Voltage sag detection method and device based on Teager energy operator and morphological filtering

Abstract

The invention provides a voltage sag detection method and device based on a Teager energy operator and morphological filtering, and relates to the technical field of power system sag protection and electric energy quality monitoring. The method comprises the following steps: firstly, converting a three-phase voltage signal into a voltage component under a two-phase static coordinate system through Clark conversion; secondly, a virtual instantaneous voltage power modulus containing a fundamental wave offset mechanism is constructed, and sag characteristic components can be effectively extracted; then, a morphological filtering combination filter is used for conducting denoising processing on the differential power signals, high-frequency impulse noise interference is eliminated, and pure characteristic signals are obtained; and finally, calculating an energy spectrum of the pure characteristic signal by using a Teager energy operator, and carrying out voltage sag detection in combination with a self-adaptive dynamic threshold, so as to judge the starting and ending moments of the voltage sag. According to the invention, the voltage sag can be accurately detected within 1ms, and the anti-noise performance is excellent.

Description

Technical Field

[0001] This invention relates to the field of power system voltage sag protection and power quality monitoring technology, specifically to a sub-millisecond voltage sag detection method and device based on the Teager energy operator and morphological filtering. Background Technology

[0002] Voltage sag is one of the most common power quality problems in modern industry. It is defined as a drop in the effective value of the supply voltage to 10%-90% of the rated value within a short period of time (typically 0.5ms-1min), before returning to normal. For sensitive loads such as semiconductor manufacturing and precision machining, even millisecond-level voltage drops can cause huge economic losses. Therefore, developing rapid voltage sag detection methods has significant engineering value. Existing detection technologies mainly have the following shortcomings: The root mean square (RMS) method requires at least half a cycle (10ms) of data window, which is too slow to meet the requirements of fast switching. Although the wavelet transform method has high accuracy, it has a large computational load and is greatly affected by the fundamental frequency fluctuation, making it difficult to implement in real time on low-cost embedded chips. In a three-phase unbalanced power grid system, the detection method based on instantaneous power theory may be subject to large second harmonic oscillation interference due to the influence of negative sequence components, resulting in inaccurate detection results. Summary of the Invention

[0003] The purpose of this invention is to provide a voltage sag detection method and device based on the Teager energy operator and morphological filtering, which solves the technical problems of existing detection methods such as large detection delay, easy malfunction in strong noise environment, and poor accuracy in three-phase unbalanced scenarios. It can accurately detect voltage sag within 1ms and has excellent noise resistance.

[0004] To achieve the above objectives, in a first aspect, the present invention provides a voltage sag detection method based on the Teager energy operator and morphological filtering. The three-phase voltage signals of the power line are acquired in real time, and Clark transformation is performed on the three-phase voltage signals to obtain... α-β Voltage components in a coordinate system; A virtual instantaneous voltage power modulus is constructed based on the voltage component, and a differential power signal is constructed based on the virtual instantaneous voltage power modulus; A morphological filtering combined operational filter is used to denoise the differential power signal to obtain a clean feature signal; The energy spectrum is obtained by operating on the pure feature signal using the discrete Teager energy operator; An adaptive dynamic threshold is set, and voltage sag detection of power lines is performed based on the relationship between the adaptive dynamic threshold and the energy spectrum.

[0005] According to the voltage sag detection method based on the Teager energy operator and morphological filtering provided by the present invention, the virtual instantaneous voltage power modulus is: , in, This represents the virtual instantaneous voltage power modulus at the nth sampling point. These represent the nth sampling point. Axis voltage component.

