Flexible DC converter valve temperature prediction method based on voltage centralized measurement

Abstract

The invention provides a voltage centralized measurement-based temperature prediction method for a flexible DC converter valve. The method comprises the following steps of collecting a bridge arm AC side voltage and a sub-module capacitor voltage through a master-slave sensor structure; classifying and correcting the voltage data based on nine types of sampling moments; solving the total loss of one cycle according to the obtained voltage data; constructing a thermal network model, and calculating thermal resistance; and calculating the junction temperature of the IGBT according to the calculated total loss and thermal resistance of one period.

Description

Technical Field

[0001] The present invention relates to the technical field of flexible direct current power transmission, and in particular to a method for predicting temperature of a flexible direct current converter valve based on centralized voltage measurement. Background Art

[0002] Flexible direct current transmission technology is widely used in renewable energy grid connection and grid interconnection due to its advantages of strong controllability and fast response speed, but the junction temperature monitoring of flexible direct current converter valves is still a key challenge for system reliability. Traditional methods rely on a large number of temperature sensors or complex thermal simulation models, which have problems such as high hardware cost, sensor susceptibility to interference and failure, difficulty in data synchronization, and large errors in temperature rise estimation under dynamic conditions. Although existing improvement schemes try to reduce the number of sensors or simplify the model, they fail to accurately correct the dynamic changes of sub-module capacitor voltage, ignore the details of transient loss of switches, or use fixed thermal resistance parameters, resulting in insufficient temperature prediction accuracy. Therefore, a temperature prediction method that takes into account low cost, high reliability and dynamic adaptability is needed to achieve accurate mapping of loss and temperature rise through voltage data drive. Summary of the invention

[0003] In order to solve the above technical problems, the present invention provides a method for predicting the temperature of a flexible DC converter valve based on centralized voltage measurement, which can realize the prediction of the temperature of the flexible DC converter valve.

[0004] The technical solution adopted by the present invention is as follows:

[0005] A method for predicting temperature of a flexible direct current converter valve based on centralized voltage measurement comprises the following steps: collecting bridge arm AC side voltage and submodule capacitor voltage through a master-slave sensor structure; classifying and correcting voltage data based on nine types of sampling moments; solving the total loss of one cycle according to the obtained voltage data; constructing a thermal network model and calculating thermal resistance; and calculating the junction temperature of the IGBT according to the calculated total loss of one cycle and thermal resistance.

[0006] The u m_au is the AC side output voltage of the upper bridge arm of phase A measured by the main sensor, u m_auFs is the DC capacitance voltage value of an FBSM measured by the slave sensor.

[0007] The specific steps of classifying and correcting voltage data based on nine types of sampling moments are as follows:

[0008] At the first sampling moment, the output voltage of the bridge arm AC side is at a level. Since the output voltage of the bridge arm AC side is generated by only one submodule that is put into operation, the capacitor voltage u of the submodule is ci(k) is accurate, that is, equal to the measured voltage u of the main sensor m_au(k) , other submodules are kept bypassed, so their capacitor voltage u cj(k) is unchanging.

[0009] At the second sampling moment, the submodule measured by the slave sensor remains switched on, and the other submodule has just been switched on. The AC side output voltages of the bridge arm measured by the master sensor at sampling moments k-1 and k are u m_au (k-1) and u m_au (k). Change in voltage on the AC side of the bridge arm Δu m (k) is:

[0010] Δu m (k) = u m_au (k)-u m_au (k-1) (1)

[0011] Δu m (k) contains the last submodule voltage u ci (k) and the sum of the voltage changes of the submodule capacitors that are kept in operation ΣS j (k)Δu cj (k), maintain the voltage change of the submodule capacitor Δu cj (k) is considered to be equal to the capacitance voltage change Δu measured from the sensor cs (k):

[0012] Δu cj (k) = Δu cs (k) = u cs (k)-u cs (k),j≠i,s (2)

[0013] u cs (k) and u cs (k-1) is the submodule capacitor voltage measured by the slave sensor at sampling time k and k-1. The submodule capacitor voltage u without changing the operating state can be calculated from equation (2) and the following equation: cj (k).

