Failure detection system, detection circuit and failure detection method

Abstract

A failure detection system, a detection circuit, and a failure detection method. The failure detection system includes a power conversion circuit including a plurality of switches, and a detection circuit operably coupled to the power conversion circuit. The detection circuit is to: capture first, second, and third current signals (I Uav , Vav , Wav ), where the first current signal is provided by a first pair of the plurality of switches, the second current signal is provided by a second pair of the plurality of switches, and the third current signal is provided by a third pair of the plurality of switches; convert the first, second, and third current signals into two quadrature components; generate a first plot and a second plot based on the two quadrature components; and detect an anomaly associated with at least one of the plurality of switches based on comparing the first plot and the second plot.

Description

Technical Field

[0001] One embodiment disclosed herein relates to a failure detection system, detection circuit, and failure detection method, and particularly to a failure detection system, detection circuit, and failure detection method for a power conversion circuit. Background Technology

[0002] Modern UPS systems incorporate sophisticated power conversion technologies to ensure continuous operation during power disturbances, protecting sensitive electronic devices from voltage fluctuations, surges, and power outages. Some advancements are the result of improvements in power semiconductor devices, such as insulated-gate bipolar transistors (IGBTs), which offer high efficiency and fast switching characteristics. Summary of the Invention

[0003] One embodiment of this disclosure provides a failure detection system comprising a power conversion circuit and a detection circuit. The power conversion circuit includes a plurality of switches, each having three terminals. The detection circuit is operatively coupled to the power conversion circuit and is used to perform the following steps: Acquiring a first current signal (I... Uav ), a second current signal (I) Vav and a third current signal (I) Wav The first current signal is provided by a first pair of switches, the second current signal is provided by a second pair of switches, and the third current signal is provided by a third pair of switches. The first and third current signals are converted into two quadrature components. A first plot and a second plot are generated based on the two quadrature components. An anomaly associated with at least one of the switches is detected by comparing the first plot and the second plot.

[0004] Another embodiment of this disclosure provides a detection circuit, which includes an interface circuit, a conversion circuit, and a driving circuit. The interface circuit is used to receive a first current signal (I0). Uav ), a second current signal (I) Vav and a third current signal (I) WavThe first current signal is provided by a first pair of switches coupled to a power supply voltage, the second current signal is provided by a second pair of switches coupled to the power supply voltage, and the third current signal is provided by a third pair of switches coupled to the power supply voltage. A conversion circuit is used to perform the following steps: (i) converting the first to third current signals into two quadrature components; (ii) generating a first plot and a second plot based on the two quadrature components respectively; and (iii) identifying an anomaly associated with at least one of the first to third pairs of switches by comparing the first plot and the second plot. A drive circuit is used to adjust one or more of a plurality of control signals controlling the first to third pairs of switches respectively based on the identified anomaly.

[0005] Another embodiment of the present disclosure provides a failure detection method, which includes the following steps: Receiving a first current signal (I0). Uav ), a second current signal (I) Vav and a third current signal (I) Wav The first current signal is provided by a first pair of switches coupled to a power supply voltage, the second current signal is provided by a second pair of switches coupled to the power supply voltage, and the third current signal is provided by a third pair of switches coupled to the power supply voltage. These first and third current signals are converted into two quadrature components. A first plot and a second plot are generated based on these two quadrature components. An anomaly associated with at least one of the first and third pairs of switches is identified by comparing the first and second plots. Based on the identified anomaly, one or more of a plurality of control signals controlling the first and third pairs of switches are adjusted respectively. Attached Figure Description

[0006] When accompanied by Figure 1 When reading this document, the various aspects of one embodiment of the present disclosure can be best understood from the following embodiments. It should be noted that, according to standard industry practice, the various features are not drawn to scale. In fact, for clarity of explanation, the dimensions of the various features may be arbitrarily increased or decreased.

[0007] Figure 1 A schematic diagram of an example system for detecting semiconductor device failure according to some embodiments is shown;

[0008] Figure 2 The following are examples of embodiments shown. Figure 1 Examples of simulated amplitude and / or measured amplitude plots of the current waveform of the system;

[0009] Figure 3 An example flowchart of a method for detecting semiconductor device failure according to some embodiments is shown;

[0010] Figure 4 The illustration shows, according to some embodiments, components including those joined to a heat sink. Figure 1 Example structure of power conversion circuit of example system;

[0011] Figure 5 Illustrations based on some embodiments Figure 4 Block diagram of the thermal model of the instance structure in the example;

[0012] Figure 6 The diagram illustrates the relationship between [various embodiments] and [other embodiments]. Figure 4 Example diagram showing the relationship between the current signal associated with the instance structure and the temperature of the junction surface;

[0013] Figures 7 to 10 The following are examples of methods for analysis. Figure 1 Bode plots of the health status of one or more insulated gate bipolar transistors (IGBTs) in the power conversion circuit of an example system.

[0014] Figure 11 The diagram illustrates the use of some embodiments by Figure 1 This provides an overview of analytical techniques used by example systems to detect failures in semiconductor devices.

[0015] [Symbol Explanation]

[0016] 100: System

[0017] 101: Power Conversion Circuit

[0018] 102: Detection Circuit

[0019] 103: Power Supply

[0020] 104: Switch

[0021] 104A: Switch

[0022] 104B: Switch

[0023] 104C: Switch

[0024] 104D: Switch

[0025] 104E: Switch

[0026] 104F: Switch

[0027] 106: Interface Circuit

[0028] 108: Conversion Circuit

[0029] 110: Current controller

[0030] 112: Conversion Circuit

[0031] 114: Pulse Width Modulation Circuit

[0032] 116: Drive circuit

[0033] 118: Interface Circuit

[0034] 120: Conversion Circuit

[0035] 122: Voltage controller

[0036] 200A: Drawing

[0037] 200B: Drawing

[0038] 200C: Drawing

[0039] 202: Analog Waveform

[0040] 204: Measurement Waveform

[0041] 206: Measurement Waveform

[0042] 300: Method

[0043] 302: Operation

[0044] 304: Operation

[0045] 306: Operation

[0046] 308: Operation

[0047] 400: Structure

[0048] 402: Chip

[0049] 404: Base plate

[0050] 406: Heatsink

[0051] 408: Star symbol

[0052] 500: Thermal Model

[0053] 600: Figure

[0054] 602: Boundary

[0055] 604: Data point

[0056] 606: Data point

[0057] 700: Bodetu

[0058] 800: Bodetu

[0059] 900: Bodetu

[0060] 1000: Bodetu

[0061] 1100: Steps

[0062] 1102: Steps

[0063] 1104: Steps

[0064] 1106: Steps

[0065] 1108: Steps

[0066] 1110: Steps

[0067] 1112: Functionality Matching

[0068] 1114: Technology

[0069] A: Node

[0070] B: Node

[0071] C: Node

[0072] DCB: Direct Copper Bonding Layer

[0073] f1~f3: Harmonic frequencies

[0074] i * od Orthogonal components

[0075] i * oq Orthogonal components

[0076] Ibus: Bus current

[0077] I capa Current

[0078] I capb Current

[0079] I capc Current

[0080] I LA Current

[0081] I LB Current

[0082] I LC Current

[0083] I oA Current

[0084] I oB Current

[0085] I oC Current

[0086] i od Orthogonal components

[0087] i oq Orthogonal components

[0088] L: Indicator

[0089] NTC: Negative Temperature Coefficient

[0090] P V Total power

[0091] R thCH Thermal resistance

[0092] R thCHA Thermal resistance

[0093] R thCNTC Thermal resistance

[0094] R thJC Thermal resistance

[0095] R thJNTC Thermal resistance

[0096] S1~S6: Pulse Width Modulation Signals

[0097] T1~T6: Switches

[0098] T Ambient Ambient temperature

[0099] T Case : Casing temperature

[0100] T Heatsink Radiator temperature

[0101] T Junction : Joint surface temperature

[0102] T NTC Temperature measurement

[0103] V ab Voltage signal

[0104] V bc Voltage signal

[0105] V ca Voltage signal

[0106] vconA Voltage control signal

[0107] v conB Voltage control signal

[0108] v conC Voltage control signal

[0109] v cond Orthogonal components

[0110] v conq Orthogonal components

[0111] v * cd : Voltage

[0112] v * cq : Voltage

[0113] v d : Voltage

[0114] v od : Voltage

[0115] v oq : Voltage

[0116] Z A Load

[0117] Z B Load

[0118] Z C Load Detailed Implementation

[0119] The following disclosure provides numerous different embodiments or instances for implementing various features of the provided object. Specific examples of elements and arrangements are described below to simplify one embodiment of this disclosure. Of course, these are merely examples and not intended to be limiting. For example, in the following description, the formation of a first feature above or on a second feature may include embodiments in which the first and second features are formed in direct contact, and may also include embodiments in which additional features may be formed between the first and second features such that the first and second features are not in direct contact. Furthermore, an embodiment of this disclosure may repeat element symbols and / or letters in various instances. This repetition is for simplicity and clarity and does not in itself define the relationship between the various embodiments and / or configurations discussed.

[0120] Furthermore, for ease of description, spatial relative terms such as “below,” “under,” “lower,” “above,” “upper,” “top,” “bottom,” and similar terms may be used herein to describe the relationship between one element or feature and another, as shown in the figures. The spatial relative terms are intended to cover different orientations of the device in use or operation other than those shown in the figures. The device may be oriented in other ways (rotated 90 degrees or otherwise), and the spatial relative descriptors used herein may be interpreted accordingly.

