Dc power line carrier anti-interference device and method for distributed energy regulation

Abstract

The application provides a DC power line carrier anti-interference device and method for distributed energy regulation, comprising an A / D data acquisition module, a BP neural compensation Kalman filtering module, an interference cancellation module, a DC carrier communication module, a DC carrier communication interface, a DC interface and a power supply module. The method comprises data selection and normalization, BP neural network structure design, BP neural network training, Kalman filtering process and BP neural compensation Kalman filtering process. The application solves the problems that the statistical characteristics of the DC power line carrier channel interference are unknown, and the parameters are difficult to accurately measure and estimate, realizes joint estimation of the interference signal and the model parameters, eliminates the interference signal, increases the reliability of data transmission, and improves the anti-interference performance of the DC power line carrier.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of direct current power supply, and particularly relates to a direct current power line carrier anti-interference device and method for distributed energy regulation. BACKGROUND

[0002] Due to stable and convenient arrangement of direct current power supply cables, simple interface, low failure rate of communication equipment, and small maintenance workload, the direct current power line carrier technology can realize high reliability, low time delay, and long distance transmission of data, and is gradually applied to intelligent buildings, smart parks, and flexible load power supply. With direct current access of distributed photovoltaic and energy storage resources, the power grid dispatching center can effectively increase the utilization rate of distributed resources, improve the proportion of new energy consumption, and support the safe and stable operation of the power system. However, the topology structure of the direct current power line carrier channel is complex, and is affected by factors such as pulse interference and near-electric noise interference caused by dynamic access of a large number of renewable energy and intelligent electrical equipment, resulting in large signal attenuation and short transmission distance in the direct current power line, and even signal interruption. The traditional anti-interference device effectively filters specific frequency points or frequencies other than the frequency points to realize interference signal suppression. However, the anti-interference performance of the direct current power line carrier communication system for distributed energy regulation still faces the following challenges:

[0003] 1. The direct current power line carrier channel environment is complex and changeable, and is easily disturbed by noise with unknown statistical characteristics, resulting in a high packet loss rate of the direct current power line transmission signal.

[0004] 2. The existing anti-interference device is mainly designed for linear systems and is difficult to be applied to the nonlinear direct current power line carrier channel in the distributed energy regulation scene, and the anti-interference performance of the direct current power line carrier is low. SUMMARY

[0005] The technical problem to be solved by the application is:

[0006] 1. How to realize accurate joint estimation of the interference signal and the model parameters of the direct current power line carrier channel.

[0007] Most of the existing direct current power line carrier anti-interference methods and devices are based on the assumption that the model parameters of the interference noise are known, but in the actual direct current power line carrier channel, the parameters of the accurate noise model are difficult to obtain, resulting in low anti-interference performance of the device. Therefore, how to realize accurate joint estimation of the interference signal and the model parameters of the direct current power line carrier channel is a problem to be solved.

[0008] 2. How to improve the anti-interference performance of the power line carrier anti-interference device in the nonlinear direct current power line carrier channel.

[0009] Traditional power line carrier anti-interference device is mainly designed for linear system, while the noise components in DC power line carrier channel are complex, and the noise characteristics are difficult to be described by linear equation. Therefore, how to improve the power line carrier anti-interference device to adapt to the transmission environment of nonlinear DC power line channel is an urgent problem to be solved.

[0010] To solve the above technical problems, the application provides a DC power line carrier anti-interference device and method for distributed energy regulation, and the specific technical scheme is as follows:

[0011] The DC power line carrier anti-interference device for distributed energy regulation system comprises seven modules: an A / D data acquisition module, a BP neural compensation Kalman filter module, an interference cancellation module, a DC carrier communication module, a DC carrier communication interface, a DC interface and a power supply module.

[0012] The A / D data acquisition module is connected with the BP neural compensation Kalman filter module, the interference cancellation module is connected with the A / D data acquisition module and the BP neural compensation Kalman filter module, the DC carrier communication module is connected with the interference cancellation module, and the DC carrier communication interface is connected with the DC carrier communication module; the DC interface is connected with the power supply module.