[0014] u cj (k) = u cj (k-1)+S j (k)Δu cj (k),j≠i,s (3)

[0015] The final submodule capacitor voltage u ci (k) can be calculated using the following formula:

[0016] u ci (k) = Δu m (k)-∑S j (k)Δu cj (k),j≠i,s (4)

[0017] At the third sampling moment, the submodule measured by the sensor remains switched on, another submodule is just bypassed, and the submodule capacitor voltage u is unchanged in the operating state. cj( k) can be calculated by equations (2) and (3). The voltage of the submodule that has just been bypassed at sampling time k is:

[0018] u ci (k) = u ci (k-1)+Δu eci (k) (5)

[0019] At the fourth sampling moment, the submodule measured by the slave sensor has just been put into operation, and the capacitance voltage change of other submodules that remain in operation can be calculated by the following formula:

[0020] Δu cj (k)=(Δu m (k)-u cs (k)) / ∑S j (k),j≠i,s (6)

[0021] Capacitor voltage u cj (k) can be calculated by equation (3) and equation (6). All submodule capacitor voltages are relatively accurate because the submodule capacitor voltages with switching state changes are directly measured from the sensor, i.e. u ci (k) is equal to u cs (k).

[0022] At the fifth sampling moment, the submodule measured by the slave sensor has just been bypassed, and the capacitance voltage change of other submodules that remain switched on can be calculated by the following formula:

[0023] Δu cj (k)=(Δu m (k)+u cs (k)) / ∑S j (k),j≠i,s (7)

[0024] Capacitor voltage u cj (k) can be calculated by equation (3) and equation (7). All submodule capacitor voltages are relatively accurate because the submodule capacitor voltages with switching state changes are directly measured from the sensor, i.e. u ci (k) is equal to u cs (k).

[0025] At the sixth sampling moment, the submodule measured by the slave sensor remains bypassed, and another submodule has just been put into operation. The capacitance voltage change Δu of the other submodules that remain in operation eci(k) can be calculated by the following formula. Their capacitance voltage u cj (k) can be calculated by the following formula and formula (3):

[0026]

[0027] The capacitor voltage u of the submodule just put into operation ci (k) can be calculated by the following formula:

[0028] u ci (k) = Δu m (k)-∑S j (k)Δu ecj (k),j≠i,s (9)

[0029] At the seventh sampling moment, the submodule measured by the sensor remains bypassed, and another submodule has just been bypassed. Similarly, the capacitance voltage change of the other submodules that remain put into operation is Δu eci (k) can be calculated by equation (8). Their capacitor voltage u cj (k) can be calculated by equation (3) and equation (8).

[0030] The voltage u of the submodule capacitor just bypassed ci (k) can be calculated by the following formula and formula (5):

[0031]

[0032] At the eighth sampling moment, the submodule measured by the slave sensor remains switched on, and another submodule has just been switched on. ci (k) can be calculated by the following formula:

[0033] u ci (k) = Δu m (k)-u cs (k) (11)

[0034] At the ninth sampling moment, the submodule measured by the slave sensor remains switched on, and another submodule is just bypassed. The capacitor voltage u of the other submodule that remains switched on cj (k) can be calculated by the following formula:

[0035] u cj (k) = Δu m (k)-u cs (k) (12)

[0036] The capacitor voltage of the submodule that has just been bypassed can be calculated using equations (5) and (10).

[0037] The total loss for one cycle is calculated by the following formula:

[0038]

[0039] P all =P T +P on +P off (16)

[0040] Where P T ,P on ,P off They are the conduction loss, turn-on loss and turn-off loss of one cycle of IGBT respectively; P all is the total loss for one cycle.

[0041] The thermal resistance is calculated by the following formula:

[0042]

[0043] In the formula, R i is the thermal resistance value of the i-th order in the thermal network model; t is the operating time, τ i is the time constant.

[0044] The junction temperature of the IGBT is calculated by the following formula:

[0045] T j =P all ·Z j-c +T c (18)

[0046] In the formula, Z j-c is the thermal impedance from junction to case.

[0047] Beneficial effects of the present invention:

[0048] The present invention can predict the temperature of the flexible direct current converter valve by centrally measuring the temperature of the flexible direct current converter valve. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a flow chart of a method for predicting temperature of a flexible DC converter valve based on centralized voltage measurement according to an embodiment of the present invention;

[0050] Figure 2 This is a sensor configuration diagram of a hybrid flexible direct current converter valve and MS-MT based on HBSM and FBSM according to an embodiment of the present invention;

[0051] Figure 3 This is a flowchart of the operation of MS-MT according to an embodiment of the present invention;

[0052] Figure 4This is a thermal network model of an IGBT according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] like Figure 1 As shown, the method for predicting temperature of a flexible direct current converter valve based on centralized voltage measurement according to an embodiment of the present invention comprises the following steps:

[0055] S1, collects the bridge arm AC side voltage and submodule capacitor voltage through the master-slave sensor structure.

