WO2021014282A1

WO2021014282A1 - Overheating detection in an electric drive based on monitoring of thermal behaviour of semiconductor modules of the electric drive

Info

Publication number: WO2021014282A1
Application number: PCT/IB2020/056672
Authority: WO
Inventors: Anurag Nandwana; Abhilash PANI; Arpit Sisodia
Original assignee: Abb Schweiz Ag
Priority date: 2019-07-24
Filing date: 2020-07-16
Publication date: 2021-01-28

Abstract

Examples of overheating detection in an electric drive are disclosed. In an example, measurements of operating parameters of the electric drive are obtained. One of a junction temperature and a casing temperature for the one or more semiconductor modules are estimated based on a data model of the electric drive and the measurements of the operating parameters. The data model is generated from historical data of measurements of the operating parameters of the electric drive. An index associated with overheating of the one or more semiconductor modules is determined. The index is continuously to detect the overheating of the one or more semiconductor modules. In response to the index being higher than a threshold value for the index, the overheating is detected and an indication is generated for an operator to undertake preventive action.

Description

OVERHEATING DETECTION IN AN ELECTRIC DRIVE BASED ON MONITORING OF

THERMAL BEHAVIOUR OF SEMICONDUCTOR MODULES OF THE ELECTRIC DRIVE

TECHNICAL FIELD

[0001] The present invention relates, in general to monitoring of semiconductor modules in electric drives. More specifically, the present invention relates to detection of overheating in an electric drive based on monitoring of thermal behaviour of semiconductor modules of the electric drive.

BACKGROUND

[0002] Generally, an electric drive is used to control operation of an electric motor. In some examples, the electric drive may be used to control operation of other electrical machines. The electric drive consists of various electrical components that provide controlled power output used to run the electric motor. The major electrical components in the electric drive include power electronic converters, filter circuits, and controller circuits. Various stresses on electric drive may result in gradual degradation of above mentioned electrical components resulting in failure of the electric components which consequently may result in a failure of the electric drive and therefore failure of the electric motor operation. Such breakdowns may result in unplanned downtime of entire downstream process and might incur large costs to the business.

[0003] Power electronic converters of an electric drive is a crucial and expensive component and failure of the power electronic converter may lead to failure of the electric drive. Failures in power electronic converter may be mainly because of failures in semiconductor devices, such as Insulated-Gate Bipolar Transistors (IGBTs), Metal Oxide Semiconductor Field Effect Transistors (MOSFETs) present inside the power electronic converters.

BRIEF DESCRIPTION OF DRAWINGS

[0004] The following detailed description references the drawings, wherein: [0005] Fig. 1A illustrates a block diagram of a condition monitoring system for detecting overheating in an electric drive based on monitoring of thermal behaviour of one or more semiconductor modules of the electric drive, according to an example;

[0006] Fig. 1 B illustrates a sectional view of one or more semiconductor modules enclosed within a casing in an electric drive, according to an example;

[0007] FIG. 2 illustrates a block diagram of a condition monitoring system, according to an example;

[0008] Fig. 3 illustrates a graph depicting actual values of junction temperature and estimated values of junction temperature plotted over a period of time, according to an example;

[0009] Fig. 4 illustrates a graph depicting number of overheating instants (number of anomalies) per day as detected by the condition monitoring system, according to an example; and

[0010] Figs. 5A and 5B illustrate a method for detecting overheating in an electric drive based on monitoring of thermal behaviour of one or more semiconductor modules of the electric drive, according to an example.

DETAILED DESCRIPTION

[0011] Conventionally, casing temperatures of the semiconductor modules in the electric drive are measured for the purpose of monitoring the operation of the electric drive. As an example method, two temperature measurements can be been taken from two different location of a baseplate of an IGBT module or casing of power electronics device. Two ageing indicators may be calculated based on the temperature measurements at these locations. These two aging indicators indicates two different kind of degradation. These indicators may be compared with predefined threshold to indicate aging of switches in the IGBT module. However, this method is designed for steady state operation and does not consider the fluctuation of temperature of the cooling system of the IGBT module. Moreover, getting a steady state operation for electric drive is rare as load on the electric drive and switching frequency changes frequently. Thus, this method may have a limited practical application. [0012] In another method an online monitoring of IGBT may be performed where, the IGBT to be tested is compared with a test IGBT by inputting driving voltage. Short-circuit current is measured for both to be tested and test IGBT and compared to detect aging of IGBT. Since, the test IGBT and to be tested IGBT are in a bridge circuit therefore a design change is necessary.

