A method for detecting abnormality of DC voltage measurement of MMC-HVDC system

Abstract

The application provides a kind of detection method for the abnormal detection of DC voltage measurement of MMC-HVDC system, comprising the following steps: S100, obtaining the DC voltage, DC current and the upper bridge arm voltage and lower bridge arm voltage of each phase of MMC-HVDC system;S200, establishing the DC current differential equation of MMC-HVDC system;S300, after discretization of DC current differential equation, it is expressed by standard form of Kalman filter;S400, using Kalman filter to observe the state of DC current and obtaining the observed value of DC current;S500, according to the observed value of DC current, determine whether the DC voltage measurement is abnormal or not.It has the advantages of high reliability, simple detection logic, short time consumption, and can also adapt to complex MMC-HVDC system, easy to industrial application.

Description

Technical Field

[0001] This invention relates to the field of high voltage direct current transmission technology, and in particular to a method for detecting abnormal DC voltage measurement in an MMC-HVDC system. Background Technology

[0002] Modular multilevel converters (MMCs) offer advantages such as modular design, good scalability, low switching losses, and high power quality, making them promising candidates for high-voltage direct current (HVDC) transmission. MMC-based flexible direct current transmission systems (MMC-HVDC) have become a representative solution to the challenges of long-distance, high-capacity power transmission, and are widely used in renewable energy grid integration, asynchronous grid interconnection, and isolated power supply.

[0003] Safety issues in MMC-HVDC systems have always been a key concern for both academia and industry. DC voltage measurement devices, as crucial equipment in high-voltage DC measurement systems, are essential for the safe and stable operation of DC transmission systems. DC voltage measurement data is used for the control and protection of HVDC systems; abnormal data measurements can not only cause voltage fluctuations in the DC system but may also lead to malfunctions in control and protection systems, forcing DC system shutdowns and resulting in incalculable economic losses. In recent years, several HVDC projects already in operation in China have experienced DC system malfunctions due to abnormal DC voltage measurements, seriously threatening the normal operation of the DC system.

[0004] If a DC voltage measurement anomaly occurs, timely alerting of operators for inspection and handling can significantly reduce the impact on the HVDC system. Therefore, rapid detection technology for DC voltage measurement anomalies is of great importance to HVDC systems. Research on methods for detecting DC voltage measurement anomalies in MMC-HVDC systems is relatively limited; how to quickly and accurately detect DC voltage measurement anomalies is a pressing issue that needs to be addressed for MMC-HVDC systems. Summary of the Invention

[0005] The purpose of this invention is to provide a method for detecting abnormal DC voltage measurement in an MMC-HVDC system, which can quickly and accurately detect abnormal DC voltage measurement.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A method for detecting abnormal DC voltage measurement in an MMC-HVDC system, characterized in that:

[0008] The steps include: S100, acquiring the DC voltage, DC current, and upper and lower bridge arm voltages of the MMC-HVDC system;

[0009] S200. Establish the DC current differential equation of the MMC-HVDC system;

[0010] S300. After discretizing the DC current differential equation, it is expressed in the standard form of Kalman filtering.

[0011] S400: Use a Kalman filter to observe the state of DC current and obtain DC current observation values;

[0012] S500: Determine whether the DC voltage measurement is abnormal based on the DC current observation value.

[0013] Preferably, in step S200, the DC current differential equation of the MMC-HVDC system is established as follows:

[0014]

[0015] Among them, u dc i dc These represent DC voltage and current, respectively; L0 and R0 represent the bridge arm inductance and resistance, respectively; and j = a, b, c represents the three phases a, b, and c.

[0016] Preferably, in step S300, the DC current differential equation is discretized to obtain the discretized DC current differential equation:

[0017]

[0018] The Kalman filter state observation equation is established using the discretized DC current differential equation:

[0019]

[0020] Where k represents the MMC at time k, A, B, and C are system parameters, Y is the observed variable, H is the observed parameter, and w and v are the process noise and measurement noise of the observation system, respectively.

