the sampling step, it is necessary to increase the corresponding pseudo-measurement data. In theory, the higher the accuracy of pseudo-measurement data generation, the better the improvement of estimation effect. However, increasing the interpolation factor r will improve the estimation accuracy more than the optimization of the interpolation algorithm. Therefore, in order to reduce the calculation time, this step preferably uses linear interpolation to obtain pseudo-measurement data.

[0041] Linear interpolation refers to the measurement value z at time k-1. k-1 and the measured value z at time k k To estimate the null value at the i-th moment, i is between k-1 and k, so the estimated value z at the i-th moment is i for:

[0042]

[0043] In the above formula, t k ,t k-1 ,t i Represents k, k-1, and i moments.

[0044] Step S102: filtering the acquired multi-source measurement data and pseudo-measurement data using an unscented Kalman filter algorithm to obtain a preliminary state estimation value.

[0045] Preferably, the filtering using the unscented Kalman filter algorithm in this step to obtain a preliminary state estimate may include four parts: a time update step, a state prediction step, a measurement update step and a state update step.

[0046] Time update step: Using unscented transformation, the Sigma sampling point set is obtained according to the estimated value of the state variable and the covariance matrix at time k. The expressions of the sampling points and sampling point weights are respectively:

[0047]

[0048] Where n is the dimension of the state variable, is the sigma sampling point set, is the estimated value of the state variable at time k, P k is the covariance matrix, λ is the scaling factor, α is the UT transformation coefficient, k is the default coefficient, generally k=3-n, is the initial value of the state quantity weight, is the initial value of the covariance weight, is the i-th parameter of the state quantity weight, is the i-th parameter of the covariance weight, β is the regularization coefficient, and β=2 is generally taken.

[0049] State prediction step: The mean and covariance matrix of the one-step prediction value of the Sigma sampling point set are calculated using the following equations (4) and (5).

[0050]

[0051] in, is the mean of the one-step prediction values ​​at sigma point, P k+1|k is the covariance matrix, is the state transfer function, Q k is the covariance matrix of the process noise.

[0052] Measurement update step: Generate a new set of Sigma sampling points based on the mean of the one-step prediction value, substitute it into the measurement equation to obtain the new predicted measurement value, and further obtain the error covariance matrix and cross covariance matrix, as shown in the following equations (6)-(8):

[0053]

[0054] in, is the mean of the real measurement value of the sigma point one step, h is the measurement equation, is the new Sigma sampling point set, is the error covariance matrix, is the cross covariance matrix, and R is the covariance matrix of the measurement noise.

[0055] State update step: Use the error covariance matrix and the cross covariance matrix to calculate the Kalman gain at time k+1 to obtain the state estimate and covariance matrix of the system at the next time. This step can be specifically performed using the following formula (9):

[0056]

[0057] In the above formula, K k+1 is the Kalman gain at time k+1, is the estimated state value at the next moment, P k+1 is the covariance at the next moment.

[0058] Repeat the above four steps until the number of cycles is greater than the interpolation factor and the data sampling is completed to obtain the preliminary state estimate.

[0059] Step S103: Establishing an actual power grid mathematical model and a simplified model, and calculating the difference between the two models as a compensation amount.

[0060] Since the actual power grid mathematical model takes into account the on and off states of the converter, the dimension of the state space matrix is ​​large and the model is highly nonlinear, which leads to divergence problems when performing Kalman filter estimation. In view of this situation, it is generally chosen to replace each micro-source part with an equivalent one to obtain a simplified model for dynamic state estimation. The simplified model is as follows: Figure 2 As shown in Figure 2, the error between the estimated result and the true value mainly comes from the gap between the mean value and the ripple, which corresponds to the modeling error in the process noise. At this time, the negative effect of the process noise may even be greater than the filtering effect, resulting in poor estimation results.

[0061] Step S104: superimposing the compensation amount onto the preliminary state estimation value to obtain a final state estimation value.

[0062] In order to improve the accuracy of bus voltage estimation, the embodiment of the present application uses the difference between the solution results of the model before simplification (i.e., the actual power grid mathematical model) and the simplified model as the compensation correction for the simplified model, and superimposes it on the preliminary estimation result of the Kalman filtering algorithm to construct an error compensation model, thereby improving the model accuracy and reducing the impact of process noise.

