Partial Discharge Classification in High Voltage Equipment

What is a Partial Discharge (PD)?

Partial discharge (PD) is a type of electrical discharges that does not bridge electrodes and it affects only a portion of insulating material while the rest of it is not affected.

PDs can take place in solids, liquids, and gases. The important thing about PDs it their detrimental impact on the health of insulation system. PDs cause the accelerated aging of electrical equipment. There are three types of discharges:

(1) Internal discharge that happens in the cavities and inclusions within the dielectric material.

Internal Discharge

(2) Surface discharge that happens at the surface of dielectric material, usually in the presence of moisture and contamination.

Surface discharge

(3) Corona that happens in the regions where electric field is extremely inhomogeneous and high, usually around the metal tips and edges.

Corona

 

What is it important to investigate PD?

PD phenomenon has been studied since early, mid-19th century as it has been known as the deterioration factor in cables, transformers, motor windings, and almost all the electrical equipment.

Recently, another industry, Aviation Industry, has urged the study of PD phenomenon.

In recent decades, extensive research has been conducted on the electrification of commercial aircraft to reduce the dependency on mechanical, hydraulic, and pneumatic systems and replace them with electrical systems.  A primary goal of this path is to make the power density of the more/all-electric aircraft (MEA/AEA) closer to that of conventional aircraft.

The problem is that the technologies and improvements for achieving higher power density escalate the electrical tension on the insulation system by enhancing the risk of partial discharge. One of these technologies is the wide bandgap semiconductors which generates fast-rise, high-frequency voltage pulse. Another one is the higher voltage level required for electric aircraft which increases the probability of PD inception.

Besides, the harsh environmental conditions at high altitudes have proved as negative factors for insulation systems.

What we aim to do here is to make use of the measurements of the propagated signals in electrodes. It has been shown that this signals can be indicator of PD activities. We want to see if we can classify the PD types based on the short-term recording of this signal.

Therefore, the goal of this study is to classifying the PD type based on short-term behavior of insulation in a matter of second. This allows taking emergency actions like quick switching to backup equipment.

How does our input data look like?

In the Figure 1, you can see a generic format of raw data which has rows of time instants and the columns of cycles of data which are usually a 50 or 60-Hz depending on the frequency of power grid. The time instants are scaled based on a single period of voltage. Meaning that for 50Hz signal, all the time instants are between zero and 20 milliseconds.

Figure 1 – Raw data structure

If we plot the amplitude of signal versus the time instants, we have a figure called phase-resolved partial discharge pattern (see Figure 2). As you can see, this pattern is quite different for each type of discharge.

(a) corona
(b) surface discharge
(c) internal discharge

Figure 2 – Discharge types: (a) corona, (b) surface, (c) internal discharge

 

Now, let’s see how we can prepare this data for classification methods.

When we are doing short-term analysis in which we have only seconds of data, it is very important to generate meaning datasets that can properly train the classification methods.

For the introduced batch of data, we follow four steps to generate the training, validation, and test data.

-Step 1: We randomly choose 90%, and 10% of cycles for training and testing, respectively.

-Step 2: Sort the cycles of data based on phase (time instants) order

-Step 3: Assuming that the arrays of should have n_f features, we categorize the time instants into n_f groups (see Figure 3).

-Step 4: Finally, each array of training data is formed by choosing a random time instant from each group, and also choosing a random cycle (similar procedure for test data).

Figure 3 – Partitioning of sorted batch of data

The measurements are performed for 4 seconds for each discharge type. There are 200 cycles of data for each discharge type. A cycle of data is a discrete measurement over a period of the 50-Hz voltage signal. Therefore, the time instants in a cycle are between 0 and 20 milliseconds. Moreover, time instants are phase-resolved meaning that they are scaled for the single cycle of 0-20 milliseconds. At each time instant, the signal is measured in volts.

