Do you want to become a member in one of the most unique Organisations in the world? Then, take a look at the positions offered in the Industrial Control and Systems Group! We are is always looking for students, trainees, Graduates and Staff who would like to join.
Graduates:
SCADA Applications Engineer (Origin)
You will participate in the development of the technical supervisory layer of different CERN facilities (e.g. Cryogenics, HVAC, gas, electrical networks…) and in the optimization of the management of such control systems. You will integrate the SCADA Application Service (SAS) team and participate in other related team activities.
The main tasks will be the following:
- Design, develop and maintenance of supervisory control applications (e.g. Technical Infrastructure monitoring)
- Install, upgrade and configure SCADA systems to monitor and control various industrial processes (e.g. Cryogenics, HVAC, gas...)
- Development of tools (e.g. custom scripts) to enhance system functionality and performance
- Management and maintenance of the facilities hosting such tools
- Collaborate with project managers and other stakeholders to understand system requirements and develop solutions that align with their needs.
- Assist on the general schedule to upgrade all the existing CERN WinCC OA projects and perform system upgrades ensuring minimal disruption to operations
- Testing and validation activities to make smooth upgrades to the software versions (e.g. WinCC OA, UNICOS)
- Provide support to CERN end-users within the SCADA/WinCC OA domains
Industrial machines automation engineer (Origin)
You will participate in the development and commissioning of control systems of machines devoted to assemble superconducting magnets. The main tasks will be the following:
- Gathering the requirements of the control systems and compiling all the relevant documents (e.g. specifications, functional analysis, etc).
- Designing control systems for the machines to be consolidated.
- Programming the PLC, configuring local operation panels and the SCADA (Supervisory Control and Data Acquisition) layers of the control systems.
- Commissioning by managing the tests offline and in the field.
- Archive all the produced materials.
Field of interest:
- Industrial automation, control engineering
The main tasks will be the following:
-
- Defining and managing requirements
- Configuration and programming PLCs and fieldbus systems / Programming PLCs.
- Programming/Software development: TIA portal in SCL.
- Configuration, programming and using Supervisory Control And Data Acquisition (SCADA) systems.
- Testing procedures (Factory and Site Acceptance Tests).
HL-LHC Cryogenics refrigerators controls (Quest)
You will participate in the development and commissioning of new HL-LHC project cryogenics plants control systems. The scope of the project covers the control systems for two new cryogenic systems (comprising industrial equipment like cold boxes, compressors and transfer lines), of the LHC accelerator cryogenics system.
Detailed description of Activity:
The main tasks will be the following:
- Gathering the requirements of the control systems and compiling all the relevant documents (e.g. specifications, functional analysis, etc).
- Designing control systems for the new cryogenics refrigerators industrial installations following the CERN standards (e.g. UNICOS) and employing advanced tools for generating the code automatically.
- Programming PLC and configure the SCADA (Supervisory Control and Data Acquisition) of the control systems.
- Commissioning by managing the tests offline and in the field.
The main tasks will be the following:
-
- Defining and managing requirements
- Configuration and programming PLCs and fieldbus systems / Programming PLCs.
- Programming/Software development: IEC-61131-3 standard, Schneider Unity Pro (EcoStruxure Control Expert)
- Configuration, programming and using Supervisory Control And Data Acquisition (SCADA) systems (WinCC OA)
- Testing procedures (Factory and Site Acceptance Tests).
Real Time control systems development and support (Quest)
he successful candidate will integrate a small team in charge of developing real time software applications highly focused on industrial systems and using the standard CERN frameworks (i.e. FESA, UNICOS). The scope will vary from interlocks and protection systems, motion (stepping motors), cryogenics and magnet alignment.
The main tasks will include:
- Develop new features and make improvements in the ecosystem tools for developing Cryogenics control systems
- Support and extend the existing software running on computers interfacing with the hardware, built with the FESA framework.
- Identify requirements for new software features in collaboration with the users and prepare the specifications to meet the needs.
- Design, develop, test, deploy, document and maintain new software on the front-end computers
- Commissioning of the installations to ensure a smooth transition from the RUN3 of the LHC accelerator
- Provide first line support to the users
- The concerned Real-Time developments (FESA) are:
- Participate in all RT development within the CE section FESA team
- CRYOGENICS: new request and expected variations of devices (e.g. level sensors signal conditioners)
- FRAS: integration and commissioning of the new instrumentation within the FRAS Supervision based on UNICOS.
- Others: PIC project and CRYO project: Ensuring the proper functioning of the new development
- Participate in all RT development within the CE section FESA team
- Additionally the candidate will work in specific software developments focused in the integration of PLCs in all the phases of development, test and commissioning by developing and maintaining interfaces to the APIs of the PLCs.
For Quests (Experienced Graduates), please apply via this dedicated website.
For Origins (Entry-Level Graduates), please apply via this website.
Students:
Post on "Safe Neural Networks"
Neural Networks (NNs) have been found useful in many fields related to machine learning and modelling aspects. Deep neural networks, are being deployed in domains where trustworthiness is a big concern. As an illustrative example, recent catastrophic failures caused by autonomous vehicles highlighted the importance to ensure that these systems need to be correct. The area of verification of neural networks has grown rapidly over the past years and aims to provide methods to verify automatically that a neural network meets its intended specifications.
The use of NNs in reinforcement learning has challenged the use of the existing advanced control algorithms with a more pragmatic approach which simplifies the cost of such complex feedback controllers by releasing the needs of building the models employed in the controllers. However NNs do not provide any formal guarantee of their behaviour and therefore their appliance in safety critical systems is limited as robustness and stability are of a paramount importance. Moreover, machine learning systems are often inscrutable black boxes: they are effective predictors but lack human-understandable explanations. This problem of limited reliability of the NNs is the reason the research community expressed an increase interest in their verification. Formal methods seem a reasonable tool to make this verification.
At CERN, deep neural networks are already widely used in different domains, from analysing data from the collisions, computer vision to detect anomalies in the accelerator complex to name some examples. Practical experience acquired during their application to real machines (e.g. AWAKE, LINAC4 trajectory steering)1 has proved the feasibility of their application to real accelerator control problems. Lasts particle accelerator domain conferences (i.e. ICALEPCS, IPAC) have shown the importance of these new technologies highlighting many different real applications even in critical feedback control applications.
However, their use in safety-critical applications is constrained due to the difficulty to give safety guarantees for the neural network. This research project will focus on providing safety guarantees in order to be able to deploy neural networks in safety-critical applications, making use of both, the expertise on formal methods developed at CERN2,3 with previous Ph.D. thesis and the tools and knowledge provided by specialised groups in universities with recognised experts in the field (e.g the Verification of Autonomous Systems Lab4 from Imperial College London).
Most of the existing methods and tools for the verification of NNs are not scalable and they need enormous computational resources. The use of formal methods seem to be a good approach to give a answer to the verification problem. Formal methods for analysis of probabilistic systems have an important role to play in efficient generation of influence-directed explanations.
The following questions will be addressed:
- Are formal methods an adequate approach to verify NNs?
- Can we provide a general-purpose solution?
- Could we design a verification tool based on formal methods to the current state of the art NNs?
- Can we provide a comparable solution to the stability and robustness of feedback controllers?
Please apply via SmartRecruiters.
Trainees:
We are offering traineeships variying between one to six months on various topics on technology, frameworks, engineering and further fields of interest. To know more, please apply via Smartrecruiters and contact us then via e-mail (be.ics.secretariat@cern.ch).