[0006] According to the voltage sag detection method based on the Teager energy operator and morphological filtering provided by the present invention, the differential power signal is: , in, S imp (n) represents the differential power signal at the nth sampling point; p inst (n-1) represents the virtual instantaneous voltage power modulus at the (n-1)th sampling point; t represents time. According to the voltage sag detection method based on the Teager energy operator and morphological filtering provided by the present invention, the pure feature signal is: , in, This represents the pure feature signal at the nth sampling point; Indicates the opening operation; This indicates the closing operation; This represents the erosion operation; This represents the expansion operation; g For the structuring element, g = [1, 1, 1]. According to the voltage sag detection method based on the Teager energy operator and morphological filtering provided by the present invention, the expression for the corrosion operation is as follows: , in, k This is the current sampling point number being processed; N This represents the total number of sampling points; i Here are the indexes of the structure elements, i = 0, 1, 2. According to the voltage sag detection method based on the Teager energy operator and morphological filtering provided by the present invention, the expression for the expansion operation is as follows: , in, k This is the current sampling point number being processed; N This represents the total number of sampling points; i Here are the indexes of the structure elements, i = 0, 1, 2. According to the present invention, a voltage sag detection method based on the Teager energy operator and morphological filtering is provided, with the energy spectrum as follows: , in, E TEO (n) represents the energy spectrum of the nth sampling point. This represents the energy of the pure feature signal at the nth sampling point. This represents the pure feature signal at the (n-1)th sampling point. This represents the pure feature signal at the (n+1)th sampling point.

[0007] According to the voltage sag detection method based on the Teager energy operator and morphological filtering provided by the present invention, the adaptive dynamic threshold is: , Where K is the reliability coefficient; L is the sliding window length; Indicates the first i Background noise energy at each sampling time; S normal ( i ) indicates the first i The characteristic signal sample values ​​of the power grid under normal operating conditions at each sampling time. According to the present invention, a voltage sag detection method based on the Teager energy operator and morphological filtering is provided, which detects voltage sags in power lines based on the relationship between an adaptive dynamic threshold and the magnitude of the energy spectrum, including: When E TEO (n)>T th When a voltage dip occurs in the power line, the current sampling time is recorded as the start time of the voltage dip; when E TEO (n)≤T th When at least 3 sampling points have been collected, the voltage sag is determined to have ended, and the current sampling time is recorded as the voltage sag termination time. Secondly, the present invention provides a voltage sag detection device based on the Teager energy operator and morphological filtering, comprising: The conversion unit is used to acquire the three-phase voltage signals of the power line in real time, perform Clark conversion on the three-phase voltage signals, and obtain... α-β Voltage components in a coordinate system; The construction unit is used to construct a virtual instantaneous voltage power modulus based on voltage components, and to construct a differential power signal based on the virtual instantaneous voltage power modulus; The processing unit is used to perform denoising on the differential power signal using morphological filtering combined operational filters to obtain a clean feature signal; The computation unit is used to perform operations on the pure feature signal using the discrete Teager energy operator to obtain the energy spectrum; The detection unit is used to set an adaptive dynamic threshold and to detect voltage sags in power lines based on the relationship between the adaptive dynamic threshold and the energy spectrum.

[0008] This invention has at least the following beneficial effects: 1. An improved instantaneous disturbance power model was constructed: It abandons the traditional dependence of instantaneous power on current signals, constructing a virtual instantaneous voltage power modulus based solely on voltage components, and optimizing feature extraction through a "fundamental cancellation mechanism." Specifically, by calculating the differential increments of instantaneous active and reactive power, the abrupt change characteristics of voltage sags are highlighted; simultaneously, the steady-state component is subtracted to eliminate fundamental frequency oscillation interference caused by three-phase imbalance, ensuring that the extracted feature signal contains only sag information, significantly improving purity. This approach retains the sensitivity of instantaneous power to voltage changes while solving the second harmonic oscillation problem in three-phase imbalance scenarios, eliminating the need for additional negative-sequence component separation algorithms and simplifying the calculation process.

[0009] 2. Introduction of Morphological Filtering: To address the critical weakness of TEO (Transient Oscillator) in noise sensitivity, a morphological filter combining "opening operation + closing operation" is introduced before TEO calculation. The opening operation (erosion followed by dilation) specifically removes positive spike noise from the signal, while the closing operation (dilation followed by erosion) specifically filters out negative spike noise. This approach can more thoroughly suppress high-frequency pulse interference without compromising the transient change characteristics.