[0021] Preferably, step S400 includes the following steps:

[0022] S410. Calculate the predicted value of the DC current at time k using the observed value of the DC current at time k-1.

[0023] S420. Calculate the state error covariance at time k.

[0024] S430, Calculate the Kalman gain;

[0025] S440. The predicted value of DC current is corrected using Kalman gain to obtain the observed value of DC current at time k.

[0026] S450, correct the state error covariance at time k.

[0027] Preferably, in step S410, the observed value of the DC current at time k-1 is used. and the DC voltage u obtained at time k dc (k) and bridge arm voltage u pj (k), u nj (k) Calculate the predicted value of the DC current at time k. The calculation formula is:

[0028]

[0029] Preferably, in step S420, the formula for calculating the state prediction error covariance P(k|k-1) at time k is:

[0030] P(k|k-1)=A(k)P(k-1)A T (k)+Q(k);

[0031] Where P(k-1) is the optimal state error covariance at time k-1, and Q(k) is the covariance of the process noise w(k).

[0032] Preferably, in step S430, the Kalman gain K(k) is calculated using the following formula:

[0033]

[0034] Where R(k) is the covariance of the measurement noise v(k).

[0035] Preferably, in step S440, the observed value of the DC current at time k The calculation formula is:

[0036]

[0037] Among them, i dc (k) represents the actual measured value of the DC current.

[0038] Preferably, in step S450, the optimal state error covariance P(k) at time k is calculated:

[0039] P(k) = [1-K(k)]P(k|k-1).

[0040] Preferably, step S500 includes the following steps:

[0041] S510. Compare the DC current observation value obtained through the Kalman filter with its actual measurement value to obtain the observation residual;

[0042] S520. Set a deviation threshold and a time threshold, compare the observed residual with the deviation threshold, and determine that the DC voltage measurement is abnormal when the observed residual is greater than the deviation threshold and the duration exceeds the time threshold.

[0043] In step S520, the deviation threshold I thr The setting is to satisfy:

[0044]

[0045] Where max() is the function to find the maximum value, r1, r2, and r3 are the observation residuals of the MMC-HVDC system under normal operating condition, AC side fault condition, and DC side fault condition, respectively, and r noise The observation residuals are caused by system noise. This represents the minimum observation residual corresponding to an anomaly in DC voltage measurement.

[0046] Time threshold T thr The setting is to satisfy:

[0047] T d <T thr <T ref ;

[0048] Where T d T is the duration of the system disturbance. ref To control the processing time of the protection system for faults.

[0049] The method for detecting abnormal DC voltage measurements in an MMC-HVDC system according to the present invention employs a Kalman filter to observe the DC current and determines whether the DC voltage measurement is abnormal based on the observation results. This method has high reliability, can adapt to complex MMC-HVDC systems, achieves rapid detection of abnormal DC voltage measurements, and can identify the type of abnormality, which is beneficial for operators to confirm specific fault points. It also has the advantages of requiring no additional hardware resources, having simple detection logic, and being easy to integrate into the MMC-HVDC control and protection system, making it suitable for industrial applications. Attached Figure Description

[0050] Figure 1 This is a schematic diagram of the topology of the MMC-HVDC system;

[0051] Figure 2 This is a flowchart of the method for detecting abnormal DC voltage measurement in the MMC-HVDC system of the present invention;

[0052] Figure 3 This is a schematic diagram of the current deviation under abnormal operating conditions without DC voltage measurement in an embodiment of the present invention;

