Thesis worker - Look Ahead Energy - Optimal Control Of An Integrated Thermal Management System

Job description

Introduction Thesis work is an excellent way to get closer to Scania and build relationships for the future. Many of today's employees began their Scania career with their degree project. Background Thermal management consumes a significant amount of auxiliary energy in a battery-electric vehicle, particularly during cold starts, fast charging, high-load driving, and overnight cab heating. Conventional rule-based controllers generally react to current temperatures and fixed thresholds. Such control can activate a high-power heater or fan shortly before a limit is reached, even when a lower-energy action could have been started earlier. Heavy-duty vehicles increasingly have access to look-ahead information, such as route elevation, expected vehicle speed, traction power, charging plans, ambient temperature, and driver schedule. This creates opportunities to: • spread heating or cooling over a longer period; • use PE or battery waste heat before using resistive heating; • operate the heat pump when its COP is favorable; • pre-cool before an uphill segment or fast-charging event; • preheat the battery before charging; • charge the cab coolant circuit with PE waste heat for later battery heating during a planned stop; • reduce or delay heat-pump operation before the next vehicle start; • avoid excessive fan operation in noise-sensitive situations; • prevent temperature-limit violations and associated power derating. An important opportunity is to use the cab coolant circuit as a short-term thermal buffer. If a future battery-heating demand is predicted, PE waste heat can be transferred through PeToCab to increase the cab-circuit coolant temperature, even when there is no cabin heating demand. During a subsequent stop, the stored heat can be transferred through CabToBattery to prevent excessive battery cooling and reduce or delay heat-pump operation before the next start. The challenge is that the thermal circuits are physically coupled. A decision that benefits one subsystem can reduce the thermal margin or available heat of another. For example, extracting heat from the battery can support cab heating but may reduce battery charging performance, while transferring condenser heat to the PE circuit may require additional fan power to reject that heat to ambient. A system-level optimizer is therefore needed to coordinate these interactions over time. Objective The thesis is expected to provide: a consistent graph, valve, and mode representation of the integrated thermal architecture; a validated reduced-order model suitable for look-ahead control; a hybrid predictive thermal-energy management strategy; a quantified comparison of heat recovery, cab-circuit thermal buffering, electric heating, heat-pump operation, and heat rejection; an assessment of the value and required accuracy of different forecast signals; real-time feasibility results and recommendations for production implementation; identified architecture or instrumentation limitations that prevent optimal coordination. The thesis will be considered successful if it: predicts the temperatures needed for supervisory decisions with validated accuracy; maintains safety-critical constraints whenever the problem is physically feasible and invokes the fallback strategy when the optimization is infeasible; demonstrates lower thermal-management energy than the reference controller in at least the scenarios where look-ahead information has actionable value; quantifies when charging the cab coolant circuit through PeToCab and discharging it through CabToBattery reduces subsequent heat-pump energy; explains any scenarios with no benefit or degraded performance; quantifies sensitivity to forecast errors and model mismatch; demonstrates a credible route to real-time execution or clearly identifies the simplifications required to achieve it. Job description Develop and evaluate a machine-learning framework to identify five health states in a heavy-duty vehicle’s active cooling refrigerant system. The thesis will compare two ways of estimating conditions before the expansion valve—an observer-based estimate and direct sensor measurements—across TXV and EXV systems. You will prepare and qualify test data, establish reliable fault labels, and develop features based on the system’s physical behaviour. You will compare different classification approaches and assess their ability to distinguish faults such as low refrigerant charge, expansion-valve underfeeding or overfeeding, and compressor degradation. The work will test how well the models generalize to unseen hardware units and test sessions. You will also calibrate their confidence and define when a diagnosis should be reported as inconclusive. The results will provide recommendations for further validation toward production use. Education/program/focus Indicate education, program or focus: Number of students: 1 Start date for the thesis work: 1st of February 2027 Estimated time required: 5 month Contact persons and supervisors Contact Name: Göran Lissel Function: Head of TGRBTEA – Electrification Management control – Thermal & Aux Phone: +46 8 553 51937 Mobile: +46 70 086 0984 e-Mail: Supervisor Name: Karim Mohammed Razaul Function: Software Development Engineer – Thermal & Aux Mobile: +46767225137 e-Mail: Application: Your application must include a CV, personal letter and transcript of grades A background check might be conducted for this position. We are conducting interviews continuously and may close the recruitment earlier than the date specified. Publication date from - to. 2026-10-22 – 2026-11-22