[0010] 3. Adaptive Dynamic Threshold Mechanism: This mechanism overcomes the drawbacks of fixed thresholds, which are prone to false triggering and missed detections. It dynamically adjusts the judgment threshold based on the background noise level during normal grid operation. By statistically analyzing the noise amplitude of the TEO output under normal conditions using a sliding window, and combining this with the reliability coefficient K and the sliding window length L, the adaptive dynamic threshold T is calculated in real time. th It can automatically adapt to different noise environments, raising the threshold to avoid false triggering under strong noise and lowering the threshold to ensure false negatives under weak noise, thus significantly improving the robustness of the method. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0012] In the attached diagram: Figure 1 This is a flowchart of the voltage sag detection method based on the Teager energy operator and morphological filtering of the present invention. Figure 2 (a) is a schematic diagram of the corrosion and expansion operations of the present invention. Figure 2 (b) is a schematic diagram of the noise reduction effect of the combined operational filter of the present invention; Figure 3 (a), (b), (c), and (d) are respectively the original three-phase voltage input waveform, the improved instantaneous power differential signal waveform, the characteristic signal after morphological filtering, and the TEO energy spectrum and detection result in the simulation experiment of this invention. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0014] The following detailed description of some embodiments of the present invention will be provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0015] This invention introduces the Teager Energy Operator (TEO) into voltage sag detection. The TEO has extremely high instantaneous feature extraction capability, requiring only 3 sampling points to calculate energy, enabling sub-millisecond-level detection. Simultaneously, it combines improved instantaneous power and morphological filtering methods to achieve both "sub-millisecond-level response" and "strong noise immunity."

[0016] Please see Figure 1 This invention provides a fast voltage sag detection method based on the Teager energy operator and morphological filtering, comprising the following steps: Step 1: Coordinate Transformation. Real-time acquisition of three-phase voltage signals from power lines. u a , u b , u c For three-phase voltage signals u a , u b , u c Perform the Clark transform to obtain α-β Voltage components in coordinate system u α , u β The expression is: .

[0017] It should be noted that, as Figure 1 As shown, the present invention does not use a phase-locked loop during coordinate transformation, thus achieving faster speed and facilitating rapid detection of voltage sags in subsequent operations.

[0018] Step 2: Improve instantaneous power calculation. The traditional formula for instantaneous power calculation is: It is necessary to know the current. , Therefore, it cannot be calculated in scenarios where only voltage monitoring is used. To address this, the present invention is based on the voltage component u. α u β Construct a virtual instantaneous voltage power modulus, where the virtual instantaneous voltage power modulus at the nth sampling point (the current sampling point) is: , in, Represents the nth sampling point Axis voltage component.

[0019] When the power line is in steady state, the virtual instantaneous voltage power modulus It is a constant (the square of the amplitude). When a voltage dip occurs in a power line... Instantaneous changes, virtual instantaneous voltage power modulus A step change will occur, forming a distinct transient characteristic. To adapt to the Teager energy operator, this invention removes the DC component and retains only the abrupt change component. Therefore, the differential power signal constructed based on the virtual instantaneous voltage power modulus is as follows: ; in, S imp (n) represents the differential power signal at the nth sampling point, which is used to characterize voltage jumps by using the differential power signal of the virtual instantaneous voltage power modulus; p inst (n-1) represents the virtual instantaneous voltage power modulus of the (n-1)th sampling point (the previous sampling point); t represents time.

[0020] By constructing a differential power signal, it is possible to suppress second-harmonic oscillations caused by negative sequence.

[0021] Step 3, Morphological filtering. Directly apply... S imp (n) When performing TEO operations, it is easily affected by high-frequency pulse noise in the power grid, leading to misjudgments. Therefore, this invention uses a morphological filter combined with an operational filter, with the preferred approach being "open-closed cascade". The differential power signal is denoised by opening and closing operations to obtain a clean feature signal:

[0022] in, This represents the pure feature signal at the nth sampling point; Indicates the opening operation; This indicates the closing operation; This represents the erosion operation; This represents the expansion operation; g The structuring element is expressed as: g=[1,1,1]. Choosing a structuring element with a length of 3 can match the minimum 3-point sampling requirement of TEO, while minimizing the additional phase delay caused by filtering.