[0013] The A / D data acquisition module converts the collected analog signal into a digital signal.

[0014] The BP neural compensation Kalman filter module generates an error compensation of the optimal filter estimation value.

[0015] The interference cancellation module subtracts the filter compensation output from the received original signal to reduce or cancel the noise in the original signal.

[0016] The DC carrier communication module and the DC carrier communication interface transmit the photovoltaic panel light intensity, energy storage battery charging and discharging power distributed energy regulation data to the power grid regulation center through the power line.

[0017] The DC interface converts the input voltage into a fixed output voltage.

[0018] The power supply module supplies low-voltage DC power to the entire device.

[0019] The DC carrier communication module comprises a signal modulation unit, a push-pull amplification circuit and a signal sending unit.

[0020] The signal modulation unit modulates the distributed energy regulation data onto a plurality of orthogonal sub-channels, converts it into a low-speed sub-data stream and transmits it in parallel; the push-pull amplification circuit copies and improves the transmission power of the entire system; and the signal sending unit realizes the external transmission of the distributed energy regulation data.

[0021] The anti-interference method of the DC power line carrier anti-interference device for distributed energy regulation comprises data selection and normalization, BP neural network structure design, BP neural network training, Kalman filtering process and BP neural compensation Kalman filtering process. The specific process is described as follows.

[0022] S1, data selection and normalization

[0023] The S-shaped function is set as the hidden layer activation function of the BP neural network. First, the input data is normalized, including decimal normalization, max-min normalization and standard deviation normalization.

[0024] S2, BP neural network structure design

[0025] A three-layer neural network structure is adopted, the number of input layer neurons of the neural network is set to 4, the number of output layer neurons is set to 1, and the number of hidden layer neurons is determined to be 10 according to the number of input layer and output layer neurons. The learning rate of the neural network is set to 0.99.

[0026] S3, BP neural network training

[0027] The BP neural network receives historical sample data information from the Kalman filter, repeatedly trains the network parameters, and stops training when the system preset training times are met or the prediction error accuracy requirement is met.

[0028] S4, Kalman filtering process

[0029] Gaussian white noise is adopted to model the background noise in the distributed energy regulation scene. The narrowband noise is represented as a low-frequency sinusoidal modulation signal, that is:

[0030] i(k)=w(k)sin(2πf c ) (1)

[0031] wherein w(k) represents the amplitude of the signal, which changes slowly; f c is the frequency of the carrier.

[0032] During the distributed energy regulation process, the transmitter transmits the spread spectrum signal, background noise interference and narrowband noise interference to the regulation center through the power line. Therefore, the signal received by the regulation center is represented by the following formula:

[0033] r(k)=s(k)+n(k)+i(k) (2)

[0034] wherein r(k) is the received signal, s(k) is the spread spectrum signal, and n(k) is the background noise.

[0035] Under the condition that the noise model parameters are unknown, Kalman filter is used to estimate the harmonic interference signal and its model parameters. The state equation is defined as:

[0036] X(k) = FX(k-1) + e(k) (3)

[0037] wherein X(k) = [i(k), i(k-1), a1(k-1)] T , e(k) is process noise, is the state transition matrix, wherein a2 = -1 is a model parameter, a1(k) = a1(k-1), is the interference frequency.

[0038] The measurement equation is represented as:

[0039] z(k) = HX(k) + v(k) (4)

[0040] wherein H = [1, 0, 0], v(k) = s(k) + n(k) represents measurement noise.

[0041] The Kalman prediction equation is as follows:

[0042] X(k|k-1) = FX(k-1|k-1) (5)

[0043] P(k|k-1) = ΦP(k-1)Φ T + Q (6)

[0044] wherein X(k-1|k-1) represents the optimal estimation of the Kalman filter at the k-1th time, X(k|k-1) represents the predicted state vector at the kth time, P(k|k-1) is the covariance matrix of X(k|k-1), and Q is the process noise matrix.

