Method for optimizing reliability of torque control system of switched reluctance motor

Abstract

The invention discloses a method for optimizing the reliability of a torque control system of a switched reluctance motor, which comprises the following steps of: firstly, dividing an original driving system into three parts, namely a motor body, a detection unit and a power converter, respectively analyzing the types of faults which can be generated by the three parts, and carrying out dynamic quantitative analysis on the running state of the motor; motor states under different faults are evaluated, and a rotating speed fluctuation fault criterion and a current over-limit fault criterion which are suitable for a torque control strategy are obtained; secondly, device junction temperatures at different positions on the power converter are quickly and accurately obtained through a cauer thermal model, so that the failure rate of the device can be calculated; then, in order to simplify the algorithm and improve the accuracy of the system, an original driving system is divided into three parts, namely a motor body, a detection unit and a power converter, a k-out-of-n: G model is adopted to perform reliability analysis on the three parts, and then the three parts are combined with an RBD model to calculate the reliability of the whole system; and finally, an online optimization strategy is provided, and after a fault occurs, the running state of the motor is improved and the reliability of the system is improved by adjusting running parameters such as a turn-on angle and a turn-off angle.

Description

Technical Field

[0003] This invention proposes a method for optimizing the reliability of a switched reluctance motor torque control system. This method takes into account the different thermal stresses of the power converter and the strong fault tolerance of the motor itself, enabling a more accurate assessment of the reliability of the drive system. Furthermore, it further improves the reliability of the switched reluctance motor drive system through online real-time optimization control strategies, facilitating the timely maintenance of the drive system. Background Technology

[0005] To achieve sustainable development of resources and the environment, countries are increasingly emphasizing the research and utilization of new energy sources such as solar, wind, and tidal energy. Among these, wind energy, due to its abundant reserves and clean, renewable nature, has seen rapid technological development, leading to the widespread application of wind turbines. Switched reluctance motors, with their robust structure, high fault tolerance, and low manufacturing cost, have been widely used in aerospace systems, electric vehicles, and wind power generation, among other fields.

[0006] In switched reluctance generator control systems, the asymmetrical half-bridge power converter (AHBPC) is widely used due to its high reliability and flexible control methods. Because each phase of the switched reluctance motor is independently controlled when driven by the AHBPC, they do not affect each other. Therefore, even if one phase fails, the other phases can still be stably controlled. Furthermore, it can be simplified or modified according to specific application conditions and actual needs to achieve the required control strategy. Therefore, the asymmetrical half-bridge power converter is widely used to drive switched reluctance motors.

[0007] Reliability assessment is crucial for ensuring the high performance and safe operation of switched reluctance motor control systems. Commonly used assessment models include Reliability Block Diagram (RBD), Fault Tree Model, k-out-of-n:G model, and Markov model. When assessing more complex systems, k-out-of-n:G model and Markov model are typically used. However, when using the Markov model to calculate system reliability, the formulas are lengthy and the calculation process is complex, and computational errors are easily introduced. Therefore, this approach uses the k-out-of-n:G model for reliability analysis, and finally combines it with the RBD model to calculate the overall system reliability.

[0008] When a switched reluctance motor adopts a torque control strategy, its operating state differs from that of traditional current chopping control and angle position control strategies. Therefore, the fault criteria used for switched reluctance motor drive systems employing torque control strategies will also differ. The operating state of the motor will change after a fault occurs at a certain point. Thus, appropriate fault criteria are proposed for the torque control strategy. In addition, to improve the reliability of the drive system, an online method for improving system reliability is adopted. By adjusting the system's operating parameters online under fault conditions, the operating state of the motor after a fault can be effectively improved, thereby enhancing the system's reliability. Summary of the Invention

[0010] In view of the shortcomings or improvement needs of the existing technology, this invention proposes a method for optimizing the reliability of the torque control system of a switched reluctance motor.