[0013] According to another method an on state collector emitter voltage drop and voltage of solder layer of IGBT may be measured periodically and compared with the base value when IGBT was not in use. A deviation of these measurement from base value beyond failure standards may be monitored which indicates poor condition of IGBT. However, measurement of collector emitter voltage drop and voltage of solder layer of IGBT are complex. Furthermore, this method is offline and the measurements have to be taken periodically and compared with base values.

[0014] The approaches of the present invention, discloses a system and method for health assessment of semiconductor devices present within the power electronic converters in an electric drive. The present invention enables real-time health assessment of semiconductor devices within the electric drive and thereby helps to anticipate and plan for potential downtime. Additionally, preventive measures can be taken after identification of health degradation of the semiconductor devices to avoid replacement of the entire electric drive. Thus, the cost of maintenance of the electric drives may be reduced and unplanned downtime due to electric drive failure may be avoided.

[0015] According to an example of the present invention, methods and systems for detecting overheating in an electric drive based on monitoring of thermal behavior of one or more semiconductor modules of the electric drive is disclosed. The one or more semiconductor modules are enclosed within a casing, where each semiconductor module comprises a plurality of semiconductor switches. The method of the present invention includes obtaining measurements of operating parameters of the electric drive with one or more sensors of the electric drive. Each the one or more sensors is a current sensor, a voltage sensor, a temperature sensor and a soft sensor. The operating parameters of the electric drive include current of the electric drive, switching frequency, a measured casing temperature, an ambient temperature, and a coolant temperature.

[0016] One of a junction temperature and a casing temperature for the one or more semiconductor modules may be estimated based on a data model of the electric drive and the measurements of the operating parameters. The junction temperature is representative of an average of one or more junction temperatures associated with the one or more semiconductor modules and the casing temperature is representative of an average temperature of the surface of the casing. The data model is generated from historical data of measurements of the operating parameters of the electric drive and interrelation between the operating parameters being established based on thermal relationship between components of the electric drive.

[0017] The method further includes determining an index associated with overheating of the one or more semiconductor modules, where the index is a difference between one of the estimated junction temperature and a reference junction temperature and the estimated casing temperature and a reference casing temperature.

[0018] The index is continuously monitored to detect the overheating of the one or more semiconductor modules. In response to the index being higher than a threshold value for the index overheating of the one or more semiconductor modules is detected. In response to detecting the overheating, an indication is generated for an operator to undertake preventive action to reduce the overheating

[0019] Thus, the present invention enables detection of overheating of one or more semiconductor modules in an electric drive and enables the operator to undertake preventive action before failure of the electric drive thereby avoiding downtime. Also, the method of the present invention is executed in real-time while the electric drive is in operation, thereby facilitating real-time health monitoring of the electric drive. Further, the present invention does not require complex measurements of on state collector emitter voltage drop and voltage of solder layer of IGBT for health monitoring, as may required for other health monitoring techniques. The present invention also takes into account the fluctuation of temperature of the cooling system of the semiconductor module for health monitoring of the electric drive. Also, no additional components or modifications to the circuitry of the electric drive is required for implementation of the techniques of the present invention.

[0020] The following detailed description refers to the accompanying drawings. Wherever possible, the same reference numbers are used in the drawings and the following description to refer to the same or similar parts. While several examples are described in the description, modifications, adaptations, and other implementations are possible. Accordingly, the following detailed description does not limit the disclosed examples. Instead, the proper scope of the disclosed examples may be defined by the appended claims.

[0021] Fig. 1A illustrates a block diagram of a condition monitoring system 100 for detecting overheating in an electric drive based on monitoring of thermal behaviour of one or more semiconductor modules of the electric drive. In an example, the condition monitoring system 100 or system 100 may be used for monitoring overheating in converters or inverters. In an example, the system 100 may be an edge device, a personal computer, a cloud server, a gateway, or the like.