[0121] Typically, the reliability and efficiency of electrical systems can be enhanced or ensured by using uninterruptible power supplies (UPS), variable frequency drives (VFDs), or other components or devices. Such electrical systems may rely on power elements such as insulated-gate bipolar transistors (IGBTs) to manage and distribute power. For example, an IGBT may be a semiconductor device that forms part of a UPS and / or VFD unit. IGBTs combine the high input impedance of a metal-oxide-semiconductor field-effect transistor (MOSFET) with the low on-state conduction losses of a bipolar junction transistor (BJT) for a variety of applications, including but not limited to high-voltage or high-current applications. Structurally, an IGBT may include a four-layer (PNPN) configuration with three terminals (e.g., gate, collector, and emitter). The gate may be referred to as the gate terminal, contact, or electrode. The collector and emitter may each be referred to as the source / drain (S / D) terminals or electrodes. IGBTs can be operated by controlling the current flow between the S / D terminals (e.g., between the collector and emitter) using a voltage applied to the gate.

[0122] While IGBTs are implemented to support the operation of various electrical systems, they may lack preventative maintenance strategies and rely on hardware detection mechanisms. These mechanisms can be considered lagging indicators, responding to faults after they occur rather than preventing them. For example, some maintenance methods trigger fault indicators after the electrical system experiences a problem, rather than detecting faults in IGBT operation. In such cases, reliance on reactive measures can pose risks to the operation and safety of the electrical system.

[0123] In another instance, some detection mechanisms may lack the ability to analyze or detect component aging and changes that could lead to component (e.g., IGBT) failure. Some systems that rely solely on rapid hardware detection and protection mechanisms may lack the ability to ensure long-term reliability. For example, overvoltage and overcurrent detection mechanisms can trigger the shutdown of at least a portion of the electrical system via hardware circuitry, which can interrupt operation and cause downtime. In some cases, other detection mechanisms, such as certain temperature detection mechanisms, can trigger shutdown. Predicting and preventing failure before it occurs through reactive fault response can be challenging due to a lack of insight into the gradual degradation of components.

[0124] To address these challenges, technical solutions and methods integrate and deploy models (e.g., formulas or mechanisms) with various detection mechanisms to detect or predict faults / failures of one or more components (such as, but not limited to, IGBTs in electrical systems, such as power conversion circuits). These systems and methods may involve components used for at least one of current analysis (or calculation), temperature analysis, or voltage spectrum analysis. Current analysis may involve or utilize current transformation to detect potential anomalies in power components (e.g., IGBTs) by analyzing the amplitude and (phase) angle from current measurements. Temperature analysis utilizes datasets to establish standard values ​​and tolerances by collecting and analyzing current and temperature data. Voltage spectrum analysis records the voltage spectrum (e.g., Bode plot) and calculates frequency deviations under component variations. Systems and methods may utilize these components independently or in combination, such as at least one or a combination of current analysis, temperature analysis, and / or voltage spectrum analysis techniques. By implementing the techniques discussed herein, systems and methods can enable early IGBT health index methods to monitor and assess IGBT health (or operating conditions) before failure. Systems and methods can intercept anomalies before system or component failure, allowing / enabling preemptive inspections, component replacements, or other remedial measures for electrical systems to reduce downtime, improve the safety and reliability of electrical systems, and enhance operational efficiency.

[0125] Figure 1 A schematic diagram of an example system 100 for detecting semiconductor device failure according to some embodiments is shown. Elements or circuitry of system 100 may be a single device or part of a plurality of devices operatively coupled to each other. System 100 may include various elements, such as, but not limited to, power conversion circuitry 101 and detection circuitry 102. Elements of system 100 (e.g., elements of detection circuitry 102) may be operated or feature or function to detect or determine potential faults in power conversion circuitry 101.

[0126] Power conversion circuit 101 may be included, correspond to, or implemented in one or more electrical systems (such as a UPS or VFD). Power conversion circuit 101 may include or be composed of one or more hardware elements for performing or carrying out one or more functions discussed herein. For example, power conversion circuit 101 may be operable to convert power from one form to another, such as from alternating current (AC) to direct current (DC), or vice versa. Power conversion performed by power conversion circuit 101 may be implemented to ensure proper operation and reliability of electrical systems.

[0127] As an example, in the context of a UPS, the power conversion circuit 101 may include elements such as a rectifier, converter, or filter. The power conversion circuit 101 (e.g., a rectifier) ​​converts input AC power from a power source into DC power for charging the UPS battery. During a power outage, the power conversion circuit 101 (e.g., a converter) converts stored DC power from the battery back into AC power to supply connected devices, thereby ensuring a continuous and uninterrupted power supply.

[0128] In another example, within the context of VFD, power conversion circuit 101 can partially control or support the control of motor speed by changing the frequency of the power supplied to the motor. Power conversion circuit 101 may include, for example, a rectifier, a DC link, and / or a converter. Power conversion circuit 101 (e.g., a rectifier) ​​can convert input AC power to DC power, which can then be smoothed and stored, for example, in a DC link. Power conversion circuit 101 (e.g., a converter) can convert the DC power back to AC power with variable frequency and voltage, which can be used to control the speed and torque of the motor. It should be noted that power conversion circuit 101 can be used for other purposes or implemented in various other applications, not limited to those discussed herein.

[0129] The power conversion circuit 101 may include a power supply 103 (e.g., a power source), multiple switches 104, inductors, capacitors, or other components. The power conversion circuit 101 may be electrically or operably coupled to one or more loads (e.g., Z...). A Z B and Z C The power supply 103 can provide the power or voltage required to operate the components of the power conversion circuit 101. In some cases, the power conversion circuit 101 may include an additional power supply. In other cases, the power supply 103 may be an external power supply separate from the power conversion circuit 101.

[0130] In some configurations, the power conversion circuit 101 may include six switches 104 (e.g., switches 104A to 104F). The switches 104 may be labeled T1 to T6, such as at least... Figure 1 As shown. Switch 104 may be referred to as transistor 104. Power conversion circuit 101 may include more or fewer numbers of switches 104, such as not limited to six switches 104, to function as a power converter. Switch 104 may include, correspond to, or be implemented using IGBTs. Switch 104 (e.g., IGBT) may regulate the current via power conversion circuit 101.

[0131] Individual switch 104 may include three terminals, for example, similar to a metal-oxide-semiconductor (MOS) transistor. The three terminals may include a gate terminal, a first S / D terminal (e.g., collector), and a second S / D terminal (e.g., emitter). Switch 104 can be used to provide a three-phase current (e.g., I0). oA I oB and I oC Switch 104 can be controlled to turn on and off by individual gate signals, thereby regulating the current flowing through the circuit. The gate signals can be provided by one or more elements of the detection circuit 102 or other circuits in system 100.

[0132] Turning on switch 104 may involve applying a predetermined voltage level at the gate terminal of switch 104. By turning on, switch 104 allows current to flow between the S / D terminals. For example, turning off switch 104 may involve stopping or preventing the application of voltage at the gate terminal of switch 104, or in some cases, applying a relatively low voltage level (below the predetermined voltage level) at the gate terminal. When turned off, switch 104 may not allow current to flow between the S / D terminals.

[0133] Switch 104 can be turned on and off to convert the input power supplied from power supply 103 into a desired output form. Switch 104 can be arranged in a configuration including three bridge arms (extending from each individual node A to C at least to the load impedance Z). A To Z CIn a three-phase bridge configuration, one or more switches 104 are operable to generate three-phase currents, which may be spaced 120 degrees apart. Each of the three bridge arms may correspond to a phase of the three-phase system (e.g., phases A through C). Each bridge arm is associated with a pair of switches. Switches 104A through 104F may be referred to as T1 through T6, respectively. For example, T1 and T2 (e.g., the first pair of switches 104) may be electrically connected at node “A” to operably form the first phase bridge arm. T3 and T4 (e.g., the second pair of switches 104) may be electrically connected at node “B” to operably form the second phase bridge arm. T5 and T6 (e.g., the third pair of switches 104) may be electrically connected at node “C” to operably form the third phase bridge arm. T1, T3, and T5 may be referred to as high-side switches. T2, T4, and T6 may be referred to as low-side switches. Current in phase AC may flow through nodes A through C, respectively.

[0134] In operation, for phase A current, when T1 is on and T2 is off, the current (e.g., I) oA The current can flow from the positive DC bus (of power supply 103) through T1 and through the load Z. A And it returns to the negative DC bus. This generates a positive half-cycle for phase A. When T1 is off and T2 is on, current can flow from the negative DC bus through T2 and through the load Z. A And return to the positive DC bus, thus generating a negative half-cycle for phase A. For phase B current, when T3 is on and T4 is off, the current (e.g., I) oB It can flow from the positive DC bus through T3, and through the load Z. b The current flows back to the negative DC bus, thus generating a positive half-cycle for phase B. When T3 is off and T4 is on, current can flow from the negative DC bus through T4 and through the load Z. b It then returns to the positive DC bus, thus generating a negative half-cycle for phase B.

[0135] In further operation, for phase C, when T5 is on and T6 is off, the current (I) oC It can flow from the positive DC bus through T5, and through the load Z. c And it returns to the negative DC bus, thus generating a positive half-cycle for phase C. When T5 is off and T6 is on, current can flow from the negative DC bus through T6 and through the load Z. c And it returns to the positive DC bus, thus generating a negative half-cycle for phase C. It should be noted that the load Z... A Z B and Z c This may include or correspond to any resistive, inductive, or capacitive element driven by the power conversion circuit 101. The load may receive current from the switch and represent the load impedance in a three-phase system. The value of the load impedance can affect the current and voltage characteristics of the power conversion circuit 101.

[0136] The indicators (e.g., L) of the power conversion circuit 101 may be used at least to store energy in the form of a magnetic field and / or smooth current. For example, each inductor may act as a filter to filter current ripple and provide a relatively more stable output. The capacitors (e.g., C) of the power conversion circuit 101 may be used at least to store energy in the form of an electric field and smooth voltage. For example, each capacitor may act as another filter to filter current ripple and provide a relatively more stable output voltage. In some configurations, the power conversion circuit 101 may include other types of switches for three-phase systems.