[0011] This invention first employs a three-phase asymmetrical half-bridge power converter. Using this power converter can effectively improve the power generation efficiency of the switched reluctance generator, and when one phase fails, the other phases of the motor can still be stably controlled, thus improving the reliability of the system.

[0012] Secondly, the original drive system is divided into three parts: the motor body, the detection unit, and the power converter. The types of faults that may occur in each of these three parts are analyzed, the motor's operating status is dynamically and quantitatively analyzed, the motor status under different faults is evaluated, and fault criteria suitable for torque control strategies are obtained.

[0013] Then, this invention proposes to divide the original drive system into three parts: the motor body, the detection unit, and the power converter, and to perform reliability analysis on each part using the k-out-of-n:G model. Finally, the reliability of the entire system is calculated by combining these three parts with the RBD model.

[0014] Finally, in order to improve the reliability of the drive system, this invention proposes a method for improving system reliability online. By adjusting the system's operating parameters online under fault conditions, the operating state of the motor after a fault can be effectively improved, thereby enhancing the system's reliability. Attached Figure Description

[0016] Figure 1 This is a diagram showing the operating mode of the switched reluctance generator drive system of the present invention.

[0017] Figure 2 This is the current path diagram after a fault occurs in the asymmetric half-bridge power converter of the present invention.

[0018] Figure 3 Yes, it's a package diagram. Figure 4This is a Cauer thermal model diagram of the power converter of the present invention.

[0019] Figure 5 It is the RBD reliability model for switched reluctance motors.

[0020] Figure 6 This is the SRM ontology k-out-of-n:G model of the present invention. Figure 7 This is the detection unit k-out-of-n:G model of the present invention. Figure 8 This is the improved k-out-of-n:G model of the power converter in this invention.

[0021] Figure 9 This is the k-out-of-n:G model diagram of the present invention.

[0022] Figure 10 This is a flowchart of the online system reliability improvement process of the present invention.

[0023] Figure 11 This is the k-out-of-n:G model of the power converter after online optimization according to the present invention. Detailed Implementation

[0025] The following description, in conjunction with the accompanying drawings, provides further examples of the invention.

[0026] The operating mode of the switched reluctance generator drive system of the present invention is as follows: Figure 1 As shown, due to the repetitive nature of the control methods, for the sake of simplicity, only phase A will be used as an example here.

[0027] The voltage equation under excitation mode is shown below:

[0028]

[0029] In the formula U d R is the bus voltage, R is the motor internal resistance, and i a and L a These are the phase current and phase inductance of the motor, respectively, and ω is the angular velocity of the motor. At this time, the phase current of phase A increases rapidly.

[0030] The voltage equation for the 0-voltage freewheeling mode is:

[0031]

[0032] At this point, the current in phase A gradually decreases.

[0033] The voltage equation for the negative voltage demagnetization mode is:

[0034]

[0035] At this time, the current of phase A is also decreasing, but the rate of decrease is greater than that of the 0-voltage freewheeling mode.

[0036] When an upper MOSFET-open circuit (UMOC) occurs in phase A, the current path in the original excitation mode will suddenly change as follows: Figure 2 (a) shows the current path where the A-phase current changes from a rapid increase to a gradual decrease.

[0037] When an upper MOSFET short circuit (UMSC) occurs in phase A, the current path in the original zero-voltage freewheeling mode will become as follows: Figure 2 (b) Current path, at which point the phase A current changes from a gradual decrease to a rapid increase.

[0038] When a lower MOSFET-open circuit (LMOC) occurs in phase A, the current path in the original excitation mode will suddenly change as follows: Figure 2 (c) The current path under which the phase A current in the excitation mode changes from a rapid increase to a gradual decrease.