[0045] The Kalman update equation is as follows:

[0046]

[0047] X(k|k) = X(k|k-1) + K(k)·[Z(k) - HX(k|k-1)] (8)

[0048] P(k) = [I - K(k)H]·P(k|k-1) (9)

[0049] wherein K(k) is the Kalman filter gain, X(k|k) is the optimal filter estimation value at the current time, and P(k) is the updated covariance matrix.

[0050] S5, BP neural compensation Kalman filtering process

[0051] The trained BP neural network predicts the filter compensation value at the current time according to the input Kalman filter parameters capable of affecting the filter error, and the input parameters of the BP neural network are defined as:

[0052] (1) the estimated value error X(k|k-1)-X(k-1|k-1);

[0053] (2) the Kalman filter gain K(k);

[0054] (3) the error between the observation value and the estimated value Z(k)-HX(k|k-1);

[0055] (4) the optimal estimated error covariance matrix P(k).

[0056] The estimated error compensation amount output by the BP neural network is added to the optimal estimated value of the Kalman filter at the current time, and the filter compensation of the optimal estimated value of the Kalman filter is obtained.

[0057] The present application has the beneficial effects:

[0058] (1) The present application is aimed at the problems of pulse interference and near-electric appliance noise interference caused by the dynamic access of a large number of distributed energy and intelligent electrical equipment in the DC power line carrier channel, considers the difficulty in accurately measuring the current interference statistical characteristics, proposes a DC power line carrier anti-interference method based on BP neural compensation Kalman filtering to realize joint estimation of interference signals and model parameters, and overcomes the problem of low anti-interference performance caused by the inability of existing devices to accurately estimate interference parameters.

[0059] (2) The DC power line carrier anti-interference device designed for distributed energy regulation utilizes BP neural network to compensate and correct the output of Kalman filtering, obtains error compensation of the optimal filter estimation value, overcomes the bottleneck that the interference estimation value of the existing device deviates greatly from the true value and is difficult to apply to nonlinear systems, and makes it adapt to the DC power line carrier transmission environment with complex channel characteristics in the distributed energy regulation scene. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1 DC power line carrier anti-interference device for distributed energy regulation;

[0061] Figure 2 Flowchart of DC power line carrier anti-interference method for distributed energy regulation;

[0062] Figure 3 Principle diagram of DC power line carrier anti-interference method for distributed energy regulation. DETAILED DESCRIPTION

[0063] The specific technical solutions of the present invention will be described with reference to the embodiments.

[0064] To address the challenges of unknown interference statistics and difficulty in accurate measurement, as well as significant deviations between estimated and actual values, in DC power line carrier channels under distributed energy regulation scenarios, this invention designs a method such as... Figure 1 The diagram illustrates a DC power line carrier anti-interference device for distributed energy regulation systems, and proposes a method such as... Figure 2 The DC power line carrier anti-interference method shown is designed for distributed energy regulation, which effectively improves the anti-interference performance of DC power line carrier devices in distributed energy regulation environments.

[0065] like Figure 1 As shown, the DC power line carrier anti-interference device for distributed energy regulation systems proposed in this invention includes seven modules: an A / D data acquisition module, a BP neural compensation Kalman filter module, an interference cancellation module, a DC carrier communication module, a DC carrier communication interface, a DC interface, and a power supply module.

[0066] The A / D data acquisition module converts the acquired analog signals into digital signals; the BP neural compensation Kalman filter module generates error compensation for the optimal filter estimate; the interference cancellation module subtracts the filter compensation output from the received original signal to reduce or cancel noise in the original signal; the DC carrier communication module and DC carrier communication interface transmit distributed energy regulation data such as photovoltaic panel illuminance and energy storage battery charging and discharging power to the power grid control center via power lines; the DC interface converts the input voltage into a fixed output voltage; and the power supply module provides low-voltage DC power to the entire device. Details are as follows.

