Model-based dynamic compensation of load cell response in weighing machines affected by environmental vibrations

This work proposes and validates experimentally a model-based scheme for the compensation of environmental vibrations affecting load cell measurements in automatic weighing machines. Weighing machines are often adversely affected by low frequency vibrations which may arise from the coupled effects of machine frame flexibility and the excitation induced by internal (inertial) or external (impact) vibration sources. These vibrations are generally named “environmental vibrations”, since they seem to arise from the environment around the machine. Environmental vibrations have a detrimental effect on load cell responses and can in turn deteriorate machine performances. It is usually ineffective to try overcoming this problem by low-pass filtering load cell measurements: low cut-off frequencies usually downgrade machine speed by both introducing delay and increasing filtered signal rise time. Since automatic weighing machines need to operate at ever increasing speed, alternative approaches must be investigated. In this work it is suggested to make use of the mechanical models of the weighing machine and the load cells to process supplementary accelerometer measurements and compute an effective compensation of the effect of environmental vibrations on load cell response. The technique is here applied to a multi-head weighing machine in order to prove its effectiveness and implementability in industrial devices with real-time controllers. ► Environmental vibrations badly affect load cell measurements in weighing machines. ► We propose a dynamic-model-based scheme for environmental vibration compensation. ► The method takes advantage of accelerometers and a multibody model of the system. ► The method has been implemented in an industrial machine with a real-time controller. ► We report experimental results proving the effectiveness of the method.