[0039] When a lower MOSFET-Short Circuit (LMSC) occurs in phase A, the current path in the original negative voltage demagnetization mode will become as follows: Figure 2 (d) The current path under this condition will change the rapid decrease of the A-phase current in the negative voltage demagnetization mode to a gradual decrease. The power converter section may also experience upper diode-open circuit (UDOC), upper diode-short circuit (UDSC), lower diode-open circuit (UDOC), lower diode-short circuit (UDSC), capacitor-open circuit (COC), and capacitor-short circuit (CSC) faults.

[0040] The main detection units of the motor are current sensors and position sensors. Therefore, the faults of these two parts can be divided into the following five types: Position Sensor Zero Output (PSZO), Position Sensor Constant Output (PSSCO), Current Sensor Zero Output (CSZO), Current Sensor Constant Output (CSCO), and Current Sensor Gain Output (CSGO).

[0041] Motor body faults are categorized into winding open circuit faults (WOC) and winding short circuit faults (WSC). Therefore, based on the above analysis, there are a total of 14 fault types in the switched reluctance motor drive system. Generally, any fault in the motor is considered a failure, but due to the strong fault tolerance of switched reluctance motors, they can still operate stably after some faults. Therefore, this invention performs dynamic quantitative analysis of the motor's operating state and evaluates the motor's state under different fault conditions.

[0042] Torque stability is one of the important operating indicators of electric vehicles. Furthermore, according to the mechanical motion equations, we can see that the motor torque is also closely related to torque. The mechanical motion equations are shown below:

[0043]

[0044] In the formula, J and D are the moment of inertia and coefficient of friction of the motor, respectively, and T L This represents the load torque of the motor.

[0045] Therefore, this invention provides a quantitative analysis of torque fluctuations in a motor after different faults occur, following the adoption of a torque strategy.

[0046] This invention quantitatively analyzes the torque fluctuations of a motor after different faults occur following the adoption of a torque strategy. The resulting fault criterion for torque fluctuation is as follows: when the torque fluctuation exceeds 5%, the system is considered to have failed; when the torque fluctuation is below 5%, the system is considered to be operating reliably. Simultaneously, the maximum current that components can withstand during motor operation is limited. Excessive operating current can easily lead to component breakdown. Therefore, the fault criterion for phase current is: when the operating current exceeds 30A, the system is in a failed state; when the operating current is below 30A, the system is in a safe operating state.

[0047] The failure rate of a device can be obtained by the following formula:

[0048]

[0049] In the formula λ b π represents the fundamental failure rate of the device. i The stress coefficient of the device (usually including the applied coefficient π) A quality coefficient π Q Environmental coefficient π E and temperature coefficient π T PF represents the probability of device i failing (short circuit / open circuit).

[0050] Therefore, the failure rate of MOSFETs can be expressed as follows, as shown in the table:

[0051]

[0052] In the formula T j It is the junction temperature of the device.

[0053] The failure rate of a diode can be expressed as:

[0054]

[0055] Equations (7) and (8) show that the failure rate of power electronic devices MOSFETs and diodes is closely related to their junction temperature. Therefore, it is particularly important to obtain the junction temperature of these two devices quickly and accurately. Taking a MOSFET as an example, the package diagram of this device is shown below. Figure 3 As shown, each power device consists of a MOSFET chip layer, an equivalent bonding layer, a substrate, a thermal pad, and a heatsink. Existing fast junction temperature calculation models typically use the Cauer model. This model primarily uses RC circuit models of different orders to equivalently fit the thermal path model of the MOSFET or diode. Therefore, this chapter also uses the Cauer model to fit the thermal path model of the power device, as shown below. Figure 4 As shown.

[0056] Here, Z jc For RC fitting of MOSFET devices, Z-axis is used for RC fitting of thermal conductive sheets. ch The thermal resistance of the heat sink is represented by Z. ha And Z jc Z ch and Z ha All of these can be obtained through curve fitting:

[0057]

[0058] In the formula r i It is the thermal resistance of the Cauer network, c i t is the heat capacity of the Cauer network, and t is the time unit.