[0067] 1. A / D data acquisition module

[0068] The A / D data acquisition module is connected to the BP neural compensation Kalman filter module. Through four steps—sampling, holding, quantization, and encoding—it converts the acquired analog signal into a digital signal that can be further stored, processed, and analyzed. Sampling transforms the continuously changing analog quantity into a time-discrete analog quantity that reflects the characteristics of the original signal. Holding keeps the sampled value constant during the A / D conversion time, ensuring A / D conversion accuracy and eliminating conversion errors. Quantization normalizes the sampled level to a discrete digital level that approximates it. Encoding represents each quantized level using binary code.

[0069] 2. BP Neural Compensation Kalman Filter Module

[0070] Considering the complex communication environment of the DC power line carrier channel, the electromagnetic interference generated by the frequent switching of electrical equipment, and the nonlinear characteristics of the interference, the application sets a BP neural compensation Kalman filter module after the A / D data acquisition module, adopts a DC power line carrier anti-interference method based on BP neural compensation Kalman filtering, the principle is as shown in Figure 3 The fitting prediction function of the BP neural network is used to compensate and correct the related parameters of the Kalman filter, more accurate optimal estimation value filter error compensation is obtained, accurate joint estimation of the interference signal and the model parameters is realized, the interference signal is eliminated in the interference cancellation module, the performance of the device in eliminating nonlinear interference noise and the reliability of the DC power line data transmission are effectively improved, so that the DC power line anti-interference device designed by the application is flexibly applied to the distributed energy regulation DC power line carrier communication channel anti-interference scene.

[0071] The DC power line carrier anti-interference method for distributed energy regulation mainly includes five steps of data selection and normalization, BP neural network structure design, BP neural network training, Kalman filtering process and BP neural compensation Kalman filtering process, and the specific process is as follows:

[0072] S1、Data selection and normalization

[0073] The application sets the S-type function as the hidden layer activation function of the BP neural network. Considering that when the input data is too large or too small, the gradient of the function tends to zero, and the weight update is very slow, in order to avoid the difference between the input data and the neuron threshold being too large and causing the utility of the neural network to decrease, the input data is first normalized, including decimal normalization, max-min normalization and standard deviation normalization.

[0074] S2、BP neural network structure design

[0075] The application adopts a three-layer neural network structure, sets the number of input layer neurons of the neural network to 4, the number of output layer neurons to 1, and the number of hidden layer neurons to 10 according to the number of input layer and output layer neurons. The learning rate of the neural network is set to 0.99.

[0076] S3、BP neural network training

[0077] The BP neural network receives historical sample data information from the Kalman filter, repeatedly trains the network parameters until the system preset training times are met or the prediction error accuracy requirement is met, and then the training stops.

[0078] S4、Kalman filtering process

[0079] In the distributed energy regulation scene, the interference existing in the direct current power line carrier channel is mainly background noise and narrowband interference caused by harmonics. The background noise mainly comes from the random access of electrical equipment in the distributed energy regulation, and its power spectral density decreases with the increase of frequency. The background noise in the distributed energy regulation scene is modeled by using Gaussian white noise in the application. The narrowband noise is a kind of noise with very narrow frequency band, and its frequency bandwidth is much smaller than the center frequency. It can be expressed as a low frequency sinusoidal modulation signal, that is

[0080] i(k)=w(k)sin(2πf c ) (1)

[0081] Wherein, w(k) represents the amplitude of the signal, which changes slowly; f c is the frequency of the carrier.

[0082] In the distributed energy regulation process, the transmitting end transmits spread spectrum signals, background noise interference, narrowband noise interference and the like to the regulation center through the power line. Therefore, the signal received by the regulation center can be expressed by the following formula

[0083] r(k)=s(k)+n(k)+i(k) (2)

[0084] Wherein, r(k) is the received signal, s(k) is the spread spectrum signal, and n(k) is the background noise.

