Joint Compression Research Articles

High-precision State of Charge Estimation of Lithium-ion Batteries Based on Joint Compression Factor Particle Swarm Optimization-Forgetting Factor Recursive Least Square-Adaptive Extended Kalman Filtering

Accurate state of charge (SOC) estimation is an important basis for battery energy management and the applications of lithium-ion batteries. In this paper, an improved compression factor particle swarm optimization-forgetting factor recursive least square (CFPSO -FFRLS) algorithm is proposed, in which the forgetting factor is optimized to identify more accurate parameters for high-precision SOC estimation of lithium-ion battery. In order to improve the SOC estimation accuracy, a dual noise update link is introduced to the traditional extended Kalman filter (EKF), which enhances the algorithm’s ability to adapt to noise by updating the process and measurement noises in real time. The experimental results of parameter identification and SOC estimation show that the CFPSO-FFRLS algorithm proposed significantly improves the accuracy of parameter identification, and the joint CFPSO-FFRLS-AEKF algorithm can accurately estimate the SOC of lithium-ium battery under different working conditions. Under HPPC, BBDST and DST working conditions, the mean absolute errors of SOC estimation are 1.14%, 0.78% and 1.1%, which are improved by 42.71%, 65.79% and 39.56% compared with FFRLS-EKF algorithm, and the root mean square errors are 1.18%, 0.99% and 1.11%, improved by 44.86%, 65.98% and 51.74%, respectively.

A Consequence of Failure of Mid Span Compression Joint on the Kelanitissa-Kolonnawa Line

Mid-span compression joints in overhead transmission line conductors need to give not only reliability, but also electrical and mechanical properties equivalent to the original conductor. The use of an inappropriate joint can lead to disastrous consequences. An example of this is the complete blackout of the Sri Lankan power system in October 2009 through the use of an inadequate size joint. This paper is based on investigations which revealed that the initiating cause of the blackout, was the failure of a previously replaced inadequate mid-span straight-through joint between ZEBRA and ZTACIR conductors. Inadequate line protection allowed it to progress to a complete blackout. A detailed analysis of the failure is presented and provides information to avoid such occurrences in the future.

Mechanical Properties of Deep Variable Dip Joint Rock Mass in Reservoir Area under Wet and Dry Conditions

The mechanical property of deep complex jointed rock mass is a hot topic in rock mechanics. In order to grasp the deformation and damage rules of through-going variable dip joint rock masses, the triaxial compression test and joint surface morphology scan test were conducted on cylindrical specimens in dry and wet conditions under 0, 5, 10, and 20 MPa water pressure. Through these tests, the deterioration law of rock samples under different pore water pressure in dry and wet conditions was studied. The effect of pore water pressure on the strength of saturated jointed rock sample is nonlinear. The imposed pore water pressure can significantly increase the axial deformation of rock samples, and higher pore water pressure can facilitate the deformation deterioration of samples. When the pore pressure is high, the rock samples show the characteristics of sliding shear failure. Under different pore pressure, the compressive strength of dry jointed rock is significantly higher than that of saturated jointed rock, and the saturated rock is more susceptible to sliding in the joint plane than dry rock. Dry jointed rock samples have stronger deformation ability than saturated jointed rock samples. The change rate of the morphologic parameters and the distribution of the failure cracks indicate that the stress concentration is evident in the middle of the joint plane.

Accelerating distributed machine learning with model compression and graph partition

The rapid growth of data and parameter sizes of machine learning models makes it necessary to improve the efficiency of distributed training. It is observed that the communication cost usually is the bottleneck of distributed training systems. In this paper, we focus on the parameter server framework which is a widely deployed distributed learning framework. The frequent parameter pull, push, and synchronization among multiple machines leads to a huge communication volume. We aim to reduce the communication cost for the parameter server framework. Compressing the training model and optimizing the data and parameter allocation are two existing approaches to reducing communication costs. We jointly consider these two approaches and propose to optimize the data and parameter allocation after compression. Different from previous allocation schemes, the data sparsity property may no longer hold after compression. It brings additional opportunities and challenges for the allocation problem. We also consider the allocation problem for both linear and deep neural network (DNN) models. Fixed and dynamic partition algorithms are proposed accordingly. Experiments on real-world datasets show that our joint compression and partition scheme can efficiently reduce communication overhead for linear and DNN models.