[0059] Depend on Figure 4 As can be seen, the Cauer junction temperature model not only needs to fit the thermal resistance and thermal capacity of different parts of the device, but also needs to calculate the device's losses.

[0060] Switching losses and conduction losses constitute the total losses of a MOSFET. The conduction loss of a MOSFET can be expressed as:

[0061]

[0062] In the formula R ds It is the on-resistance, i d It is the on-current of the MOSFET, P M_con It is the MOSFET conduction loss, t con This is the duty cycle of the MOSFET.

[0063] The switching loss of a MOSFET can be expressed as:

[0064]

[0065] In the formula P M_sw It is the switching loss of the MOSFET, E on It is the energy generated when the MOSFET is turned on, E off This refers to the energy generated when the MOSFET is turned off. These parameters can be obtained by building a device simulation model in LTspice software, while R... ds This information can be obtained from the relevant component datasheet. Therefore, the junction temperature of device i can be obtained using the following formula:

[0066]

[0067]

[0068] In the formula P loss For device power loss, T rt It is room temperature.

[0069] Based on the established thermal circuit model, the junction temperatures of MOSFETs and diodes can be obtained accurately and quickly. Then, based on the junction temperatures of the devices and formulas (7) and (8), the failure rate of the power electronic devices can be calculated.

[0070] The switched reluctance motor drive system employing a torque control strategy consists of three parts: the motor body, the power converter, and the detection unit. The static reliability of the system is assumed to be such that failure will occur due to any fault; therefore, this chapter uses the RBD model to calculate the system's static reliability. The RBD model used is as follows: Figure 5 As shown.

[0071] Therefore, the static reliability R1(t) of the system can be calculated as follows:

[0072]

[0073] In the formula R SRM (t), R PC (t) and R DT (t) represent the individual reliability of the motor body, power converter, and detection components, respectively; N is the total number of faults in the entire drive system; λ i It is the failure rate of device i.

[0074] By substituting the device failure rate, the static reliability of the system can be obtained as follows:

[0075]

[0076] To effectively reduce the complexity of calculating the reliability of the SRM drive system, this invention employs a k-out-of-n:G model applicable to multi-state reliability assessment to evaluate the reliability of the motor body, detection unit, and power converter, and uses an RBD model to calculate the reliability of the entire system. In this invention, to further simplify the algorithm, a level-two fault state is identified as a failure state. The k-out-of-n:G reliability models for each part of the system are as follows: Figure 6 As shown in Figures 7 and 8.

[0077] When using the k-out-of-n:G model to calculate the reliability of a system, we assume that if k phases in an n-order system are operating stably, the system is still considered to be operating safely and reliably. The k-out-of-n:G model can be calculated using the following formula:

[0078]

[0079] Therefore, after calculating the reliability of the motor body, power converter, and detection unit in the system using the k-out-of-n:G model, the system reliability can be obtained by formula (14):

[0080]

[0081] However, by optimizing the operation of each state individually, the overall system reliability is improved by enhancing the reliability of each state separately. The basic k-out-of-n:G model is as follows: Figure 9 As shown.

[0082] Taking a power converter as an example, this invention proposes an online optimization method for system reliability to improve system reliability. This method calculates and optimizes the reliability of each fault state separately, ultimately improving system reliability. The flowchart of this method is as follows: Figure 10 As shown.