[0137] The power conversion circuit 101 may include sensors or other types of monitoring devices to obtain measurement data for analysis. For example, the power conversion circuit 101 may include current sensors, voltage sensors, temperature sensors, etc. A current sensor may monitor the input and output currents associated with switch 104. A temperature sensor may monitor the temperature of one or more components of the power conversion circuit 101. A voltage sensor may monitor the voltage level (e.g., DC or AC voltage level) in the power conversion circuit 101. Sensor data generated by the sensors may be provided in real time to detection circuit 102 and other circuits or devices. For example, the sensor data may be used, at least in part, to detect or determine any anomalies within the power conversion circuit 101 or to initiate at least one action to prevent failure.

[0138] The power conversion circuit 101 is operatively coupled to the detection circuit 102 to detect potential faults, such as those caused by an open or short circuit in at least one of the switches 104. The detection circuit 102 can be used to detect faults before the power conversion circuit 101 fails or shuts down, minimizing downtime, improving safety, and increasing the efficiency and reliability of the power conversion circuit 101 (or other circuits that have been tested or analyzed).

[0139] Detection circuitry 102 may include or be composed of one or more hardware elements. Detection circuitry 102 may be circuitry located remotely from power conversion circuitry 101. Detection circuitry 102 may be operatively coupled to power conversion circuitry 101 via one or more wired or wireless connections. In some cases, detection circuitry 102 and power conversion circuitry 101 may be part of a device. Detection circuitry 102 may include one or more elements for performing fault detection, including at least one interface circuit 106, 118, at least one conversion circuit 108, 112, 120, at least one current controller 110, at least one pulse width modulation (PWM) circuit 114, at least one drive circuit 116, at least one voltage controller 122, or other elements for detecting potential faults in power conversion circuitry 101, such as those related to the operation of at least one of the switches 104. In some configurations, the elements of detection circuitry 102 may be individual or separate elements used to jointly perform the fault / failure detection operations discussed herein. In some other configurations, the elements of the detection circuit 102 may represent the function of a single element used for fault detection.

[0140] The detection circuit 102 may include more or fewer components, and is not limited to a combination of Figure 1 The components provided are as follows. For example, to use current data or waveforms for fault detection (e.g., current analysis), the detection circuit 102 may not include or may not utilize the interface circuit 118, the conversion circuit 120, and / or the voltage controller 122. In another example, to use voltage data or waveforms for fault detection (e.g., voltage spectrum analysis), the detection circuit 102 may not include or may not utilize the interface circuit 106, the conversion circuit 108, and / or the current controller 110. In a further example, to use temperature data for fault detection (e.g., temperature analysis), the detection circuit 102 may include one or more additional components to interface with (and receive data from) the power conversion circuit 101 or to process the temperature data for fault or anomaly detection.

[0141] Interface circuit 106 can be used to interface with or communicate with one or more circuits, devices, or components to receive signals or data for processing, such as current signals, voltage signals, etc. In some cases, interface circuit 106 can provide data to other circuits within system 100. For example, interface circuit 106 can receive three-phase current values ​​or signals (e.g., I0) from power conversion circuit 101. oa I ob and I oc Each current signal can be provided by a separate pair of switches 104. For example, the first current signal (I... oa The first pair of switches (e.g., T1 and T2) can provide the second current signal.ob This can be provided by a second pair of switches (e.g., T3 and T4). The third current signal (I) oc The current signal may be provided by a third pair of switches (e.g., T5 and T6). In some cases, the current signal may be interchangeably represented with the current value. The current value may be a measured current value or an analog current value. Interface circuit 106 may receive measured current values ​​from one or more sensors (e.g., current sensors) that monitor the three-phase current in the power conversion circuit 101 at the three arms (associated with node AC). In some cases, interface circuit 106 may include one or more sensors for capturing or measuring current values. Interface circuit 106 may receive current values ​​measured during operation of power conversion circuit 101 for at least a predetermined time period.

[0142] Interface circuit 106 can receive simulated current values ​​for power conversion circuit 101. Interface circuit 106 can receive simulated current values ​​from an external device (e.g., a simulation device) that performs the simulation of power conversion circuit 101. Simulated values ​​(e.g., current values, voltage values, or other information) can be based on a specific configuration of the simulated circuit. In this case, the simulation can be based on a three-phase system with six switches 104, where current values ​​can be obtained at each individual bridge arm (each bridge arm associated with an individual phase spaced 120 degrees apart). For example, the simulation can provide the current value at each bridge arm of the three-phase system. The simulated values ​​can represent, for example, a predicted (or expected) current flowing through each bridge arm of power conversion circuit 101 based on the circuit configuration and the state or condition of one or more components.

[0143] In some configurations, interface circuitry 106 can obtain analog current values ​​from power conversion circuitry 101 when switch 104 operates as expected (e.g., normal or fault-free operation). Interface circuitry 106 can receive multiple analog current values ​​during operation. Operation of power conversion circuitry 101 may involve alternatively activating (e.g., turning on) or deactivating (e.g., turning off) individual switches 104. Interface circuitry 106 can provide analog values ​​of power conversion circuitry 101 under normal conditions to conversion circuitry 108.

[0144] In some configurations, the interface circuit 106 may obtain an analog current value from the power conversion circuit 101, wherein at least one switch 104 is faulty, for example, at least one of the switches 104 is experiencing or in an open-circuit or short-circuit state. In this case, the analog current value may be referred to as, for example, a fault analog current value, an open-circuit analog current value, or a short-circuit analog current value.

[0145] Interface circuit 106 can receive or capture open-circuit analog current values ​​by simulating open circuits on individual switches 104. When a switch is open, current may not be conducted from one S / D terminal to another. The open-circuit state of individual switches 104 can be simulated by blocking or not applying voltage at the gate (e.g., no gate signal, therefore no current flows through the individual switches). For example, during the operation of power conversion circuit 101 in the simulation, individual switches 104 can be set to be open for a relatively short or brief period. When individual switches are set to be open, three-phase current values ​​can be captured. For example, using six switches 104, interface circuit 106 can receive at least six sets of analog current values.

[0146] Interface circuit 106 can receive or capture simulated short-circuit current values ​​by simulating short circuits on individual switches 104. To simulate a short circuit, a short-circuit path (e.g., a wire or connection) can be provided in the simulation, connecting one S / D terminal of each individual switch to another part of the circuit. For example, to simulate a short circuit in T1, a short-circuit path can be placed from one S / D terminal to another. In another example, a short-circuit path can be placed from the collector (or drain) of T1 to the emitter (or source) of T2 or another part of the power conversion circuit 101 to simulate a short circuit in T1. The simulation of the short circuit can be applied to other switches 104. Interface circuit 106 can receive multiple sets of three-phase current values ​​associated with short circuits in individual switches 104.

[0147] In some other cases, a short circuit in the switch can be simulated by keeping the switch in the on state throughout the operation of the power conversion circuit 101. Other non-limiting techniques or configurations can be implemented for simulation. The current value (e.g., simulated or measured) can be used to determine potential faults or anomalies within the power conversion circuit 101. The interface circuit 106 can process the received current value and provide the processed current value as input to the conversion circuit 108. For example, processing the current value may involve applying a data filter or converting the data format. In some cases, the interface circuit 106 can forward the current value to the conversion circuit 108 without processing.

[0148] The conversion circuit 108 can receive (analog or measured) current signals or data from the interface circuit 106. The conversion circuit 108 can convert three-phase current values ​​(e.g., first to third current signals) into two quadrature components (or values), for example, i oq (For example, I) Qav ) and i od (For example, I) Dav The transformation circuit 108 can use the example formula (1) to calculate or derive two orthogonal components from each set of current values.

[0149] (1)

[0150] Individual sets of current values ​​can be captured during different startup or deactivation periods of switch 104 in power conversion circuit 101. For the purpose of providing an example, at least six sets of current values ​​(e.g., measured or simulated) may exist, each for the individual startup of a particular switch. Quadrature components (e.g., i od andi oq A vector can be a pair of vectors that are perpendicular to each other in a given space. Orthogonal components can refer to the parts of a signal that do not interfere with each other.

[0151] The conversion circuit 108 can process additional information from the interface circuit 106. The conversion circuit 108 can perform other processing techniques, not limited to determining the quadrature component. The conversion circuit 108 can provide the quadrature component to the current controller 110 for processing and determining the presence of potential faults or anomalies. For example, when the power conversion circuit 101 operates as expected (e.g., normal operation without faults), the conversion circuit 108 can generate, calculate, or otherwise drive the quadrature component from the analog current signal. When at least one of the switches 104 is in an open-circuit state, the conversion circuit 108 can derive the quadrature component from the analog current signal. When at least one of the switches 104 is in a short-circuit state, the conversion circuit 108 can derive the quadrature component from the analog current signal. The conversion circuit 108 can derive the quadrature component from the measured current signal during operation of the power conversion circuit 101.

[0152] The current controller 110 can receive quadrature components from the conversion circuit 108. The current controller 110 can process the quadrature components to derive and generate an amplitude or oscillation of a current signal (e.g., I0). Sav ) and the phase angle associated with the amplitude (e.g., The vector of ). For example, the current controller 110 can derive the amplitude and phase angle during the operation of the power conversion circuit 101 using or based on example formula (2). The output of applying example formula (1) can be used as the input of example formula (2).

[0153] (2) ,

[0154] The current controller 110 can iteratively process two quadrature components from various states (e.g., analog or real) of the power conversion circuit 101. For example, the current controller 110 can generate amplitude and phase angle (e.g., under normal operation, open circuit, or short circuit conditions) from the quadrature component associated with the analog current signal. In another example, the current controller 110 can generate amplitude and phase angle from the quadrature component associated with the measured current signal.

[0155] The current controller 110 can generate tables based on amplitude and phase angle derived from analog values. For the purpose of providing examples, the current controller 110 can generate a first table for simulating an open circuit (e.g., example table 1) and a second table for simulating a closed circuit (e.g., example table 2). The values ​​in the tables are provided as example values ​​and should not be limited to those presented herein. The tables may include amplitude, phase angle, and phase range associated with individual amplitudes and phase angles.