[0022] Fig. 1A shows electric drives 102-1 , 102-2, 102-3. 102-N, collectively referred to as electric drives 102. The system 100 may be communicatively coupled to an electric drive, converter, or inverter, through a remote monitoring device. As shown in Fig. 1A, the system 100 is coupled to the electric drives 102 via a remote monitoring device 104. A data communication between the remote monitoring device 104 and the system 100 may be established through an ethernet connection or via a private network or the internet.

[0023] The description hereinafter is explained in respect of a single electric drive referenced as 102. In an example, the remote monitoring device 104 may be connected directly to a panel bus of the electric drive 102, if the electric drive 102 has an Assistant control panel. In such an example, the data link connections may be made through RJ45 connectors. In another example, the electric drive 102 may have a control unit where the data link connections may be made through fiber optic links. The remote monitoring device 104 may obtain the measurements of different parameters from the electric drive 102.

[0024] In an example, the electric drives 102 include sensors (not shown) for measurements of different parameters. The sensors may be a current sensor, a voltage sensor, a temperature sensor and a soft sensor. The sensors may measure operating parameters of the electric drive 102. The operating parameters include current of the electric drive 102, switching frequency, a measured casing temperature, an ambient temperature, and a coolant temperature.

[0025] Each of the electric drives 102 has one or more semiconductor modules 106. Each semiconductor module includes a plurality of semiconductor switches (not shown). The semiconductor modules 106 of an electric drive may be enclosed within a casing. The one or more semiconductor modules of an electric drive are enclosed within a casing (not shown in Fig. 1A). Fig. 1B illustrates a sectional view of a collection of one or more semiconductor modules enclosed within a casing in an electric drive, such as the electric drive 102. In Fig. 1B, the one or more semiconductor modules is represented as a block 108 which are encapsulated in a casing 110. The casing 110 may also be referred as a base plate. The casing 110 may have fins on its outer surface through which a coolant fluid may be passed.

[0026] The condition monitoring system 100 includes a temperature estimation module 112 and a fault detection module 114. The temperature estimation module 112 and the fault detection module 114 may be implemented as either software installed within the condition monitoring system 100, or as hardware in the form of electronic circuitry. In an example, the temperature estimation module 112 and the fault detection module 114 may be coupled with a processor of the system 100. The present invention is capable of detecting overheating of the electric drive 102 and thereby provide an indication to the operator to take protective measure consequently preventing downtime and higher degradation of the semiconductor modules in the electric drive 102.

[0027] In an example, the system 100 includes a transceiver which receives measurements of the operating parameters of the electric drive 102 from the remote monitoring device 104. The measurements may be obtained through one or more sensors of the electric drive 102, where each of the one or more sensors is a current sensor, a voltage sensor, a temperature sensor and a soft sensor. In an example, the remote monitoring device 104 may transmit the measurements of the operating parameters to the system 100.

[0028] The temperature estimation module 112 coupled to the processor estimates one of a junction temperature and a casing temperature for the one or more semiconductor modules based on a data model of the electric drive 102 and the measurements of the operating parameters. The junction temperature is representative of an average of one or more junction temperatures associated with the one or more semiconductor modules. With reference to Fig. 1 B, the junction temperature may be an average temperature of a surface the one or more semiconductors modules 108, where the surface is between the one or more semiconductor modules 108 and the casing 110. The junction temperature is referenced as Tj in Fig. 1B. The casing temperature is representative of an average temperature of the surface of the casing. With reference to Fig. 1 B, the casing temperature may be an average temperature of a surface the casing 110, where the surface is between the casing 110 and the coolant. The casing temperature is referenced as Tc in Fig. 1 B.

[0029] The data model is generated from historical data of measurements of the operating parameters of the electric drive 102 and interrelation between the operating parameters being established based on thermal relationship between components of the electric drive 102. The details of the data model and estimation based on the data model are elaborated later in the description with reference to Fig. 2. [0030] The fault detection module 114 coupled to the processor determine an index assodated with overheating of the one or more semiconductor modules, where the index is a difference between one of the estimated junction temperature and a reference junction temperature and the estimated casing temperature and a reference casing temperature. The fault detection module 114 continuously monitors the index to detect the overheating of the one or more semiconductor modules. In response to the index being higher than a threshold value for the index, overheating of the one or more semiconductor modules is detected and in response to detecting the overheating, an indication is generated for an operator to undertake preventive action to reduce the overheating. Thereby, the present invention detects overheating and enables the operator to carry out/schedule maintenance activity. These and other aspects are further described in conjunction with Figs. 2 to 5B.