Effect of Arthrodesis Device Type and Trajectory on Subtalar Joint Compression

The use of subtalar arthrodesis procedures has been widely implemented to relieve hindfoot issues after failure of conservative treatments; however, fusion failures persist in some patients with certain risk factors. Currently, surgeons utilize cannulated screws in these arthrodesis procedures to immobilize the subtalar joint. Recent clinical studies have demonstrated improved fusion outcomes in at-risk patients using sustained dynamic compression devices in the tibiotalocalcaneal complex. These devices utilize pseudoelastic nitinol which enables sustained dynamic compression when faced with postoperative bone resorption, joint settling, and bone relaxation. While the clinical success of these devices has been established in the tibiotalocalcaneal complex, the ability of sustained dynamic compression devices to apply joint compression in the subtalar joint has not been quantified. As such, the goals of this study were to (1) compare the ability of static compression devices and sustained dynamic compression devices to apply joint compression and (2) assess the impact of device trajectory on joint compression. A custom mechanical testing fixture was utilized to test the compression applied across the subtalar joint by one sustained dynamic compression device (in anterior and posterior trajectories) as compared to two cannulated screws (in both parallel and diverging trajectories). Testing revealed the sustained dynamic compression devices generated 53% greater compression as compared to the static compression devices, despite single versus dual device usage, respectively. Additionally, both types of devices applied joint compression forces in an insertion trajectory-independent manner. These data illustrate the ability of a single SDC device to maintain significantly improved joint compressive forces as compared to two static cannulated screws, regardless of insertion trajectory. These SDC devices may be of particular interest for at risk patients or in revision cases.

Unsourced Random Access Over Frequency-Selective Channels

We propose and study unsourced random access over frequency-selective channels via orthogonal frequency division multiplexing, employed to mitigate the channel effects. The decoder utilizes a compressed sensing-based joint activity detection and channel estimation algorithm coupled with an algorithm treating interference as noise and successive interference cancellation. Our results demonstrate that the proposed method offers a competitive performance with grant-based frequency division multiple access; and that, the performance loss with estimated channel state information is less than 2 dB when the number of active users is not very large.

Exploring the modification factors of exercise therapy on biomechanical load in patients with knee osteoarthritis: a systematic review and meta-analysis.

The objective of this systematic review and meta-analysisis to clarify the effect of exercise therapy on the first peak knee adduction moment (KAM), as well as other biomechanical loads in patients with knee osteoarthritis (OA), and identify physical characteristics that influence differences in biomechanical load after exercise therapy. The data sources are PubMed, PEDro, and CINAHL, from study inception to May 2021. The eligibility criteria include studies evaluating the first peak (KAM), peak knee flexion moment (KFM), maximal knee joint compression force (KCF), or co-contraction during walking before and after exercise therapy in patients with knee OA. The risk of bias was independently assessed by two reviewers usingPEDro and NIH scales. Among 11 RCTs and nine non-RCTs, 1119 patients with knee OA were included (average age: 63.7 years). As the results of meta-analysis, exercise therapy tended to increase the first peak KAM (SMD 0.11; 95% CI: -0.03-0.24), peak KFM (SMD 0.13; 95% CI: -0.03-0.29), and maximal KCF (SMD 0.09; 95% CI -0.05-0.22). An increased first peak KAM was significantly associated with a larger improvement in knee muscle strength and WOMAC pain. However,the quality of evidence regarding the biomechanical loads was low-to-moderate according to the GRADE approach.The improvement in pain and knee muscle strength may mediate the increase in first peak KAM, suggesting difficulty in balancing symptom relief and biomechanical load reduction. Therefore, exercise therapy may satisfy both aspects simultaneously when combined with biomechanical interventions, such as a valgus knee brace or insoles. Registration: PROSPERO (CRD42021230966).

Peak knee joint moments accurately predict medial and lateral knee contact forces in patients with valgus malalignment

Compressive knee joint contact force during walking is thought to be related to initiation and progression of knee osteoarthritis. However, joint loading is often evaluated with surrogate measures, like the external knee adduction moment, due to the complexity of computing joint contact forces. Statistical models have shown promising correlations between medial knee joint contact forces and knee adduction moments in particularly in individuals with knee osteoarthritis or after total knee replacements (R2 = 0.44–0.60). The purpose of this study was to evaluate how accurately model-based predictions of peak medial and lateral knee joint contact forces during walking could be estimated by linear mixed-effects models including joint moments for children and adolescents with and without valgus malalignment. Peak knee joint moments were strongly correlated (R2 > 0.85, p < 0.001) with both peak medial and lateral knee joint contact forces. The knee flexion and adduction moments were significant covariates in the models, strengthening the understanding of the statistical relationship between both moments and medial and lateral knee joint contact forces. In the future, these models could be used to evaluate peak knee joint contact forces from musculoskeletal simulations using peak joint moments from motion capture software, obviating the need for time-consuming musculoskeletal simulations.

