A Fault Diagnosis Method for DC Microgrids Based on Minimal Sensors

Abstract

This invention discloses a fault diagnosis method for DC microgrids based on minimal sensors. This method shares converter output voltage and current sampling data with the controller of the distributed power source. It detects different faults by constructing a feature quantity based on the multi-step differential cumulative sum of the converter output voltage and current. It classifies inter-pole faults, positive-to-ground faults, and negative-to-ground faults based on the error of the output voltage estimation value. Finally, it uses non-real-time communication between adjacent sources to determine the line where the fault is located. This method can detect and classify different faults without additional sensors, has high accuracy in diagnosing high-resistance faults, and offers good reliability and economy without requiring real-time communication.

Description

Technical Field

[0001] This invention belongs to the field of DC microgrid fault diagnosis technology, specifically a DC microgrid fault diagnosis method based on minimal sensors. Background Technology

[0002] With the increasing prominence of global energy shortages and environmental pollution, renewable energy generation such as wind and solar power has received growing attention. Microgrids are an effective way to integrate renewable energy generation, overcoming the volatility and intermittency of renewable energy sources and maximizing their power supply efficiency. Based on the form of power transmission, microgrids can be divided into AC microgrids and DC microgrids. DC microgrids have advantages such as simple control, low power loss, and fewer conversion stages, and have broad development prospects. However, the difficulties in fault protection for DC microgrids have become a significant factor restricting their development. On the one hand, due to the difference in power transmission methods, traditional grid fault protection methods are difficult to apply to DC microgrids; on the other hand, DC microgrids are highly electronic systems with low inertia and weak overcurrent tolerance, resulting in rapid development and propagation of line faults, posing challenges to the speed and accuracy of fault diagnosis.

[0003] Existing fault diagnosis methods for DC microgrids can be mainly divided into methods based on single-end line information and methods based on double-end line information. Methods based on single-end line information identify faults by detecting the electrical state at one end of the line, such as detecting single-end current, current change rate, power, or current-limiting reactor voltage. These methods rely solely on local information for fault detection, requiring no communication and offering high reliability. However, they are susceptible to short-circuit impedance and struggle to detect high-resistance faults. Methods based on double-end line information detect faults by comparing differences in electrical quantities such as current and power at both ends of the line. They typically exhibit high sensitivity and selectivity; however, they rely on real-time communication between the two ends of the line and are significantly affected by communication delays. Furthermore, for detecting grounding faults in bipolar DC microgrids, existing methods require the simultaneous installation of corresponding measuring devices on both the positive and negative poles, increasing the cost of fault diagnosis.

[0004] The current patent application status is as follows:

[0005] Patent No.: CN202110783338.9, Patent Title: A Fault Component Identification Method Applicable to Fault Self-Clearing DC Distribution Network. This invention proposes a fault component identification method for DC distribution networks using fault self-clearing converters combined with fast disconnect switches. Based on the fault characteristics of each bus at different fault locations, virtual voltage characteristics of the bus are constructed to identify the faulty line. Based on the switching process of some sub-modules in the hybrid converter, auxiliary identification criteria are constructed based on the capacitance charge transfer amount to complete the identification of the faulty bus. This invention can accurately identify faulty components and accurately measure fault distance, has strong resistance to transition resistance, and reduces the investment in line sensors.

[0006] This invention identifies faulty lines by constructing virtual voltage characteristics of each bus based on the fault characteristics of each bus at different fault locations in a DC distribution network. The calculation of these characteristics requires real-time communication to obtain voltage information from multiple buses and output current information from multiple converters. Furthermore, the proposed fault bus identification criterion based on capacitance charge transfer aims to distinguish between bus faults and faults at the end of bus-related lines. This requires coordination of switching sub-modules within the converter and opening and closing of all bus-related line disconnect switches, resulting in a long fault identification time and a significant impact on the system's power supply reliability.

[0007] The fault characteristic quantities constructed in this invention only require the output voltage and current of the local converter for calculation, and fault detection does not require communication. The fault classification method based on the converter output voltage estimation error proposed in this invention distinguishes between inter-pole faults, positive-pole grounding faults, and negative-pole grounding faults, and no disconnecting switches or circuit breakers need to be activated before the fault is cleared.