[0053] Figure 4 This is a schematic diagram of the current deviation after an abnormality occurs in the DC voltage measurement in an embodiment of the present invention. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this invention clearer, the following detailed description of the method for detecting abnormal DC voltage measurements in the MMC-HVDC system of this invention, in conjunction with the accompanying drawings and embodiments, provides further illustrative information. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0055] To facilitate understanding, a brief introduction to the MMC-HVDC system will be given first, such as... Figure 1 As shown, this MMC-HVDC system consists of a three-phase, six-arm bridge. Each arm has an identical structure, consisting of an arm inductor L0, an arm resistor R0, and N identical submodules (SM) cascaded together. The submodules employ a half-bridge structure, each containing two power switches S1 and S2, two diodes D1 and D2 connected in anti-parallel to the power switches, and a DC capacitor C. When S1 is turned on and S2 is turned off, the submodule is in the active state, and the output voltage is the capacitor voltage u. C When control S1 is off and S2 is on, the submodule is in a disconnected state, and the output voltage is 0. Under the action of MMC voltage equalization control, the capacitor voltages of each bridge arm submodule are approximately equal.

[0056] like Figure 2 As shown, a method for detecting abnormal DC voltage measurement in an MMC-HVDC system is characterized by:

[0057] The steps include: S100, acquiring the DC voltage, DC current, and upper and lower bridge arm voltages of the MMC-HVDC system;

[0058] S200. Establish the DC current differential equation of the MMC-HVDC system;

[0059] S300. After discretizing the DC current differential equation, it is expressed in the standard form of Kalman filtering.

[0060] S400: Use a Kalman filter to observe the state of DC current and obtain DC current observation values;

[0061] S500: Determine whether the DC voltage measurement is abnormal based on the DC current observation value.

[0062] It should be noted that in step S100, the DC voltage and DC current are obtained directly through a measuring device, and the MMC bridge arm voltage is calculated using the following formula:

[0063]

[0064] In the formula, the subscripts j = a, b, c represent the three phases a, b, c, and u pj u nj These represent the voltages of the upper and lower bridge arms, respectively, n pj n nj These represent the number of sub-modules deployed in the upper and lower bridge arms, respectively. These represent the average capacitor voltages of the upper and lower bridge arm submodules, respectively.

[0065] Specifically, in step S200, the DC current differential equation of the MMC-HVDC system is established as follows:

[0066]

[0067] Among them, u dc i dc These represent DC voltage and current, respectively; L0 and R0 represent the bridge arm inductance and resistance, respectively; and j = a, b, c represents the three phases a, b, and c.

[0068] Further, in step S300, the DC current differential equation is discretized to obtain the discretized DC current differential equation:

[0069]

[0070] The Kalman filter state observation equation is established using the discretized DC current differential equation:

[0071]

[0072] Where k represents the MMC at time k, A, B, and C are system parameters, Y is the observed variable, H is the observed parameter, and w and v are the process noise and measurement noise of the observation system, respectively. In actual operation, the process noise and measurement noise can be assumed to be Gaussian white noise.

[0073] As one possible implementation, step S400 includes the following steps:

[0074] S410. Calculate the predicted value of the DC current at time k using the observed value of the DC current at time k-1.

[0075] S420. Calculate the state error covariance at time k.

[0076] S430, Calculate the Kalman gain;

[0077] S440. The predicted value of DC current is corrected using Kalman gain to obtain the observed value of DC current at time k.

[0078] S450, correct the state error covariance at time k.

[0079] Specifically, in step S410, the observed value of the DC current at time k-1 is used. and the DC voltage u obtained at time k dc (k) and bridge arm voltage u pj (k), u nj (k) Calculate the predicted value of the DC current at time k. The calculation formula is:

[0080]

[0081] In step S420, the formula for calculating the state prediction error covariance P(k|k-1) at time k is:

[0082] P(k|k-1)=A(k)P(k-1)A T (k)+Q(k);

[0083] Where P(k-1) is the optimal state error covariance at time k-1, and Q(k) is the covariance of the process noise w(k).

[0084] In step S430, the Kalman gain K(k) is calculated using the following formula:

[0085]

[0086] Where R(k) is the covariance of the measurement noise v(k).