[0031] FIG. 2 illustrates a block diagram of a condition monitoring system 200, in accordance with an example. The system 200 includes processors) 202 and a memory 204. The processor(s) 202 may be a single processing unit or a number of units, all of which could include multiple computing units. The processor(s) 202 may be implemented as one or more microprocessor, microcomputers, digital signal processors, central processing units, state machines, logic circuitries, and/or any devices that manipulate signals based on operational instructions. Among other capabilities the processor(s) 202 are adapted to fetch and execute processor-readable instructions stored in the memory 204 to implement one or more functionalities. In an example, the system 200 may be a doud server having processing capabilities distributed over different nodes of the cloud server.

[0032] The memory 204 may be coupled to the processor(s) 202. The memory 206 may indude any computer-readable medium known in the art induding, for example, volatile memory, such as Static Random-Access Memory (SRAM) and Dynamic Random-Access Memory (DRAM), and/or non-volatile memory, such as Read Only Memory (ROM), Erasable Programmable ROMs (EPROMs), flash memories, hard disks, optical disks, and magnetic tapes. In an example, the system 200 may be a cloud server having data storage capabilities distributed over different storage devices or databases of a cloud network which are accessible by the cloud server.

[0033] The system 200 includes interface(s) 206. The interface(s) 206 may include a variety of software and hardware enabled interfaces. The interface(s) 206 may enable the communication and connectivity between the system 200 and other components of the network, such as the remote monitoring device 104. The interface(s) 206 may facilitate multiple communications within a wide variety of protocols and may also enable communication with one or more computer enabled terminals or similar network components.

[0034] The system 200 further includes module(s) 208. The module(s) 208 may be implemented as a combination of hardware and programming (for example, programmable instructions) to implement a variety of functionalities of the module(s) 208. In examples described herein, such combinations of hardware and programming may be implemented in several different ways. For example, the programming for the module(s) 208 may be executable instructions. Such instructions in turn may be stored on a non-transitory machine-readable storage medium which may be coupled either directly with the system 200 or indirectly (for example, through networked means). In case implemented as a hardware, the module(s) 208 may include a processing resource (for example, either a single processor or a combination of multiple processors), to execute such instructions. In the present examples, the processor-readable storage medium may store instructions that, when executed by the processing resource, implement module(s) 208. In other examples, module(s) 208 may be implemented by electronic circuitry.

[0035] In an example, the module(s) 208 include the temperature estimation module(s) 112. In addition, the module(s) 208 may further include the fault detection module 114, and other module(s) 210. The other module(s) 214 may implement functionalities that supplement applications or functions performed by the system 200 or any of the module(s) 208. In addition, the system 200 may further include a transceiver 212 and other component(s) 214. Such other component(s) 214 may include a variety of other electrical components that enable functionalities of managing and monitoring the electric drive.

[0036] In operation, the system 200 may obtaining measurements of operating parameters of the electric drive 102 with one or more sensors of the electric drive 102 through the transceiver 212. The system 200 may receive the measurements from the remote monitoring device, such as the remote monitoring device 104.The measurements may be taken by one or more sensors, such as a current sensor, a voltage sensor, a temperature sensor and a soft sensor. In an example, the operating parameters include current of the electric drive, switching frequency, a measured casing temperature ambient temperature, and coolant temperature.

[0037] The temperature estimation module 112 may use a data model for estimation of one of the junction temperature and the casing temperature based on the obtained measurements of the operating parameter. In an example, the temperature estimation module 112 may be trained using the data model which is generated from historical data of measurements of the operating parameters of the electric drive and interrelation between the operating parameters being established based on thermal relationship between components of the electric drive. Through the following equations the data model for estimation of junction temperature is elaborated. The equations below show a relationship between various temperatures and measurements at“t-1”. The invention is not restricted to the measurements of the previous instance. For example, the junction temperature can be estimated with measurements at "t" or "t-n", or a combination of these, and such variations will be apparent to those skilled in the art.

[0038] A similar data model may be used for estimation of the casing temperature. The change in junction temperature of a semiconductor module at any instant depends upon the heat generated by switches in the module and heat lost to ambient and coolant system through the casing. Following equation models the change in junction temperature D7,(0 at time instant t.

