A method for operating an injection valve by ascertaining an opening time and/or closing time of the injection valve on the basis of a sensor signal. The method includes: providing an analysis point time series by sampling a sensor signal of a sensor of the injection valve; using a nonlinear, data-based first submodel in order to obtain a first model output on the basis of the analysis point time series; using a linear, data-based second submodel in order to obtain a second model output on the basis of the analysis point time series; ascertaining the opening time and/or closing time as a function of the first and second model outputs.

A computer-implemented method for training a data-based time determining model for determining an opening or closing time of an injection valve based on a sensor signal. The method includes: providing an unlabeled analysis point time series by sampling the sensor signal of a sensor of the injection valve; training the data-based time determining model to assign a time specification which represents a specific opening or closing duration to an analysis point time series, the training process being carried out using a first shifting function to time-shift the analysis point time series and a second shifting function in order to time-shift the time specification. A consistency loss function is used for the training process.

An electronic control unit for a vehicle with a combustion engine and a method of fuel analysis are provided. At least one dynamic torque sensor value from a high pressure pump of the vehicle and at least one additional sensor value including at least one pressure sensor value and/or at least one timing value are used to determine whether a combustible fuel type currently in use is known, unknown, or similar to a known fuel type. In each case, the operation of the combustion engine is optimized using specific parameter configurations for the fuel injectors of the vehicle. The specific parameter configurations are either retrieved from a database, or are generated using artificial intelligence methods.

A system comprising

a distribution grid (2) for a fuel,

combustion engines (3), which are coupled with the distribution grid (2) and are configured to combust the fuel, and

a computer system (4) comprising data connections (5) to the combustion engines (3) and a data storage device (6), wherein the computer system (4) is configured to receive engine operation parameters stemming from an operation of the combustion engines (3) at a first time and/or during a first time period via the data connections (5) and geographical data of the combustion engines (3) are stored in the data storage device (6), wherein

the computer system (4) has a processor (7) which is configured to compute a prediction for at least one characteristic parameter of the fuel at a second time and/or during a second time period later than the first time and/or the first time period and with respect to a geographical location, and

the computation of the prediction being based on the geographical data and the engine operation parameters of the combustion engines (3).

A method of generating vehicle control data is provided. The method is executed using a processor and a storage device and includes: storing first data that prescribe a relationship between a state of a vehicle and an action variable that indicates an action related to an operation of an electronic device; acquiring a detection value from a sensor that detects the state of the vehicle; operating the electronic device; calculating a reward, on the basis of the acquired detection value; in a case where a predetermined condition is met, updating the first data using, as inputs to update mapping determined in advance, the state of the vehicle, a value of the action variable, and the reward; and in a case where the state of the vehicle does not meet the predetermined condition, obtaining second data by adapting the relationship between the state of the vehicle and the action variable.

A system for training a deep learning system to detect engine knock with accuracy associated with high fidelity knock detection sensors despite using data from a low fidelity knock detection sensor. The system includes an engine, a high fidelity knock detection sensor, a low fidelity knock detection sensor, and an electronic processor. The electronic processor is configured to receive first data from the high fidelity knock detection sensor. The electronic processor is also configured to receive second data from the low fidelity knock detection sensor. The electronic processor is further configured to map the first data to the second data, train the deep learning system, using training data including the mapped data, to determine a predicted peak pressure using data from the low fidelity knock detection sensor, receive third data from the low fidelity knock detection sensor, and using the third data, determine the predicted peak pressure.

A method and arrangement for ascertaining an amount of substance emitted as a result of the operation of a functional unit of a vehicle includes transmitting signals from a signal source, which are generated independently of the amount of substance to be ascertained, as input data, to a data processing apparatus. The data processing apparatus contains at least one neural network as a trained model for processing the input data. The method includes generating output data representing the emitted amount of substance in the data processing apparatus using the at least one neural network.

The invention relates to a method for the operational analysis of an engine and/or for calibrating a controller of the engine, in particular an internal combustion engine, wherein run-up occurs of test points defined by values of a plurality of predetermined operating parameters and selected from a multidimensional test space using a statistical experiment design, whereby at least one operating parameter is in each case changed from one test point to the next test point in a plurality of steps in the run-up of the test points, wherein operational measurements are performed at measurement points resulting from a respective increment and at the actual test points, whereby measurement data from the operational measurements for the analysis and calibration of the controller are output and continuously stored, as well as a corresponding system.

A neural network method, comprising: modeling an environment; implementing a policy based on the modeled environment, to perform an action by an agent within the environment, having at least one estimated dynamic parameter; receiving an observation and a temporally-associated cost or reward based on operation of the agent in the environment controlled according to the policy; and updating the policy, dependent on the received observation and the temporally-associated cost or reward, to improve the policy to optimize an expected future cumulative cost or reward. The policy may represent a set of parameters defining an artificial neural network having a plurality of hierarchical layers and having at least one layer which receives inputs representing aspects of the received observation indirectly from other neurons, and produce outputs to other neurons which indirectly implement the policy, the plurality of hierarchical layers being trained according to respectfully distinct training criteria.

A storage device stores a first mapping that receives, as an input, first input variables including a previously estimated value for a fuel temperature variable, a pump variable on a state of a fuel pump, a first engine variable on a state of an engine, and an outside air temperature variable on an outside air temperature, and outputs the fuel temperature variable. Further, an execution device is configured to acquire the first input variables and estimate the fuel temperature variable by applying the acquired first input variables to the first mapping. Therefore, it is possible to estimate the fuel temperature variable by applying the first input variables to the first mapping even without providing a temperature sensor.