The remaining properties of raw datasets and the final training and test datasets are as follows:

  • The number of time instants in the dataset of each discharge type is different and is follows:  Corona (2405), Surface (795), and Internal (3521).
  • The number of training arrays for each type of discharge is n_s=15000.
  • The candidate numbers of features are n_f=64, 128, 256.
  • The number of testing arrays for each type of discharge is n_t=1000.
  • For n_f=256, the training and testing datasets will have shapes of (45000×256) and (3000×256), respectively.
  • The labels for corona, surface, and internal discharges are 0, 1, and 2, respectively.
Which classification methods do we use? Results?

Based on the definition of the problem, this is a supervised learning problem, and the candidate methods are as follows:

(1) K-Nearest Neighbor (KNN)

(2) Logistic Regression

(3)Gaussian Naïve Bayes

(4) Random Forest

(5) Bagging Trees

(6) Adaptive Boosting

Table 1 – Summary of results

The accuracy and computation time of these classification methods are shown in Table 1. For all candidates of , the boosting method presents the highest accuracy. However, most computation time also belongs to this method. If we compare the data duration (which is 4 seconds), for the optimal case of , it is about 100 times the data duration. For the time-sensitive application of insulation health for electric aircraft, it does not look promising. Similarly, decision tree bagging offers high accuracy, but also high computation time. Among all the methods, logistic regression has the poorest performance by far, probably due to the fact that it cannot handle Non-linear problems because of its linear decision surface.

Also, KNN does not perform well. Although at first glance, it might seem like a good choice for the assessment of short-term behavior, KNN does not perform well in the presence of noise. Furthermore, KNN can suffer from skewed class distributions. When a certain class is very frequent in the training set, it will tend to dominate the majority voting.

Random Forest and Gaussian Naïve Bayesian classifiers are among the quickest methods with acceptable accuracy. But our choice would definitely be Gaussian Naïve Bayes Classifier; it offers almost 80% accuracy in less than one-tenth of a second which is a bright choice for the case of electric aircraft. In the case of a danger threatening the insulation system, and thus, the whole electrical equipment, Switches can act in less than a second and protect the entire system against a potential fault. Despite its simplicity, Gaussian Naïve Bayes is known to have a promising performance for small datasets which is confirmed in this study.

<i class=”fa fa-arrow-right” aria-hidden=”true”></i>

Final Remarks:
  • Aviation industry aims to electrify aircraft to achieve emission reduction of 50% set by NASA.
  • Enhancement of the power density (weight) → enhances the electric tension on insulation systems.
  • Partial discharge is an indicator of insulation system’s health.
  • This study aims to use short-term response of insulation systems → To predict the type of discharge and take proper mitigating actions in a matter of seconds.
  • Bagging and Boosting methods  offer highest accuracies → Also, highest computation time
  • Gaussian Naïve Bayes classifier offer nearly as accurate as ensemble methods’ accuracy → In less than a second which  makes it a promising solution for aeronautic applications.

 

Additional Files

Through following links, you have access to the project paper, presentation slides, codes, and data.

Identification of the Challenges and Remedies in the Supervisor- Graduate Student Relationship

Introduction:

Nowadays, young scholars are influenced by experienced professionals in terms of scientific advisory and mentoring. In higher education institutions, graduate students need guidance from their supervisors to take correct steps toward achieving excellence in scientific areas. In the relationship between the supervisor and the student, both sides ought to be familiar with the purpose of this relationship and be prepared to take steps toward that. However, the different perceptions of the issues by the supervisor and the student may lead to a conflict [1].

The extent of complications is dependent upon the dimension of the relationship. The simplest relationship is the one that has the least financial or commercial interest [2]. In this case, the expectations of the supervisor are restricted to the student’s thesis. On the other hand, any additional research or teaching assistantship requires more consideration by both supervisor and the student as it adds to the complexity of the relationship through mingling the employment and supervision. The role of supervisors is bolder since a student has to trust the supervisor, is less experienced in their relationship, and has less power [2].

The outcome of this relationship is important not only in the way it affects the professional life of a student but also in the way it influences the future of academia. A student at higher levels of education can become a faculty member in the future. In this case, this person unavoidably takes the behavior of his/her advisor as one of the role models. Now, what happens if the supervisor is passing some inappropriate characteristics? Therefore, it is crucial to have well-educated faculty members not only in their scientific side but also in their supervisory role.

