A distribution network data adjustment method and system

Abstract

The present invention provides a method for adjusting distribution network data, comprising: obtaining historical measurement data and real-time measurement data; predicting the measurement values ​​of the historical measurement data to obtain a priori estimates, and obtaining a priori errors based on the real-time measurement data; calculating a Kalman gain for the priori errors; and calculating a posterior estimate and a posterior variance based on the Kalman gain, the priori estimates, and the priori errors. This method effectively improves the accuracy of the predicted estimates and addresses the problem of reduced filtering effectiveness caused by inaccurate predicted estimates.

Description

Technical Field

[0001] The present invention relates to the field of power systems, and in particular to a distribution network data adjustment method and system. Background Art

[0002] In recent years, the demand for electricity in production and daily life has grown year by year, and the load on distribution networks has increased. Users have increasingly stringent requirements for power supply quality. To monitor the operation of the power system, distribution network terminals collect state variables such as current, voltage, and switch position. Therefore, the quality of distribution network data is crucial to its safe and stable operation. Low-quality power system monitoring data will not produce accurate results, even with the most precise processing logic and efficient parallel strategies.

[0003] The data of the distribution network measurement system mainly depends on the measurement error generated when the measurement system is sampled. The error sources mainly include: (1) Measuring instruments: First, the error in the measuring instrument itself, and second, the error caused by the mismatch of instrument accuracy. (2) Observers or self-reading devices: The identification ability of observers and self-reading devices has certain limitations, so there will be certain errors in reading the measured values ​​and even when the equipment is installed, which has a significant impact on the observation results. (3) External conditions: The external conditions during observation, such as temperature, humidity, wind speed, etc., will have a certain impact on the observation results and even the observed objects. Therefore, as long as the observation measurement occurs in actual engineering, there will be errors of one kind or another. Therefore, the errors generated by the measurement in the distribution network are inevitable.

[0004] Traditional surveying and adjustment has stringent requirements for the properties of the measured system. The measured system or object must be unbiased, which means that after multiple measurements, the measured quantity must converge to a fixed value. In other words, the measured system is a static system. However, the distribution network is a time-varying dynamic system. Its true value will fluctuate over multiple measurements based on real-time power output and load changes. This makes it impossible to collect a large amount of error information and lacks sufficient sensitivity to capture system fluctuations and sags. This leads to inaccurate prediction estimates, which in turn degrades the filtering effect. Summary of the Invention

[0005] The technical solution provided by the present invention is:

[0006] A distribution network data adjustment method, comprising:

[0007] Obtain historical measurement data and real-time measurement data;

[0008] Predicting the historical measurement data to obtain a priori estimated value, and obtaining a priori error based on the real-time measurement data;

[0009] Performing Kalman gain calculation on the prior error;

[0010] The posterior estimate and posterior variance are calculated based on the Kalman gain, the prior estimate, and the prior error.

[0011] Preferably, the historical data includes:

[0012] Node voltage amplitude, phase angle difference, branch active power, reactive power flow measurement, node injected active power, reactive power measurement, branch current measurement or node injected current.

[0013] Preferably, the performing measurement value prediction on the historical measurement data to obtain a priori estimated values ​​includes:

[0014] Processing the distribution network historical data through adaptive filtering;

[0015] The distribution network data information processed by adaptive filtering is added to the judgment of statistical errors to filter out some errors;

[0016] The data after filtering out some errors is used to predict the measured value to obtain a priori estimated value.

[0017] Preferably, the data after filtering out some errors by adding the distribution network data information processed by adaptive filtering to the judgment of statistical errors is shown in the following formula:

[0018]

[0019] Where σ is the standard deviation of U(k), U(k) is the voltage measurement value of the distribution network node at time k, and y(k) is the processed data information output by the adaptive filter; is the transpose of the estimated coefficient; the estimated coefficient w ( k ) is calculated as follows:

[0020]

[0021] Where g is the control factor; e(k) is the error;

[0022] The calculation formula of the error e(k) is as follows:

[0023] .