[0085] Most of the traditional filtering methods assume that the noise model parameters are known, and cannot be widely applied to the actual distributed energy regulation scene. Under the premise that the noise model parameters are unknown, the Kalman filter is used to jointly estimate the harmonic interference signal and its model parameters in the application. The state equation is defined as

[0086] X(k)=FX(k-1)+e(k) (3)

[0087] Wherein, X(k)=[i(k),i(k-1),a1(k-1)] T , e(k) is the process noise, is the state transition matrix, wherein a2=-1 is the model parameter, a1(k)=a1(k-1), is the interference frequency.

[0088] The measurement equation is represented as

[0089] z(k)=HX(k)+v(k) (4)

[0090] Wherein, H=[1,0,0], v(k)=s(k)+n(k) represents the measurement noise.

[0091] The Kalman prediction equation is as follows

[0092] X(k|k-1) = FX(k-1|k-1) (5)

[0093] P(k|k-1) = ΦP(k-1)Φ T + Q (6)

[0094] where X(k-1|k-1) denotes the optimal estimation of the Kalman filter at the k-1th time, X(k|k-1) denotes the predicted state vector at the kth time, P(k|k-1) is the covariance matrix of X(k|k-1), and Q is the process noise matrix.

[0095] The Kalman update equation is as follows:

[0096]

[0097] X(k|k) = X(k|k-1) + K(k) · [Z(k) - HX(k|k-1)] (8)

[0098] P(k) = [I - K(k)H] · P(k|k-1) (9)

[0099] where K(k) is the Kalman filter gain, X(k|k) is the optimal filter estimation value at the current time, and P(k) is the updated covariance matrix.

[0100] S5, BP neural compensation Kalman filtering process

[0101] The trained BP neural network predicts the filter compensation value at the current time according to the input Kalman filter parameters that can affect the filter error, and the input parameters of the BP neural network are defined as:

[0102] (1) the estimation error X(k|k-1) - X(k-1|k-1);

[0103] (2) the Kalman filter gain K(k);

[0104] (3) the error between the observation value and the estimation value Z(k) - HX(k|k-1);

[0105] (4) the optimal estimation error covariance matrix P(k).

[0106] The estimation error compensation amount ΔX(k) = X(k) - X(k|k) output by the BP neural network is added to the optimal estimation value of the Kalman filter at the current time, to obtain the filter compensation of the optimal estimation value of the Kalman filter.

[0107] 3. Interference cancellation module

[0108] The interference cancellation module is connected with the A / D data acquisition module and the BP neural compensation Kalman filter module, receives the original signal output by the A / D data acquisition module and the filter compensation of the optimal estimation value after being corrected by the BP neural compensation Kalman filter module, reduces or cancels the noise in the original signal by subtracting the filter compensation of the optimal estimation value from the original signal, and realizes the anti-interference of the direct current power line carrier system.

[0109] 4. DC carrier communication module

[0110] The DC carrier communication module is connected with the interference cancellation module, and comprises a signal modulation unit, a push-pull amplification circuit and a signal sending unit. The signal modulation unit modulates the distributed energy regulation data onto a plurality of orthogonal sub-channels by using the orthogonal frequency division multiplexing (OFDM) technology, converts the distributed energy regulation data into low-speed sub-data streams for parallel transmission, and effectively improves the transmission quality of the distributed energy regulation data by using the advantages of high transmission rate and strong anti-interference ability of the OFDM.

[0111] 5. DC carrier communication interface

[0112] The DC carrier communication interface is connected with the DC carrier communication module, is used for transmitting the output signal of the DC carrier communication module to the power line, and further filters out interference signals, improves the loading efficiency of the carrier signal and suppresses the peak voltage based on the composite coupling technology combining the electromagnetic coupling technology and the resistance-capacitance coupling technology. Since the power line is connected with distributed power sources with different impedances, the DC carrier communication interface has strong electrical isolation function, and guarantees the operation safety of the communication system and the power grid. In addition, the DC carrier communication interface supports the national grid 1376.2 communication module interface protocol, and provides communication support for the high-proportion new energy, controllable load, energy storage and other distributed energy participating in the power grid regulation.

