www.atzonline.de elektronik 04 April 2013 Volume 8 Offprint from ATZelektronik 4/2013 Springer Automotive Media Springer Fachmedien Wiesbaden GmbH for Bosch Engineering Designing vehicle power nets A single simulation tool from initial requirements to series production
Development Vehicle Electrical System Designing vehicle power nets A single simulation tool from initial requirements to series production The complexity of vehicle electrical and electronic systems, components, and functions is growing as they become increasingly networked with each other and with the internet. Vehicle electrical systems developers can manage this complexity now and in the future by employing simulation as a central tool in designing powerful and reliable vehicle electrical systems. Bosch Engineering offers a powerful simulation tool to support the design of vehicle power nets from initial requirements to final series approval, now being used also for hybrid and electric drive train development. 2
AutHors Roman Lahmeyer is Simulation Team Leader at Bosch Engineering GmbH in Abstatt (Germany). Christoph Prehl is Project Leader for Vehicle Power Net Simulations at Bosch Engineering GmbH in Abstatt (Germany). Dongdao Zhang is a Systems Development Engineer at Bosch Engineering GmbH in Abstatt (Germany). Designing vehicle electrical systems using simulation The number of vehicle electrical and electronic components and systems is growing constantly, as is the degree to which they are networked. This increases the complexity of the vehicle as an integrated overall system. This complexity needs to be managed, and the ability to optimise the design and control of complex vehicle power nets is becoming a key competence in vehicle development. One approach that Bosch Engineering GmbH finds effective is to design vehicle power nets using simulation, supporting the complete vehicle development process and beyond system boundaries from the powertrain and efficient auxiliary units to safety, driver assistance, and comfort systems, as well as how they interact in an overall vehicle. The use of simulation in designing vehicle power nets is based on power ful new software. This can be used to design power nets both for conventional vehicles with an internal combustion engine and for engines with alternative fuels, as well as in developing electric and hybrid vehicle drive trains. V-Model simulation of electrical systems We use the V-model as the basis for designing vehicle power nets. This model, which grew out of the waterfall model and first appeared in 1979, uses a V-shape to present the various process steps in vehicle power net development, 1. Each cycle begins with an analysis of requirements. This is followed by a concept and layout phase, during which the future power net s fundamental architectures are defined. Next, specific components are chosen and the functional requirements for energy and vehicle management systems are defined. Once all requirements, functions, and components have been finalised, the design of the power net is determined and approved. The next step is to verify the model and overall system with numerous measurements and simulations. In the final step before approval for series production, typical use-case and particular worst-case scenarios are validated. Using simulations to design vehicle power nets is a particularly effective way to support vehicle development providing a single simulation tool that covers all development processes. In the past, many different simulation tools were used. Some of these helped in the first steps of the V-model to design the power net, but did not contain the necessary detail for all process steps up to approval for series production. Other software products were aimed primarily at the second half of the V-model, for use in detailed simulations, comprehensive testing, and verification. Modern simulation tools can be used across the entire V-model, from requirements analysis to approval for series production. This means that on the one hand they meet customers need to have a rough outline of system design drawn up quickly and without laborious measurements of the final components. On the other hand, customers can use these tools to carry out specific, detailed simulations of their desired architecture all the way through to approval for series production. One tool from requirements analysis to series approval Bosch Engineering has developed a new, expanded tool for designing vehicle power nets that meets the criteria for comprehensive support across all process steps in the V-model, 2. Users first define their requirements for the vehicles power net and choose components whose weight, size, and power class mean they are suited to the desired power net architecture. To this end, the simulation tool contains modular sets of specially determined key parameters. Each of these can be freely adjusted by scaling the related sets of stored parameters. For instance, alternator size can be selected in increments of 10 A. This approach makes it possible to design a vehicle power net quickly and simply. Once the rough design has been determined, the level of detail increases. Step by step, the components and partial systems that must be built in are finalised and any electrical energy management (EEM) functions present are tested. All it takes is a click of the mouse to replace the scaled parameters of the virtual components with the exact parameters of actual components and add them in to the model vehicle power net. The software features a comprehensive library of components parameters for various 3