[0156] A phase range can represent the range of phase angles associated with a particular switch. For example, for six switches 104, there are six phase ranges, each with a range of 60 degrees. The phase range can be predetermined, such as the range between 30-90, 90-150, 150-210, 210-270, 270-330, and 330-30 degrees. The current controller 110 can determine which switches 104 belong to each individual phase range based on calculated phase angles (e.g., according to the type of simulated fault).

[0157] For example, Example Table 1 may represent a simulation of the current when one of the switches T1 through T6 is intentionally disconnected for a relatively short period of time (e.g., each switch is independently simulated as an open circuit). Individual rows in Example Table 1 may correspond to the case where individual switches are intentionally disconnected (e.g., a switch experiences an open-circuit fault). In this case, when T1 is disconnected (in the simulation), the current controller 110 may calculate a phase angle of 112 degrees, which falls within the range of 90 to 150 degrees. When T2 is disconnected, the current controller 110 may calculate a phase angle of 291 degrees, which falls within the range of 270 to 330 degrees. This process can be repeated for other switches, as provided in Example Table 1.

[0158] In a further example, Example Table 2 may represent the simulation of the current when one of the switches T1 through T6 is intentionally short-circuited for a relatively short period of time (e.g., each switch is simulated as short-circuited independently). Similar to Table 1, individual rows in Example Table 1 may correspond to the case where the individual switches are intentionally short-circuited (in the simulation). In this case, when T1 is short-circuited (in the simulation), the current controller 110 may calculate the phase angle as 345 degrees, which falls within the range of 330 to 30 degrees. When T2 is short-circuited, the current controller 110 may calculate the phase angle as 167 degrees, which falls within the range of 150 to 210 degrees. This process can be repeated for other switches, as provided in Example Table 2.

[0159] Based on open-circuit or short-circuit simulations, current controller 110 can determine individual switches 104 associated with individual phase ranges, such as those shown in a table for example, to provide an example. In some cases, current controller 110 can generate a mapping (e.g., rather than a table) to map individual phase ranges to individual switches 104 based on phase angles calculated, for example, when the switches are open or short-circuited. Current controller 110 can perform other techniques to connect or associate individual switches to individual phase ranges. Current controller 110 can use phase ranges to associate potential faults with at least one of the switches 104.

[0160]

[0161] Example Table 1 - Open Circuit

[0162]

[0163] Example Table 2 - Short Circuit

[0164] To identify potential faults, the current controller 110 can generate one or more plots to compare measured and analog values ​​(e.g., amplitude at various phase angles). The one or more plots can be generated using amplitudes and phase angles calculated, for example, based on analog and / or measured current signals. The generation of the one or more plots can be combined with at least... Figure 2 To describe. It should be noted that instance plots are generated as instances, and in some cases, the current controller 110 may use other techniques (in addition to or alternative to instance plots) to record or store calculations, so that analog and measured values ​​can be compared and deviations (or inconsistencies / inconsistencies) can be identified to determine potential faults.

[0165] Figure 2 The following are examples of embodiments shown. Figure 1 Examples of simulated and / or measured amplitudes of the current waveform of the system are plotted in 200A to 200C. These examples are provided as examples to illustrate potential plotting of amplitudes calculated from simulated or measured current signals and are not intended to limit the features or operations discussed herein. The x-axis and y-axis of plots 200A to 200C may represent two orthogonal components, where each point of the waveform in plots 200A to 200C may represent the amplitude at the corresponding phase angle.

[0166] The current controller 110 can generate a plot 200A including an analog waveform 202 (e.g., an ideal waveform). The analog waveform 202 can represent the amplitude and phase angle calculated based on an analog current signal from a normally operating power conversion circuit 101, for example, simulating a current signal under the assumption that the switches are operating normally. A normally operating switch can refer to a switch that is neither open nor closed and operates as expected or desired. For example, the current controller 110 can generate a plot 200A (e.g., a first plot) based on vectors derived from example formulas / equations (e.g., example formulas (1) and / or (2)), the corresponding amplitude and corresponding phase angle of which are calculated via analog current signals simulating the assumption that the switches 104 of the power conversion circuit 101 are each operating normally.

[0167] The current controller 110 can generate a plot 200B that includes a measured waveform 204. The measured waveform 204 can represent the amplitude and phase angle calculated from the measured current signal when at least one of the switches 104 is open. The plot 200B can include an analog waveform 202 for comparison with the measured waveform 204. For example, the current controller 110 can generate the plot 200B (e.g., a second plot) based on a vector derived from an example formula, the corresponding amplitude and phase angle of which are calculated via a current signal (e.g., a measured current value or signal) measured during operation of the power conversion circuit.

[0168] The current controller 110 can generate a plot 200C including a measured waveform 206. The measured waveform 206 can represent the amplitude and phase angle calculated from the measured current signal when at least one of the switches 104 is short-circuited. The plot 200C can include an analog waveform 202 for comparison with the measured waveform 206. The resulting plots 200A to 200C are provided as examples, for example, to illustrate example measurements of at least one switch experiencing an open-circuit fault or a short-circuit fault. The current controller 110 can store the plot in the memory or storage device of the detection circuit 102. In some cases, waveforms 202, 204 or waveforms 202, 206 can be generated in separate plots.

[0169] In some implementations, the current controller 110 may provide one or more plots 200A to 200C to an external device. The plots 200A to 200C may be presented and analyzed by an operator to determine any anomalies, irregularities, or inconsistencies between the measured waveform and the analog waveform. Inconsistency may refer to a difference between the measured waveform and the analog waveform. Inconsistency may indicate a potential anomaly associated with at least one of the switches 104, such as a potential failure of at least one switch 104 in the power conversion circuit 101. Alignment between the measured waveform and the analog waveform (e.g., no inconsistency) may indicate no anomaly associated with the switch 104.

[0170] In various implementations, the current controller 110 can (automatically) detect inconsistencies or differences between a measured waveform and an analog waveform. For example, the current controller 110 can compare the amplitudes of the measured waveform and the analog waveform at different phase angles. The current controller 110 can determine the difference between the amplitudes of the measured waveform and the analog waveform based on the comparison. The current controller 110 can compare the difference with a predetermined threshold value. The current controller 110 can detect inconsistencies between the measured waveform and the analog waveform based on at least one amplitude (at a certain phase angle) being greater than or equal to the predetermined threshold value. Based on the detected inconsistency, the current controller 110 can obtain the phase angle associated with the inconsistency and identify the phase range of the phase angle. The identified phase range (e.g., referring to Example Table 1 or Example Table 2, depending on the context) can represent one of the switches 104. Therefore, the current controller 110 can use the resulting plot (e.g., 200B or 200C) to detect one of the switches 104 with a potential fault.

[0171] For example, current controller 110 can compare the amplitudes of waveforms 202 and 204 of plot 200B at various phase angles. Waveform 202 may refer to a first plot (or a waveform of the first plot), and waveform 204 may refer to a second plot (or a waveform of the second plot). Current controller 110 can detect inconsistencies between the two waveforms 202 and 204 (or between the first and second plots). Inconsistencies can occur at a phase angle between the amplitudes of waveform 202 and waveform 204. For example, current controller 110 can detect an inconsistency of approximately 45 degrees based on the difference between the amplitudes of waveforms 202 and 204 being greater than or equal to a predetermined threshold.

[0172] In some cases, the current controller 110 can determine whether the amplitude difference indicates a potential open circuit or short circuit by comparing the difference with a predetermined range of differences. If the difference falls within a first range (e.g., a relatively low amplitude difference), the current controller 110 can determine that the amplitude difference indicates a potential open circuit. If the difference falls within a second range greater than the first range (e.g., a relatively high amplitude difference), the current controller 110 can determine that the amplitude difference indicates a potential short circuit.

[0173] Using the generated table or mapping (e.g., Instance Table 1 or Instance Table 2), the current controller 110 can identify the phase range of detected inconsistencies. For example, the current controller 110 can determine that an inconsistency in Instance Drawing 200B may indicate an abnormal open circuit. In such a case, the current controller 110 can use Instance Table 1. If the inconsistency occurs at approximately a 45-degree phase angle, the current controller 110 can identify T4 as a switch potentially faulty (e.g., abnormally open). In this case, the inconsistency may occur at other phase angles within the phase range where the current controller 110 identifies T4 as a switch. In other scenarios, the current controller 110 can detect anomalies or inconsistencies at other phase angles.

[0174] In another example, current controller 110 may generate plot 200C. Current controller 110 may compare the amplitudes of waveforms 202 and 206 of plot 200C at various phase angles, wherein anomalies or inconsistencies can be detected by comparing the differences between the amplitudes to predetermined thresholds (e.g., closed-circuit thresholds). Current controller 110 may determine that the difference is greater than the predetermined threshold. In this case, the difference may be within or above a second difference range, which may indicate an abnormal short circuit or switch. For example, in plot 200C, current controller 110 may detect inconsistencies at a first phase angle of approximately 60 to 70 degrees (for a first switch) and a second phase angle of approximately 300 degrees (for a second switch). For a potential short circuit, current controller 110 may utilize Example Table 2 (e.g., a mapping between individual switches and phase ranges). In this case, current controller 110 may identify the first switch associated with a 60 to 70-degree phase angle as T6 and the second switch associated with a 300-degree phase angle as T4.

[0175] In a further example, the current controller 110 can distinguish whether a deviation is caused by an open circuit or a closed circuit based on the range of differences between the amplitudes of individual waveforms. For example, inconsistencies within a first range of differences (e.g., relatively small differences, but greater than a predetermined threshold) may indicate a potential open circuit. In some cases, inconsistencies outside the first range of differences (or within a second range of differences) (e.g., relatively large differences) may indicate a potential short circuit. The current controller 110 may utilize other suitable techniques to distinguish whether a deviation is caused by an open circuit or a closed circuit. The current controller 110 may use any other suitable techniques to analyze and detect inconsistencies in the resulting plot, such as at least one of graphical analysis, point-by-point comparison, cross-correlation, etc. Inconsistencies may be identified or detected at other phase angles, not limited to the examples herein.