[0023] Erosion operation on differential power signal S imp (n) Perform shrinkage processing to eliminate the prominent part of the positive spike noise, the formula is: , Dilation operation on differential power signal S imp (n) is expanded to fill the depression portion of the negative spike noise, using the following formula: , in, k This is the current sampling point number being processed. N The total number of sampling points. i The structuring element index is (i=0, 1, 2). A schematic diagram of the erosion and dilation operations is shown below. Figure 2 As shown in (a).

[0024] Opening operations remove positive spike noise, while closing operations remove negative spike noise. This step is applied to differential power signals. S imp (n) Denoising processing can effectively smooth out non-faulty glitches and obtain a clean feature signal. S fil (n). The denoising effect of the combined filter is as follows: Figure 2 As shown in (b).

[0025] Step 4: TEO Energy Calculation. The Discrete Teager Energy Operator has extremely high analytical resolution. It is used to calculate the pure characteristic signal. S fil (n) is used to perform calculations to obtain the energy spectrum, which is the final detection index. E TEO (n), the expression is: , in, E TEO (n) represents the energy spectrum of the nth sampling point. This represents the energy of the pure feature signal at the nth sampling point. This represents the pure feature signal at the (n-1)th sampling point. This represents the pure feature signal at the (n+1)th sampling point.

[0026] At the instant a voltage sag occurs, the signal frequency and amplitude change abruptly, and the TEO's output value... E TEO (n) will exhibit a sharp pulse peak, which is the core basis for determining voltage sag.

[0027] Step 5: Adaptive Threshold Determination. Set an adaptive dynamic threshold. T th The threshold value is adjusted by real-time statistical analysis of the background noise level during normal grid operation. The formula is as follows: , Where K is the reliability coefficient, ranging from 1.5 to 3.0, with a default value of 2.0; n represents the current time (the latest nth sampling point); L is the sliding window length, ranging from 5 to 10 sampling points, with a default value of 8 sampling points, used to balance the response speed and stability of the threshold; Indicates the first i Background noise energy at each sampling time, S normal ( i ) represents the first i The characteristic signal sample values ​​of the power grid under normal operating conditions at each sampling time.

[0028] Within the initial sliding window, a preset empirical fixed threshold is used. Once the sliding window is filled with data, the system switches to adaptive mode. Voltage sag detection of the power line is performed based on the relationship between the adaptive dynamic threshold and the energy spectrum. The specific process includes: When E TEO (n)>T th When a voltage dip occurs in the power line, the current sampling time is recorded as the start time of the voltage dip; when E TEO (n)≤T th When at least 3 sampling points have been collected, the voltage sag is determined to have ended, and the current sampling time is recorded as the voltage sag termination time.

[0029] Based on the same inventive concept, another embodiment of the present invention provides a voltage sag detection device based on the Teager energy operator and morphological filtering. This device corresponds to the method of the foregoing embodiment and includes: The conversion unit is used to acquire the three-phase voltage signals of the power line in real time, perform Clark conversion on the three-phase voltage signals, and obtain... α-β Voltage components in a coordinate system; The construction unit is used to construct a virtual instantaneous voltage power modulus based on voltage components, and to construct a differential power signal based on the virtual instantaneous voltage power modulus; The processing unit is used to perform denoising on the differential power signal using morphological filtering combined operational filters to obtain a clean feature signal; The computation unit is used to perform operations on the pure feature signal using the discrete Teager energy operator to obtain the energy spectrum; The detection unit is used to set an adaptive dynamic threshold and to detect voltage sags in power lines based on the relationship between the adaptive dynamic threshold and the energy spectrum.