[0083] (1) Improvement of MOSFET open-circuit fault

[0084] As analyzed above, when the motor is at low speed, if a UMOC or LMOC fault occurs in phase A, a large phase current will be generated in phase B, exceeding the safe threshold range. This is because in the commutation zone, torque is generated jointly by phases A and B. After a phase A open-circuit fault, only phase B generates torque. However, the inductance change rate of phase B is low at this time, thus causing the phase B current to exceed the threshold. Therefore, the reliability under a UMOC or LMOC fault is as follows:

[0085]

[0086] At this point, for UMOC or LMOC faults, the operating parameters are adjusted online to gradually increase the turn-on angle of phase B. The phase current of phase B gradually decreases to within the safe threshold range. The reliability under UMOC or LMOC faults is then:

[0087]

[0088] (2) Improvement of MOSFET short-circuit fault

[0089] Similarly, when the motor is operating at medium to high speeds, if a UMSC or LMSC fault occurs in phase A, the original demagnetization mode of phase A changes to a zero-voltage freewheeling mode, reducing the current drop rate. At high speeds, the current drop time is shorter than at low speeds, which leads to an increase in the negative torque generated by the motor at high speeds, causing the motor to fall behind the reference speed. In this case, the reliability under a UMSC or LMSC fault is:

[0090]

[0091] By adjusting the turn-off angle of the faulty phase online for UMSC or LMSC faults, the faulty phase's turn-off angle is brought close to zero before the inductance drops, effectively reducing the negative torque generated by the motor. The reliability under UMSC or LMSC faults after adjustment is as follows:

[0092]

[0093] It can be seen from formulas (18), (19), (20), and (21) that selecting appropriate fault-tolerance strategies for some fault modes can effectively improve the reliability of the operating state.

[0094] After adopting an online optimized control strategy, the k-out-of-n:G model of the power converter is as follows: Figure 11 As shown. The reliability of the optimized system is expressed as:

[0095]

[0096] The reliability assessment of a system by combining RBD and k-out-of-n:G models can take into account the system's fault tolerance and thermal stress compared to static reliability models, thus providing a more accurate assessment of the drive system's reliability. Furthermore, the system's reliability can be further improved by adopting online optimization strategies.

Claims

1. To address the shortcomings or improvement needs of existing technologies, this invention proposes a method for optimizing the reliability of a switched reluctance motor torque control system. This optimization method proposes fault criteria suitable for torque control strategies in switched reluctance motor drive systems employing torque control systems. Simultaneously, it employs an online calculation and real-time reliability optimization method, using a combination of RBD and k-out-of-n:G models to calculate the dynamic reliability of the switched reluctance motor torque control system, and adjusts the system's operating parameters online under fault conditions. This effectively improves the motor's operating state after a fault, enhancing system reliability. Compared to other optimization methods, this method is faster and more accurate. The design method includes the following steps: Step 1: Given a switched reluctance motor drive system, which consists of three parts: a switched reluctance generator, a detection unit, and a three-phase asymmetrical half-bridge power converter, analyze the fault types that may occur in each of these three parts. Step 2: The operating state of the switched reluctance motor after adopting a torque control strategy differs from that of traditional current chopping control and angle position control strategies. Therefore, fault criteria need to be redefined for the switched reluctance motor drive system using a torque control strategy. Since the switched reluctance motor has strong fault tolerance, it can still operate in a stable state after some faults. Therefore, based on the motor fault types analyzed in Step 1, the motor state under different faults is evaluated. Because torque stability is an important indicator of motor operation, a quantitative analysis of torque fluctuations after different faults is performed, and appropriate torque fluctuation failure judgment criteria are formulated. Excessive current can also cause breakdown faults; therefore, a failure judgment criterion can be developed based on current. Step 3: An online optimization method for system reliability is adopted, calculating and optimizing the reliability of each fault state separately. For open-circuit faults, the on-state angle of the faulty phase is adjusted online to gradually reduce the phase current to within the safe threshold range. For short-circuit faults, the off-state angle of the faulty phase is adjusted online to bring the faulty phase close to zero before the inductance drops, effectively reducing the negative torque generated by the motor. By adopting an online optimization method for system reliability, after a system fault, selecting an appropriate fault-tolerant strategy can effectively improve the system's operating state and also improve the reliability of the drive system after a fault under the torque control strategy. Step 4: To obtain the failure rate of each device, it is necessary to calculate the junction temperature of different switching devices in the power converter. Therefore, the Cauer thermal resistance model can quickly and accurately obtain the junction temperature parameters of different devices on the power converter. Substituting the calculated junction temperature into the failure rate formula, the failure rate of the device can be calculated. Step 5: Reliability assessment is an important part of the high-performance and safe operation of the switched reluctance motor control system. Commonly used reliability assessment models include: Reliability Block Diagram (RBD) model, fault tree model, k-out-of-n:G model, and Markov model. By conducting quantitative analysis on different operating states of the switched reluctance motor system employing torque control strategy, and combining the failure judgment criteria established in step 2, this invention uses the RBD model to calculate the static reliability of the system, and uses a combination of the RBD model and the k-out-of-n:G model to calculate the dynamic reliability of the switched reluctance motor torque control system.