[0063] The steps described above can also be found in Figure 3 As shown, the dotted box shows filtering using the unscented Kalman filter algorithm, r is the interpolation factor, and k is the number of data sampling times.

[0064] The beneficial effects of the above method of the present application are described below through simulation analysis:

[0065] (1) Calculation results under different noise conditions

[0066] Figure 4 and Figure 5 The bus voltage control results of whether to consider the filtering algorithm under Gaussian white noise and Gaussian biased noise are given. It can be seen that the actual operation results of the bus voltage under the influence of noise will deviate from the reference voltage, and the more serious the noise, the worse the control effect. Especially when there is biased noise in the system, the actual operation results will have a significant deviation from the target voltage. By adding the Kalman filtering algorithm proposed in this paper, the estimation results can be made as close to the ideal state as possible, so that the controller can maintain good tracking performance when facing different measurement noises, improving the stability of the system.

[0067] In order to further demonstrate the estimation accuracy and robustness of the method proposed in this application, Table 1 gives a comparison of the evaluation indicators of the above situation under different noises, among which Me, Mae, and Mmse are used to evaluate the filtering effect of the algorithm, and Isse, Iae, and Imse are used to evaluate the tracking effect of the controller. It can be seen that when facing different noises, the use of the unscented Kalman filtering algorithm (UKF) has significantly improved the filtering effect and tracking effect compared to not using the filtering algorithm. For the DC bus voltage, under Gaussian white noise, the filtering accuracy of UKF is improved by 92.87% compared with the case where filtering is not considered; under biased Gaussian noise, the filtering accuracy of UKF is improved by 97.69% compared with the case where filtering is not considered.

[0068] Table 1

[0069]

[0070] (2) Computational performance analysis

[0071] Dynamic state estimation requires that the numerical solution speed of each prediction and correction step must be faster than the scanning speed of the PMU. For the proposed UFK algorithm, this application sets the PMU acquisition frequency to 60Hz and the total sampling to 180 times. The calculation time is 0.2025s and 1.1517s respectively, and the single calculation time is 0.0011 and 0.0064s, which is less than the single data sampling period of 0.0167s. Even if the algorithm needs to consider the compensation mechanism, the calculation time of the method proposed in this application can meet the real-time requirements of state estimation.

[0072] From the above, it can be seen that the multi-source data state estimation method based on model compensation and Kalman filtering proposed in the present invention first collects multi-source measurement data, and adds pseudo-measurements between adjacent measurement data through the interpolation algorithm, thereby shortening the sampling step and reducing the discretization error caused by differentiation. Then, by using the unscented Kalman filter algorithm, the noise is filtered out so that the filtering result is as consistent as possible with the model. Finally, through the error compensation model, the compensation correction of the simplified model is superimposed on the preliminary estimation result of the Kalman filter algorithm, thereby improving the model accuracy and reducing the impact of process noise.

[0073] like Figure 6 A flowchart of a multi-source data state estimation method based on model compensation and Kalman filtering is provided in another embodiment of the present application. The method comprises the following steps:

[0074] Step S601: Acquire multi-source measurement data in the power system.

[0075] Step S602: Calculate a measurement residual vector of multi-source measurement data, where the measurement residual vector is a difference between a measurement value and a state estimation value.

[0076] The state estimate is the estimated value of the system state variable obtained by the state estimation algorithm, and the measured value is the actual measured value obtained from various sensors and systems. The difference between the measured value and the state estimate reflects the deviation between the measured value and the model prediction value, and the residual vector contains the residual values ​​of all measurement points.

[0077] Step S603: Calculate the weighted sum of squares of the measurement residual vector and compare it with the threshold of the chi-square distribution. If the value is greater than the threshold, it is considered that there is bad data in the multi-source measurement data.

[0078] In order to determine whether there is bad data, it is necessary to perform a statistical test on the measurement residual vector. In this step, each residual value is squared, multiplied by the corresponding weight, and then summed. The weight is usually related to the accuracy of the measurement value. The higher the accuracy of the measurement value, the greater the weight. Then, the threshold of the chi-square distribution is determined according to the pre-set confidence level and degrees of freedom (usually the number of measurements). Finally, the calculated weighted sum of squares is compared with the chi-square distribution threshold. If the weighted sum of squares is greater than the threshold, it indicates that the residual is too large and there is a high possibility of bad data.

