Li-Ion Cell Safety Monitoring Using Mechanical Parameters: Part I. Normal Battery Operation

The normal operation of a 18650 Lithium-ion cells has been monitored using rectangular rosette strain gauge and a pair of piezoelectric transducers. The sensors for mechanical measurements provide information about the cell deformation mechanism and electrodes structure during the cycling. The strain gauge signal revealed three type of mechanical processes. The predominant deformation pattern during galvanostatic discharge process is an isotropic cylindrical shrinkage relevant to the extraction of lithium ions from the graphite negative electrode. In the case of low-rate discharge in cyclic voltammetry mode, the deformation pattern changes to spherical growth when the state of charge falls below 40. In contrast, the thermal shrinkage and growth of the cell corresponds to simple decrease of the cell diameter with much smaller hysteresis effect. The ultrasound interrogation is able to detect repeatable progressive change of the acoustic waveform transferred across the cell in direction of the jellyroll diameter, which depends on the state of charge and does not undergo any significant changes at different cycling rates. The impact of the state of health under 2h – rated charge/discharge cycling at 25°C reveals slow progressive drift of the strain and acoustic signals corresponding to the growth of the cell size.

Machine Learning Methods for Automatic Silent Speech Recognition Using a Wearable Graphene Strain Gauge Sensor

Silent speech recognition is the ability to recognise intended speech without audio information. Useful applications can be found in situations where sound waves are not produced or cannot be heard. Examples include speakers with physical voice impairments or environments in which audio transference is not reliable or secure. Developing a device which can detect non-auditory signals and map them to intended phonation could be used to develop a device to assist in such situations. In this work, we propose a graphene-based strain gauge sensor which can be worn on the throat and detect small muscle movements and vibrations. Machine learning algorithms then decode the non-audio signals and create a prediction on intended speech. The proposed strain gauge sensor is highly wearable, utilising graphene’s unique and beneficial properties including strength, flexibility and high conductivity. A highly flexible and wearable sensor able to pick up small throat movements is fabricated by screen printing graphene onto lycra fabric. A framework for interpreting this information is proposed which explores the use of several machine learning techniques to predict intended words from the signals. A dataset of 15 unique words and four movements, each with 20 repetitions, was developed and used for the training of the machine learning algorithms. The results demonstrate the ability for such sensors to be able to predict spoken words. We produced a word accuracy rate of 55% on the word dataset and 85% on the movements dataset. This work demonstrates a proof-of-concept for the viability of combining a highly wearable graphene strain gauge and machine leaning methods to automate silent speech recognition.

Shaping the Design Features of a Dynamometer for Measuring Resistance Biaxial Components of Symmetrical Coulters

The article lays out the methodology for shaping the design features of a strain gauge transducer, which would make it possible to study forces and torques generated during the operation of symmetrical seeder coulters. The transducers that have been known up until now cannot be used to determine forces and torques for the coulter configuration adopted by the authors. For this purpose, the design of the transducer in the form of strain gauge beams was used to ensure the accumulated stress concentration. A detailed design was presented in the form of a 3D model, along with a transducer body manufactured on its basis, including the method for arranging the strain gauges thereon. Moreover, the article discusses the methodology of processing voltage signals obtained from component loads. Particular attention was paid to the methodology of determining the load capacity of the transducer structure, based on finite element method (FEM). This made it possible to choose a transducer geometry providing the expected measurement sensitivity and, at the same time, maintaining the best linearity of indications, insignificant coupling error, and a broad measurement range. The article also presents the characteristics of the transducer calibration process and a description of a special test stand designed for this purpose. The transducer developed within the scope of this work provides very high precision of load spectrum reads, thus enabling the performance of a detailed fatigue analysis of the tested designs. Additionally, the versatility it offers makes it easy to adapt to many existing test stands, which is a significant advantage because it eliminates the need to build new test stands.

Strain gauge placement optimization methodology to measure multiaxial loads of complex structure

The use of a strain gauge to measure loads is, in some respects, similar to its use in determining stress, but a different approach is required. In load measurement, it is necessary to compile a suitably selected configuration of strain gauges, which can be used to measure often very complex loads of the structure. For designing the engine mount instrumentation for the Flying Test Bed, an optimization tool has been developed. The algorithm and the theory behind the instrumentation design are described in detail. The basic principle is to find the strain gauge configuration that eliminates the measurement error due to the noise in the measured signal as much as possible. The input for optimization is the strain response of the structure to the applied loads analyzed using the FE model. In contrast to the common strategy using purely stochastic methods, this developed tool uses a hybrid approach based on a combination of a heuristic approach with repeated deterministic local optimization. The optimization is focused on the connection of a simple uni-axial strain gauge to a quarter-bridge and a T-rosette to a half-bridge that provides temperature compensation. Furthermore, an approach is proposed that takes into account the possibility of failure of some strain gauges. The instrumentation is thus robust and allows to obtain quality data even in the event of failure of some of the strain gauges.

Processing of strain gauge control signals by the Kalman filter

One of the methods for detecting defects in the rolling surface of the wheels of freight cars is to measure the deformations of the rail under the passing train. The method is based on the analysis of a strain gauge signal. The main task of the strain gauge signal analysis is the selection of informative components and the removal (filtering) of interference. The paper presents methods of filtering diagnostic signals of strain gauge control and the selection of informative components. The useful signal component can be used to measure the mass of cars, to determine the dynamic load on the rails and to detect defects in the rolling surface of the wheels. The method of adaptive Kalman filtering and linear convolution are proposed as signal processing tools. Based on these algorithms, a software module based on the.NET Framework has been developed using the C# programming language. The algorithms were tested on the signals received when the train was moving along the active section of the track, with a strain gauge control system located on it. The computational complexity and speed of the algorithms are assessed, and the possibility of their further application in the autonomous mode of the system is investigated. The results show that the use of the Kalman filtering algorithm provides a significant performance advantage over the linear convolution algorithm.

About the dynamic error of strain gauge torque measuring devices

The dynamic error of devices belongs to the number of errors that are difficult to estimate. The mechanism for forming this error on the example of torsional torque dynamometersis briefly considered as the most common in the practice of research on the energy of agricultural machines. The limiting ratios of frequencies of external influences and natural vibrations of strain-measuring devices are given. Recommendations are made to reduce dynamic error in torque strain analysis. The present review and the accumulated experience of strain gauge research allows us to recommend some directions for reducing dynamic errors in torque measurements. In order for a strain gauge to keep up with changes in torque in its inertial characteristics, it must have the smallest torque inertia. In order to reduce the probability of high-frequency harmonics from the elastic vibrations of the strain gauge, it must be sufficiently rigid. From this point of view, strain rods, strain sprockets, etc., having a moment of inertia greater than that of the same gear parts, are irrational. Based on modern micromodules and power supplies, the system can have a small mass and have no significant effect on the inertial characteristic of the strain gauge.