[0113] 6. DC interface

[0114] The DC interface is connected with the power supply module, is used for converting the input voltage into a fixed output voltage, and effectively prevents the complex DC power line carrier communication environment in the distributed energy regulation scene from interfering with and damaging the device. The interface is composed of horizontal and vertical sockets, an insulating base, contact springs and a directional key groove. The two contact springs are positioned at the center of the base, are cross-arranged and are not connected with each other.

[0115] 7. Power supply module

[0116] The power supply module adopts an external power supply, and is used for supplying power for an A / D data acquisition module, a BP neural compensation Kalman filtering module, an interference cancellation module, a direct current (DC) carrier communication module, a DC carrier communication interface and a DC interface low-voltage DC power supply in the DC power line carrier anti-interference device for the distributed energy regulation and control system.

[0117] A storage medium having stored thereon a computer program, which, when executed by a processor, implements the steps of the above method.

[0118] An electronic device, comprising:

[0119] A memory having stored thereon a computer program;

[0120] A processor configured to execute the computer program in the memory to implement the steps of the above method.

Claims

1. A DC power line carrier anti-interference device for distributed energy regulation systems, characterized in that, include: A / D data acquisition module, BP neural compensation Kalman filter module, interference cancellation module, DC carrier communication module, DC carrier communication interface, DC interface, power supply module; The BP neural compensation Kalman filter module adopts a DC power line carrier anti-interference method based on BP neural compensation Kalman filtering. It uses the fitting and prediction function of BP neural network to compensate and correct the relevant parameters of Kalman filtering, obtains accurate optimal estimate values, and compensates for filtering errors. This achieves accurate joint estimation of interference signals and model parameters, and eliminates interference signals in the interference cancellation module. The A / D data acquisition module is connected to the BP neural compensation Kalman filter module; the interference cancellation module is connected to both the A / D data acquisition module and the BP neural compensation Kalman filter module; the DC carrier communication module is connected to the interference cancellation module; the DC carrier communication interface is connected to the DC carrier communication module; and the DC interface is connected to the power supply module. The A / D data acquisition module converts the acquired analog signals into digital signals; The BP neural compensation Kalman filter module generates error compensation for the optimal filter estimate. The interference cancellation module subtracts the filter compensation output from the received original signal to reduce or cancel the noise in the original signal; The DC carrier communication module and DC carrier communication interface will transmit distributed energy regulation data, including photovoltaic panel irradiance and energy storage battery charging and discharging power, to the power grid control center via power lines. The DC interface converts the input voltage into a fixed output voltage. The power supply module provides low-voltage DC power to the entire device.

2. The DC power line carrier anti-interference device for distributed energy regulation systems according to claim 1, characterized in that, The DC carrier communication module includes a signal modulation unit, a push-pull amplifier circuit, and a signal transmission unit. The signal modulation unit modulates the distributed energy regulation data onto several orthogonal sub-channels, converting it into low-speed sub-data streams for parallel transmission; the push-pull amplifier circuit replicates and increases the overall system's transmission power; and the signal transmission unit realizes the external transmission of the distributed energy regulation data.

3. A DC power line carrier anti-interference method for distributed energy regulation, characterized in that, The apparatus described in claim 1 or 2 includes data selection and normalization, BP neural network structure design, BP neural network training, Kalman filtering process, and BP neural compensation Kalman filtering process.

4. The DC power line carrier anti-interference method for distributed energy regulation according to claim 3, characterized in that, Specifically, the following steps are included: S1, Data Selection and Normalization Set the activation function of the hidden layer of the BP neural network to a sigmoid function; first, normalize the input data; S2, BP neural network structure design A three-layer neural network structure is adopted, with 4 neurons in the input layer and 1 neuron in the output layer. The number of neurons in the hidden layer is determined to be 10 based on the number of neurons in the input and output layers. The learning rate of the neural network is set to 0.