[0056] In one embodiment of the present invention, u m_au is the AC side output voltage of the upper bridge arm of phase A measured by the main sensor, u m_auFs is the DC capacitance voltage value of an FBSM measured by the slave sensor.

[0057] S2, classify and correct the voltage data based on nine types of sampling moments.

[0058] At the first sampling moment, the output voltage of the bridge arm AC side is at a level. Since the output voltage of the bridge arm AC side is generated by only one submodule that is put into operation, the capacitor voltage u of the submodule ci(k) is accurate, that is, equal to the measured voltage u of the main sensor m_au(k) , other submodules are kept bypassed, so their capacitor voltage u cj(k) is unchanging.

[0059] At the second sampling moment, the submodule measured by the slave sensor remains switched on, and the other submodule has just been switched on. The output voltages of the bridge arm AC side measured by the master sensor at sampling moments k-1 and k are u m_au (k-1) and u m_au (k). Change in voltage on the AC side of the bridge arm Δu m (k) is:

[0060] Δu m (k) = u m_au (k)-u m_au (k-1) (1)

[0061] Δu m (k) contains the last submodule voltage u ci (k) and the sum of the voltage changes of the submodule capacitors that are kept in operation ΣSj (k)Δu cj (k), maintain the voltage change of the submodule capacitor Δu cj (k) is considered to be equal to the capacitance voltage change Δu measured from the sensor cs (k):

[0062] Δu cj (k) = Δu cs (k) = u cs (k)-u cs (k),j≠i,s (2)

[0063] u cs (k) and u cs (k-1) is the submodule capacitor voltage measured by the slave sensor at sampling time k and k-1. The submodule capacitor voltage u without changing the operating state can be calculated from equation (2) and the following equation: cj (k).

[0064] u cj (k) = u cj (k-1)+S j (k)Δu cj (k),j≠i,s (3)

[0065] The final submodule capacitor voltage u ci (k) can be calculated using the following formula:

[0066] u ci (k) = Δu m (k)-∑S j (k)Δu cj (k),j≠i,s (4)

[0067] The third type of sampling moment, the submodule measured by the sensor remains switched on, the other submodule has just been bypassed, and the submodule capacitor voltage u of the operating state has not changed cj( k) can be calculated by equations (2) and (3). The voltage of the submodule that has just been bypassed at sampling time k is:

[0068] u ci (k) = u ci (k-1)+Δu eci (k) (5)

[0069] At the fourth sampling moment, the submodule measured by the slave sensor has just been put into operation, and the capacitance and voltage changes of other submodules that remain in operation can be calculated by the following formula:

[0070] Δu cj (k)=(Δu m(k)-u cs (k)) / ∑S j (k),j≠i,s (6)

[0071] Capacitor voltage u cj (k) can be calculated by equation (3) and equation (6). All submodule capacitor voltages are relatively accurate because the submodule capacitor voltages with switching state changes are directly measured from the sensor, i.e. u ci (k) is equal to u cs (k).

[0072] At the fifth sampling moment, the submodule measured by the slave sensor has just been bypassed, and the capacitance voltage change of other submodules that remain switched on can be calculated by the following formula:

[0073] Δu cj (k)=(Δu m (k)+u cs (k)) / ∑S j (k),j≠i,s (7)

[0074] Capacitor voltage u cj (k) can be calculated by equation (3) and equation (7). All submodule capacitor voltages are relatively accurate because the submodule capacitor voltages with switching state changes are directly measured from the sensor, i.e. u ci (k) is equal to u cs (k).