5

Where C₁ ,C₂ and C₃ depend on the material, interface properties and the interfacial area. The junction temperature at a particular time instant“t” may be referred to as 7,(0 as expressed in equation 2.

[0039] S represents the heat generation from a semiconductor device.

Cond-loss is conduction loss in the semiconductor device and is dependent on ON-state current I² and ON-state resistance R_on. R_on has positive temperature coefficient and hence E_Cond-loss will increase with operating temperature.

E _{switching -loss} is the switching loss and depends on switching frequency of the semiconductor device.

[0040] Therefore, based on the above, the junction temperature at any instant depends on operating parameters of the electric drive, such as, current of the electric drive, switching frequency, a measured casing temperature and ambient temperature. Putting it into an equation, junction temperature at time instant t may be expressed as,

25

[0041] A linear model can be used to model 7,(0 as,

In case of measurement of the ambient temperature T_A is not available, K₆ x T_A(t - 1) term in previous equation (7) can be approximated as a constant intercept term 7} (intercept). Thus, 5

[0042] The coefficients of above equation (8) are unknown and can be estimated using coefficient identification method. Based on the above equations, and the historical data of the operating parameters of the electric drive for the (t-1)^th instant, the temperature estimation module 112 may be trained to estimate the junction temperature and an estimation error E

based on error in training.

Where is the estimated junction temperature

Although estimation of junction temperature is illustrated in the above example, the temperature estimation module 112 can estimate the casing temperature using a similar technique. In an example, the drive operating parameters used to estimate the casing temperature include current of the electric drive, switching frequency, coolant temperature, and ambient temperature.

[0043] The fault detection module 114 then determines an index associated with overheating of the one or more semiconductor modules. In an example, the index is a difference between the estimated junction temperature and a reference junction temperature. Thus, in one example, the index may be expressed as,

Where

represents a reference junction temperature.

represents junction temperature estimated by the trained model. E represents the error term in estimation.

In anpther example, the index is a difference between the estimated casing temperature and a reference casing temperature. In an example, the reference casing temperature is based on measurements of one or more sensors positioned on the casing. Thus, in one example, the index may be expressed as,

Where

T_c represents a reference casing temperature.

represents casing temperature estimated by the trained model.

E represents the error term in estimation.

[0044] In an example, the operating parameters, the reference junction temperature, and the reference casing temperature are based on sensor data received from one of a data collector unit of the electric drive and soft sensors. In an example, the reference junction temperature may be a predefined temperature determined based on historical data of the operating parameters of the electric drive.

[0045] The fault detection module 114 continuously monitors the index to detect the overheating of the one or more semiconductor modules. In response to the index being higher than a threshold value for the index, overheating of the one or more semiconductor modules is detected. In an example, if the value of the index (I) is positive, overheating may be detected.

[0046] In response to detecting the overheating, the transceiver 212 may generate an indication for an operator to undertake preventive action to reduce the overheating. In an example, a sound alarm may be generated for the operator. On hearing the sound alarm the operator may schedule maintenance activity and check for blockage of fans in the electric drive or may also check for ageing of the semiconductor components and thereby undertake a corrective action. In another example, the indication may be generated by displaying a message in a Human Machine Interface (HMI) coupled to the electric drive, where the message prompts the operator to undertake the preventive action. In an example, such a message may be generated by the system 100 and may be displayed in the remote monitoring device, such as the remote monitoring device 104.

[0047] Fig. 3 illustrates a graph 300 depicting actual values of junction temperature and estimated values of junction temperature plotted over a period of time. The Y-axis of the graph of Fig. 3 represents average junction temperature associated with one or more of the semiconductor modules and the X-axis represents time. The graph 300 shows the change of the junction temperature on a particular date at different time intervals. The dots represent the actual value of the junction temperature and lines represent the estimated value of the junction temperature. From Fig. 3, it may be noted that the actual value has shifted up from the estimated value for the particular date (Feb 26) as evident from Error! Reference source not found.3. In such a situation, an overheating is detected by the condition monitoring system 100 or 200 as explained in earlier.