[0008] Patent No.: CN 201811362494.2, Patent Title: Fault Detection Method for Ring DC Distribution Network Based on Voltage Prediction, belonging to the field of safe operation of DC distribution networks. The DC distribution network has protection devices installed at both ends, including circuit breakers, sensors, relays, and reactors. Its key feature is that it predicts the voltage value of the line side at the next moment based on the sampled voltage values ​​at any three sampling time points on the line side, and uses the difference between the predicted voltage value and the measured voltage value at the time of the fault as the basis for protection action. This invention solves the problems of weak speed of traditional DC distribution network protection and over-reliance on communication and the damping state of the circuit. This invention relies only on the voltage value of the line side, effectively avoiding the over-reliance on the damping state of the circuit at the time of the fault when setting protection using elements such as current, the first and second derivatives of current, and the first derivative of voltage.

[0009] It identifies faults by the difference between the predicted voltage and the measured voltage at the time of the fault. Its voltage prediction method is based on the second derivative of voltage. The premise that this fault detection method can quickly and accurately identify faults is that there is a very large voltage drop at the moment of the fault. However, when the fault impedance is large, the voltage drop is slow. In this case, it is difficult to guarantee the speed and accuracy of fault detection using this method.

[0010] The fault detection method proposed in this invention is achieved by calculating the multi-step differential cumulative sum of the converter output voltage and current, which significantly amplifies fault characteristics and enables rapid and accurate detection of different types of faults even when the fault impedance is large. The output voltage estimation method of this invention is based on the dynamic analysis of faults in the converter output capacitor and aims to achieve fault classification, which is significantly different from the voltage prediction method of the aforementioned invention. Summary of the Invention

[0011] To address the above issues, this invention proposes a fault diagnosis method for DC microgrids based on minimal sensors. This method requires no real-time communication, has high reliability, can accurately detect high-resistance faults, and can detect and classify inter-pole faults, positive grounding faults, and negative grounding faults in bipolar DC microgrids without the need for additional sensors.

[0012] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0013] A fault diagnosis method for DC microgrids based on minimal sensors, characterized by comprising the following steps:

[0014] Step 1) Measure the output current and voltage of the distributed power source and calculate the corresponding fault characteristic quantities SI and SV;

[0015] The calculation steps for fault characteristic quantities SI and SV are as follows:

[0016] First, calculate the multi-step differential of the output current and voltage:

[0017] Δi(k)=i(k)-i(kn i (1)

[0018] Δv(k)=v(k)-v(kn v (2)

[0019] In the formula, i(k) and v(k) are the sampled values ​​of the output current and voltage of the distributed power source at time k, respectively, and n i and n v These are the sampling intervals for current and voltage, respectively;

[0020] Then, the multi-step differential sliding window cumulative sum of the output current and voltage is calculated:

[0021]

[0022]

[0023] The window length of the current multi-step differential sliding window cumulative sum is n. i The window length of the voltage multi-step differential sliding window cumulative sum is n. v ;

[0024] Step 2) Compare the calculated fault characteristic quantities SI and SV with the preset threshold SI. th SV th The comparison is performed, and if SI or SV exceeds the threshold, the line connected to the converter is considered to be faulty.

[0025] Step 3) Estimate the output voltage of the converter and classify the faults based on the error between the estimated voltage value and the actual measured value;

[0026] Step 4) Distributed communication is carried out between adjacent distributed generation units to send fault detection results to each other. When two adjacent units detect a fault at the same time, it can be determined that the fault is located on the line connected to the two adjacent units. Then, the faulty line is isolated according to the identified fault type.

[0027] As a further improvement to the present invention, step 3) is as follows:

[0028] First, the converter output voltage is estimated based on the steady-state values ​​of the output voltage and current before the fault and the measured current value after the fault:

[0029]

[0030] Among them, v o (k0), i o (k0) represent the steady-state values ​​of the converter output voltage and current before the fault, respectively. s Where C is the sampling period, and C is the capacitance on the output side of the converter;

[0031] Then, the difference between the estimated output voltage and the measured value is calculated:

[0032]

[0033] Finally, based on the voltage estimation error Δv o (k) The magnitude of the error is used to determine the type of fault. If the absolute value of the estimation error is less than the threshold, an inter-pole fault is considered to have occurred; if the estimation error is greater than the threshold, a positive-pole grounding fault is considered to have occurred; if the estimation error is less than the negative threshold, a negative-pole grounding fault is considered to have occurred, as shown in the following formula:

[0034]

[0035] Among them, ε v This is the preset error threshold.