[0079] Step S604: when bad data is detected, a normalized residual of each multi-source measurement data is calculated, wherein the normalized residual is the measurement residual divided by its standard deviation.

[0080] In order to identify specific bad data, it is also necessary to calculate the normalized residual of each measurement. The standard deviation here refers to the standard deviation of each measurement, which can be determined based on historical data or the accuracy information of the measuring instrument. The residual of each measurement is divided by its standard deviation to obtain the normalized residual. The normalized residual eliminates the differences in dimension and accuracy between different measurements, making it easier to compare.

[0081] Step S605: remove the multi-source measurement data corresponding to the normalized residual with the largest absolute value from the multi-source measurement data.

[0082] The measurement value corresponding to the normalized residual with the largest absolute value is considered bad data because it has the largest deviation from the state estimate. Of course, it is also possible to set the absolute value greater than the set threshold to be considered bad data. The identified bad data is removed from the multi-source measurement data set and is no longer used for subsequent state estimation calculations.

[0083] Step S606: Generate pseudo measurement data between adjacent measurement data using interpolation method.

[0084] Step S607: filtering the acquired multi-source measurement data and pseudo-measurement data using an unscented Kalman filter algorithm to obtain a preliminary state estimation value.

[0085] Step S608: Establishing the actual power grid mathematical model and the simplified model, and calculating the difference between the two models as the compensation amount.

[0086] Step S609: superimposing the compensation amount onto the preliminary state estimation value to obtain a final state estimation value.

[0087] The above steps S606-S609 are the same as the above steps S101-S104, and will not be further described here.

[0088] From the above, it can be seen that in the power system, the measurement data obtained from various sources, such as PMS3.0, distribution automation system, electricity consumption information collection system and marketing SG186 system, constitute multi-source measurement data. However, due to factors such as sensor failure and communication interference, there may be errors or outliers in these multi-source measurement data, that is, bad data. These bad data will seriously affect the accuracy of state estimation. In order to improve the reliability of state estimation, this embodiment pre-processes the multi-source measurement data, identifies and eliminates bad data, so that the final state estimation result is more accurate.

[0089] like Figure 7 The structure diagram of a multi-source data state estimation device based on model compensation and Kalman filtering provided by an embodiment of the present application is shown, and the device includes: a data acquisition unit 710, an interpolation unit 720, a filtering unit 730, a model compensation unit 740 and a state estimation unit 750, which are connected in sequence. Among them:

[0090] The data acquisition unit 710 is used to acquire multi-source measurement data in the power system.

[0091] The interpolation unit 720 is used to generate pseudo measurement data between adjacent measurement data by using an interpolation method.

[0092] The filtering unit 730 is used to filter the acquired multi-source measurement data and pseudo-measurement data using an unscented Kalman filter algorithm to obtain a preliminary state estimation value.

[0093] The model compensation unit 740 is used to establish an actual power grid mathematical model and a simplified model, and calculate the difference between the two models as a compensation amount.

[0094] The state estimation unit 750 is used to add the compensation amount to the preliminary state estimation value to obtain a final state estimation value.

[0095] Preferably, the above-mentioned multi-source measurement data comes from at least one of PMS3.0, distribution automation system, electricity consumption information collection system, and marketing SG186 system.

[0096] Preferably, the interpolation unit 720 is specifically used to estimate the null value at the i-th moment according to the measured value at the k-1 moment and the measured value at the k moment by using a linear interpolation method, wherein i is between k-1 and k.

[0097] Preferably, the filtering unit 730 uses an unscented Kalman filtering algorithm to perform filtering, and obtaining a preliminary state estimate includes: using an unscented transformation to obtain a Sigma sampling point set based on the state variable estimate and covariance matrix at time k; calculating the mean and covariance matrix of the one-step prediction value of the Sigma sampling point set; generating a new Sigma point set based on the mean of the one-step prediction value, substituting it into the measurement equation to obtain a new predicted measurement value, and further obtaining an error covariance matrix and a cross-covariance matrix; using the error covariance matrix and the cross-covariance matrix to calculate the Kalman gain at time k+1, and obtaining the state estimate and covariance matrix of the system at the next moment; repeating the above steps until the number of cycles is greater than the interpolation factor and the data sampling is completed to obtain a preliminary state estimate.

[0098] Preferably, the simplified model is a model obtained by equivalently replacing each micro-source part of the actual power grid mathematical model.

