A method, device and terminal equipment for predicting reactive power of DC arc furnace

Abstract

The present invention is applicable to the technical field of load forecasting, and provides a method for predicting reactive power of a DC electric arc furnace, the method comprising: obtaining monitoring data of the DC electric arc furnace, the monitoring data comprising a power supply voltage time domain waveform array and a feeder current time domain waveform array obtained by synchronous sampling; performing fast Fourier transform on the monitoring data to obtain a frequency domain array, the frequency domain array comprising a voltage amplitude array, a voltage phase array, a current amplitude array, and a current phase array; calculating the reactive power array of the DC electric arc furnace based on the frequency domain array; and performing rolling prediction calculation on the reactive power of the DC electric arc furnace based on an autoregressive sliding average model and a reactive power array. The method for predicting reactive power of a DC electric arc furnace provided by the present invention can perform rolling prediction on the reactive power of the DC electric arc furnace based on the monitoring data obtained in real time and a preset autoregressive sliding average model, thereby improving the accuracy and reliability of power prediction.

Description

Technical Field

[0001] The present invention belongs to the technical field of load prediction, and in particular relates to a method, a device and a terminal device for predicting reactive power of a DC arc furnace. Background Art

[0002] The DC arc furnace is a typical random load. During the smelting process of the DC arc furnace, the impact reactive power generated will have a serious impact on the voltage fluctuation and flicker of the connected power grid.

[0003] Traditionally, in order to suppress the impact of DC arc furnace load on the power quality of the distribution network, a TCR-type SVC static VAR compensation device can be configured in the medium voltage system of the DC arc furnace. However, the SVC device's suppression effect on the voltage fluctuation and flicker of the arc furnace system is closely related to the response time of the TCR. Since the TCR uses thyristor phase-controlled rectification, when the response time is greater than 10ms, the flicker improvement rate is generally less than 50%, resulting in frequent excessive flicker when the DC arc furnace users are connected to the public connection point, affecting the application of the DC arc furnace. In addition, reactive power predictive control can be used to make up for the defects in the TCR response time and enhance the improvement effect of SVC on voltage fluctuations and flicker, but the prediction time of this method is long and cannot meet the needs of practical applications. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a method, an apparatus and a terminal device for predicting reactive power of a DC arc furnace, which can improve the accuracy of reactive power prediction of a DC arc furnace.

[0005] A first aspect of an embodiment of the present invention provides a method for predicting reactive power of a DC arc furnace, comprising:

[0006] Acquiring monitoring data of a DC arc furnace, wherein the monitoring data includes a power supply voltage time domain waveform array and a feeder current time domain waveform array obtained by synchronous sampling;

[0007] Performing a fast Fourier transform on the power supply voltage time domain waveform array and the feeder current time domain waveform array to obtain a frequency domain array; the frequency domain array includes a voltage amplitude array, a voltage phase array, a current amplitude array, and a current phase array;

[0008] Calculate the reactive power array of the DC arc furnace based on the frequency domain array;

[0009] Based on the autoregressive moving average model and the reactive power array, a rolling prediction calculation is performed on the reactive power of the DC arc furnace.

[0010] A second aspect of an embodiment of the present invention provides a device for predicting reactive power of a DC arc furnace, comprising:

[0011] A monitoring data acquisition module, used to acquire monitoring data of a DC arc furnace, wherein the monitoring data includes a power supply voltage time domain waveform array and a feeder current time domain waveform array obtained by synchronous sampling;

[0012] A fast Fourier transform module, used for performing fast Fourier transform on the power supply voltage time domain waveform array and the feeder current time domain waveform array to obtain a frequency domain array; the frequency domain array includes a voltage amplitude array, a voltage phase array, a current amplitude array, and a current phase array;

[0013] A reactive power array calculation module, used for calculating the reactive power array of the DC arc furnace based on the frequency domain array;

[0014] The prediction module is used to perform rolling prediction calculation on the reactive power of the DC arc furnace based on the autoregressive sliding average model and the reactive power array.

[0015] A third aspect of an embodiment of the present invention provides a terminal device, including 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.

[0016] A fourth aspect of an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method described above are implemented.

[0017] A fifth aspect of an embodiment of the present invention provides a computer program product. When the computer program product is run on a terminal device, the electronic device executes the steps of any one of the methods described in the first aspect.