A Joint Channel Estimation and Compression Method Based on GAN in 6G Communication Systems

Due to the increasing popularity of communication devices and vehicles, the channel environment becomes more and more complex, which makes conventional channel estimation methods further increase the pilot overhead to maintain estimation performance. However, it declines the throughput of communication networks. In this paper, we provide a novel two-stage based channel estimation method by using generative adversarial networks (GANs) to handle this problem in orthogonal frequency division multiplexing (OFDM) systems. Specifically, the first stage aims to learn the mapping from a low-dimensional latent variable to the real channel sample. During the second stage, an iterative algorithm method is designed to find the optimal latent variable by matching the pilot channels of a real channel and generated channel. Then, the data channels are recovered based on the learned mapping relationship between the latent variable and the real channel sample. The simulation results show that our proposed method can achieve a performance gain of more than 2 dB with a pilot reduction by 75% when SNR is 10 dB, by comparing with the widely used Wiener filter interpolation method. In addition, as the low-dimensional latent variable can be obtained simultaneously, it can also be used for reducing the feedback overhead.

Patient-Reported and Radiographic Outcomes After Revision Sacroiliac Joint Fusion.

Sacroiliac joint fusion (SIJF) has been established as an effective treatment for sacroiliac joint dysfunction. However, failure necessitating revision has been reported in up to 30% of cases. Little is known regarding outcomes of revision SIJF. We retrospectively reviewed all revision SIJF at a single academic center between 2017 and 2020. Revision surgery was performed using the principles of joint decortication, bone grafting, compression, and rigid internal fixation. Outcomes were assessed at 6 months and 1 year after surgery using the Oswestry Disability Index (ODI), Numeric Pain Rating Scale (NPRS), and Single Assessment Numeric Evaluation (SANE) scale. Fusion was assessed using computed tomography at 12 months postoperatively. Eighteen revision SIJFs in 13 patients were included. The mean age was 55.8 years (range 35-75). Mean body mass index was 27.9 (range 21.7-36.7). Sixty-two percent of the patients were women. The indications for revision were pseudarthrosis without fixation failure in 14 cases (77.8%), hardware failure (loosening) in 3 cases (16.7%), and continued pain after partial fusion in 1 case (5.6%). ODI and NPRS scores demonstrated significant statistical and clinical improvements at all timepoints. Mean (SD) ODI scores improved from 53.8 (19.9) preoperative to 37.5 (19.8) at 6 months and 32.9 (21.7) at 12 months. Improvement in ODI was found in 15 joints (83.3%), and the minimal clinically important difference (MCID) was achieved in 12 joints (66.7%). Mean (SD) NPRS scores improved from 6.5 (1.4) preoperative to 3.2 (2.8) at 6 months and 3.4 (2.6) at 12 months. Improvement in NPRS was also identified in 17 joints (94.4%), and 10 joints (55.6%) achieved MCID for NPRS. Mean (SD) SANE score was 72.0% (30.8) at 6 months and 70.0% (33.8) at 12 months. There were no radiographic lucencies, implant subsidence, or implant fractures at final follow-up. We identified an 88.9% fusion rate with definitive bridging bone across the sacroiliac joint. Utilizing a principles-based technique of joint decortication, compression, and rigid internal fixation, revision SIJF showed an improvement in patient-reported outcomes as well as high rate of fusion at 12 months. The most common indications for revision SIJF are symptomatic pseudarthrosis and implant loosening. This is the largest series of revision SIJF to date.

Tibial tubercle osteotomy to unload the patellofemoral joint

The tibial tubercle osteotomy (TTO) is a versatile tool for unloading patellar and trochlear focal chondral defects as well as treating patellofemoral arthritis with or without patellofemoral instability. The TTO indirectly functions to address chondral defects in the patellofemoral joint (PFJ) by unloading compressive forces in the affected area and preventing further chondral deterioration and pain. Since its inception, the TTO technique has undergone iterative modification and refinement. The Fulkerson anteromedialization has persisted as the workhorse of modern TTO procedures with good surgical results, though it is not without limitations. Recently, the use of novel surgical systems such as the multidirectional tibial tubercle transfer as well as concomitant soft tissue balancing procedures have been implemented to address challenges posed by the classic Fulkerson technique and improve TTO customizability and reproducibility. In addition, TTO has increasingly been used in conjunction with concomitant procedures such as autologous chondrocyte implantation and osteochondral allograft transfer in the PFJ, to protect the restored cartilage from repeat injury. Though the anteriorization TTO has long been supported to help alleviate pathologic joint compression, recent advancements in surgical guides as well as combined procedures involving multiplane TTO, cartilage restoration, and soft tissue balancing allow for customization of treatment that more thoroughly addresses the complex anatomy and pathology in the PFJ. This article aims to provide an overview of the rationale and recent advancements in the use of TTO to unload and treat chondral defects in the PFJ.

C2SP-Net: Joint Compression and Classification Network for Epilepsy Seizure Prediction.