[0099] Preferably, the above device also includes a preprocessing unit, which is used to calculate the measurement residual vector of the multi-source measurement data, the measurement residual vector is the difference between the measurement value and the state estimation value; calculate the weighted square sum of the measurement residual vector and compare it with the threshold of the chi-square distribution. If the value is greater than the threshold, it is considered that there is bad data in the multi-source measurement data; when the bad data is detected, calculate the normalized residual of each multi-source measurement data, the normalized residual is the measurement residual divided by its standard deviation; the multi-source measurement data corresponding to the normalized residual with the largest absolute value is eliminated from the multi-source measurement data.

[0100] The multi-source data state estimation device based on model compensation and Kalman filtering proposed in the present invention first collects multi-source measurement data, and adds pseudo-measurements between adjacent measurement data through an interpolation algorithm, thereby shortening the sampling step and reducing the discretization error caused by differentiation. Then, by using the unscented Kalman filter algorithm, the noise is filtered out so that the filtering result is as consistent as possible with the model. Finally, through the error compensation model, the compensation correction of the simplified model is superimposed on the preliminary estimation result of the Kalman filter algorithm, thereby improving the model accuracy and reducing the influence of process noise. Therefore, the present invention greatly improves the state estimation accuracy under conditions of complex noise, bad data, and load mutations, so that the bus voltage remains stable near the reference value.

[0101] Figure 8 is a schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 8The electronic device shown is a general data processing device, which includes a general computer hardware structure, which at least includes a processor 801 and a memory 802. The processor 801 and the memory 802 are connected via a bus 803. The memory 802 is suitable for storing one or more instructions or programs executable by the processor 801. The one or more instructions or programs are executed by the processor 801 to implement the steps in the multi-source data state estimation method based on model compensation and Kalman filtering.

[0102] The processor 801 may be an independent microprocessor or a collection of one or more microprocessors. Thus, the processor 801 executes the commands stored in the memory 802 to execute the method flow of the embodiment of the present invention as described above to realize the processing of data and the control of other devices. The bus 803 connects the above-mentioned multiple components together, and at the same time connects the above-mentioned components to the display controller 804 and the display device and the input / output (I / O) device 805. The input / output (I / O) device 805 may be a mouse, a keyboard, a modem, a network interface, a touch input device, a somatosensory input device, a printer, and other devices known in the art. Typically, the input / output (I / O) device 805 is connected to the system via an input / output (I / O) controller 806.

[0103] The memory 802 may store software components such as an operating system, a communication module, an interaction module, and an application program. Each of the modules and applications described above corresponds to a set of executable program instructions that implement one or more functions and methods described in the embodiments of the invention.

[0104] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the multi-source data state estimation method based on model compensation and Kalman filtering are implemented.

[0105] An embodiment of the present invention further provides a computer program product, comprising a computer program / instruction, which, when executed by a processor, implements the steps of the above-mentioned multi-source data state estimation method based on model compensation and Kalman filtering.

[0106] From the above, it can be seen that the multi-source data state estimation method and device based on model compensation and Kalman filtering proposed in the present invention first collects multi-source measurement data, and adds pseudo-measurements between adjacent measurement data through an interpolation algorithm, thereby shortening the sampling step and reducing the discretization error caused by differentiation. Then, by using the unscented Kalman filter algorithm, the noise is filtered out so that the filtering result is as consistent as possible with the model. Finally, through the error compensation model, the compensation correction of the simplified model is superimposed on the preliminary estimation result of the Kalman filter algorithm, thereby improving the model accuracy and reducing the impact of process noise. Therefore, the present invention greatly improves the state estimation accuracy under conditions of complex noise, bad data, and load mutations, so that the bus voltage remains stable near the reference value.

[0107] The preferred embodiments of the present invention are described above with reference to the accompanying drawings. Many features and advantages of these embodiments are clear from this detailed description, and therefore the claims are intended to cover all such features and advantages of these embodiments that fall within their true spirit and scope. In addition, since many modifications and changes are easily conceivable to those skilled in the art, it is not intended to limit the embodiments of the present invention to the precise structure and operation illustrated and described, but all suitable modifications and equivalents falling within its scope may be covered.

[0108] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take 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.