[0024] Preferably, the calculation formula of the prior estimate is as follows:

[0025]

[0026] Where, is the smoothed value at time k; is the smoothed trend;

[0027] Among them, the The calculation formula is as follows:

[0028]

[0029] described The calculation formula is as follows:

[0030]

[0031] Where, and is a constant parameter less than 1; x ( k | k-1 ) for the use of k-1 Time and k-1 Information before the moment k A priori estimate of the moment.

[0032] Preferably, the calculation formula of the prior error is as follows:

[0033]

[0034] Where, is the prior error; U ( k )for k The measured voltage value of the distribution network node at the moment; B k is the state transfer matrix, is the first-order term of the Taylor expansion of the prior estimation formula, Q k is the system error matrix.

[0035] Preferably, the calculation formula of the Kalman gain is as follows:

[0036]

[0037] Where: J ( k ) is the Kalman gain; D △ is the variance matrix of the measurement error; is the transpose of the state transfer matrix.

[0038] Preferably, the calculation formula of the posterior estimate is as follows:

[0039]

[0040] Where, is the posterior estimate, M ( k ) is the difference between the measured value and the prior estimate;

[0041] The difference between the measured value and the prior estimate isM ( k ) is calculated as follows:

[0042]

[0043] Where, U ( k )for k The voltage measurement value of the distribution network node at the moment.

[0044] Preferably, the calculation formula of the posterior variance is as follows:

[0045]

[0046] in, is the posterior variance.

[0047] A distribution network data adjustment system, comprising:

[0048] Acquisition module: used to obtain historical measurement data and real-time measurement data;

[0049] Prediction module: used to predict the measurement value of the historical measurement data to obtain a priori estimated value, and obtain a priori error based on the real-time measurement data;

[0050] A first calculation module: configured to calculate the Kalman gain of the prior error;

[0051] The second calculation module is used to calculate the Kalman gain and prior estimation, prior error, posterior estimation and posterior variance.

[0052] Preferably, the prediction module includes: a historical data processing submodule and a calculation submodule;

[0053] The historical data processing submodule is used for:

[0054] Processing the distribution network historical data through adaptive filtering;

[0055] The distribution network data information processed by adaptive filtering is added to the judgment of statistical errors to filter out some errors;

[0056] Performing measurement value prediction on the data after filtering out some errors to obtain a priori estimated values;

[0057] The calculation submodule is used to calculate the prior estimation and the prior error;

[0058] The calculation formula of the prior estimate is as follows:

[0059]

[0060] Where, is the smoothed value at time k; is the smoothed trend;

[0061] Among them, the The calculation formula is as follows:

[0062]

[0063] described The calculation formula is as follows:

[0064]

[0065] Where, and is a constant parameter less than 1; x ( k | k-1 ) for the use of k-1 Time and k-1 Information before the moment k A priori estimate of the moment;

[0066] The calculation formula of the prior error is as follows:

[0067]

[0068] Where, is the prior error; U ( k )for k The measured voltage value of the distribution network node at the moment; B k is the state transfer matrix, is the first-order term of the Taylor expansion of the prior estimation formula, Q k is the system error matrix.

[0069] Preferably, the first calculation module specifically includes:

[0070]

[0071] Where: J ( k ) is the Kalman gain; D △ is the variance matrix of the measurement error; is the transpose of the state transfer matrix.

[0072] Preferably, the second calculation module specifically includes:

[0073] The calculation formula of the posterior estimate is as follows:

[0074]

[0075] Where, is the posterior estimate, M ( k ) is the difference between the measured value and the prior estimate;

[0076] The difference between the measured value and the prior estimate is M ( k ) is calculated as follows:

[0077]

[0078] Where, U ( k )for k The measured voltage value of the distribution network node at the moment;

[0079] The calculation formula of the posterior variance is as follows:

[0080]

[0081] in, is the posterior variance.