[0176] The current controller 110 can provide external devices with anomaly detection results, notifications, or alarms indicating any switch that may experience a fault and the type of fault (e.g., open circuit or short circuit). The current controller 110 can determine one or more actions in response to the detection of an anomaly. Actions may include at least one of the following: adjusting the operation of at least one of the switches 104 (e.g., the identified switch), initiating a fault recovery process, or shutting down the power conversion circuit 101 to prevent potential or further damage. In some cases, actions may include sending an analysis report or summary in response to completion, sending an alarm in response to the detection of an anomaly, or providing other instructions to operator devices or external devices.

[0177] In further embodiments, actions may include terminating operation, actively shutting down, or disconnecting devices that utilize or include the power conversion circuit 101. Actions may be predetermined by the operator or administrator and pre-configured for the detection circuit 102. The current controller 110 may determine one or more actions to be taken based on, but not limited to, at least one of the type of anomaly (e.g., open circuit or short circuit), the number of switches experiencing an anomaly, or the specific switches experiencing an anomaly. In some embodiments, after an anomaly is detected, one or more components may be replaced or repaired; for example, a notification may inform the operator to perform maintenance on one or more components of system 100.

[0178] The current controller 110 can communicate with other elements of the detection circuit 102 to perform actions. The current controller 110 can at least partially provide commands to control the operation of one or more switches 104. For example, the current controller 110 can send two orthogonal components (e.g., v0, v0) to the conversion circuit 112, indicating the desired voltage amplitude (or magnitude) and phase angle. conq and v cond The conversion circuit 112 can convert the quadrature components into voltage control signals (e.g., v) using at least example formula (2). conA v conB and v conC The conversion circuit 112 can provide control signals to the PWM circuit 114. The voltage control signals can control one or more of the switches 104 used for phases A, B, and C, respectively.

[0179] PWM circuit 114 can receive control voltage V from conversion circuit 112 conA v conB and v conCThe PWM circuit 114 can use a control voltage to generate a PWM signal (e.g., S1 to -S6) that controls the switching of the switches 104 of the power conversion circuit 101. The PWM circuit 114 can adjust the duty cycle (or other operation) of the PWM signal based on the control voltage to achieve a desired action, such as a desired output current, voltage, phase, etc. For example, if an open-circuit fault is detected, the PWM circuit 114 can modify the PWM signal to reduce stress on the affected switches, for example, by reducing the voltage or frequency. If a short-circuit fault is detected, the PWM circuit 114 can adjust the PWM signal to prevent further damage to the circuit, for example, by terminating the operation of one or more switches 104. The PWM circuit 114 can transmit the generated PWM signal to the drive circuit 116.

[0180] The drive circuit 116 can receive PWM signals from the PWM circuit 114. The drive circuit 116 can use these signals to control the gate drive devices of switches T1 to T6 in the power conversion circuit 101. For example, the drive circuit 116 can operate as an interface unit for controlling one or more elements of the power conversion circuit 101 or communicating with it. In some cases, the drive circuit 116 can be electrically coupled to the gate of individual switches 104 to control switching operations. The drive circuit 116 can operate the switches according to the PWM signals, including implementing corrective actions determined by the current controller 110. For example, the drive circuit 116 can turn off a switch experiencing a short circuit to prevent further damage or adjust the switching frequency to mitigate the effects of an open circuit. For example, the drive circuit 116 can provide feedback to the current controller 110 regarding the condition of the switches 104. In some configurations, the predetermined action may involve terminating the operation of all switches 104 in response to the detection of an anomaly, allowing the operator to initiate a maintenance procedure.

[0181] The detection circuit 102 may include other elements for performing other features or functions. For example, the detection circuit 102 may include interface circuitry 118, conversion circuitry 120, and voltage controller 122. Interface circuitry 118 may receive voltage signals (e.g., voltage data or values) measured or simulated in power conversion circuitry 101. Voltage signals (e.g., V...) ab V bc and V ca This can be associated with a three-phase system, where the first voltage signal (e.g., V) ab The second voltage signal (e.g., V) can be captured between nodes a and b downstream of inductor L. bc The third voltage signal (e.g., V) can be captured between nodes b and c downstream of inductor L. caThe voltage signal can be captured between nodes a and c downstream of inductor L. In some embodiments, the voltage signal can be used in conjunction with the current signal discussed herein, for example, to detect a potential fault in at least one of the switches 104 of the power conversion circuit 101. In some other cases, the voltage signal can be used for other purposes. The conversion circuit 120 can process the voltage signal for the voltage controller 122. Processing the voltage signal can include obtaining one or more quadrature components and other derivatives of the voltage value. The voltage controller 122 can use the processed voltage value to regulate the output voltage of the power conversion circuit 101. For example, regulation can involve, but is not limited to, maintaining the output voltage within a specified range, compensating for voltage fluctuations, or ensuring stable operation of the system at least one of these.

[0182] Voltage controller 122 can generate control signals based on processed voltage values ​​to achieve desired voltage regulation. These control signals can be sent to components such as PWM circuit 114, drive circuit 116, or in some cases to current controller 110 to send control signals to PWM circuit 114 or drive circuit 116. PWM circuit 114 can adjust the duty cycle of the PWM signal to regulate the output voltage. Drive circuit 116 can manage the gate drive of switches T1-T6 to implement the regulation. Other types of commands can be generated and communicated to PWM circuit 114 or drive circuit 116.

[0183] The detection circuit 102 may include a memory unit for storing analog values, measurements, signals, generated plots, generated tables, or other information received, acquired, or otherwise generated by the detection circuit 102 or the power conversion circuit 101. The memory unit may provide storage for data used during fault detection. The memory unit may store one or more actions to be performed when an anomaly is identified. The memory unit may store time intervals for when to perform analysis. The stored data may be accessed by one or more elements of the detection circuit 102 (or by other authorized driving devices or circuits) for analysis and comparison. The memory unit may store the results of the fault detection process, including identified faults and corresponding switches.

[0184] The detection circuit 102 can communicate with external systems or devices to report detected faults. This communication can be achieved via various interfaces, such as wired or wireless communication channels. The detection circuit 102 can send alarms or notifications to external systems, indicating the presence of a fault and specific switches that cause or may cause potential problems related to the operation or health of the power conversion circuit 101. Communication can be part of corrective actions to resolve detected faults.

[0185] The detection circuit 102 can be designed to operate in real time, continuously monitoring the operation of the power conversion circuit 101. Real-time monitoring enables the detection circuit 102 to responsively identify and report any faults, thereby ensuring the reliability and efficiency of the power conversion circuit 101. In some cases, one or more elements of the detection circuit 102 can operate at predefined intervals to detect potential faults. The detection circuit 102 can use various analysis techniques, such as current analysis, temperature analysis, and / or voltage spectrum analysis, to detect faults in the switch 104. The detection circuit 102 can be used in other circuits, not limited to the power conversion circuit 101 used for fault detection.

[0186] Figure 3 An example flowchart of a method 300 for detecting semiconductor device failure according to some embodiments is shown. Method 300 may be performed by one or more circuits or devices of at least system 100, such as in combination with… Figure 1 As described. For example, one or more operations of method 300 may be performed by detection circuit 102 to detect potential failure, malfunction, or anomaly of the IGBT (or switch) implemented in power conversion circuit 101. Therefore, the following discussion of method 300 may be referenced. Figures 1 to 2 Some of the element symbols used in at least one of them are taken as non-limiting examples. Furthermore, method 300 is merely an example and is not intended to limit any embodiment of this disclosure. Therefore, it should be understood that... Figure 3 Additional operations are provided before, during, and after Method 300, and this document may only briefly describe some of these other operations. Method 300 may be performed simultaneously and / or with the exception of... Figure 3 Any order other than the one described herein.

[0187] Method 300 may begin with operation 302 of receiving (or capturing) a current signal provided by multiple switches (e.g., switch 104, e.g., IGBT). One or more elements of the detection circuit (e.g., 102) (such as interface circuitry (e.g., 106)) may receive or capture a first current signal (I0) from the power conversion circuit (e.g., 101). Uav ), second current signal (I) Vav ) and the third current signal (I Wav The detection circuitry may be operatively coupled to the power conversion circuitry. The power conversion circuitry may include multiple switches (e.g., 104). Each of the switches may have three terminals, including, for example, a gate terminal, a first S / D terminal, and a second S / D terminal. Operation of method 300 may be performed or executed by at least the detection circuitry.

[0188] A first current signal may be provided by a first pair of switches (e.g., T1 and T2). A second current signal may be provided by a second pair of switches (e.g., T3 and T4). A third current signal may be provided by a third pair of switches (e.g., T5 and T6). The first pair of switches may include a first switch (e.g., T1 or 104A) and a second switch (e.g., T2 or 104B) operably forming a first phase bridge arm. The first bridge arm may be connected to node A in a power conversion circuit, such as by combining at least Figure 1 As described herein, the second pair of switches may include a third switch (e.g., T3 or 104C) and a fourth switch (e.g., T4 or 104D) operably forming a second phase bridge arm. The second bridge arm may be connected to node B in the power conversion circuit. The third pair of switches may include a fifth switch (e.g., T5 or 104E) and a sixth switch (e.g., T6 or 104F) operably forming a third phase bridge arm. The third bridge arm may be connected to node C in the power conversion circuit. Individual switches discussed herein may include or be implemented using an insulated-gate bipolar transistor (IGBT). In some cases, the switches may be implemented using other semiconductor devices or other types of transistors.

