In one embodiment, a method is provided. The method includes receiving a plurality of signals representative of an engine noise transmitted via a plurality of noise sensors, wherein the noise sensors are disposed in a grid about an engine. The method further includes receiving a knock sensor signal representative of an engine noise transmitted via a knock sensor. The method additionally includes deriving a combustion event based on the knock sensor signal, and deriving an engine condition based on the plurality of signals and the combustion event. The method also includes communicating the engine condition.

Monitoring of a machine is performed by an acoustic monitor which acquires, through an acoustic sensor, acoustic signals from a vicinity of a machine, while the machine is operative. A processor calculates a frequency spectrum of a segment of the acquired acoustic signals, determines boundaries of a frequency band to be analyzed and extracts, from the calculated frequency spectrum, a base frequency window in the determined boundaries, and one or more harmonics windows of harmonics of the determined boundaries. For each of the base and harmonic windows a weight based on a distribution of values of frequencies in the windows is determined and a parameter of operation of the machine is calculated as a function of a weighted sum of the base and harmonic windows. The operation of the machine is evaluated responsive to the calculated parameter.

A monitoring device and a method for detecting mechanical torsional oscillations in a rotor shaft of a turbine-generator system. Specifically, the monitoring device and a method for detecting mechanical torsional oscillations in a turbine-generator system having a plurality of, such as two, three, four, five, six or more, turbines that are interconnected by means of a common turbine shaft.

An engine health diagnostic apparatus is provided for analysing health of a reciprocating internal combustion engine. The apparatus comprises feature generation circuitry for processing vibration sensor data received from a vibration sensor detecting vibration at a component of the reciprocating internal combustion engine 4 and generating a feature vector indicating multiple features of the sensor data. Processing circuitry processes the feature vector using a trained classification model which is defined by model parameters characterising a decision boundary of healthy operation learnt from a training set of feature vectors captured during healthy operation of the engine. The model generates an engine health indication providing a quantitative indication of deviation of the feature vector from the decision boundary of healthy operation.

An engine health diagnostic apparatus is provided for analysing health of a reciprocating internal combustion engine. The apparatus comprises feature generation circuitry for processing vibration sensor data received from a vibration sensor detecting vibration at a component of the reciprocating internal combustion engine 4 and generating a feature vector indicating multiple features of the sensor data. Processing circuitry processes the feature vector using a trained classification model which is defined by model parameters characterising a decision boundary of healthy operation learnt from a training set of feature vectors captured during healthy operation of the engine. The model generates an engine health indication providing a quantitative indication of deviation of the feature vector from the decision boundary of healthy operation.

Embodiments are disclosed for enhancing engine sound. An example method for a vehicle comprises acquiring a signal including harmonic content generated by an engine of the vehicle, upmixing the signal into a plurality of channels for a given number of engine orders, adjusting an order filter for each engine order of the given number of engine orders based on operating conditions of the engine, filtering each channel of the plurality of channels with the corresponding order filter, mixing the filtered channels into a mono output, and outputting the mono output to at least one speaker in the vehicle. The mono output is delayed based on a position of the at least one speaker such that an occupant of the vehicle perceives the mono output as originating from the engine.

Embodiments are disclosed for enhancing engine sound. An example method for a vehicle comprises acquiring a signal including harmonic content generated by an engine of the vehicle, upmixing the signal into a plurality of channels for a given number of engine orders, adjusting an order filter for each engine order of the given number of engine orders based on operating conditions of the engine, filtering each channel of the plurality of channels with the corresponding order filter, mixing the filtered channels into a mono output, and outputting the mono output to at least one speaker in the vehicle. The mono output is delayed based on a position of the at least one speaker such that an occupant of the vehicle perceives the mono output as originating from the engine.

The present disclosure relates to a fault diagnosis method of a reciprocating machinery based on a keyphasor-free complete-cycle signal. The method includes the following steps: 1) building a complete-cycle vibration signal image library; 2) training an image recognition model; 3) acquiring a complete-cycle data on a keyphasor-free basis; 4) building an automatic feature extraction model; and 5) inputting a hidden layer feature of an autoencoder into a support vector machine (SVM) classifier to obtain a diagnosis result. By using a deep cascade convolutional neural network (CNN), the present disclosure achieves the goal of complete-cycle data acquisition on a keyphasor-free basis, solves the problems that traditional intelligent fault diagnosis relies on a keyphasor signal and real-time diagnosis fails due to insufficient installation space. In addition, by using an autoencoder for automatic feature extraction, the present disclosure avoids manual feature selection, reduces labor costs.

The present invention relates to a drive for an electric application such as an electric motor, said drive including at least one microphone for registering noise signals occurring at the drive, wherein the microphone is connectable to a computing device for analysing the registered noise signals. The registered noise signals may be used for a maintenance process of the drive and/or a fine-tuning process of a drive control method of the drive. The present invention also relates to a maintenance process, in particular a predictive maintenance process for a corresponding drive. Furthermore, the present invention relates to a process for fine tuning a drive control method of a corresponding drive.

Methods and systems are disclosed for operating an engine that includes a knock control system that may determine contributions of individual noise sources to an engine background noise level. The contributions of the individual noise sources may be the basis for establishing the presence or absence of knock in one or more engine cylinders.