[0036] Beneficial effects:

[0037] The present invention proposes a fault diagnosis method for DC microgrids based on minimal sensors, which can achieve rapid and accurate detection of high-resistance faults and classify inter-pole faults, positive-pole grounding faults and negative-pole grounding faults. In addition, the method does not require real-time communication, has anti-communication delay capability, and does not require additional voltage and current sensors, thus having advantages in reliability and economy. Attached Figure Description

[0038] Figure 1 This is a flowchart of the present invention;

[0039] Figure 2 This is a schematic diagram of the DC microgrid structure of the present invention;

[0040] Figure 3 This is a schematic diagram of the current fault characteristic quantity detection of the present invention;

[0041] Figure 4 This is a schematic diagram of voltage fault characteristic quantity detection according to the present invention;

[0042] Figure 5 A schematic diagram of the discharge circuit for the converter output capacitor under different fault conditions;

[0043] Figure 6 This is a schematic diagram of the inter-pole fault classification results of the present invention;

[0044] Figure 7 This is a schematic diagram of the positive grounding fault classification results of the present invention;

[0045] Figure 8 This is a schematic diagram of the negative electrode grounding fault classification results of the present invention;

[0046] Figure 9 This is a schematic diagram of the distributed non-real-time communication fault location method of the present invention. Detailed Implementation

[0047] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:

[0048] Example 1:

[0049] This embodiment and the following embodiments all use... Figure 2 The DC microgrid shown is used as an example for illustration. Figure 2The illustrated DC microgrid is a ring-type bipolar DC microgrid with a voltage level of 400V (±200V). It includes one photovoltaic (PV) power generation unit, three energy storage units, and two DC load units. Each distributed power source and load is connected to the bus via a DC / DC converter. The PV power generation unit employs a maximum power point tracking (MPPT) control strategy to fully utilize renewable energy, while the energy storage units use a droop control strategy to regulate the system voltage. The switching frequency of the converter in the system is set to 20kHz, and the voltage and current sampling frequency is also set to 20kHz.

[0050] when Figure 2 When an inter-pole fault or a positive-to-ground fault occurs on line DE of the DC microgrid shown, the converter output current of the load unit and energy storage unit at both ends of the line will rise rapidly. However, the magnitude of the current change will decrease as the fault resistance increases. In order to amplify the fault characteristics and suppress noise interference, a multi-step differential current accumulation is used to construct the fault characteristic quantity:

[0051] Δi(k)=i(k)-i(kn i (1)

[0052]

[0053] against Figure 2 For the system shown, considering its fault dynamic characteristics, protection time requirements, and sampling frequency, the interval n of the multi-step differential current is determined. i Set to 4, cumulative sum and sliding window length to 4, detection threshold SI th Set it to 10. Figure 3 The results of fault detection using multi-step differential current accumulation are shown when line DE experiences inter-electrode faults (PP) and positive ground faults (PPG), with a fault resistance of 10Ω. As can be seen from the figure, both high-resistance inter-electrode faults and positive ground faults can be detected within 0.1ms at both ends of the faulty line.

[0054] When a negative ground fault occurs in the line, the positive line is less affected, and the positive output current of the converter changes slowly, making it difficult to use for rapid fault detection. To avoid adding an extra sensor, this method uses the converter output voltage for negative ground fault detection. Because the voltage drop is slow, a multi-step differential method is used to capture fault characteristics, and these characteristics are amplified through sliding window accumulation.

[0055] Δv(k)=v(k)-v(kn v (3)

[0056]

[0057] against Figure 2For the system shown, considering its fault dynamics, protection time requirements, and sampling frequency, the voltage multi-step differential interval n is... v Set the cumulative and sliding window length to 20, and the detection threshold SV. th Set it to -20. Figure 4 The results of fault detection using voltage multi-step differential summation are shown when a 10Ω negative ground fault (NPG) occurs on line DE. As can be seen from the figure, a high-resistance negative ground fault can be detected at both ends of the faulty line within 1ms.