[0189] The current signal (e.g., first to third current signals) may include measurement signals captured by one or more sensors. Interface circuitry may receive sensor data corresponding to the current signal or value. In some cases, the current signal may include an analog signal derived from a simulation of the configuration for the power conversion circuitry. The simulation may involve simulating the current signal for the power conversion circuitry under conditions where all switches operate as expected or normally. In some cases, the simulation may involve simulating the current signal for the power conversion circuitry, where individual switches are in an open or short-circuit state for a relatively short period of time.

[0190] Method 300 can continue to operation 304, which converts the current signal into two quadrature components. One or more elements of the detection circuit (e.g., conversion circuit 108) can convert the first to third current signals into two quadrature components. The conversion circuit can use formulas or equations to perform the conversion. For example, the first of the two quadrature components (I... Dav It can be represented as or derived as: The second of the two orthogonal components (I) Qav It can be represented as or derived as: .

[0191] Then, method 300 can proceed to operation 306, which generates the first and second plots. One or more elements of the detection circuit (e.g., current controller 110) can generate the first and second plots, each based on two orthogonal components or a vector derived from the two orthogonal components. The vector derived from the two orthogonal components can be derived using a formula. For example, the current controller can generate a plot with amplitude (I) based on the following equation.Sav ) and phase angle ( The vector of ) , Amplitude and phase angle can represent one of multiple vectors to produce separate plots.

[0192] The plot may include or correspond to a waveform formed from vectors generated based on two orthogonal components. The current controller may generate a first plot (e.g., 200A or waveform 202) based on a first vector derived from an equation, wherein the corresponding first amplitude and corresponding first phase angle of the first vector can be calculated via first to third signals simulated under the assumption that the switches of the power conversion circuit are operating normally. For example, the first plot may represent a vector or value expected when measuring the power conversion circuit under normal operating conditions.

[0193] The current controller can generate a second plot (e.g., 200B or 200C, or waveform 204 or waveform 206) based on a second vector derived from the equation, wherein the corresponding second amplitude and corresponding second phase angle of the second vector can be calculated via first to third signals measured during operation of the power conversion circuit. In this case, the first plot can represent a vector or value measured by one or more sensors of the power conversion circuit during operation of the switch.

[0194] Method 300 may continue to operation 308, which detects anomalies based on a comparison of the first and second plots. One or more elements of the detection circuitry (e.g., current controller 110) may detect anomalies associated with at least one of the switches based on a comparison of the first and second plots (e.g., a comparison between an analog waveform and a measured waveform). The current controller may determine an anomaly in response to identifying an inconsistency between the first and second plots. For example, an inconsistency may refer to a difference between a first amplitude of the first plot and a second amplitude of the second plot at a certain phase angle. For a difference to be considered inconsistent or an anomaly, the difference may be greater than or equal to a predetermined threshold or range.

[0195] In some cases, based on the identified inconsistency, the current controller can determine that the anomaly is caused by an abnormal disconnection of one of the switches (e.g., at least one switch is in an open-circuit state). The inconsistency may occur between the first phase angle and the second phase angle. In other cases, based on the identified inconsistency, the current controller can determine that the anomaly is caused by an abnormal short circuit of one of the switches (e.g., at least one switch is in a short-circuit state).

[0196] In some cases, detection circuitry (e.g., a current controller) can identify one of the switches as open based on a relatively small offset between the plotted amplitudes. For example, if the inconsistency or offset falls within a first difference range (e.g., indicating a relatively small offset), the current controller can determine that the switch may be experiencing an open-circuit state. A relatively small offset indicates a relatively small deviation between the measured amplitude and the expected amplitude of the switch operating under normal conditions. This deviation can indicate that the switch is not conducting current normally.

[0197] In some cases, a current controller may identify one of the switches as short-circuited based on a relatively large offset between the plotted amplitudes. For example, if the inconsistency or offset is above a first difference range or falls within a second difference range greater than the first range (e.g., indicating a relatively large offset, such as two or three times larger than the offset for an open-circuit state), the current controller may determine that the switch is likely experiencing a short-circuit state. A relatively large offset when the switch is operating under normal conditions may indicate a relatively large deviation between the measured amplitude and the expected amplitude. This deviation may indicate that the switch is conducting current when it should not, potentially causing or indicating a short circuit. Detection circuitry may utilize or employ other techniques, such as, to distinguish between at least one switch experiencing an open-circuit state or a short-circuit state.

[0198] After identifying an anomaly, one or more elements of the detection circuit (e.g., drive circuit 116) may adjust one or more of the control signals of the respective gates controlling the first to third pairs of switches based on the identified anomaly. For example, the drive circuit may shut down or disable the switch (or an associated pair of switches) identified as potentially having an anomaly. The drive circuit may adjust the switching frequency of at least one switch associated with the anomaly. The drive circuit may perform other actions to control one or more switches. The control of the drive circuit may be predetermined or pre-configured. For example, the drive circuit may control one or more switches based on the type of anomaly or fault and / or the number of switches with an anomaly. In some embodiments, the detection circuit may provide a report to an external device (e.g., an operator device) including the results of the analysis. In some configurations, in response to the detection of an open circuit or short circuit associated with one of the switches, the detection circuit may disable or terminate the operation of the power conversion circuit and notify the operator to initiate a maintenance procedure. Other actions may be performed to prevent further damage to the system or circuitry, but are not limited to those discussed herein.

[0199] Example implementation of temperature method

[0200] Based on the current signal flowing through each of switches T1 to T6, the total power supply (P) can be calculated using the following formula / equation (3). 损失 ):

[0201] (3) .

[0202] In example formula (3), This can represent or indicate the switching power of each of switches T1 to T6. The second and third terms of example formula (3) can together represent the conduction loss of each of switches T1 to T6. Furthermore, It can represent the equivalent resistance of each of switches T1 to T6. It can represent the current signal flowing through each of switches T1 to T6. It can represent the phase of each of switches T1 to T6. It can represent the critical voltage of each of switches T1 to T6.

[0203] In some configurations, in addition to or in lieu of the current analysis, techniques, or methods discussed above, the detection circuit 102 may rely on the total power (P). 损失 ) and measured temperature (T) NTC To determine or estimate the mating surface temperature (T) 接合面 Based on the estimated contact surface temperature, the detection circuit 102 can determine whether any of the switches T1 to T6 is likely to malfunction or has already malfunctioned. For example, Figure 4 An example structure 400 is shown, according to some embodiments, including a power conversion circuit 101 (embodied as chip 402) coupled to a heat sink 406. Figure 5 The combination shown according to some embodiments is at least Figure 4 The description of the thermal model 500 of structure 400 is a block diagram.

[0204] exist Figure 4 In this configuration, heat sink 406 can be formed as the bottom layer, providing thermal management by dissipating heat from components or materials above it. The heat sink can be coupled to the base plate 404 of the component, for example, Figure 1The system 100 has a baseplate including a power conversion circuit 101 (embodied as a chip 402). The baseplate 404 can serve as a mechanical support, aiding in heat dissipation. The baseplate 404 can be manufactured or positioned above a heat sink 406. One or more direct copper bonded (DCB) layers can be formed or manufactured above the baseplate 404. For example, the DCB layers may comprise a ceramic substrate bonded with copper on both sides. For example, the DCB layers may provide electrical insulation and thermal conductivity between at least the chip 402 and the baseplate 404. The chip 402 and a negative temperature coefficient (NTC) thermistor can be formed above the DCB layers. The chip 402, including or corresponding to the power conversion circuit 101, may correspond to a semiconductor device, for example, including at least an IGBT. The chip 402 and / or the NTC can be bonded to the heat sink 406 via the baseplate 404 and the DCB layers using at least one suitable bonding technique, such as soldering, applying adhesive, etc. The NTC thermistor can be positioned adjacent to the chip 402. The NTC thermistor can monitor the temperature of chip 402 or the area surrounding chip 402. The NTC thermistor can, for example, provide feedback, including thermal data for thermal management, to detection circuit 102 or other circuits authorized to receive thermal data for analysis. Although Figure 4 Not shown in the figure, but the detection circuit 102 may be electrically or communicatively coupled to the chip 402 (e.g., power conversion circuit 101) to use the temperature data for analysis.

[0205] The temperature of the mating surface can be indicated Figure 4 The star symbol shown (e.g., 408) or represented by it. The junction temperature can reflect the actual temperature when the power conversion circuit 101 is operating. The measured temperature can refer to the temperature that can be measured by any suitable method, such as via at least one temperature sensor and other types of sensors or measurements. During the operation of the power conversion circuit 101, the detection circuit 102 can calculate the total power (which corresponds to...) Figure 5 P in v Furthermore, using a set of data corresponding to the acquired measured temperature, the power conversion circuit 101 can be based on... Figure 5 The thermal model shown is used to estimate the temperature of the mating surface. For example, in Figure 5 In the middle, multiple thermal resistance values ​​(R) thJNTC , R thJC , R thHA It can be based on various experimental or measurement data (T) 外壳 T 散热器 T 环境 (to model or otherwise provide)

[0206] Based on this principle and various experimental or measurement data, a tolerable boundary can be set / configured to define the relationship between the interface temperature and the current signal (e.g., current value) flowing through at least one of the switches 104, such as... Figure 6 As shown. Therefore, when operating the power conversion circuit 101, the detection circuit 102 can monitor (or plot) the relationship between the current signal and the junction surface temperature in real time, and determine whether the plot is surrounded by or within a boundary. The plot may include data points or vectors, correspond to data points or vectors, or be represented by data points or vectors. If the plot is surrounded by (or within) a boundary, the detection circuit 102 can determine that a particular switch is not prone to failure. On the other hand, if the plot is outside the boundary, the detection circuit 102 can determine that a particular switch is prone to failure.

[0207] For example, Figure 6 Instructions and instructions according to some embodiments are shown. Figure 4 Figure 600 illustrates the relationship between the current signals of one or more switches (e.g., one or more of T1 to T6) of the power conversion circuit 101 associated with the example structure and the temperature of the junction surface. Figure 600 may include one or more boundaries 602 indicating the acceptable operating range of the switches. Boundaries 602 may be established at least based on the principles discussed above and / or experimental or measurement data.