[0082] Compared with the prior art, the present invention has the following beneficial effects:

[0083] The present invention provides a method for adjusting distribution network data, comprising: obtaining historical measurement data and real-time measurement data; predicting the measurement values ​​of the historical measurement data to obtain a priori estimates, and obtaining a priori errors based on the real-time measurement data; calculating a Kalman gain for the priori errors; and calculating a posterior estimate and a posterior variance based on the Kalman gain, the priori estimates, and the priori errors. This method effectively improves the accuracy of the predicted estimates and addresses the problem of reduced filtering effectiveness caused by inaccurate predicted estimates. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 A schematic flow chart of a distribution network data adjustment method according to the present invention;

[0085] Figure 2 A flow chart of an embodiment of a method for adjusting distribution network data according to the present invention; DETAILED DESCRIPTION

[0086] In order to better understand the present invention, the present invention is further described below with reference to the accompanying drawings and examples.

[0087] Example 1:

[0088] The present invention provides a distribution network data adjustment method as follows Figure 1 Shown, including:

[0089] S1. Obtain historical measurement data and real-time measurement data;

[0090] S2. Predicting the historical measurement data to obtain a priori estimated value, and obtaining a priori error based on the real-time measurement data;

[0091] S3, performing Kalman gain calculation on the prior error;

[0092] S4. Calculate the posterior estimate and posterior variance based on the Kalman gain, the prior estimate, and the prior error.

[0093] like Figure 2 The specific process is as follows:

[0094] S1. Obtain historical measurement data and real-time measurement data;

[0095] Step 1: Obtain the measured value of the distribution network voltage amplitude and initialize the state variables.

[0096] The choice of the initial value of the state variable mainly affects the accuracy of the estimated output at the beginning of the cycle. However, the strong robustness of the Kalman filter will enable the entire filtering system to converge to a more satisfactory result even if the initial value is less accurate.

[0097] Assume that there is a measurement phase of the node voltage amplitude ,in U ( i ) is the i The voltage measurement value of discrete sampling time, we need to estimate the first n-1 historical redundant measurements k The true value of a state quantity at a point in time.

[0098] Initialize the control factor of adaptive filtering g , the parameters in the two-parameter exponential smoothing method α and β .

[0099] S2. Predicting the historical measurement data to obtain a priori estimated value, and obtaining a priori error based on the real-time measurement data;

[0100] Step 2: Adaptive filtering

[0101] The adaptive filter performs the first optimization on the input voltage amplitude measurement data. The flow chart is as follows: Figure 2 As shown in Figure 2, the processing steps follow equations (1) to (3). Since only historical data is used, the adaptive filter cannot respond quickly to real-time changes. Therefore, the 3σ principle is added to equation (3) to determine whether the abnormal value of the sudden jump in voltage amplitude is an error or a system fluctuation.

[0102] (1)

[0103] (2)

[0104] (3)

[0105] in w ( k ) is the estimated coefficient; g is the control factor; e ( k ) is the error, σ is U ( k ), U ( k )for k The measured voltage value of the distribution network node at the moment, y ( k ) is the output of the adaptive filter.

[0106] Step 3: Make predictions based on the two-parameter exponential smoothing method to obtain the prior estimate and prior error.

[0107] Exponential smoothing (also known as linear extrapolation) is a simple short-term load forecasting method with the advantages of requiring few variables and fast computation, making it suitable for online operations. The forecasting function primarily relies on a two-parameter exponential smoothing method, which allows the forecasting system to receive relatively accurate historical input data while maintaining a fast response.

[0108] The output of step 2 y ( k ) is processed as the input of this step, and the processing steps follow equations (4)-(6).

[0109] (4)

[0110] (5)

[0111] (6)

[0112] Among them, the parameters and is a constant less than 1; x ( k | k-1 ) for the use of k-1 Time and k-1 Information before the moment k A priori estimate of the moment;

[0113] Then calculate the prior error

[0114] (7)

[0115] Here U ( k ) is the measured phaseU 1 in U ( k ). B k is the state transfer matrix, is the first-order term of the Taylor expansion of the state equation (Equation (4), Q k is the system error matrix, which can be considered as a constant matrix.

[0116] Step 4: Calculate the posterior estimate and posterior error and update

[0117] S3, performing Kalman gain calculation on the prior error;

[0118] First, the prior error calculated in step 3 P ( k | k-1 ) into the formula to calculate the Kalman gain,

[0119] (8)

[0120] in, J ( k ) is the Kalman gain, D △ is the variance matrix of the measurement error.