[0058] Example 2:

[0059] This embodiment illustrates a fault classification method based on output voltage estimation. Figure 5 The discharge circuits of the converter output capacitor under different fault conditions are illustrated. Due to the slow response of the converter after a fault, the feed current i into the capacitor... in The change is relatively slow, while the voltage across the capacitor remained stable before the fault, and the currents before and after the capacitor were in a dynamic equilibrium. Therefore:

[0060] i in (t)≈i in (t 0- )≈i o (t 0- (5)

[0061] According to Kirchhoff's current law, the discharge current of the capacitor can be estimated as:

[0062]

[0063] For inter-pole faults, the output voltage can be estimated using the initial values ​​of the converter output voltage and current before the fault, and the sampled values ​​of the current after the fault.

[0064]

[0065] Discretize, and we get:

[0066]

[0067] For inter-electrode faults, the estimated output voltage obtained using the above formula has a very small error compared to the actual value. However, for ground faults, due to the different discharge circuits, the estimated value will deviate significantly from the actual value. Specifically, the capacitance value in the fault circuit of a positive ground fault is twice that of an inter-electrode fault, therefore its estimated voltage will be smaller than the actual value. In the case of a negative ground fault, the estimated capacitor discharge current i is obtained by using the current measured at the positive terminal. o (t) is less than the actual discharge current of the negative capacitor, so its estimated voltage value will be greater than the actual value. Therefore, the fault type can be determined based on the voltage estimation error:

[0068]

[0069] in, ...

[0070] Figures 6-8 The voltage estimation curves for line DE under inter-pole fault, positive-to-ground fault, and negative-to-ground fault are shown, with an error threshold ε. v Set to 1.5, as shown in the figure, the line can accurately classify different types of faults at both ends.

[0071] Example 3:

[0072] This embodiment describes a fault location method based on distributed non-real-time communication. For ring or mesh DC microgrids, the output side of each converter is connected to multiple lines. The fault diagnosis result of a single converter alone cannot determine the faulty line. Therefore, distributed communication between adjacent units is used to determine the faulty line. Figure 9 As shown.

[0073] When a unit detects a fault and identifies its type, it sends the diagnostic results to multiple adjacent units. At the same time, it receives fault diagnostic results from adjacent units. When two units send and receive the same fault diagnostic results, it can be determined that the fault is located on the line between the two units. The unit then randomly sends corresponding signals to the short circuits at both ends of the line to isolate the faulty line.

[0074] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other way. Any modifications or equivalent changes made based on the technical essence of the present invention shall still fall within the scope of protection claimed by the present invention.

Claims

1. A least sensor based DC microgrid fault diagnosis method, characterized in that, The method comprises the following steps: Step 1) measuring the output current and voltage of the distributed power supply and calculating corresponding fault characteristic quantities SI and SV; The calculation steps of the fault characteristic quantities SI and SV are as follows: First, the multi-step difference of the output current and voltage is calculated: （1） （2） In the formula, and are respectively the sampling values of the output current and voltage of the distributed power supply at time k, and are respectively the sampling interval numbers of the current and voltage. Then, the multi-step difference sliding window cumulative sum of the output current and voltage is calculated: （3） （4） Wherein, the window length of the current multi-step differential sliding window cumulative sum is , and the window length of the voltage multi-step differential sliding window cumulative sum is ; Step 2) comparing the calculated fault feature quantities SI, SV with preset threshold values SI th , SV th , and if SI or SV exceeds the threshold values, considering that a fault has occurred in the line connected to the transformer; Step 3) estimating the output voltage of the transformer, and classifying the fault according to the error between the voltage estimation value and the actual measured value; The specific steps of step 3) are as follows: First, the output voltage of the transformer is estimated according to the steady-state values of the output voltage and current before the fault and the current measurement value after the fault: （5） wherein , are the steady state values of the converter output voltage, current before the fault, respectively, is the sampling period, C is the size of the capacitor on the output side of the converter. Then, the output voltage estimation value is subtracted from the actual measured value: （6） Finally, the fault type is determined according to the voltage estimation error size, if the absolute value of the estimation error is less than a threshold value, it is considered that the inter-pole fault occurs; if the estimation error is greater than the threshold value, it is considered that the positive pole ground fault occurs; if the estimation error is less than a negative threshold value, it is considered that the negative pole ground fault occurs, as shown in the following formula: （7） wherein ε v is a preset error threshold; Step 4) distributed communication is performed between adjacent distributed power generation units, and the fault detection results are transmitted to each other; when a fault is detected by both of the adjacent units, it is determined that the fault is located on the line connected by the two units, and then the fault line is isolated according to the type of the fault identified.

Images