[0018] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: the embodiments of the present invention provide a method for predicting reactive power of a DC electric arc furnace, the method comprising: obtaining monitoring data of the DC electric arc furnace, the monitoring data comprising a power supply voltage time domain waveform array and a feeder current time domain waveform array obtained by synchronous sampling; performing fast Fourier transform on the monitoring data to obtain a frequency domain array, the frequency domain array comprising a voltage amplitude array, a voltage phase array, a current amplitude array, and a current amplitude array; calculating the reactive power array of the DC electric arc furnace based on the frequency domain array; and performing rolling prediction calculation on the reactive power of the DC electric arc furnace based on an autoregressive sliding average model and a reactive power array. The method for predicting reactive power of a DC electric arc furnace provided by the embodiments of the present invention can perform rolling prediction on the reactive power of the DC electric arc furnace based on the monitoring data obtained in real time and a preset autoregressive sliding average model, thereby improving the accuracy and reliability of power prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 paying creative labor.

[0020] Figure 1 It is a schematic diagram of the implementation flow of the reactive power prediction method of a DC arc furnace provided by an embodiment of the present invention;

[0021] Figure 2 Schematic diagram of an application scenario of a DC arc furnace reactive power prediction method provided by an embodiment of the present invention;

[0022] Figure 3 It is another implementation flow diagram of the method for predicting reactive power of a DC arc furnace provided by an embodiment of the present invention;

[0023] Figure 4 It is a schematic diagram of reactive power variation trend of the reactive power prediction method of a DC arc furnace provided by an embodiment of the present invention;

[0024] Figure 5 Schematic diagram of the application effect of the reactive power prediction method of a DC arc furnace provided by an embodiment of the present invention;

[0025] Figure 6 It is another schematic diagram of the application effect of the reactive power prediction method of a DC arc furnace provided by an embodiment of the present invention;

[0026] Figure 7 is another schematic diagram of the application effect of the reactive power prediction method of a DC arc furnace provided by an embodiment of the present invention;

[0027] Figure 8 It is a structural schematic diagram of a reactive power prediction device for a DC arc furnace provided in an embodiment of the present invention;

[0028] Fig. 9 It is a schematic diagram of a terminal device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0029] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0030] In order to illustrate the technical solution of the present invention, a specific embodiment is provided below for illustration.

[0031] Figure 1 The following is a schematic diagram showing the implementation process of the DC arc furnace reactive power prediction method provided by the embodiment of the present invention. Figure 1 The DC arc furnace reactive power prediction method provided by the embodiment of the present invention may include steps S101 to S104.

[0032] S101: Acquire monitoring data of a DC arc furnace, where the monitoring data includes a power supply voltage time domain waveform array and a feeder current time domain waveform array obtained by synchronous sampling.

[0033] In some embodiments, during the normal smelting process of the DC arc furnace, monitoring data is acquired, which includes a time domain waveform array of the power supply voltage and feeder current at the DC arc furnace access point.

[0034] Figure 2 A schematic diagram of obtaining the access point power supply voltage and feeder current in an embodiment of the present invention is shown.

[0035] See also Figure 2 , during the normal smelting process of the DC arc furnace, the voltage time domain waveform array u(k) is obtained on the bus PT at the high voltage side of the arc furnace transformer, i.e., the voltage access point, and the current time domain waveform array i(k) is obtained on the medium voltage feeder CT of the DC arc furnace, where k = 1, 2, ..., is the number of points in the corresponding time domain waveform array. The voltage time domain waveform array and the current time domain waveform array are obtained by synchronous sampling, and the number of sampling points per cycle is N, and the value of N can include 64, 128, 256 or 512.

[0036] S102: Performing fast Fourier transform on the power supply voltage time domain waveform array and the feeder current time domain waveform array to obtain a frequency domain array; the frequency domain array includes a voltage amplitude array, a voltage phase array, a current amplitude array, and a current phase array.

[0037] In some embodiments, S102 includes: performing fast Fourier transform on the power supply voltage time domain waveform array and the feeder current time domain waveform array based on the window width and sliding step of the fast Fourier transform; wherein the window width is a single cycle length, the sliding step is a half cycle length, and the cycle length is the cycle length of the power supply voltage time domain waveform and the feeder current time domain waveform.