[0075] At the sixth sampling moment, the submodule measured by the slave sensor remains bypassed, and another submodule has just been put into operation. The capacitance voltage change of the other submodules that remain in operation is Δu eci (k) can be calculated by the following formula. Their capacitance voltage u cj (k) can be calculated by the following formula and formula (3):

[0076]

[0077] The capacitor voltage u of the submodule just put into operation ci (k) can be calculated by the following formula:

[0078] u ci (k) = Δu m (k)-∑S j (k)Δu ecj (k),j≠i,s (9)

[0079] At the seventh sampling moment, the submodule measured by the slave sensor remains bypassed, and another submodule has just been bypassed. Similarly, the capacitance voltage change of the other submodules that remain switched on is Δu eci(k) can be calculated by equation (8). Their capacitor voltage u cj (k) can be calculated by equation (3) and equation (8).

[0080] The voltage u of the submodule capacitor just bypassed ci (k) can be calculated by the following formula and formula (5):

[0081]

[0082] At the eighth sampling moment, the submodule measured by the slave sensor remains switched on, and another submodule has just been switched on. The capacitor voltage u of the submodule just switched on ci (k) can be calculated by the following formula:

[0083] u ci (k) = Δu m (k)-u cs (k) (11)

[0084] At the ninth sampling moment, the submodule measured by the slave sensor remains switched on, and another submodule has just been bypassed. The capacitor voltage u of the other submodule that remains switched on cj (k) can be calculated by the following formula:

[0085] u cj (k) = Δu m (k)-u cs (k) (12)

[0086] The capacitor voltage of the submodule that has just been bypassed can be calculated by equations (5) and (10). The sensor configuration diagram of the hybrid flexible direct current converter valve and MS-MT based on HBSM and FBSM is shown in Figure 2 As shown; the operation flow chart of MS-MT is as follows Figure 3 shown.

[0087] S3, solve the total loss for one cycle based on the obtained voltage data.

[0088] In one embodiment of the present invention, the total loss in one cycle is calculated by the following formula:

[0089]

[0090] P all =P T +P on +P off (16)

[0091] Where P T ,P on ,P offThey are the conduction loss, turn-on loss and turn-off loss of one cycle of IGBT respectively; P all is the total loss for one cycle.

[0092] S4, build a thermal network model and calculate thermal resistance.

[0093] In one embodiment of the present invention, the thermal resistance is calculated by the following formula:

[0094]

[0095] In the formula, R i is the thermal resistance value of the i-th order in the thermal network model; t is the operating time, τ i is the time constant. The thermal network model of IGBT is as follows Figure 4 shown.

[0096] S5, calculating the junction temperature of the IGBT based on the calculated total loss in one cycle and the thermal resistance.

[0097] In one embodiment of the present invention, the junction temperature of the IGBT is calculated by the following formula:

[0098] T j =P all ·Z j-c +T c (18)

[0099] In the formula, Z j-c is the thermal impedance from junction to case.

[0100] According to the above-mentioned embodiment of the present invention, the flexible direct current converter valve temperature prediction method based on centralized voltage measurement realizes the centralized measurement of the bridge arm AC side voltage and the sub-module capacitor voltage through master-slave sensors, and combines the dynamic classification correction mechanism of nine types of sampling moments, and establishes a complete mapping relationship from electrical loss to thermal resistance conduction through multi-dimensional loss model and dynamic thermal network coupling calculation, and finally realizes the online real-time prediction of IGBT junction temperature, which significantly improves the reliability and dynamic response capability of flexible direct current converter valve temperature monitoring.

Claims

1. A method for predicting temperature of a flexible direct current converter valve based on centralized voltage measurement, characterized in that: The following steps are involved: The bridge arm AC side voltage and submodule capacitor voltage are collected through the master-slave sensor structure; Classify and correct voltage data based on nine types of sampling moments; Solve the total loss of one cycle based on the obtained voltage data; Build a thermal network model and calculate thermal resistance; The junction temperature of the IGBT is calculated based on the calculated total loss in one cycle and thermal resistance.

2. The method for predicting temperature of flexible direct current converter valve based on centralized voltage measurement according to claim 1, characterized in that: The u m_au is the AC side output voltage of the upper bridge arm of phase A measured by the main sensor, u m_auFs is the DC capacitance voltage value of an FBSM measured by the slave sensor.

