Abstract: A fuel supply system for an active pre-combustor, including a cylinder assembly. The cylinder assembly comprises a cylinder head, a cylinder body, and a piston. The cylinder head, the cylinder body and the piston form a main combustor. The fuel supply system for the active precombustor further includes a precombustion chamber assembly and a plunger air pump assembly. The pre-combustor assembly is communicated with the main combustor. The plunger air pump assembly is communicated with the pre-combustor assembly. The plunger air pump assembly can actively mix air with fuel into a mixed fuel and deliver it into the pre-combustor assembly (20). The fuel supply system for the active pre-combustor supplies the mixed fuel to the pre-combustor on the basis of a reciprocating stroke opposite to the piston by adding the plunger air pump assembly on the pre-combustor assembly, and is safe, reliable, and efficient.

Abstract: A fuel supply system for an active pre-combustor, including a cylinder assembly. The cylinder assembly comprises a cylinder head, a cylinder body, and a piston. The cylinder head, the cylinder body and the piston form a main combustor. The fuel supply system for the active precombustor further includes a precombustion chamber assembly and a plunger air pump assembly. The pre-combustor assembly is communicated with the main combustor. The plunger air pump assembly is communicated with the pre-combustor assembly. The plunger air pump assembly can actively mix air with fuel into a mixed fuel and deliver it into the pre-combustor assembly (20). The fuel supply system for the active pre-combustor supplies the mixed fuel to the pre-combustor on the basis of a reciprocating stroke opposite to the piston by adding the plunger air pump assembly on the pre-combustor assembly, and is safe, reliable, and efficient.

Abstract: Various examples relate to an apparatus, device, method, and computer program for determining a placement of an execution of a computer program, and to an apparatus, device, method, and computer program for training at least one machine-learning model. The apparatus for determining the placement of an execution of a computer program comprises processing circuitry that is configured to generate a graph representation of a computer program, generate, using a first machine-learning model, a vector embedding of the graph representation of the computer program, and to determine, based on an output of a second machine-learning model, a placement of an execution of the computer program on one or more hardware devices of a heterogenous plurality of hardware devices of a computer system, with the vector embedding and information on the load of the hardware devices being provided as input to the second machine-learning model.

Abstract: The present invention a mixed fuel suitable for diesel engines, which consists of the following components by volume percentage: 35% to 37.5% of diesel, 35% to 37.5% of ethanol and 25% to 30% of PODE, wherein the diesel is petroleum diesel having a cetane number not less than 48; the purity of the ethanol by mass percentage is not less than 99.5%; and, the PODE is a mixture of two fuels shown by molecular formulae CH3O(CH2O)3CH3 and CH3O(CH2O)4CH3. The use of this fuel is advantageous for emission reduction.

Abstract: The present invention discloses a cold start operation device and control method for a diesel engine in a plateau region under a low ambient temperature, allowing a starter to drive a piston to compress the low-temperature air, by using the heated air as the fresh inlet air to introduce it into the cylinder, so that the in-cylinder temperature at the compressing TDC is increased to obtain the ignition temperature, and finally achieving the cold start of diesel engines in plateau regions. The present invention avoids complicated preheating devices, and has simple structure and reliable performance, can achieve effective cold start in different low ambient temperature environments.

Abstract: Systems, apparatuses and methods may provide for technology that generates a series of time-stamped object graphs based on object trajectory histories derived from external object data for a plurality of external objects, such as vehicles. The technology may also generate, via a first neural network such as a graph attention network, a series of relational object representations based on the series of time-stamped object graphs, and determine, via a second neural network such as a long short-term memory network, predicted object trajectories for the plurality of external objects based on the series of relational object representations. The technology may also modify behavior of an autonomous vehicle based on the predicted object trajectories and real-time perceptual error information.