[0208] Within the boundary (or an acceptable range), at least one of the switches 104 is expected to operate reliably without any failure. Outside the boundary, at least one of the switches 104 may be considered to be prone to failure or currently failing. Figure 600 may show real-time data points 604, 606, each representing the current signal and contact temperature of at least one individual switch. For illustrative purposes, data point 604 may be associated with one of the switches 104, and data point 606 may be associated with the other of the switches 104. For example, data point 606 may fall within boundary 602. Detection circuit 102 may determine that the corresponding switch for data point 606 is operating normally. Detection circuit 102 may identify data point 604 as falling outside boundary 602. In this case, detection circuit 102 may identify the switch associated with data point 604 as potentially failing or prone to failure.

[0209] The detection circuit 102 can use at least plotting 600 to monitor the performance of switches 104 within the power conversion circuit 101 (or other circuitry or apparatus). By analyzing the positions of data points 604, 606 relative to boundary 602, the detection circuit 102 can detect early signs of potential failure associated with at least one of the switches 104. For example, if data point 604 gradually moves towards the edge of boundary 602, the detection circuit 102 can interpret this indication as a potential failure of the individual switches under stress. The detection circuit 102 can take proactive measures, such as adjusting operating conditions or scheduling maintenance, to prevent complete failure of components or the system. This real-time monitoring and analysis ensures the reliability and efficiency of the power conversion circuit 101, minimizes downtime, and extends the lifespan of the switches 104.

[0210] Example implementation of the voltage spectrum method

[0211] In some configurations, in addition to or instead of the current and / or temperature analysis discussed above, the detection circuit 102 may rely on a voltage spectrum (or frequency response diagram) based on a Bode plot to determine whether any of the switches T1 to T6 is likely to malfunction or has already malfunctioned. Figures 7 to 10 The following are examples illustrating the analysis of at least [specific combinations]. Figure 1 Bode plots 700 to 1000 depict the health status of one or more switches 104 (e.g., IGBTs) in the power conversion circuit 101 of the described example system 100. Plots 700 to 1000 may be used by at least detection circuit 102 (or one or more elements of detection circuit 102) to detect potential faults in the power conversion circuit 101.

[0212] Each of the graphs 700-1000 may be a graphical representation of the frequency response of a system (e.g., power conversion circuit 101), illustrating the changes in the amplitude and phase of the signal relative to frequency. Graphs 700 to 1000 may be referred to as frequency response graphs, including voltage spectra for analysis performed by at least the detection circuit 102.

[0213] For example, the detection circuit 102 may store the first wave Testo 700 (e.g., Figure 7 As shown in the figure, this ensures that all switches T1 to T6 operate normally. In other words, the first wave diagram 700 can represent the frequency response when the switch 104 of the power conversion circuit 101 operates normally (without malfunction). As shown, when all switches T1 to T6 operate normally, there may be no spikes in any of the main frequencies f1, f2, and f3 (corresponding to the 1st, 2nd, and 3rd harmonics, respectively).

[0214] In certain situations, the second Porto 800 (e.g., Figure 8This could include spikes at the first to third harmonic frequencies f1, f2, and f3. In response to detecting spikes at the first to third harmonic frequencies, the detection circuit 102 can determine that switch T1 may be experiencing an open-circuit state. In another scenario, the third wave Testo 900 (e.g., Figure 9 This could include a spike at the second harmonic frequency f2. In this case, the detection circuit 102 can determine that switches T1 and T2 may be experiencing an open circuit state based on the spike detected at the second harmonic frequency f2.

[0215] In yet another scenario, the fourth Potevio 1000 (for example, Figure 10 The spikes may include those at the first and third harmonic frequencies (e.g., f1 and f3). In response to the detection of spikes at these harmonic frequencies, the detection circuit 102 may determine that switches T1 and T3 may be experiencing an open-circuit state. The detection circuit 102 may detect spikes based on the amplitude (or change in amplitude) in the plot being greater than or equal to a predetermined threshold value. The threshold value may be modified or configured by the operator of system 100. The predetermined threshold value may vary depending on the configuration of the power conversion circuit 101 or the specifications of the switches 104. For the purpose of providing examples, example table 3 may include example amplitudes at the first to third harmonic frequencies during different states of one or more switches 104.

[0216]

[0217] Example Table 3

[0218] In some cases, the detection circuit 102 may be deployed with graphical analysis techniques (or other types of analysis techniques) to detect one or more spikes at harmonic frequencies in the plot 800 to 1000. In some embodiments, one or more Bode plots (e.g., 700 to 1000) may include more than three harmonic frequencies (or in some cases, fewer than three harmonic frequencies). In some configurations, at least one of the different harmonic frequencies having the amplitude of a spike or individual spikes may be configured or predefined to represent at least one switch 104 experiencing a failure or malfunction and / or the type of failure associated with at least one switch 104.

[0219] In some cases, detection circuit 102 can determine which of the switches 104 may be experiencing an open-circuit or short-circuit state based on the pattern of one or more spikes within Bode plots 700 to 1000. The spike patterns may be predetermined or simulated and stored in memory. Detection circuit 102 may utilize other analyses or historical data to detect whether at least one of the switches 104 is potentially experiencing a fault state or the type of fault experienced by at least one of the switches 104. It should be noted that figures 700 to 1000 are provided as examples and are not intended to limit other possible implementations or variations that may be made without departing from at least the features or functions described herein.

[0220] Figure 11 The diagram illustrates the use of some embodiments by Figure 1 An overview of the analytical techniques used by the instance system 100 for detecting semiconductor device failures 1100. Figure 11 The operation can be performed or carried out by one or more components of the instance system 100, such as combining at least Figure 1 The power conversion circuit 101 and detection circuit 102 are described. For example, at step 1100, one or more elements of the detection circuit 102 (e.g., interface circuit 106, conversion circuit 108, current controller 110, drive circuit 116, etc.) may perform the various operations discussed herein for protecting IGBT models from failure based on at least one analysis technique (e.g., current analysis, thermal or temperature analysis, and / or voltage spectrum analysis).

[0221] Step 1100 for protecting a failed IGBT (e.g., at least one switch 104) model may involve operation 1102. Operation 1102 may include, for example, monitoring output current, NTC thermistor readings, and (e.g., DC) bus voltage at 1104. Detection circuitry 102 may collect data for analyzing the health status of the IGBT (e.g., switch 104 of power conversion circuitry 101).

[0222] At step 1106, the detection circuit 102 can predict the transition values ​​from voltage, current, and / or temperature, and compare these values ​​with standard values ​​(e.g., values ​​representing normal operation of switch 104). For example, the transitions from voltage, current, and / or temperature may refer to values ​​or data obtained during voltage spectrum analysis, current analysis, and / or temperature analysis. If the predicted values ​​deviate significantly from the standard values, the detection circuit 102 can identify or mark the IGBT as potentially malfunctioning.

[0223] At step 1106, the detection circuit 102 predicts the values ​​of voltage, current, and temperature, and compares them with standard values. This step involves using a predictive algorithm to estimate the future values ​​of these parameters based on historical data and current operating conditions. The predicted values ​​can be compared with predefined standard values ​​to identify any potential problems. If the predicted value deviates significantly from the standard value, the detection circuit 102 can mark the IGBT as potentially malfunctioning.

[0224] Step 1106 may involve functional matching 1112, including techniques 1114 for data processing, such as neural networks, machine learning, big data mining, and / or artificial intelligence (AI). It should be noted that the detection circuit 102 may utilize any suitable data processing techniques, machine learning models, or AI algorithms to process the input to produce an output (e.g., whether at least one of the switches 104 is potentially experiencing any malfunction, such as an open-circuit or short-circuit state). Such techniques can be used to improve the accuracy and reliability of the analysis. Data can be collected from a variety of sources, including at least one of UPS / VFD systems (e.g., operating modes, overload or over-protection functions), process areas (e.g., location, time, device, batch, carrier, or transmission priority), or interface area sensors (e.g., voltage, current, or temperature), to name a few. A comprehensive data collection approach ensures that the detection circuit 102 can access all desired information for accurate analysis. The detection circuit 102 can utilize the data for various types of analysis, including but not limited to current analysis, thermal analysis, and / or voltage spectrum analysis discussed herein.

[0225] For example, current analysis may involve checking for intensity anomalies in each phase at least during each cycle and tracking the extent of the anomalies to identify which phase of the IGBT (e.g., a switch) is experiencing a problem. In another example, thermal or temperature analysis may involve using at least a large database from device operation to align with the boundaries of standard conditions and scenarios, and defining mode offsets to predict potential problems and schedule preventative maintenance. In a further example, voltage spectrum analysis may involve calculating the voltage spectrum under normal operating conditions and comparing the calculated spectrum with potential deviations at each harmonic order. Detected anomalies may serve as indicators of potential failure of at least one of the IGBTs (e.g., one of the switches 104 in power conversion circuit 101).

[0226] At step 1108, the detection circuit 102 may analyze feedback from key points to generate an IGBT health index. For example, based on analysis using at least one of the analytical techniques (e.g., analysis of current, temperature, or voltage data), the detection circuit 102 may calculate a health index representing the overall state of the IGBTs within the system.

[0227] At step 1110, the detection circuit 102 may check the pre-warning multiple times (e.g., five times). If the pre-warning conditions are met, the detection circuit 102 may display a warning, provide a notification, or otherwise send an alarm to a remote device or external system (such as a supervisory control and data acquisition (SCADA) system). Otherwise, the pre-warning may be cleared. The detection circuit 102 may perform checks to ensure that any potential problems are promptly identified and resolved to prevent further damage. Furthermore, the checks reduce false positives.

[0228] The detection circuit 102 can repeat operation 1102, for example, from steps 1104 to 1110. The detection circuit 102 can repeat operation 1102 at predetermined time intervals or continuously in response to a trigger from the operator. Data monitoring and analysis can be performed in real time to provide timely intervention and maintenance to ensure the reliable operation of the power conversion circuit 101 and other systems, not limited to the power conversion circuit 101.