[0038] In some embodiments, preprocessing the monitoring data includes performing a fast Fourier transform (FFT) on the time domain waveforms of the power supply voltage and the feeder current to obtain the amplitude and phase arrays of the fundamental voltage and fundamental current of the DC arc furnace, thereby obtaining the reactive power array of the DC arc furnace.

[0039] Specifically, the analysis interval of the reactive power data of the DC arc furnace should match the prediction time scale. When used for static var compensator (SVC) prediction control, the prediction should be made at least 10ms ahead, so the analysis interval of the reactive power array should not exceed 10ms. On this basis, when performing FFT analysis on the sampled voltage and current waveform data, the FFT window width is taken as a single cycle, and slides once every half cycle, so as to obtain the voltage, current amplitude and phase array with an analysis interval of 10ms, and then calculate the reactive power array of the DC arc furnace.

[0040] S103: Calculate the reactive power array of the DC arc furnace based on the frequency domain array.

[0041] In some embodiments, S103 includes: calculating the reactive power array of the DC arc furnace based on a reactive power formula.

[0042] The reactive power formula includes:

[0043]

[0044]

[0045]

[0046] Q m (n) = Q m,A (n)+Q m,B (n)+Q m,C (n)

[0047] Among them, U A (n), U B (n), U C (n) are the effective values ​​of the fundamental voltages of phase A, phase B and phase C at time n in the voltage amplitude array; are the initial phases of the fundamental voltages of phase A, phase B and phase C at time n in the voltage phase array; I A (n), I B (n), I C (n) are the effective values ​​of the fundamental currents of phase A, phase B and phase C at time n in the current amplitude array; are the initial phases of the fundamental currents of phase A, phase B and phase C at time n in the current amplitude array; Q m,A (n), Q m,B (n), Q m,C (n) are the measured values ​​of the fundamental reactive power of phase A, phase B and phase C at time n; Q m (n) is the measured value of the three-phase total fundamental reactive power at time n in the reactive power array.

[0048] S104: Based on the autoregressive moving average model and the reactive power array, a rolling prediction calculation is performed on the reactive power of the DC arc furnace.

[0049] In some embodiments, the specific calculation method for predicting the reactive power of the DC arc furnace is regression prediction.

[0050] In some embodiments, reactive power prediction of a DC arc furnace is performed based on an autoregressive moving average model (ARMA).

[0051] In some embodiments, before S104, the DC arc furnace reactive power prediction method provided by the embodiment of the present invention may further include: constructing an autoregressive moving average model.

[0052] In some embodiments, constructing the autoregressive moving average model may include steps S201 to S203.

[0053] S201: Obtain historical data of reactive power array.

[0054] S202: Determine the order of the autoregressive moving average model based on the Akaike Information Criterion and historical data.

[0055] In some embodiments, the Akaike Information Criterion (AIC) and the DC arc furnace historical smelting data Q m The order of the above reactive power prediction model is identified.

[0056] Specifically, the AIC corresponding to different model orders is calculated, and p and q corresponding to the minimum AIC value are selected to determine the optimal prediction ARMA (p, q).

[0057] In some embodiments, the orders include autoregressive term orders and moving average term orders.

[0058] In some embodiments, S202 specifically includes: establishing a DC arc furnace reactive power prediction expression of an autoregressive moving average model. Calculating Akaike information criterion values ​​corresponding to different model orders of the DC arc furnace reactive power prediction expression based on the Akaike information criterion. The order with the smallest Akaike information criterion value is used as the order of the autoregressive moving average model.

[0059] The reactive power prediction expression ARMA(p, q) of DC arc furnace includes:

[0060]

[0061] Among them, Q p(n) is the reactive power prediction value at time n, Q p (ni) is the reactive power prediction value at time ni, Q m (ni) is the measured value of reactive power at time ni, is a constant term, is the autocorrelation coefficient, θ i is the moving average coefficient, p is the order of the autoregressive term of the forecast model, and q is the order of the moving average term of the forecast model.

[0062] S203: Determine the prediction model parameters of the autoregressive moving average model based on the loss function and historical data.

[0063] In some embodiments, the prediction model parameters include a constant term, an autocorrelation coefficient, and a moving average coefficient.

[0064] In some embodiments, S203 specifically includes: calculating the test prediction value of the DC arc furnace based on historical data, calculating the loss value of the historical data and the test prediction value based on the loss function, and using the prediction model parameter with the smallest loss value as the prediction model parameter of the autoregressive moving average model.