The tensions in the student-supervisor relationship usually comprehended differently by the two parties; Students refer to reasons like bad advising while supervisors attribute the unsuccessful experience to the character of the student [3]. Faculty members usually look for skillful, motivated, and self-driven students, and those who lack these characteristics are more prone to have trouble with their advisers.

In what follows, we overview some of the major issues associated with the supervisor-student relationship. Moreover, we take a look at the mitigating actions that could be done by the student, faculty member, and university administration.

 

Factors affecting the student-supervisor relationship from the student standpoint:

  • Injustice:

In a research group, members may (mostly unintentionally) compare themselves to one another. If one feels that the privileges are distributed unevenly, the roots of dissatisfaction begins to grow for the student. Some of these privileges can be the frequency and quality of feedback, stipend, recommendations [2]. In my experience, students usually do not express their complaints about injustice in a research group unless it becomes considerably influential in their research path. If they do so, the response of the supervisor is crucial. Supervisors with a sense of leadership would act to prevent the growth of any misunderstanding (which might have mistakenly reflected as an unjust advisory) or try to reduce any bias that they might have. If the student’s complaint is accompanied by a harsh objection from the supervisor, the relationship is less likely to have a happy ending.

 

  • Lack of Scientific Support:

To guide the students in the right direction and help them to improve their research work, they need feedback from their supervisors [4]. Especially, depending upon the student’s personality and background, some may need to frequently visit their supervisor to start moving the wheels of their research. However, professors have many obligations and sometimes, other responsibilities of the faculty member in the teaching or the research side (such as a grant deadline) may lead to days of late responses [4]. If this becomes a frequent habit, the supervision of the students becomes less effective and the students’ progress slows down.

The supervision does not have a global norm; in other words, faculty members usually have their own way of advising their students. Some advisors believe in spending more time for the student in the early stage and less as the research finds its path. Some others take help from their postdoctoral students to find more time for other aspects of their professional careers [1]. In all these cases that students are deprived of consistent supervision, they may suffer from poor support that results in slow progress or even a complete halt.

 

  • Misbehavior:

Motivating students to keep working hard is very effective to help them persist in their research direction. While it is an accepted fact in academic society, the supervisors take different approaches toward it. In one of the academic laboratories I have worked with, a globally-renowned professor was famous for his nasty comments about students’ performance. Statements like “you are stupid” was not much of a surprise in the research group meetings. He believed that toughness is a key feature to success and a student who flourishes under severe pressure is the one who can conquer the tall summits. This old-fashioned habit, however, can also destroy so many talents by generating hatred in their minds about not only their research but also science and academia. Psychologists have shown us that positive feedbacks and praise at the right time and the right place can be more efficient and drive the students to move forward [2].

Another type of misbehavior is the abnormally high pressure imposed on graduate students by giving additional tasks [1]. Many graduate students are lucky enough to have financial support throughout their studies by various types of assistantships. While graduate students should majorly focus on their thesis, a heavy burden of such tasks hinders them to move toward the main academic goals.

 

  • Harassment:

According to [5], 30-40% of female students undergo some sort of sexual harassment during their college or university studies. Needless to mention that this does not limit to the female students, there is no doubt that so many cases are buried in the students’ memories. In fact, some suggest that unless it is a very serious situation, students should refrain from taking action against the supervisor, especially if their research is going in the correct direction. This is due to the negative consequences that it might have for the student and the fear from the fact that the university administration may not intervene to the benefit of the student [2]. Personally, I believe this is not a valid decision to make. We are fortunate that, in today’s world, the dignity of people is more cherished more than in the past centuries. This gives us the courage to raise our sound and speak up if we are being annoyed by someone’s misbehavior. Therefore, we should no longer let such sort of harassment find any place in academia.