[0087] Preferably, in step S440, the observed value of the DC current at time k The calculation formula is:

[0088]

[0089] Among them, i dc (k) represents the actual measured value of the DC current.

[0090] Preferably, in step S450, the optimal state error covariance P(k) at time k is calculated:

[0091] P(k) = [1-K(k)]P(k|k-1).

[0092] As one possible implementation, step S500 includes the following steps:

[0093] S510. Compare the DC current observation value obtained through the Kalman filter with its actual measurement value to obtain the observation residual;

[0094] S520. Set a deviation threshold and a time threshold, compare the observed residual with the deviation threshold, and determine that the DC voltage measurement is abnormal when the observed residual is greater than the deviation threshold and the duration exceeds the time threshold. The setting of the time threshold can avoid false detection caused by actual system disturbances.

[0095] In step S520, the deviation threshold I thr The setting is to satisfy:

[0096]

[0097] Where max() is the function to find the maximum value, r1, r2, and r3 are the observation residuals of the MMC-HVDC system under normal operating condition, AC side fault condition, and DC side fault condition, respectively, and r noise The observation residuals are caused by system noise. This represents the minimum observation residual corresponding to an anomaly in DC voltage measurement.

[0098] Time threshold T thr The setting is to satisfy:

[0099] T d <T thr <T ref ;

[0100] Where T d T is the duration of the system disturbance. ref To control the processing time of the protection system for faults.

[0101] Specifically, in step S510, the DC current observation value obtained through the Kalman filter... Compared with its actual measured value i dc By comparing (k), the observation residual r(k) is obtained:

[0102]

[0103] In S520,

[0104] After detecting an abnormality in DC voltage measurement, the type of abnormality is further determined. If the observed value of DC current is greater than the actual value, the measurement abnormality is considered to be too high; if the observed value of DC current is less than the actual value, the measurement abnormality is considered to be too low.

[0105] To verify the effectiveness of the proposed DC voltage measurement anomaly detection method, a dual-terminal MMC-HVDC simulation model was built in Matlab / Simulink. The main parameters of the simulation model are shown in Table 1. The MMC employs a classic PI dual-closed-loop control strategy, and the detection range for DC voltage measurement anomalies is 2% above the normal value.

[0106] Table 1 Main parameters of the simulation model

[0107]

[0108] Figure 3 (a), (b), and (c) show the current deviation waveforms of the MMC-HVDC system under normal operation, AC-side ground fault, and DC-side short-circuit fault conditions, respectively, with no abnormal DC voltage measurement. Both the AC-side ground fault and the DC-side short-circuit fault occur within 2 seconds. Figure 3 It can be seen that the current deviation of the system under normal operation and AC side ground fault conditions is close to zero. Although the DC side short circuit fault causes a large current deviation, it will not lead to false detection because the current deviation does not reach the deviation threshold. Complex system operating conditions will not affect the accuracy of fault detection. Figure 4 (a) and (b) show the current deviation waveforms when the MMC-HVDC system experiences abnormally high and low DC voltage measurements, respectively, during normal operation. The figures show Δu... dc This indicates the measurement deviation of the DC voltage. (From...) Figure 4 As can be seen, the current deviation exceeds the deviation threshold, and the DC voltage measurement anomaly will be detected by this invention. Furthermore, the sign of the current deviation indicates whether the measurement is too high or too low. The simulation results fully verify the effectiveness and reliability of the DC voltage measurement anomaly detection method proposed in this invention.

[0109] The method proposed in this invention uses a Kalman filter to observe the DC current state of an MMC-HVDC system. Since abnormal DC voltage measurements will lead to large observation residuals, a suitable deviation threshold is set. The relationship between the observation residual and the deviation threshold is used to achieve rapid detection of DC voltage measurement anomalies. The method proposed in this invention has the advantages of high reliability, simple detection logic, and short processing time. It can also adapt to complex MMC-HVDC systems and is easy to apply in industrial settings.