[0065] Specifically, the loss function can be calculated using the Euclidean distance between the model prediction value and the sample measured value. The expression of the loss function can include:

[0066]

[0067] Among them, L is the loss value, N is the number of test prediction values, and Q m (n) is historical data, Q p (n) is the test prediction value.

[0068] The reactive power prediction method for a DC arc furnace provided in an embodiment of the present invention applies the ARMA model, which can improve the accuracy of reactive power rolling prediction.

[0069] In some embodiments, the parameter training process may include: based on the loss function and the DC arc furnace historical smelting data Q m The parameters of the above reactive power prediction model are trained.

[0070] By minimizing the loss function L, the prediction model parameters corresponding to the minimum value of the loss function are used as the prediction model parameters, and finally the trained parameters are obtained, that is, the constant term in the reactive power prediction model. Autocorrelation coefficient Moving average coefficient θ i The value of .

[0071] In some embodiments, after S104, the method for predicting the reactive power of a DC arc furnace provided by the embodiments of the present invention may further include: calculating the adjusted coefficient of determination of the rolling prediction result based on the adjusted coefficient of determination formula. Determine whether the adjusted coefficient of determination is greater than or equal to a preset judgment threshold. If the adjusted coefficient of determination is greater than or equal to the preset judgment threshold, continue with the rolling prediction calculation.

[0072] The rolling prediction coefficient formula includes:

[0073]

[0074]

[0075] Where adjR 2 is the adjusted coefficient of determination, p is the order of the autoregressive term of the prediction model, q is the order of the moving average term of the prediction model, m is the total length of the sample, is the average value of the historical data.

[0076] If the evaluation result does not meet the preset evaluation index threshold, re-optimize the parameters of the prediction model and then perform the prediction.

[0077] Specifically, the optimal value of the dynamic evaluation index adjR 2 is 1, and the larger the value, the better the fitting effect. Before performing the dynamic evaluation, an evaluation threshold D can be set. When adjR 2 ≥ D, it indicates that the result predicted by the model is credible, and the prediction model can be used to continue the prediction. When adjR 2 < D, re-train the model parameters for prediction.

[0078] In a specific example, the evaluation threshold D = 0.4 is set.

[0079] In a specific example, the evaluation threshold D = 0.9 is set. The higher the value of the evaluation threshold, the more reliable the prediction effect.

[0080] In some embodiments, in order to implement the dynamic evaluation of the dynamic evaluation index adjR 2 , a dynamic evaluation time interval can be set. Specifically, the dynamic evaluation time interval can be selected according to the reactive power analysis interval, for example, 50 to 200 times the prediction time interval. That is, when the analysis interval is 10 ms, the time interval for each evaluation can be 1 second, 2 seconds, etc.

[0081] The method for predicting the reactive power of a DC arc furnace provided by the embodiments of the present invention uses the adjusted coefficient of determination to dynamically evaluate the prediction result, which can further ensure the accuracy and effect of the reactive power rolling prediction.

[0082] Figure 3The specific implementation flow diagram of the DC arc furnace reactive power rolling prediction method provided by the embodiment of the present invention is shown. Figure 3 In some embodiments, the three-phase voltage and current waveform data are first obtained, and the data is processed based on a sliding time window. Specifically, the length of the time window is one cycle, and the sliding step is half a cycle. The voltage and current waveform data are fast Fourier transformed to obtain the measured values ​​of the fundamental reactive power of phases A, B, and C to form a reactive power time series. Model identification and model training are performed based on the measured value data in the reactive power time series to obtain model parameters and finally form a prediction model. Furthermore, the current reactive power time series is obtained, and an autoregressive sequence and a residual sequence are constructed to perform residual calculations. The calculated color data is input into the pre-stored model to obtain the current predicted value. Continuously perform rolling predictions based on current data to achieve continuous and reliable prediction of reactive power of DC arc furnaces.

[0083] Traditionally, the medium- and long-term photovoltaic power generation available electricity forecast based on the ARIMA model can achieve the forecast of 12 months of medium- and long-term available electricity. Specifically, such forecasting methods can include obtaining the equivalent installed capacity of the reporting period; obtaining a higher forecast value through historical data statistical methods, and predicting the number of effective light resource hours based on the ARIMA model; predicting the medium- and long-term photovoltaic available electricity. Specifically, the model of the differential autoregressive moving average model ARIMA is specifically recorded as ARIMA (p, d, q). The model building process of ARIMA analysis includes judging whether the input time series is stable, model identification of the processed stable data, parameter estimation, model testing and model prediction. However, the above method can only predict the medium- and long-term photovoltaic available electricity for users, but it cannot realize the ultra-short reactive rolling forecast for DC arc furnaces, and on the other hand, it cannot realize the dynamic evaluation of the forecast effect.