[0121] Calculate new interest again

[0122] (9)

[0123] M ( k ) is the new information, that is, the difference between the measured value and the predicted value (prior estimate).

[0124] S4. Calculate the posterior estimate and posterior variance based on the Kalman gain, the prior estimate, and the prior error.

[0125] Finally, the posterior estimate is calculated

[0126] (10)

[0127] in, x ( k | k ) is the posterior estimate, that is, k Estimate of the true value at a time point

[0128] And the posterior variance calculation

[0129] (11)

[0130] in P ( k | k) is the posterior variance.

[0131] Finally, the output of the voltage amplitude adjustment of the distribution network nodes is Equation (10) and Equation (11). They represent the estimated value of the true value and the error of the estimated value respectively.

[0132] Similarly, branch active and reactive power flow measurement P ij 、 Q ij ; Node injection active and reactive power measurement P i 、 Q i ;Branch current measurement I ij and node injection current I i This method can also be used to adjust and predict other data.

[0133] Example 2:

[0134] Based on the same inventive concept, the present invention also provides a distribution network data adjustment system, which is characterized by comprising:

[0135] Acquisition module: used to obtain historical measurement data and real-time measurement data;

[0136] Prediction module: used to predict the measurement value of the historical measurement data to obtain a priori estimated value, and obtain a priori error based on the real-time measurement data;

[0137] A first calculation module: configured to calculate the Kalman gain of the prior error;

[0138] The second calculation module is used to calculate the Kalman gain and prior estimation, prior error, posterior estimation and posterior variance.

[0139] The prediction module includes: a historical data processing submodule and a calculation submodule;

[0140] The historical data processing submodule is used for:

[0141] Processing the distribution network historical data through adaptive filtering;

[0142] The distribution network data information processed by adaptive filtering is added to the judgment of statistical errors to filter out some errors;

[0143] Performing measurement value prediction on the data after filtering out some errors to obtain a priori estimated values;

[0144] The calculation submodule is used to calculate the prior estimation and the prior error;

[0145] The calculation formula of the prior estimate is as follows:

[0146]

[0147] Where, is the smoothed value at time k; is the smoothed trend;

[0148] Among them, the The calculation formula is as follows:

[0149]

[0150] described The calculation formula is as follows:

[0151]

[0152] Where, and is a constant parameter less than 1; x ( k | k-1 ) for the use of k-1 Time and k-1 Information before the moment k A priori estimate of the moment;

[0153] The calculation formula of the prior error is as follows:

[0154]

[0155] Where, is the prior error; U ( k )for k The measured voltage value of the distribution network node at the moment; B k is the state transfer matrix, is the first-order term of the Taylor expansion of the prior estimation formula, Q k is the system error matrix.

[0156] The first calculation module specifically includes:

[0157]

[0158] Where: J ( k ) is the Kalman gain; D △ is the variance matrix of the measurement error; is the transpose of the state transfer matrix.

[0159] The second calculation module specifically includes:

[0160] The calculation formula of the posterior estimate is as follows:

[0161]

[0162] Where, is the posterior estimate, M ( k ) is the difference between the measured value and the prior estimate;

[0163] The difference between the measured value and the prior estimate is M ( k ) is calculated as follows:

[0164]

[0165] Where, U ( k )for k The measured voltage value of the distribution network node at the moment;

[0166] The calculation formula of the posterior variance is as follows:

[0167]

[0168] in, is the posterior variance.

[0169] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0170] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0171] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, 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.

[0172] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work 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 The function specified in one or more boxes.

[0173] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0174] The above are merely embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are included in the scope of the claims of the present invention to be approved.