[0229] In one embodiment of this disclosure, an apparatus is disclosed. This apparatus includes: a power conversion circuit comprising a plurality of switches, each of the plurality of switches having three terminals; and a detection circuit operatively coupled to the power conversion circuit and configured to: acquire a first current signal (I0). Uav ), second current signal (I) Vav ) and the third current signal (I Wav The first current signal is provided by a first pair of the plurality of switches, the second current signal is provided by a second pair of the plurality of switches, and the third current signal is provided by a third pair of the plurality of switches; these first to third current signals are converted into two orthogonal components; a first plot and a second plot are generated based on these two orthogonal components; and an anomaly associated with at least one of the plurality of switches is detected by comparing the first plot and the second plot.

[0230] In some embodiments, the first pair of switches includes a first switch and a second switch operably forming a first phase bridge arm, the second pair of switches includes a third switch and a fourth switch operably forming a second phase bridge arm, and the third pair of switches includes a fifth switch and a sixth switch operably forming a third phase bridge arm.

[0231] In some embodiments, one of the two orthogonal components (I) Dav This is expressed as: The second of the two orthogonal components (I) Qav This is expressed as: .

[0232] In some embodiments, the detection circuit is further configured to generate an amplitude (I) based on the following equation. Sav ) and a phase angle ( A vector. , .

[0233] In some embodiments, the detection circuit is further configured to generate the first plot based on a first vector derived from the equation, wherein a corresponding first amplitude and a corresponding first phase angle of the first vector are calculated via first to third signals simulated under the assumption that the switches of the power conversion circuit are operating normally.

[0234] In some embodiments, the detection circuit is further configured to generate the second plot based on a second vector derived from the equation, wherein a corresponding second amplitude and a corresponding second phase angle of the second vector are calculated via the first to third signals measured during operation of the power conversion circuit.

[0235] In some embodiments, the detection circuitry is further configured to determine the anomaly in response to identifying an inconsistency between the first drawing and the second drawing.

[0236] In some embodiments, the inconsistency occurs between the first amplitude and the second amplitude.

[0237] In some embodiments, based on the identified inconsistency, the detection circuit is further used to determine that the anomaly is caused by an abnormal disconnection of one of the switches.

[0238] In some embodiments, the inconsistency occurs between the first phase angle and the second phase angle.

[0239] In some embodiments, based on the identified inconsistency, the detection circuit is further used to determine that the anomaly is caused by an abnormal short circuit in one of the switches.

[0240] In some embodiments, each of the switches includes an insulated gate bipolar transistor.

[0241] In another embodiment of this disclosure, a detection circuit is disclosed. This detection circuit includes: an interface circuit for receiving a first current signal (I0). Uav ), second current signal (I) Vav ) and the third current signal (I WavThe first current signal is provided by a first pair of switches coupled to a power supply voltage, the second current signal is provided by a second pair of switches coupled to the power supply voltage, and the third current signal is provided by a third pair of switches coupled to the power supply voltage; a conversion circuit is used to: (i) convert the first to third current signals into two quadrature components; (ii) generate a first plot and a second plot based on the two quadrature components; and (iii) identify an anomaly associated with at least one of the first to third pairs of switches based on a comparison of the first plot and the second plot; and a drive circuit is used to adjust one or more of a plurality of control signals controlling the first to third pairs of switches respectively based on the identified anomaly.

[0242] In some embodiments, each of the first to third pairs of switches includes an insulated gate bipolar transistor.

[0243] In some embodiments, one of the two orthogonal components (I) Dav This is expressed as: The second of the two orthogonal components (I) Qav This is expressed as: .

[0244] In some embodiments, the conversion circuit is further configured to generate an amplitude (I) based on the following equation. Sav ) and a phase angle ( A vector. , .

[0245] In some embodiments, the conversion circuit is further configured to perform the following steps: Generating a first plot based on a first vector derived from the equation, wherein a corresponding first amplitude and a corresponding first phase angle of the first vector are calculated via first to third signals simulated under the assumption that the first to third pairs of switches in a power conversion circuit are each operating normally. Generating a second plot based on a second vector derived from the equation, wherein a corresponding second amplitude and a corresponding second phase angle of the second vector are calculated via first to third signals measured during operation of the power conversion circuit. Determining an anomaly in response to identifying an inconsistency between the first plot and the second plot exceeding a threshold value.

[0246] In another embodiment of this disclosure, a method for detecting a semiconductor device failure is disclosed. This method includes: receiving a first current signal (I0). Uav ), second current signal (I) Vav ) and the third current signal (I WavThe first current signal is provided by a first pair of switches coupled to the power supply voltage, the second current signal is provided by a second pair of switches coupled to the power supply voltage, and the third current signal is provided by a third pair of switches coupled to the power supply voltage; these first to third current signals are converted into two quadrature components; a first plot and a second plot are generated based on these two quadrature components; an anomaly associated with at least one of the first to third pairs of switches is identified by comparing the first plot and the second plot; and one or more of the multiple control signals controlling the first to third pairs of switches are adjusted based on the identified anomaly.

[0247] In some embodiments, each of the first to third pairs of switches includes an insulated gate bipolar transistor.

[0248] In some embodiments, one of the two orthogonal components (I) Dav This is expressed as: The second of the two orthogonal components (I) Qav This is expressed as: The method further includes generating a vector for constructing each of the first and second drawings, the vector having an amplitude (I) based on the following equation. Sav ) and a phase angle ( ). , .

[0249] As used herein, the terms “about” and “approximately” generally indicate a value of a given quantity that may vary based on a particular technology node associated with the subject semiconductor device. Based on a particular technology node, the term “about” indicates a value of a given quantity that varies, for example, within 10 to 30% of this value (e.g., +10%, ±20%, or ±30% of this value).

[0250] The foregoing has outlined features of several embodiments to enable those skilled in the art to better understand the appearance of an embodiment of this disclosure. Those skilled in the art should understand that they can readily use an embodiment of this disclosure as the basis for designing or modifying other processes and structures to achieve the same objectives and / or advantages as the embodiments described herein. Those skilled in the art should also recognize that such equivalent constructions do not depart from the spirit and scope of an embodiment of this disclosure, and that various changes, substitutions, and modifications can be made herein without departing from the spirit and scope of an embodiment of this disclosure.

Claims

1. A failure detection system, characterized in that, Include: A power conversion circuit includes a plurality of switches, each of which has three terminals; and A detection circuit, operatively coupled to the power conversion circuit and used for: A first current signal, a second current signal, and a third current signal are acquired, wherein the first current signal is provided by a first pair of switches, the second current signal is provided by a second pair of switches, and the third current signal is provided by a third pair of switches. These first to third current signals are converted into two orthogonal components; A first plot and a second plot are generated based on these two orthogonal components; and An anomaly associated with at least one of the switches is detected by comparing the first drawing and the second drawing.

2. The failure detection system as described in claim 1, characterized in that, The first pair of switches includes a first switch and a second switch operably forming a first phase bridge arm; the second pair of switches includes a third switch and a fourth switch operably forming a second phase bridge arm; and the third pair of switches includes a fifth switch and a sixth switch operably forming a third phase bridge arm.

3. The failure detection system as described in claim 1, characterized in that, Among the two orthogonal components, the first one I Dav It is expressed as: ;and Among the two orthogonal components, the second one I Qav It is expressed as: ， Where I Uav This indicates the first current signal, I Vav This indicates the second current signal and I. Wav This indicates the third current signal.

4. The failure detection system as described in claim 3, characterized in that, The detection circuit is further used to generate an amplitude I based on the following equation. Sav and a phase angle A vector: , 。 5. The failure detection system as described in claim 4, characterized in that, The detection circuit is further used to generate the first plot based on a first vector derived from the equation, wherein a corresponding first amplitude and a corresponding first phase angle of the first vector are calculated by simulating the first to third signals under the assumption that the switches of the power conversion circuit are operating normally.

6. The failure detection system as described in claim 5, characterized in that, The detection circuit is further used to generate the second plot based on a second vector derived from the equation, wherein a corresponding second amplitude and a corresponding second phase angle of the second vector are calculated via the first to third signals measured during the operation of the power conversion circuit.

7. The failure detection system as described in claim 6, characterized in that, The detection circuit is further used to determine the anomaly in response to identifying an inconsistency between the first drawing and the second drawing.

8. The failure detection system as described in claim 7, characterized in that, The inconsistency occurs between the first amplitude and the second amplitude.

9. A detection circuit, characterized in that, Include: An interface circuit is used to receive a first current signal, a second current signal and a third current signal, wherein the first current signal is provided by a first pair of switches coupled to a power supply voltage, the second current signal is provided by a second pair of switches coupled to the power supply voltage, and the third current signal is provided by a third pair of switches coupled to the power supply voltage. A conversion circuit is used to: (i) convert the first to third current signals into two quadrature components; (ii) generate a first plot and a second plot based on the two quadrature components respectively; and (iii) identify an anomaly associated with at least one of the first to third pairs of switches by comparing the first plot and the second plot. and A drive circuit is used to adjust one or more of the multiple control signals of the first to third pairs of switches respectively based on the identified anomaly.

10. A failure detection method, characterized in that, Include: Receive a first current signal, a second current signal and a third current signal, wherein the first current signal is provided by a first pair of switches coupled to a power supply voltage, the second current signal is provided by a second pair of switches coupled to the power supply voltage, and the third current signal is provided by a third pair of switches coupled to the power supply voltage; These first to third current signals are converted into two orthogonal components; A first plot and a second plot are generated based on these two orthogonal components; An anomaly associated with at least one of the first to third pairs of switches is identified by comparing the first and second drawings. and Based on the identified anomalies, one or more of the multiple control signals of the first to third pairs of switches are adjusted respectively.

Images