[0048] Fig. 4 illustrates a graph 400 depicting number of overheating instants (number of anomalies) per day as detected by the condition monitoring system 100 or 200. The Y-axis of the graph of Fig. 3 represents number of overheating instances in a day which may correspond to the number of positive values of the index that are recorded in a day and the X-axis represents time in days. The instances in the graph identified as H are instances which resulted in trip due to overheating. On 26^th Feb 2018, there was a failure in semiconductor modules and from the plot it is evident that near to failure, anomalous events increased significantly towards 26th Feb 2018. Higher number of anomalous events were also recorded near trip instants.

[0049] From the graph 400 it may be also understood that there is a positive trend in number of anomalous event as we move from October to February as denoted by the dotted line D. Due to this gradual trend the operator may identify a time interval where though actual temperature of the junction has deviated from estimated junction temperature, there is still some time available before a failure occurs. This time interval provides an opportunity for the operator to schedule fault preventive maintenance activity. [0050] Figs. 5A and 5B illustrates a method 500 for detecting overheating in an electric drive based on monitoring of thermal behavior of one or more semiconductor modules of the electric drive, according to an example. The one or more semiconductor modules are enclosed within a casing, and each semiconductor module includes a plurality of semiconductor switches. The method 500 may be executed by a system, such as the condition monitoring system 100 or 200. The method 500 can be implemented by processing resource(s) or electrical control systems through any suitable hardware, programmable instructions, or combination thereof. In an example, step(s) of the method 500 may be performed by hardware or programming modules, such as the temperature estimation module 112 and fault detection module 210. Further, although the method 500 is described in context of the aforementioned system 100, other suitable systems may be used for execution of the method 500. It may be understood that processes involved in the method 500 can be executed based on instructions stored in a non-transitory computer-readable medium. The non-transitory computer-readable medium may include, for example, digital memories, magnetic storage media, such as a magnetic disks and magnetic tapes, hard drives, or optically readable digital data storage media.

[0051] Referring to Fig. 5, at block 502, measurements of operating parameters of the electric drive with one or more sensors of the electric drive are obtained. Each of the one or more sensors is a current sensor, a voltage sensor, a temperature sensor and a soft sensor. In an example, a reference casing temperature is based on measurements of one or more sensors positioned on the casing. In another example, the operating parameters, the reference junction temperature, and the reference casing temperature are based on sensor data received from one of a data collector unit of the electric drive and soft sensors.

[0052] At block 504, one of a junction temperature and a casing temperature for the one or more semiconductor modules is estimated based on a data model of the electric drive and the measurements of the operating parameters. The data model is generated from historical data of measurements of the operating parameters of the electric drive and interrelation between the operating parameters being established based on thermal relationship between components of the electric drive. The junction temperature is representative of an average of one or more junction temperatures associated with the one or more semiconductor modules and the casing temperature is representative of an average temperature of the surface of the casing. In an example, the operating parameters used to estimate the casing temperature include current of the electric drive, switching frequency, coolant temperature, and ambient temperature. In another example, the operating parameters used to estimate the junction temperature include current of the electric drive, switching frequency, a measured casing temperature and ambient temperature.

[0053] At block 506, an index associated with overheating of the one or more semiconductor modules is determined. The index is a difference between one of the estimated junction temperature and a reference junction temperature and the estimated casing temperature and a reference casing temperature.

[0054] The connection point A at the end of Fig. 5A signifies that the illustration of the figure is continued to the next drawing.

[0055] With reference to Fig. 5B, at block 508, the index is continuously monitored to detect the overheating of the one or more semiconductor modules. At block 510, the overheating of the one or more semiconductor modules is detected, in response to the index being higher than a threshold value for the index. In an example, the threshold value may be 0. At block 512, in response to detecting the overheating, an indication for an operator may be generated to undertake preventive action to reduce the overheating. In an example, generating the indication may include one of generating a sound alarm for the operator and displaying a message in a Human Machine Interface (HMI) coupled to the electric drive, where the message prompts to undertake the preventive action. In an example, the preventive action may include checking/carrying out maintenance activity at a fan of the electric drive or checking of electronic components of the electric drive. [0056] Although implementations of present invention have been described in language specific to structural features and/or methods, it is to be noted that the present invention is not necessarily limited to the specific features or methods described. Rather, the specific features and methods are disclosed and explained in the context of a few implementations for the present invention.