[0109] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as 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 a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0110] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0111] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0112] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A multi-source data state estimation method based on model compensation and Kalman filtering, characterized in that: The method comprises: Acquire multi-source measurement data in the power system and generate pseudo measurement data between adjacent measurement data using interpolation method; The obtained multi-source measurement data and pseudo-measurement data are filtered using an unscented Kalman filter algorithm to obtain a preliminary state estimation value; Establish the actual power grid mathematical model and simplified model, and calculate the difference between the two models as the compensation amount; The compensation amount is added to the preliminary state estimation value to obtain a final state estimation value.

2. The multi-source data state estimation method based on model compensation and Kalman filtering as claimed in claim 1, characterized in that: The multi-source measurement data comes from at least one of PMS3.0, distribution automation system, electricity consumption information collection system, and marketing SG186 system.

3. The multi-source data state estimation method based on model compensation and Kalman filtering as claimed in claim 1, characterized in that: The generating pseudo measurement data between adjacent measurement data by using interpolation method comprises: The null value at the i-th moment is estimated based on the measured value at the k-1 moment and the measured value at the k-th moment using the linear interpolation method, where i is between k-1 and k.

4. The multi-source data state estimation method based on model compensation and Kalman filtering as claimed in claim 1, characterized in that: The filtering using the unscented Kalman filter algorithm to obtain a preliminary state estimate includes: Using unscented transformation, the Sigma sampling point set is obtained according to the estimated value of the state variable and the covariance matrix at time k; Calculate the mean and covariance matrix of the one-step prediction value of the Sigma sampling point set; Generate a new Sigma point set according to the mean of the one-step prediction value, substitute it into the measurement equation to obtain a new predicted measurement value, and further obtain an error covariance matrix and a cross covariance matrix; The Kalman gain at time k+1 is calculated using the error covariance matrix and the cross covariance matrix to obtain the state estimation value and covariance matrix of the system at the next time; Repeat the above steps until the number of cycles is greater than the interpolation factor and the data sampling is completed to obtain the preliminary state estimate.

5. The multi-source data state estimation method based on model compensation and Kalman filtering as claimed in claim 1, characterized in that: The simplified model is a model obtained by equivalently replacing each micro-source part of the actual power grid mathematical model.

6. A multi-source data state estimation device based on model compensation and Kalman filtering, characterized in that: The device comprises: A data acquisition unit, used for acquiring multi-source measurement data in the power system; An interpolation unit, used for generating pseudo measurement data between adjacent measurement data by using an interpolation method; A filtering unit, used for filtering the acquired multi-source measurement data and pseudo-measurement data using an unscented Kalman filter algorithm to obtain a preliminary state estimation value; Model compensation unit, used to establish the actual power grid mathematical model and simplified model, and calculate the difference between the two models as the compensation amount; The state estimation unit is used to add the compensation amount to the preliminary state estimation value to obtain a final state estimation value.

7. The multi-source data state estimation device based on model compensation and Kalman filtering as claimed in claim 6, characterized in that: The multi-source measurement data comes from at least one of PMS3.0, distribution automation system, electricity consumption information collection system, and marketing SG186 system.

8. The multi-source data state estimation device based on model compensation and Kalman filtering as claimed in claim 6, characterized in that: The interpolation unit is specifically used to estimate the null value at the i-th moment according to the measured value at the k-1 moment and the measured value at the k moment by using a linear interpolation method, wherein i is between k-1 and k.

9. The multi-source data state estimation device based on model compensation and Kalman filtering as claimed in claim 6, characterized in that: The filtering unit uses the unscented Kalman filter algorithm to filter, and obtains the preliminary state estimation value including: Using unscented transformation, the Sigma sampling point set is obtained according to the estimated value of the state variable and the covariance matrix at time k; Calculate the mean and covariance matrix of the one-step prediction value of the Sigma sampling point set; Generate a new Sigma point set according to the mean of the one-step prediction value, substitute it into the measurement equation to obtain a new predicted measurement value, and further obtain an error covariance matrix and a cross covariance matrix; The Kalman gain at time k+1 is calculated using the error covariance matrix and the cross covariance matrix to obtain the state estimation value and covariance matrix of the system at the next time; Repeat the above steps until the number of cycles is greater than the interpolation factor and the data sampling is completed to obtain the preliminary state estimate.

10. The multi-source data state estimation device based on model compensation and Kalman filtering as claimed in claim 6, characterized in that: The simplified model is a model obtained by equivalently replacing each micro-source part of the actual power grid mathematical model.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images