[0084] The reactive power prediction method of a DC arc furnace provided by the embodiment of the present invention can establish a prediction model of a DC arc furnace, namely an ARMA prediction model, based on the autoregression and moving average theory, and use the Akaike information criterion, loss function and historical smelting data of the DC arc furnace to determine the model order and train the parameters of the prediction model, and dynamically evaluate the prediction results based on the adjustment of the determination coefficient, so as to realize the rolling prediction and real-time evaluation of the reactive power of the DC arc furnace 10ms in advance, provide support for the optimization control of the dynamic reactive compensation device and the electrode adjustment system of the DC arc furnace, and the prediction results are accurate. On the other hand, the prediction model provided by the embodiment of the present invention has a simple structure, clear physical meaning, and low calculation amount, which is convenient for practical application.

[0085] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0086] Taking a specific application scenario as an example, the DC arc furnace reactive power prediction method provided by an embodiment of the present invention is further explained.

[0087] In this example, the electric arc furnace is a 130-ton DC electric arc furnace. Measurements are taken on the high-voltage side of the transformer of the DC electric arc furnace. A single cycle is used as the FFT calculation window, and a half-cycle sliding window is used for analysis. The analysis interval is 10ms.

[0088] Figure 4 A schematic diagram of the reactive power variation trend of the DC arc furnace in this example is shown.

[0089] use Figure 4 The reactive power data between 30 seconds and 140 seconds shown in the figure are combined with the Akaike Information Criterion AIC method to identify the model. Specifically, given that the upper limit of the order is p=4, q=4, the corresponding model when AIC reaches the minimum value is ARMA(1,1), that is, p=1, q=1.

[0090] The optimal order prediction model for DC arc furnace is determined to include:

[0091]

[0092] The reactive power between 30 seconds and 140 seconds is used to train the above model parameters. The corresponding model parameters when the loss function is minimum are: k1=0.9423、k2=0.9802. The optimal prediction model includes:

[0093] Q p (n) = 4.8441 + 0.9423Q m (n-1)-0.9802[Q p (n-1)-Q m (n-1)]

[0094] Based on the model order and parameters obtained through training, a predictive analysis is performed on the data from the next 140 seconds to 200 seconds.

[0095] Figure 5 A comparison chart of the predicted reactive power value and the measured value of the DC arc furnace in the interval from 140 seconds to 200 seconds in this example is shown.

[0096] Figure 6 A comparison chart of the predicted reactive power value and the measured value of the DC arc furnace in the interval from 199 seconds to 200 seconds in this example is shown.

[0097] It can be seen that the DC arc furnace reactive power prediction method provided by the embodiment of the present invention can accurately and reliably realize reactive power prediction.

[0098] The prediction results are dynamically evaluated using the adjusted determination coefficient, and the evaluation threshold of the adjusted determination coefficient is set to D = 0.90. The 100 sets of data predicted in the previous period are dynamically evaluated once every second.

[0099] Figure 7 The figure shows the variation curves of 60 evaluation indicators obtained in the reactive power prediction process in the interval from 140 seconds to 200 seconds in this example. Figure 7 It can be seen from the data shown that the prediction effect evaluation indexes of the DC arc furnace reactive power prediction method provided by the embodiment of the present invention are all greater than the evaluation threshold value of 0.90, indicating that the prediction model is available.

[0100] Figure 8 A schematic structural diagram of a DC arc furnace reactive power prediction device provided in an embodiment of the present invention is shown.

[0101] See also Figure 8 The DC arc furnace reactive power prediction device 80 provided in the embodiment of the present invention may include a monitoring data acquisition module 810, a fast Fourier transform module 820, a reactive power array calculation module 830, and a prediction module 840.

[0102] The monitoring data acquisition module 810 is used to acquire monitoring data of the DC arc furnace, wherein the monitoring data includes a power supply voltage time domain waveform array and a feeder current time domain waveform array obtained by synchronous sampling.