3. The method for predicting temperature of flexible DC converter valve based on centralized voltage measurement according to claim 2 is characterized in that: The specific steps of classifying and correcting voltage data based on nine types of sampling moments are as follows: At the first sampling moment, the output voltage of the bridge arm AC side is at a level. Since the output voltage of the bridge arm AC side is generated by only one submodule that is put into operation, the capacitor voltage uci(k) of the submodule is accurate, that is, equal to the measured voltage um_au(k) of the main sensor. The other submodules are kept in bypass, so their capacitor voltages ucj(k) are unchanged. At the second sampling moment, the submodule measured by the slave sensor remains switched on, and the other submodule has just been switched on. The AC side output voltages of the bridge arm measured by the master sensor at sampling moments k-1 and k are u m_au (k-1) and u m_au (k). Change in voltage on the AC side of the bridge arm Δu m (k) is: （1） Δu m (k) contains the last submodule voltage u ci (k) and the sum of the voltage changes of the submodule capacitors that are kept in operation ΣS j (k) Δu cj (k), maintain the voltage change of the submodule capacitor Δu cj (k) is considered to be equal to the capacitance voltage change Δu measured from the sensor cs (k): （2） u cs (k) and u cs (k-1) is the submodule capacitor voltage measured by the slave sensor at sampling time k and k-1. The submodule capacitor voltage u without changing the operating state can be calculated from equation (2) and the following equation: cj (k): （3） The final submodule capacitor voltage u ci (k) can be calculated using the following formula: （4） At the third sampling moment, the submodule measured by the sensor remains switched on, another submodule is just bypassed, and the submodule capacitor voltage u is unchanged in the operating state. cj( k) can be calculated by equations (2) and (3). The voltage of the submodule that has just been bypassed at sampling time k is: （5） At the fourth sampling moment, the submodule measured by the slave sensor has just been put into operation, and the capacitance voltage change of other submodules that remain in operation can be calculated by the following formula: （6） Capacitor voltage u cj (k) can be calculated by equations (3) and (6). All submodule capacitor voltages are relatively accurate because the submodule capacitor voltages with switching state changes are directly measured from the sensor, i.e., u ci (k) is equal to u cs (k) At the fifth sampling moment, the submodule measured by the slave sensor has just been bypassed, and the capacitance voltage change of other submodules that remain switched on can be calculated by the following formula: （7） Capacitor voltage u cj (k) can be calculated by equation (3) and equation (7). All submodule capacitor voltages are relatively accurate because the submodule capacitor voltages with switching state changes are directly measured from the sensor, i.e., u ci (k) is equal to u cs (k) At the sixth sampling moment, the submodule measured by the slave sensor remains bypassed, another submodule has just been put into operation, and the capacitance voltage change of the other submodules that remain in operation is Δu eci (k) can be calculated by the following formula: their capacitance voltage u cj (k) can be calculated by the following formula and formula (3): （8） The capacitor voltage u of the submodule just put into operation ci (k) can be calculated by the following formula: （9） At the seventh sampling moment, the submodule measured by the sensor remains bypassed, and another submodule has just been bypassed. Similarly, the capacitance voltage change Δu of the other submodules that remain put into operation eci (k) can be calculated by formula (8), their capacitor voltage u cj (k) can be calculated by equation (3) and equation (8), The voltage u of the submodule capacitor just bypassed ci (k) can be calculated by the following formula and formula (5): （10） At the eighth sampling moment, the submodule measured by the sensor remains in operation, and another submodule has just been in operation. The capacitance voltage u of the submodule just in operation is ci (k) can be calculated by the following formula: （11） At the ninth sampling moment, the submodule measured by the slave sensor remains switched on, another submodule is just bypassed, and the capacitor voltage u of another submodule remains switched on cj (k) can be calculated by the following formula: （12） The capacitor voltage of the submodule that has just been bypassed can be calculated using equations (5) and (10).

4. The method for predicting temperature of flexible DC converter valve based on centralized voltage measurement according to claim 3 is characterized in that: The total loss for one cycle is calculated by the following formula: （13） （14） （15） （16） Where P T ,P on ,P off They are the conduction loss, turn-on loss and turn-off loss of one cycle of IGBT respectively; P all is the total loss for one cycle.

5. The method for predicting temperature of flexible direct current converter valve based on centralized voltage measurement according to claim 4 is characterized in that: The thermal resistance is calculated by the following formula: （17） In the formula, R i is the thermal resistance value of the i-th order in the thermal network model; t is the operating time, τ i is the time constant.

6. The method for predicting temperature of flexible DC converter valve based on centralized voltage measurement according to claim 5, characterized in that: The junction temperature of the IGBT is calculated by the following formula: （18） In the formula, Z j-c is the thermal impedance from junction to case.

Images