[0030] like Figure 3 As shown, to verify the effectiveness and fast response capability of the proposed method in a high-noise environment, a simulation model was built on the MATLAB / Simulink simulation platform. The simulation parameters were set as follows: system fundamental frequency f0 = 50Hz, sampling frequency f s =10kHz (i.e., sampling interval 0.1ms).

[0031] A simulated three-phase voltage symmetrical dip fault occurs in the power grid. The fault occurrence time is set to t. start =30ms, fault end time is t end =70ms, sag duration is 40ms, sag depth is set to 0.5pu (i.e., voltage drop to 50% of rated value).

[0032] To simulate the harsh electromagnetic environment of industrial sites, Gaussian white noise was superimposed on the acquired three-phase voltage signals, with the signal-to-noise ratio set to S0. NR =20dB. For example... Figure 3 As shown in (a), the original three-phase voltage input u a u b u c The waveform showed obvious spikes and random fluctuations, but step changes in amplitude could still be observed at 30ms and 70ms.

[0033] Improved power differential extraction: such as Figure 3 As shown in (b), the differential power signal S obtained after Clark transformation and the improved instantaneous power calculation proposed in this invention is... imp Although the impact characteristics at the start and end of the transient landing are highlighted, the background noise is also significantly amplified due to the differential effect, resulting in an extremely low signal-to-noise ratio. Direct detection of this signal is prone to misjudgment.

[0034] like Figure 3 As shown in (c), a morphological opening and closing combination filter is used to process the differential power signal S. imp After processing, a pure feature signal S is obtained. filIt can be seen that the combined filter effectively smooths out the random noise in the non-fault region (amplitude suppressed to near 0), while completely and steeply preserving the impact edge of the fault occurrence, verifying the excellent "noise reduction and edge preservation" characteristics of the method of the present invention.

[0035] like Figure 3 As shown in (d), the filtered signal is fed into the Teager energy operator to calculate the energy spectrum E. TEO At t=30ms (occurrence of transient sag) and t=70ms (recovery of transient sag), the energy spectrum exhibits single-pulse spikes with extremely high signal-to-noise ratio.

[0036] Combining adaptive threshold judgment logic, the system quickly locks the fault after a voltage dip. Simulation results show that the delay for detecting and issuing an alarm is only 6 sampling points (i.e., 0.6ms), fully meeting the design target of less than 0.8ms. Furthermore, no false triggers caused by noise occurred throughout the entire simulation period. Simulation results fully demonstrate that this invention can achieve sub-millisecond-level accurate detection even under strong noise interference. Simulation experiments verify that this method achieves a high signal-to-noise ratio (SNR) of S... NR Even in an environment with a voltage level of 20dB, the detection latency can still be controlled within 0.8ms (sub-millisecond level) without any false triggering.

[0037] In summary, the voltage sag detection method and apparatus based on the Teager energy operator and morphological filtering provided by this invention address the problems of malfunction and long detection delay in existing detection methods under strong noise environments, representing a novel detection architecture. First, the three-phase voltage signal is converted into voltage components in a two-phase stationary coordinate system using Clark transform. Second, a virtual instantaneous voltage power modulus incorporating a fundamental wave cancellation mechanism is constructed to effectively extract sag characteristic components. Subsequently, a morphological filtering combination filter is used to denoise the differential power signal, eliminating high-frequency pulse noise interference and obtaining a clean characteristic signal. Finally, the energy spectrum of the clean characteristic signal is calculated using the Teager energy operator, and combined with an adaptive dynamic threshold for voltage sag detection, allowing determination of the start and end times of the voltage sag. This invention can accurately detect voltage sags within 1 ms and exhibits excellent noise immunity.