[0103] The fast Fourier transform module 820 is used to perform fast Fourier transform on the power supply voltage time domain waveform array and the feeder current time domain waveform array to obtain a frequency domain array; the frequency domain array includes a voltage amplitude array, a voltage phase array, a current amplitude array, and a current phase array.

[0104] The reactive power array calculation module 830 is used to calculate the reactive power array of the DC arc furnace based on the frequency domain array.

[0105] The prediction module 840 is used to perform rolling prediction calculation on the reactive power of the DC arc furnace based on the autoregressive moving average model and the reactive power array.

[0106] The DC arc furnace reactive power prediction device 80 provided in the embodiment of the present invention can perform rolling prediction of the reactive power of the DC arc furnace based on real-time acquired monitoring data and a preset autoregressive sliding average model, thereby improving the accuracy and reliability of power prediction.

[0107] In some embodiments, the fast Fourier transform module 820 is specifically used to: perform fast Fourier transform on the power supply voltage time domain waveform array and the feeder current time domain waveform array based on the window width and sliding step of the fast Fourier transform; wherein the window width is a single cycle length, the sliding step is a half cycle length, and the cycle length is the cycle length of the power supply voltage time domain waveform and the feeder current time domain waveform.

[0108] In some embodiments, the reactive power array calculation module 830 is specifically used to calculate the reactive power array of the DC arc furnace based on the reactive power formula.

[0109] The reactive power formula includes:

[0110]

[0111]

[0112]

[0113] Q m (n) = Q m,A (n)+Q m,B (n)+Q m,C (n)

[0114] Among them, U A (n), U B (n), U C (n) are the effective values ​​of the fundamental voltages of phase A, phase B and phase C at time n in the voltage amplitude array; are the initial phases of the fundamental voltages of phase A, phase B and phase C at time n in the voltage phase array; I A (n), I B (n), I C (n) are the effective values ​​of the fundamental currents of phase A, phase B and phase C at time n in the current amplitude array; are the initial phases of the fundamental currents of phase A, phase B and phase C at time n in the current amplitude array; Q m,A (n), Q m,B (n), Q m,C (n) are the measured values ​​of the fundamental reactive power of phase A, phase B and phase C at time n; Q m (n) is the measured value of the three-phase total fundamental reactive power at time n in the reactive power array.

[0115] In some embodiments, the DC arc furnace reactive power prediction device 80 may further include a model building module for: obtaining historical data of a reactive power array, determining the order of an autoregressive moving average model based on the Akaike information criterion and historical data, and determining prediction model parameters of the autoregressive moving average model based on a loss function and historical data.

[0116] In some embodiments, the orders include autoregressive term orders and moving average term orders.

[0117] In some embodiments, the model building module is specifically used to: establish a DC arc furnace reactive power prediction expression of an autoregressive moving average model. Calculate the Akaike information criterion values ​​corresponding to different model orders of the DC arc furnace reactive power prediction expression based on the Akaike information criterion. The order with the smallest Akaike information criterion value is used as the order of the autoregressive moving average model.

[0118] The reactive power prediction expressions of DC arc furnace include:

[0119]

[0120] Among them, Q p (n) is the reactive power prediction value at time n, Q p (ni) is the reactive power prediction value at time ni, Q m (ni) is the measured value of reactive power at time ni, is a constant term, is the autocorrelation coefficient, θ i is the moving average coefficient, p is the order of the autoregressive term of the forecast model, and q is the order of the moving average term of the forecast model.

[0121] In some embodiments, the prediction model parameters include a constant term, an autocorrelation coefficient, and a moving average coefficient.

[0122] In some embodiments, the model building module is specifically used to: calculate the test prediction value of the DC arc furnace based on historical data. Calculate the loss value of the historical data and the test prediction value based on the loss function. Use the prediction model parameter with the smallest loss value as the prediction model parameter of the autoregressive moving average model.

[0123] The loss functions include:

[0124]

[0125] Among them, L is the loss value, N is the number of test prediction values, and Q m (n) is historical data, Q p (n) is the test prediction value.

[0126] In some embodiments, the DC arc furnace reactive power prediction device 80 may further include a prediction evaluation module, which is used to: calculate the adjustment determination coefficient of the rolling prediction result based on the adjustment determination coefficient formula. Determine whether the adjustment determination coefficient is greater than or equal to a preset judgment threshold, and if the adjustment determination coefficient is greater than or equal to the preset judgment threshold, continue the rolling prediction calculation.