 

Factors affecting the student-supervisor relationship from the supervisor standpoint:

  • Disloyalty and Mistrust:

Historically, the student-supervisor relationship was almost like a dictatorship [x]. In this relationship, the students were committed to following the supervisors’ commands obediently. However, this is not the case today. Part of it is due to the borderless access to knowledge (which has shrunk the professors’ charisma in the students’ minds); the other part is because the students are more concerned about their rights and the world also pushes it forward. Some faculty members are annoyed by the fact that students do not look at them as a supervisor anymore. They believe these transformations have reduced the degree of trust the students had in their supervisors. It also makes them more inclined toward changing the environment (such as supervisor or university) rather than changing and improving themselves.

 

  • Inefficient Interaction:

While many students find supervisors’ feedbacks necessary for their scientific growth, some others are upset by what they call “micromanagement”. This group of students believe in their own way of doing research and prefer not to be distracted by the supervisors’ comments. However, it is rarely accepted by the supervisors that a student evades from giving reports of the progress he/she makes. This becomes even more important if the student is also employed by the faculty member or there is a commercial benefit in the outcome of the project. The relationship, in this case, only survives if a student adheres to the guidelines of the supervisor.

 

  • Lack of Self-evaluation:

A supervisor’s role is to give some clues to the student and provide directions. To this end, in the early stages of the relationship, students may expect a high level of interaction with their supervisors. But an advisor cannot be there for the student at all times, and the students must start to become their own supervisors based on what they have been taught. This idea is promoted by many faculty members and they expect students to become more independent as they move toward the end of the program.

 

  • Lack of Professionalism:

While the supervisor should provide a medium for the students’ progress, it is also expected that the students prepare themselves for the next chapter of their career with professionalism. The supervisors desire to see the growing professionalism in their students’ actions [6]. Faculty members believe that their expectations elevate as the experience of their students enhances. Therefore, if students cannot keep up with this growth rate, they cannot be prepared for their upcoming role in academia, industry, or wherever they intend to work.

 

 

Solutions:

  • Educating supervisors:

Although supervisors are knowledgeable in their fields of expertise, this does not reflect their mastery in guiding students toward their academic goals. Therefore, they must be educated to avoid any bias in teaching, advising, and mentoring. It is also noteworthy to mention that faculty members usually do not practice “supervision” before joining an institution and they are hired mostly because of their academic achievements. Therefore, there is a gap here that must be addressed by the schools through programs that prepare students as future professors.

 

  • University Guidelines

University supervision guidelines can be a very good source for enforcing the proper practices and shine a light on the responsibilities of both sides [2]. Supervisors should be assured about the workload they should expect students. The problem is that, at least in the current era, the power imbalance makes it so that only supervisors can enforce the students’ duties and the opposite does not hold. Therefore, the administration policymakers should not only devise rules to protect the rights of the graduate students but also should advertise and inform the faculty members of these rules and their importance.

 

  • Meeting with Supervisor:

In almost every student-supervisor relationship, a situation of dissatisfaction may happen for each side; if we can resolve the issue in the early stages, we may prevent it to turn into a bigger issue and ultimately, endanger the entire relationship. Speaking to a trusted friend or an experienced consultant can be insightful before meeting with the supervisor [3].

 

  • Formal Complaint:

If the problem is not resolved in a one-on-one meeting, the intervention of the university may provide an opportunity for reconciliation. The graduate program director can be the first one to start with as he/she knows the procedure and policies [4]. If the discussion with the program director does not come helpful, the department head can be the next choice. Ultimately, the dean of graduate school and president of the university can be the student’s options. The process of seeking help can be very stressful for the students and strong documentation is essential for this purpose [4].

 

  • Changing Supervisor:

Changing a supervisor should be the last choice [4]. The least cost is slowed progress and delayed graduation due to the time needed to find and calibrate with the new supervisor. This process can also hurt the faculty members, especially, if they have financially supported the student and expected a satisfying outcome. Moreover, finding another supervisor with a common research interest is not a trivial task; in [7], the research shows that many students prefer to change their university rather than the supervisor since they fail to find supervisors that fit their research area.