99. S3, BP neural network training The BP neural network receives historical sample data from the Kalman filter and repeatedly trains the network parameters until the preset number of training iterations or the accuracy requirement of the prediction error is met, at which point the training stops. S4, Kalman filtering process A Kalman filter is used to jointly estimate the harmonic interference signal and its model parameters. S5, BP neural compensation Kalman filtering process The trained BP neural network predicts the current filter compensation value based on the input Kalman filter parameters that affect the filter error. The estimated error compensation value output by the BP neural network is added to the current Kalman filter optimal estimate value to obtain the filter compensation value of the optimal Kalman filter estimate value.

5. The DC power line carrier anti-interference method for distributed energy regulation according to claim 4, characterized in that, The normalization method in step S1 can be any one of decimal place normalization, max-min normalization, or standard deviation normalization.

6. The DC power line carrier anti-interference method for distributed energy regulation according to claim 4, characterized in that, In step S4, Gaussian white noise is used to model the background noise in the distributed energy regulation scenario; Narrowband noise is represented as a low-frequency sinusoidal modulated signal, i.e.: i(k)=w(k)sin(2πf c ) (1) Where w(k) represents the amplitude of the signal, which changes relatively slowly; f c It is the frequency of the carrier wave; In the process of distributed energy regulation, the transmitting end transmits spread spectrum signals, background noise interference, and narrowband noise interference to the control center via power lines; therefore, the signal received by the control center is represented by the following formula: r(k)=s(k)+n(k)+i(k) (2) Where r(k) is the received signal, s(k) is the spread spectrum signal, and n(k) is the background noise; Given that the noise model parameters are unknown, a Kalman filter is used to jointly estimate the harmonic interference signal and its model parameters; the state equation is defined as: X(k)=FX(k-1)+e(k) (3) Where X(k)=[i(k),i(k-1),a1(k-1)] T e(k) represents process noise. Let be the state transition matrix, where a2 = -1 is the model parameter, a1(k) = a1(k-1). For interference frequency; The measurement equation is expressed as: z(k)=HX(k)+v(k) (4) Where H = [1,0,0], and v(k) = s(k) + n(k) represents the measurement noise; The Kalman prediction equation is as follows: X(k|k-1)=FX(k-1|k-1) (5) P(k|k-1)=ΦP(k-1)Φ T +Q (6) Where X(k-1|k-1) represents the optimal estimate of the Kalman filter in the (k-1)th iteration, and X(k|k-1) represents the predicted state vector in the kth iteration. P(k|k-1) is the covariance matrix of X(k|k-1), and Q is the process noise matrix; The Kalman update equation is as follows: X(k|k)=X(k|k-1)+K(k)·[Z(k)-HX(k|k-1)] (8) P(k)=[IK(k)H]·P(k|k-1) (9) Where K(k) is the Kalman filter gain, X(k|k) is the optimal filter estimate at the current time, and P(k) is the updated covariance matrix.

7. The DC power line carrier anti-interference method for distributed energy regulation according to claim 4, characterized in that, In step S5, the input parameters of the BP neural network are defined as follows: (1) Estimation error X(k|k-1)-X(k-1|k-1); (2) Kalman filter gain K(k); (3) The error between the observed value and the estimated value is Z(k)-HX(k|k-1); (4) Optimal estimation error covariance matrix P(k).

8. The DC power line carrier anti-interference method for distributed energy regulation according to claim 7, characterized in that, The estimated error compensation amount of the BP neural network output is ΔX(k) = X(k) - X(k|k).

9. A storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 3 to 8.

10. An electronic device, characterized in that, include: A memory on which computer programs are stored; A processor for executing the computer program in the memory to implement the steps of the method according to any one of claims 3 to 8.

Images