[0127] The rolling forecast coefficient formula includes:

[0128]

[0129]

[0130] Among them, adjR 2 is the adjusted determination coefficient, p is the order of the autoregressive term of the forecast model, q is the order of the moving average term of the forecast model, m is the total length of the sample, is the average value of historical data.

[0131] Fig. 9 Schematic diagram of a terminal device provided by an embodiment of the present invention. Fig. 9 As shown, the terminal device 90 of this embodiment includes: a processor 900, a memory 910, and a computer program 920 stored in the memory 910 and executable on the processor 900, such as a DC arc furnace reactive power prediction program. When the processor 90 executes the computer program 920, the steps in the above-mentioned DC arc furnace reactive power prediction method embodiments are implemented, such as Figure 2 Alternatively, when the processor 900 executes the computer program 920, the functions of each module / unit in the above-mentioned device embodiments are implemented, for example Figure 8 Functions of modules 810 to 840 are shown.

[0132] Exemplarily, the computer program 920 may be divided into one or more modules / units, which are stored in the memory 910 and executed by the processor 900 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 920 in the terminal device 90. For example, the computer program 920 may be divided into a monitoring data acquisition module, a fast Fourier transform module, a reactive power array calculation module, and a prediction module.

[0133] The terminal device 90 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal device may include, but is not limited to, a processor 900 and a memory 910. Those skilled in the art will appreciate that Fig. 9 It is only an example of the terminal device 90 and does not constitute a limitation of the terminal device 90. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal device may also include input and output devices, network access devices, buses, etc.

[0134] The processor 900 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0135] The memory 910 may be an internal storage unit of the terminal device 90, such as a hard disk or memory of the terminal device 90. The memory 910 may also be an external storage device of the terminal device 90, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal device 90. Further, the memory 910 may also include both an internal storage unit of the terminal device 90 and an external storage device. The memory 910 is used to store the computer program and other programs and data required by the terminal device. The memory 910 may also be used to temporarily store data that has been output or is to be output.

[0136] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0137] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0138] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0139] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0140] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0141] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0142] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. . Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0143] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for predicting reactive power of a DC arc furnace, characterized in that: include: Acquiring monitoring data of a DC arc furnace, wherein the monitoring data includes a power supply voltage time domain waveform array and a feeder current time domain waveform array obtained by synchronous sampling; Performing a fast Fourier transform on the power supply voltage time domain waveform array and the feeder current time domain waveform array to obtain a frequency domain array; the frequency domain array includes a voltage amplitude array, a voltage phase array, a current amplitude array and a current phase array; Calculate the reactive power array of the DC arc furnace based on the frequency domain array; Based on the autoregressive moving average model and the reactive power array, a rolling prediction calculation is performed on the reactive power of the DC arc furnace; After performing rolling prediction calculation on the reactive power of the DC arc furnace based on the autoregressive average model and the reactive power array, the method includes: Calculate the adjusted determination coefficient of the rolling forecast results based on the adjusted determination coefficient formula; Determine whether the adjustment determination coefficient is greater than or equal to a preset determination threshold, and if the adjustment determination coefficient is greater than or equal to the preset determination threshold, continue to perform rolling forecast calculation; The adjustment coefficient of determination formula includes: Among them, adjR 2 is the adjusted determination coefficient, p is the order of the autoregressive term of the forecast model, q is the order of the moving average term of the forecast model, m is the total length of the sample, is the average value of historical data; Q m (n) represents historical data; Q p (n) represents the test prediction value; N is the number of test prediction values.

2. The method for predicting reactive power of a DC arc furnace according to claim 1, characterized in that: The fast Fourier transform is performed on the power supply voltage time domain waveform array and the feeder current time domain waveform array to obtain a frequency domain array, including: Based on the window width and sliding step of the fast Fourier transform, the power supply voltage time domain waveform array and the feeder current time domain waveform array are fast Fourier transformed; wherein the window width is a single cycle length, the sliding step is a half cycle length, and the cycle length is the cycle length of the power supply voltage time domain waveform and the feeder current time domain waveform.