 

 

 

Conclusion:

The next generation of professors, world-leaders, entrepreneurs, and all other occupations are most likely to be brought up in higher education institutions. Therefore, the current faculty members have an undeniable impact on not only the scientific career of the students but also their characters. In this study, we targeted the supervisor-student relationship from both perspectives. It was discussed that the injustice in a research group can lead to the dissatisfaction of students. Scientific support is another important factor in the relationship; students expect feedback on their work to make sure about their progress. Also, the misbehavior and harassment conducted by the supervisors endanger the health of the relationship.

From the supervisors’ perspective, lack of trust and loyalty is a major reason behind the troubles in the supervisor-student relationship. Besides, inefficient communication, self-assessment, and professionalism are other key features in a successful relationship.

Any of the issues above can ignite the conflict’s flame and lead to more severe issues. What the university administration can do to prevent such conflicts is to make things clear and set up rules for the standard relationship. Also, the administration must assure both sides that they would be treated equally against these rules. If troubles in a relationship exceed a threshold, the first remedy might be a talk between the supervisor and the student with both having the will to resolve the issue. The next step is a formal complaint to the administration level at the department level or higher. The final choice should be the change of supervisor as it is costly for both sides, especially, the student who may have slowed progress and should overcome the problem of finding another supervisor.

 

References:

[1] E. Löfström and K. Pyhältö, “Ethics in the supervisory relationship: supervisors’ and doctoral students’ dilemmas in the natural and behavioural sciences,”  Studies in Higher Education, vol. 42, no. 2, pp. 232–247, Feb. 2017, doi: 10.1080/03075079.2015.1045475.

[2] C. MacDonald and B. Williams-Jones, “Supervisor-student relations: examining the spectrum of conflicts of interest in bioscience laboratories,” Account Res, vol. 16, no. 2, pp. 106–126, 2009, doi: 10.1080/08989620902855033.

[3] S. K. Gardner, “Student and faculty attributions of attrition in high and low-completing doctoral programs in the United States,” Higher Education, vol. 58, no. 1, pp. 97–112, Jul. 2009, doi: 10.1007/s10734-008-9184-7.

[4] L. Blair, “Dealing with Student-Supervisor Problems,” in Writing a Graduate Thesis or Dissertation, L. Blair, Ed. Rotterdam: SensePublishers, 2016, pp. 121–126.

[5]  J.-C. Smeby, “Same-gender relationships in graduate supervision,” Higher Education, vol. 40, no. 1, pp. 53–67, Jul. 2000, doi: 10.1023/A:1004040911469.

[6] A. Yousefi, L. Bazrafkan, and N. Yamani, “A qualitative inquiry into the challenges and complexities of research supervision: viewpoints of postgraduate students and faculty members,” Journal of Advances in Medical Education & Professionalism, vol. 3, no. 3, pp. 91–98, Jul. 2015.

[7] C. M. Golde, “The Role of the Department and Discipline in Doctoral Student Attrition: Lessons from Four Departments,” The Journal of Higher Education, vol. 76, no. 6, pp. 669–700.

 

 

Long-term plan of higher education

There is no doubt about the role of funding agencies on the direction that Higher Ed institutions take. But does that direction lead to a bright future for humanity?

Source: https://blog.shi.com/news/educause-2018-the-future-of-it-in-higher-education/

There is no easy answer to this question; but In today’s world, we are more and more seeing the power of money in higher education institutions; we have seen cases that the faculty members who are accused of misbehaving of any sort are supported by the university administration, and the reason, many believe, lies in their ability to bring money. Although we have so many moral, yet diligent scholars who conduct valuable research in its pure form, the inclination toward cherishing the value of money more than the value of dignity and morality can endanger the bright future that we picture. The education system looks like an efficient machine that works hard to satisfy the world’s needs; however, some of the needs may not be demanded by today’s world but are crucial for long-term redemption. Here is the role of government to value those pure scientific efforts which work for the future of humanity. Are the most powerful men in the world capable of making proper decisions for that?