Claims

I/We claim:

1. A method for detecting overheating in an electric drive based on monitoring of thermal behaviour of one or more semiconductor modules of the electric drive, wherein the one or more semiconductor modules are enclosed within a casing, and wherein each semiconductor module comprises a plurality of semiconductor switches, the method comprising: obtaining measurements of operating parameters of the electric drive with one or more sensors of the electric drive, wherein each of the one or more sensors is a current sensor, a voltage sensor, a temperature sensor and a soft sensor;

estimating one of a junction temperature and a casing temperature for the one or more semiconductor modules based on a data model of the electric drive and the measurements of the operating parameters, wherein the data model is generated from historical data of measurements of the operating parameters of the electric drive and interrelation between the operating parameters being established based on thermal relationship between components of the electric drive, wherein the junction temperature is representative of an average of one or more junction temperatures associated with the one or more semiconductor modules, and the casing temperature is representative of an average temperature of the surface of the casing;

determining an index associated with overheating of the one or more semiconductor modules, wherein the index is a difference between one of the estimated junction temperature and a reference junction temperature and the estimated casing temperature and a reference casing temperature; continuously monitoring the index to detect the overheating of the one or more semiconductor modules;

detecting the overheating of the one or more semiconductor modules, in response to the index being higher than a threshold value for the index; and generating an indication for an operator to undertake preventive action to reduce the overheating, in response to detecting the overheating.

2. The method as claimed in claim 1, wherein generating the indication comprises one of:

generating a sound alarm for the operator; and

displaying a message in a Human Machine Interface (HMI) coupled to the electric drive, wherein the message prompts to undertake the preventive action.

3. The method as claimed in claim 1, wherein the reference casing temperature is based on measurements of one or more sensors positioned on the casing.

4. The method as claimed in claim 1, wherein the operating parameters used to estimate the junction temperature include current of the electric drive, switching frequency, a measured casing temperature and ambient temperature.

5. The method as claimed in claim 1, wherein the operating parameters used to estimate the casing temperature include current of the electric drive, switching frequency, coolant temperature, and ambient temperature.

6. The method as claimed in claim 1, wherein the operating parameters, the reference junction temperature, and the reference casing temperature are based on sensor data received from one of a data collector unit of the electric drive and soft sensors.

7. A condition monitoring system for detecting overheating in an electric drive based on monitoring of thermal behaviour of one or more semiconductor modules of the electric drive, wherein the one or more semiconductor modules are enclosed within a casing, the condition monitoring system being communicatively coupled to a remote monitoring device of the electric drive over a network, the condition monitoring system comprising:

a transceiver to: receive measurements of operating parameters of the electric drive from the remote monitoring device, wherein the measurements are obtained through one or more sensors of the electric drive, wherein each of the one or more sensors is a current sensor, a voltage sensor, a temperature sensor and a soft sensor;

a processor;

a temperature estimation module coupled to the processor to:

estimate one of a junction temperature and a casing temperature for the one or more semiconductor modules based on a data model of the electric drive and the measurements of the operating parameters, wherein the data model is generated from historical data of measurements of the operating parameters of the electric drive and interrelation between the operating parameters being established based on thermal relationship between components of the electric drive, wherein the junction temperature is representative of an average of one or more junction temperatures associated with the one or more semiconductor modules, and the casing temperature is representative of an average temperature of the surface of the casing; and

a fault detection module coupled to the processor to:

determine an index associated with overheating of the one or more semiconductor modules, wherein the index is a difference between one of the estimated junction temperature and a reference junction temperature and the estimated casing temperature and a reference casing temperature;

continuously monitor the index to detect the overheating of the one or more semiconductor modules; and

detect the overheating of the one or more semiconductor modules, in response to the index being higher than a threshold value for the index; and

wherein the transceiver is to:

generate an indication for an operator to undertake preventive action to reduce the overheating, in response to detecting the overheating.

8. The condition monitoring system as claimed in claim 7, wherein the reference casing temperature is based on measurements of one or more sensors positioned on the casing.

9. The condition monitoring system as claimed in claim 7, wherein the operating parameters used to estimate the junction temperature include current of the electric drive, switching frequency, a measured casing temperature and ambient temperature.

10. The condition monitoring system as claimed in claim 7, wherein the operating parameters used to estimate the casing temperature include current of the electric drive, switching frequency, coolant temperature, and ambient temperature.