The fault diagnosis for maintenance of machines operating in variable conditions requires special dedicated methods. Variable load or temperature conditions affect the vibration signal values. The article presents a new approach to diagnosing rotating machines using an artificial neural network, the training of which does not require data from the damaged machine. This is a new approach not previously found in the literature. Until now, neural networks have been used for machine diagnosis in the form of classifiers, where data from individual faults were required. A new diagnostic parameter rDPNS (Relative Differences Product of Network Statistics) as a function of the machine's shaft order was proposed as a kind of new order spectrum independent of the machine's operating conditions. The presented work analyses the use of the proposed method to diagnose misalignment and unbalance. The results of an experiment carried out in the laboratory demonstrated the effectiveness of the proposed method.

Diagnosing machines operating under variable load requires the use of special signal analysis methods. In literature there are many methods based on synchronic analysis where vibration signal is synchronized with shaft rotation of the diagnosed machine. Varying load, however, also affects the amplitude of diagnostic parameters values. The paper deals with the problem of diagnosing machines operating under variable load. The impact of load on the diagnostics parameters values has been tested on a laboratory station. Due to varying rotational speed, the diagnostic parameters were amplitudes of orders determined by means of order analysis synchronized with the shaft rotation in the diagnosed drive unit. Examined was the impact of varying load on the diagnostics effectiveness of damage, such as incorrect motor mount (soft foot) and misalignment of shafts. Single-number statistical parameters were proposed to determine the technical condition of machines operating under variable load.