1 V-model of vehicle power net development sizes and manufacturers. There is also the option to define and integrate sets of parameters for new components for example via test bench measurements. Once all components have been defined, various different vehicle power net configurations can be evaluated and compared using the simulation tool, and operating strategies and system functions can be developed. Simulation tool functions Besides selecting the optimum components for the vehicle power net, such as the battery and alternator, vehicle power net simulations allow the battery state of charge to be simulated and improved. The goal is to select components that together form a stable vehicle power net with either a positive or an even charge balance. To this end, the simulation model calculates the vehicle s fuel consumption information which is for instance used in evaluating EEM functions. Calculation of the fuel consumption is based either on test bench measurements or on application data from the engine management system. In both cases, the simulation model contains all the necessary functions, such as increased fuel consumption during engine warm-up. All parameters for internal combustion engine, battery, and various models of electrical consumer can be integrated in a temperature-dependent way. In the vehicle power net simulation, the models of electrical consumer can be set to be either passive or active consumers. In the case of passive consumers, the user chooses at the start of the simulation whether they are activated or deactivated, along with the relevant consumption values. Windshield wipers are a typical passive consumer. Since the simulation is unable to reproduce limited time periods of precipitation, the user either activates or deactivates the windshield wiper before running the simulation. This is necessary because the 2 Overview of Bosch Engineering vehicle power net development software 4
model works to a large degree as a backward simulation. That means the model does not react to driving decisions but rather to stored cycles (speed, time of switching on/off). These freely programmable cycles can for instance be taken over from prior measurements or generated synthetically. Consumers such as the engine fan are usually set to be active consumers in the vehicle power net simulation. If the temperature of the internal combustion engine rises above a defined value, the fan is either switched on or switched to a higher level. Additional consumers, for instance seat heating, can be influenced via an EEM. If the charge balance is too low, an EEM can deactivate electrical consumers that serve purely comfort functions. Vehicle power net simulations can also provide users with calculations and evaluations of the effects of EEM functions on the overall system. The simulation tool presented here features a standard EEM model, a start-stop function, and an intelligent alternator control system. Under certain predefined conditions, the start-stop function turns off the internal combustion engine when the vehicle is at a standstill in order to save fuel. The intelligent alternator control system serves primarily to save fuel and to optimise battery charge. The function increases the alternator s target voltage in overrun phases. During these phases, no fuel is required to move the vehicle forward, and the vehicle s kinetic energy can be used to increase the battery charge. Conversely, the alternator s target voltage is reduced whenever the internal combustion engine is consuming fuel and operating at low efficiency. This minimises the consumption of fuel to generate electrical energy for the vehicle. The simulation tool also allows individual EEM functions to be developed and tested. In this instance, the power net simulation is used as a physical system model; in function and software development, it is used in a model-in-the-loop or software-in-theloop environment. What is more, the simulation software can be used in system optimisation, as it features a design-of-experiments interface that allows any of the parameters to be varied and the effect on the overall system to be calculated. Connecting the vehicle power net simulation with design-of-experiments tools allows a significant reduction in the number of simulation runs, which means complex systems can be optimised more efficiently. Example simulation for layout and validation In this example of a simulation for a mid-sized vehicle with average engine power and equipment level, the dimensions of the vehicle power net s components were selected following the design and layout phase, 3. In the first simulation run, the simulation model was parameterised with a 90 A alternator and a 60 Ah battery, with the new European driving cycle as a reference. The resulting battery state of charge is even. But while simulating the exhaust-gas cycle for the purposes of calculating fuel consumption, most of the electrical consumers were deactivated. In order to design 3 Simulation results for various designs of vehicle power nets a vehicle power net, however, a typical load cycle must be taken into account in which various consumers such as interior fans, headlight, and radio are all activated. After further simulation runs in which these factors were taken into account, it was determined that the optimum alternator size for this vehicle configuration was 120 A and the optimum battery size was 80 Ah. The battery s load balance is now positive during the cycle and the battery s state of charge level rises from a starting value of 76 %. In the subsequent validation phase, the components actual behavior was reproduced in the simulation with the highest possible degree of detail. The goal of validation is to evaluate interactions within the system and to examine how the design of the vehicle power net copes with extreme conditions (e.g. in the worst-case scenario). This is also the stage at which functions for operating strategies (e.g. EEM functions) are deve- 5
loped and their effects evaluated. For validation, the exact sets of parameters for the components in the test vehicle were determined and the simulation results were subsequently compared with measurements from the vehicle. The validation results show a high degree of agreement between the simulation and actual in-vehicle behaviour, 4. Conclusion Modern simulation tools make it equally quick and simple to design a vehicle power net or to run detailed simulations for system validation and function development. That means designing vehicle power nets using simulation reduces costs compared to building actual experimental vehicles, as well as reducing the time taken for vehicle development. What is more, taking a design-of-experiments approach is a particularly efficient way to optimise vehicle power net design and achieve optimum parameterisation of EEM functions. And so the expanded Bosch Engineering software for simulation of vehicle power nets supports the entire development process following the V-model. 4 Comparison of simulation and measurement results 6