The general design considerations for feedforward artificial neural networks (ANNs) to perform motor fault detection are presented. A few noninvasive fault detection techniques are discussed, including the parameter estimation approach, human expert approach, and ANN approach. A brief overview of feedforward nets and the backpropagation training algorithm, along with its pseudocodes, is given. Some of the neural network design considerations such as network performance, network implementation, size of training data set, assignment of training parameter values, and stopping criteria are discussed. A fuzzy logic approach to configuring the network structure is presented. >

/pdf/on-the-application-and-design-of-artificial-neural-networks-3h7jp6prxh.pdf

On the application and design of artificial neural networks for motor fault detection. II

A neural network-based incipient fault detector for small and medium-size induction motors is developed. The detector avoids the problems associated with traditional incipient fault detection schemes by employing more readily available information such as rotor speed and stator current. The neural network design is evaluated in real time in the laboratory on a 3/4 hp permanent magnet induction motor. The results of this evaluation indicate that the neural-network-based incipient fault detector provides a satisfactory level of accuracy, greater than 95%, which is suitable for real-world applications. >

/pdf/a-neural-network-approach-to-real-time-condition-monitoring-15ysau1j7z.pdf

A neural network approach to real-time condition monitoring of induction motors

The use of electric motors in industry is extensive. These motors are exposed to a wide variety of environments and conditions which age the motor and make it subject to incipient faults. These incipient faults, if left undetected, contribute to the degradation and eventual failure of the motors. Artificial neural networks have been proposed and have demonstrated the capability of solving the motor monitoring and fault detection problem using an inexpensive, reliable, and noninvasive procedure. However, the major drawback of conventional artificial neural network fault detection is the inherent black box approach that can provide the correct solution, but does not provide heuristic interpretation of the solution. Engineers prefer accurate fault detection as well as the heuristic knowledge behind the fault detection process. Fuzzy logic is a technology that can easily provide heuristic reasoning while being difficult to provide exact solutions. The authors introduce the methodology behind a novel hybrid neural/fuzzy system which merges the neural network and fuzzy logic technologies to solve fault detection problems. They also discuss a training procedure for this neural/fuzzy fault detection system. This procedure is used to determine the correct solutions while providing qualitative, heuristic knowledge about the solutions. >

/pdf/using-a-neural-fuzzy-system-to-extract-heuristic-knowledge-4wfwffpodl.pdf

Using a neural/fuzzy system to extract heuristic knowledge of incipient faults in induction motors. Part I-Methodology

The principles of system protection and the proper selection, application, and coordination of components that may be required to protect industrial and commercial power systems against abnormalities that could reasonably be expected to occur in the course of system operation are presented in a in a simple, yet comprehensive, format. The principles presented apply to both new electrical system design and to the changing, upgrading, or expansion of an existing electrical distribution system.

IEEE Recommended Practice for Protection and Co-ordination of Industrial and Commercial Power Systems

A novel approach for online detection of incipient faults in single-phase squirrel-cage induction motors through the use of artificial neural networks is presented. The online incipient fault detector is composed of two parts: (1) a disturbance and noise filter artificial neural network to filter out the transient measurements while retaining the steady-state measurements, and (2) a high-order incipient fault detection artificial neural network to detect incipient faults in single-phase squirrel-cage induction motors based on data collected from the motor. Simulation results show that neural networks yield satisfactory performance for online detection of incipient faults in single-phase squirrel-cage induction motors. The neural network fault detection methodology presented is not limited to single-phase squirrel-cage motors (used as a prototype), but can also be applied to many other types of rotating machines, with the appropriate modifications. >

Methodology for on-line incipient fault detection in single-phase squirrel-cage induction motors using artificial neural networks

Motors of modern design have less tolerance to overloads than their predecessors. In addition, present-day applications tend to utilize the maximum capability of motors. Induction -motor stator insulation or rotor bars may be severely damaged by over temperature due to repeated starts, locked rotors, cyclic loads and other conditions for which existing overload devices do not provide adequate protection.

Thermal Tracking-A Rational Approach to Motor Protection

A method and apparatus for detecting motor contamination utilizes zero-sequence currents in a sensing coil to detect contamination. The detector includes a current transformer comprised of a magnetic core and a sensing coil wound about the core. A processor processes a sub-trip level current from the sensing coil to detect contamination. Three contamination detection methods, which may be used separately or in desired combination, are based on current trends, current leakage, and current clustering. An alarm is triggered when predetermined alarm criteria have been met.

Motor winding contamination detector and detection

In isolated phase bus duct, a circular bus bar is supported coaxially inside a circular housing by means of a number of annular insulators spaced along the length of the bus inside the housing. Each insulator is in the form of a collar having a mid ring portion with annular skirts on both ends. The insulator surrounds the bus and extends radially outward from it to the housing. An elastomeric ring surrounds the bus at either end of the insulator and is compressed lightly between the bus and the inner sloping surface of the inner skirt on the insulator by means of a number of tabs secured to the bus. The outer periphery of the insulator has an annular groove which contains an elastomeric ring compressed lightly between the insulator and the housing. These rings cushion the assembly and allow the bus to move towards its position of zero forces from short circuit currents. This reduces the short circuit stresses imposed on the assembly.

Bus duct assembly

An electrical current collector system includes a rotatable shaft (28); an electrically conductive slip ring (27); and a slip ring support assembly (42). The slip ring support assembly (42) comprises plural post insulators (102) spaced radially about the rotatable shaft for attaching the conductive slip ring to the rotatable shaft. The post insulators facilitate adjusting of the position of the slip ring relative to the rotatable shaft, e.g., to provide concentric positioning. The post insulators comprise an attachment/adjustment assembly (106) for connecting to the slip ring. In one embodiment, the attachment /adjustment assembly comprises a spring plate (134) which is adjustably biased to provide a preloading displacement that absorbs differential movement due to temperature changes of the slip ring an the rotatable shaft. In one example deployment, the current collector system is provided as part of a rotary transformer system (20) which, e.g., transfers electrical power between a first electrical system (e.g., first electrical grid) and a second electrical system (e.g., second electrical grid).

Slip-ring mounting assembly for high-power rotary current collector system

In an electrical substation, the use of separate towers or poles to terminate overhead lines is obviated by using bus duct for this purpose. Isolated phase or metal-clad bus duct is erected to provide not only the electrically conductive path between the overhead lines and the associated electrical equipment, but also the mechanical strength necessary in order to eliminate the towers and poles of the prior art.

Robert Henry Rehder

Papers

Thermal Tracking-A Rational Approach to Motor Protection

Motor winding contamination detector and detection

Bus duct assembly

Slip-ring mounting assembly for high-power rotary current collector system

Structure for electrical interface