3. The method for predicting reactive power of a DC arc furnace according to claim 1, characterized in that: The method of calculating the reactive power array of the DC arc furnace based on the frequency domain array includes: Calculating the reactive power array of the DC arc furnace based on a reactive power formula; The reactive power formula includes: Q m (n)=Q m,A (n)+Q m,B (n)+Q m,C (n) Among them, U A (n), U B (n), U C (n) are the effective values ​​of the fundamental voltages of phase A, phase B and phase C at time n in the voltage amplitude array; are the initial phases of the fundamental voltages of phase A, phase B and phase C at time n in the voltage phase array; I A (n), I B (n), I C (n) are the effective values ​​of the fundamental currents of phase A, phase B and phase C at time n in the current amplitude array; are the initial phases of the fundamental currents of phase A, phase B and phase C at time n in the current amplitude array; Q m,A (n), Q m,B (n), Q m,C (n) are the measured values ​​of the fundamental reactive power of phase A, phase B and phase C at time n; Q m (n) is the measured value of the three-phase total fundamental reactive power at time n in the reactive power array.

4. The method for predicting reactive power of a DC arc furnace according to claim 1, characterized in that: Before predicting and calculating the reactive power of the DC arc furnace based on the autoregressive moving average model and the reactive power array, the method includes: Obtaining historical data of the reactive power array; Determining the order of the autoregressive moving average model based on the Akaike Information Criterion and the historical data; Determining prediction model parameters of the autoregressive moving average model based on the loss function and the historical data; The orders include the autoregressive term order and the moving average term order; The step of determining the order of the autoregressive moving average model based on the Akaike information criterion and the historical data comprises: Establishing a reactive power prediction expression of a DC arc furnace based on the autoregressive moving average model; Calculate the Akaike information criterion values ​​corresponding to different model orders of the DC arc furnace reactive power prediction expression based on the Akaike information criterion; The order with the smallest Akaike information criterion value is used as the order of the autoregressive moving average model; The DC arc furnace reactive power prediction expression includes: Among them, Q p (n) is the reactive power prediction value at time n, Q p (ni) is the reactive power prediction value at time ni, Q m (ni) is the measured value of reactive power at time ni, is a constant term, is the autocorrelation coefficient, θ i is the moving average coefficient, p is the order of the autoregressive term of the forecast model, and q is the order of the moving average term of the forecast model; The prediction model parameters include a constant term, an autocorrelation coefficient and a moving average coefficient; The step of determining the prediction model parameters of the autoregressive moving average model based on the loss function and the historical data includes: Calculate the test prediction value of the DC arc furnace based on the historical data; Calculate the loss value of the historical data and the test prediction value based on the loss function; Using the prediction model parameters with the smallest loss value as the prediction model parameters of the autoregressive moving average model; The loss function includes: Among them, L is the loss value, N is the number of test prediction values, and Q m (n) is historical data, Q p (n) is the test prediction value.

5. A reactive power prediction device for a DC arc furnace, characterized in that: include: A monitoring data acquisition module, used to acquire monitoring data of a DC arc furnace, wherein the monitoring data includes a power supply voltage time domain waveform array and a feeder current time domain waveform array obtained by synchronous sampling; A fast Fourier transform module, used for performing fast Fourier transform on the power supply voltage time domain waveform array and the feeder current time domain waveform array to obtain a frequency domain array; the frequency domain array includes a voltage amplitude array, a voltage phase array, a current amplitude array, and a current phase array; A reactive power array calculation module, used for calculating the reactive power array of the DC arc furnace based on the frequency domain array; A prediction module, used for performing rolling prediction calculation on the reactive power of the DC arc furnace based on an autoregressive moving average model and the reactive power array; The DC arc furnace reactive power prediction device also includes a prediction evaluation module for: Calculate the adjusted determination coefficient of the rolling forecast results based on the adjusted determination coefficient formula; Determine whether the adjustment determination coefficient is greater than or equal to a preset determination threshold, and if the adjustment determination coefficient is greater than or equal to the preset determination threshold, continue to perform rolling forecast calculation; The adjustment determination coefficient formula includes: Among them, adjR 2 is the adjusted determination coefficient, p is the order of the autoregressive term of the forecast model, q is the order of the moving average term of the forecast model, m is the total length of the sample, is the average value of historical data; Q m (n) represents historical data; Q p (n) represents the test prediction value; N is the number of test prediction values.

6. A terminal 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 4 are implemented.

7. A computer-readable storage medium storing a computer program, 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 4 are implemented.

Images