Claims

1. A distribution network data adjustment method, characterized in that: include: Obtain historical and real-time measurement data of the distribution network; Predicting the measured value of the distribution network historical measured data to obtain a priori estimated value, and obtaining a priori error based on the real-time measured data of the distribution network; Performing Kalman gain calculation on the prior error; The distribution network historical measurement data includes: node voltage amplitude; The step of predicting the measured value of the historical measured data of the distribution network to obtain a priori estimated values ​​includes: Processing the distribution network historical measurement data through adaptive filtering; The historical measurement data of the distribution network processed by adaptive filtering is added to the judgment of statistical errors to filter out some errors; Performing measurement value prediction on the data after filtering out some errors to obtain a priori estimated values; The data after filtering out some errors by adding the historical measurement data of the distribution network processed by adaptive filtering to the judgment of statistical errors is shown in the following formula: Where σ is the standard deviation of U(k), U(k) is the voltage measurement value of the distribution network node at time k, and y(k) is the processed data information output by the adaptive filter; is the transpose of the estimated coefficient; the estimated coefficient w ( k ) is calculated as follows: Where g is the control factor; e(k) is the error; The calculation formula of the error e(k) is as follows: ； The calculation formula of the prior error is as follows: Where, is the prior error; U ( k )for k The measured voltage value of the distribution network node at the moment; B k is the state transfer matrix, is the first-order term of the Taylor expansion of the prior estimation formula, Q k is the system error matrix; Calculate the posterior estimate as follows: Where, is the posterior estimate, x ( k | k-1 ) for the use of k-1 Time and k-1 Information before the moment k A priori estimate of the time, M ( k ) is the difference between the measured value and the prior estimate; The difference between the measured value and the prior estimate is M ( k ) is calculated as follows: Where, U ( k )for k The measured voltage value of the distribution network node at the moment; Calculate the posterior variance as follows: in, is the posterior variance, J ( k ) is the Kalman gain.

2. A distribution network data adjustment method as claimed in claim 1, characterized in that, The calculation formula of the prior estimate is as follows: Where, is the smoothed value at time k; is the smoothed trend; Among them, the The calculation formula is as follows: described The calculation formula is as follows: Where, and is a constant parameter less than 1.

3. A distribution network data adjustment method as claimed in claim 1, characterized in that, The calculation formula of the Kalman gain is as follows: Where: D △ is the variance matrix of the measurement error; is the transpose of the state transfer matrix.

4. A distribution network data adjustment system, characterized in that: include: Acquisition module: used to obtain historical measurement data and real-time measurement data of the distribution network; Prediction module: used to predict the measurement value of the distribution network historical measurement data to obtain a priori estimated value, and obtain a priori error based on the real-time measurement data of the distribution network; A first calculation module: configured to calculate the Kalman gain of the prior error; The distribution network historical measurement data includes: node voltage amplitude; The prediction module includes: a historical data processing submodule and a calculation submodule; The historical data processing submodule is used for: Processing the distribution network historical measurement data through adaptive filtering; The historical measurement data of the distribution network processed by adaptive filtering is added to the judgment of statistical errors to filter out some errors; Performing measurement value prediction on the data after filtering out some errors to obtain a priori estimated values; The data after filtering out some errors by adding the historical measurement data of the distribution network processed by adaptive filtering to the judgment of statistical errors is shown in the following formula: Where σ is the standard deviation of U(k), U(k) is the voltage measurement value of the distribution network node at time k, and y(k) is the processed data information output by the adaptive filter; is the transpose of the estimated coefficient; the estimated coefficient w ( k ) is calculated as follows: Where g is the control factor; e(k) is the error; The calculation formula of the error e(k) is as follows: ； The calculation submodule specifically includes: Calculate the prior error as follows: Where, is the prior error; U ( k )for k The measured voltage value of the distribution network node at the moment; B k is the state transfer matrix, is the first-order term of the Taylor expansion of the prior estimation formula, Q k is the system error matrix; The second calculation module performs the following steps: Calculate the posterior estimate as follows: Where, is the posterior estimate, x ( k | k-1 ) for the use of k-1 Time and k-1 Information before the moment k A priori estimate of the time, M ( k ) is the difference between the measured value and the prior estimate; The difference between the measured value and the prior estimate is M ( k ) is calculated as follows: Where, U ( k )for k The measured voltage value of the distribution network node at the moment; Calculate the posterior variance as follows: in, is the posterior variance, J ( k ) is the Kalman gain.

5. A distribution network data adjustment system as claimed in claim 4, characterized in that, The first calculation module specifically includes: Where: D △ is the variance matrix of the measurement error; is the transpose of the state transfer matrix.

Images