[0038] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the embodiments disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. It should be understood that the invention is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A voltage sag detection method based on the Teager energy operator and morphological filtering, characterized in that, include: The three-phase voltage signals of the power line are acquired in real time, and the three-phase voltage signals are subjected to Clark transformation to obtain... α-β Voltage components in a coordinate system; A virtual instantaneous voltage power modulus is constructed based on the voltage components, and a differential power signal is constructed based on the virtual instantaneous voltage power modulus; The differential power signal is denoised using a morphological filtering combined operational filter to obtain a clean feature signal; The energy spectrum is obtained by operating the pure feature signal using the discrete Teager energy operator; An adaptive dynamic threshold is set, and voltage sag detection of power lines is performed based on the relationship between the adaptive dynamic threshold and the energy spectrum.

2. The voltage sag detection method based on Teager energy operator and morphological filtering according to claim 1, characterized in that, The virtual instantaneous voltage power modulus is: ， in, This represents the virtual instantaneous voltage power modulus at the nth sampling point. These represent the nth sampling point. Axis voltage component.

3. The voltage sag detection method based on Teager energy operator and morphological filtering according to claim 2, characterized in that, The differential power signal is: ， in, S imp (n) represents the differential power signal at the nth sampling point; p inst (n-1) represents the virtual instantaneous voltage power modulus at the (n-1)th sampling point; t represents time.

4. The voltage sag detection method based on Teager energy operator and morphological filtering according to claim 3, characterized in that, The pure feature signal is: ， in, This represents the pure feature signal at the nth sampling point; Indicates the opening operation; This indicates the closing operation; This represents the erosion operation; This represents the expansion operation; g For the structuring element, g = [1, 1, 1].

5. The voltage sag detection method based on Teager energy operator and morphological filtering according to claim 4, characterized in that, The expression for the corrosion operation is: ， in, k This is the current sampling point number being processed; N This represents the total number of sampling points; i Here are the indexes of the structure elements, i = 0, 1, 2.

6. The voltage sag detection method based on Teager energy operator and morphological filtering according to claim 4, characterized in that, The expression for the expansion operation is: ， in, k This is the current sampling point number being processed; N This represents the total number of sampling points; i Here are the indexes of the structure elements, i = 0, 1, 2.

7. The voltage sag detection method based on Teager energy operator and morphological filtering according to claim 4, characterized in that, The energy spectrum is: ， in, E TEO (n) represents the energy spectrum of the nth sampling point. This represents the energy of the pure feature signal at the nth sampling point. This represents the pure feature signal at the (n-1)th sampling point. This represents the pure feature signal at the (n+1)th sampling point.

8. The voltage sag detection method based on Teager energy operator and morphological filtering according to claim 7, characterized in that, The adaptive dynamic threshold is: ， Where K is the reliability coefficient; L is the sliding window length; Indicates the first i Background noise energy at each sampling time; S normal ( i ) indicates the first i The characteristic signal sample values ​​of the power grid under normal operating conditions at each sampling time.

9. The voltage sag detection method based on Teager energy operator and morphological filtering according to claim 8, characterized in that, Voltage sag detection of power lines is performed based on the relationship between the adaptive dynamic threshold and the energy spectrum, including: When E TEO (n)>T th When a voltage dip occurs in the power line, the current sampling time is recorded as the start time of the voltage dip; when E TEO (n)≤T th When at least 3 sampling points have been collected, the voltage sag is determined to have ended, and the current sampling time is recorded as the voltage sag termination time.

10. A voltage sag detection device based on the Teager energy operator and morphological filtering, characterized in that, include: The conversion unit is used to acquire the three-phase voltage signals of the power line in real time, and perform Clark conversion on the three-phase voltage signals to obtain... α-β Voltage components in a coordinate system; The construction unit is used to construct a virtual instantaneous voltage power modulus based on the voltage component, and to construct a differential power signal based on the virtual instantaneous voltage power modulus; The processing unit is used to perform denoising processing on the differential power signal using a morphological filtering combined operational filter to obtain a clean feature signal; The computation unit is used to perform operations on the pure feature signal using the discrete Teager energy operator to obtain the energy spectrum; The detection unit is used to set an adaptive dynamic threshold and perform voltage sag detection on the power line based on the relationship between the adaptive dynamic threshold and the energy spectrum.

Images