2. Based on claim 1, a method for optimizing the reliability of a switched reluctance motor torque control system is proposed. This method employs a torque control strategy for the switched reluctance motor. However, the operating state of the motor differs from traditional current chopping control and angle position control strategies. Therefore, new fault criteria need to be developed for the switched reluctance motor drive system using the torque control strategy. This method analyzes the possible fault types in three parts: the motor body, the detection unit, and the power converter. Since the switched reluctance motor has strong fault tolerance, it can still operate in a stable state after partial faults. Therefore, based on the analyzed motor fault types, the motor state under different faults is evaluated. Because this method uses a torque control strategy, torque stability is a crucial operating indicator for the motor. Therefore, by quantitatively analyzing the torque fluctuations after different faults occur after the motor adopts the torque strategy, fault criteria for torque are derived. Simultaneously, the maximum current that components can withstand during motor operation is limited; a large operating current can easily lead to component breakdown faults. Therefore, fault criteria for phase current can be derived.

3. Based on claim 2, this invention employs an online optimization method for system reliability, calculating and optimizing the reliability of each fault state separately to improve system reliability. When an open-circuit fault occurs in a MOSFET, the operating parameters are adjusted online for either an upper MOSFET-Open Circuit (UMOC) fault or a lower MOSFET-Open Circuit (LMOC) fault in phase A. Specifically, the turn-on angle of phase B is gradually increased, causing the phase current of phase B to gradually decrease to within a safe threshold range. When a short-circuit fault occurs in a MOSFET, the turn-off angle of the faulty phase is adjusted online for either an upper MOSFET-Short Circuit (UMC) fault or a lower MOSFET-Short Circuit (LMC) fault in phase A. This ensures the faulty phase approaches zero before the inductance decreases, effectively reducing the negative torque generated by the motor. Therefore, when a motor experiences a fault, selecting an appropriate fault-tolerant strategy under partial fault modes can effectively improve the reliability of that operating state.

4. Based on claim 3, this invention obtains the junction temperature of the MOSFET and diode using the Cauer temperature model. Combining the junction temperature with the device failure rate obtained from the prediction manual, the failure rate of the power electronic device can be calculated. Furthermore, to consider the fault tolerance of different parts of the system, this invention analyzes the different failure modes of the switched reluctance motor body, the detection unit, and the power converter separately, using the RBD model to calculate the static reliability of the motor body, the detection unit, and the power converter respectively; whereby the static reliability R1(t) of the system can be calculated as: In the formula R SRM (t), R PC (t) and R DT (t) represent the individual reliability of the motor body, power converter, and detection components, respectively; N is the total number of faults in the entire drive system; λ i This represents the failure rate of device i. The dynamic reliability of the motor body, detection unit, and power converter is evaluated using the k-out-of-n:G model, which is applicable to multi-state reliability assessment. The k-out-of-n:G model can be calculated using the following formula: Combining the two models allows for the calculation of the overall system reliability. Compared to static reliability models, combining the RBD and k-out-of-n:G models to evaluate system reliability can take into account the system's fault tolerance and thermal stress, thus providing a more accurate assessment of the drive system's reliability.

Images