[0110] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for detecting abnormal DC voltage measurement in an MMC-HVDC system, characterized in that: The steps include: S100, acquiring the DC voltage, DC current, and upper and lower bridge arm voltages of the MMC-HVDC system; S200. Establish the DC current differential equation of the MMC-HVDC system; S300. After discretizing the DC current differential equation, it is expressed in the standard form of Kalman filtering. S400: Use a Kalman filter to observe the state of DC current and obtain DC current observation values; S500: Determine whether the DC voltage measurement is abnormal based on the DC current observation value.

2. The method for detecting abnormal DC voltage measurement in an MMC-HVDC system according to claim 1, characterized in that: In step S200, the DC current differential equation of the MMC-HVDC system is established as follows: ； in, , These represent DC voltage and current, respectively. , These represent the inductance and resistance of the bridge arms, respectively. express , , Three phases; in, , These represent the voltages of the upper and lower bridge arms, respectively.

3. The method for detecting abnormal DC voltage measurement in an MMC-HVDC system according to claim 2, characterized in that: In step S300, the DC current differential equation is discretized to obtain the discretized DC current differential equation: ； The Kalman filter state observation equation is established using the discretized DC current differential equation: ； in, This indicates that MMC is in time, , , For system parameters, For observed variables, For observation parameters, , These are the process noise and measurement noise of the observation system, respectively.

4. The method for detecting abnormal DC voltage measurement in an MMC-HVDC system according to claim 3, characterized in that: Step S400 includes the following steps: S410, Utilization Calculation of DC current observations at time t. The predicted value of the DC current at a given time; S420, Calculation State error covariance at time t; S430, Calculate the Kalman gain; S440. The predicted value of DC current is corrected using Kalman gain to obtain... The observed value of the DC current at any given time; S450, correction The state error covariance at time t.

5. The method for detecting abnormal DC voltage measurement in an MMC-HVDC system according to claim 4, characterized in that: In step S410, using Observed value of DC current at time as well as DC voltage obtained at any time and bridge arm voltage , calculate Predicted value of DC current at any time The calculation formula is: 。 6. The method for detecting abnormal DC voltage measurement in an MMC-HVDC system according to claim 5, characterized in that: In step S420, State prediction error covariance at time 1 The calculation formula is: ； in, for The optimal state error covariance at time t. For process noise The covariance.

7. The method for detecting abnormal DC voltage measurement in an MMC-HVDC system according to claim 6, characterized in that: In step S430, the Kalman gain The calculation formula is: ； in, To measure noise The covariance.

8. The method for detecting abnormal DC voltage measurement in an MMC-HVDC system according to claim 7, characterized in that: In step S440, Observed value of DC current at time The calculation formula is: ； in, This is the actual measured value of the DC current.

9. The method for detecting abnormal DC voltage measurement in an MMC-HVDC system according to claim 8, characterized in that: In step S450, calculate optimal state error covariance at time 1 : 。 10. The method for detecting abnormal DC voltage measurement in an MMC-HVDC system according to any one of claims 1 to 9, characterized in that: Step S500 includes the following steps: S510. Compare the DC current observation value obtained through the Kalman filter with its actual measurement value to obtain the observation residual; S520. Set a deviation threshold and a time threshold, compare the observed residual with the deviation threshold, and determine that the DC voltage measurement is abnormal when the observed residual is greater than the deviation threshold and the duration exceeds the time threshold. In step S520, the deviation threshold The setting is to satisfy: ； in It is a function for finding the maximum value. , , These are the observation residuals of the MMC-HVDC system under normal operating conditions, AC side fault conditions, and DC side fault conditions, respectively. The observation residuals are caused by system noise. This represents the minimum observation residual corresponding to an anomaly in DC voltage measurement. Time threshold The setting is to satisfy: ； in The duration of the system disturbance. To control the processing time of the protection system for faults.

Images