Recent developments in brain-machine inter-face technology have rendered seizure prediction possible. However, the transmission of a large volume of electro-physiological signals between sensors and processing apparatuses and the related computation become two major bottlenecks for seizure prediction systems due to the constrained bandwidth and limited computational resources, especially for power-critical wearable and implantable medical devices. Although many data compression methods can be adopted to compress the signals to reduce communication bandwidth requirement, they require complex compression and reconstruction procedures before the signal can be used for seizure prediction. In this paper, we propose C2SP-Net, a framework to jointly solve compression, prediction, and reconstruction without extra computation overhead. The framework consists of a plug-and-play in-sensor compression matrix to reduce transmission bandwidth requirements. The compressed signal can be utilized for seizure prediction without additional reconstruction steps. Reconstruction of the original signal can also be carried out in high fidelity. Compression and classification overhead from the energy consumption perspective, prediction accuracy, sensitivity, false prediction rate, and reconstruction quality of the proposed framework are evaluated using various compression ratios. The experimental results illustrate that our proposed framework is energy efficient and outperforms the competitive state-of-the-art baselines by a large margin in prediction accuracy. In particular, our proposed method produces an average loss of 0.6% in prediction accuracy with a compression ratio ranging from 1/2 to 1/16.

Motor rehabilitation of aphasic stroke patient: the possibility of Rood's approach.

Motor rehabilitation of aphasic stroke patient: the possibility of Rood's approach.

Intra-cervical gastric conduit bulging after esophagectomy due to compression of the sternoclavicular joint: a report of two cases

Intra-cervical gastric conduit bulging after esophagectomy due to compression of the sternoclavicular joint: a report of two cases

Near-Instantaneously Adaptive Learning-Assisted and Compressed Sensing-Aided Joint Multi-Dimensional Index Modulation

Near-Instantaneously Adaptive Learning-Assisted and Compressed Sensing-Aided Joint Multi-Dimensional Index Modulation

Numerical investigation of the compression–bending stiffness of segmental joints with different types of joint surfaces

The compression–bending stiffness of segmental joints plays a crucial role in the analysis of segment lining structures, and setting reasonable values for joint compression–bending stiffness is a key determinant of the accuracy of mechanical calculations of lining structures. In this study, a three-dimensional refined numerical model of segmental joints is established, and solid elements are used to simulate important components, such as segments and bolts, which essentially reproduce the structural characteristics of the joints and accurately reflect the contact relationship between the segments. Full-scale compression–bending tests are conducted to verify the numerical model, and the compression–bending calculation and analysis of six different segmental joints are performed. The characteristic equations for calculating the compression–bending stiffness of segmental joints with different thicknesses are determined. The results show that the refined three-dimensional model can accurately reflect the compression–bending stiffness of the segmental joints, and the deviations between the numerical and the experimental results appear when damage phenomenon occur on the joint surface. Axial force can improve the compression–bending stiffness of segmental joints, and its influence increases with an increase in the bending moment and decreases with an increase in the thickness. The moment–rotation curves of the joints with different thicknesses are similar. Moreover, increasing the thickness effectively improves the compression–bending stiffness of segmental joints, and the influence increases with increasing bending moment or decreasing axial force. The moment–rotation relationship of a segmental joint has two characteristic curves, I and II, and the type of characteristic curve primarily depends on the structural characteristics of the joint surface. The characteristic equations based on the data-fitting method and the moment–rotation data of segmental joints with different thicknesses can be used to simply and accurately calculate the compression–bending stiffness of segmental joints.

Towards improving the accuracy of musculoskeletal simulation of dynamic three-dimensional spine rotations with optimizing model and algorithm.

The accuracy of musculoskeletal simulations greatly relies on model structures and optimization algorithms. This study investigated the unclarified influence of accounting for several commonly-simplified different model components and optimization criteria on spinal musculoskeletal simulations. The study constructed a full-body musculoskeletal model with passive components of functional spinal units and spinal muscles subject-specifically refined. A muscle redundancy solver was built with 15 optimization criteria. Three-dimensional spine rotations and spinal muscle activities were measured using optical motion capture and electromyogram techniques when eight healthy volunteers performed standing, flexion/extension, lateral bending, and axial rotation. The effect of the model with four different conditions of the passive components and the sensitivity of the 15 optimization criteria on simulations were investigated. Accounting for the refined passive components significantly improved the simulation accuracy. Different optimization criteria behaved distinctly for different motions. Generally minimizing the sum of squared muscle activations outperformed the others, with the highest averaged correlation coefficient (0.82) between the estimated erector spinae muscle activations and measured electromyography and with the estimated joint compression forces comparable to in vivo reference data. This study highlights the importance of passive model components and proposes a suitable optimization framework for realistic spinal musculoskeletal simulations.

Seismic performance of precast wide beam-column connections with asymmetrical anchorage reinforcement details

This study aims to investigate the seismic performances of precast concrete (PC) wide beam-column connections under reversed cyclic loading, in which the precast beam width is 1.5 times larger than that of column. Experimental program consists of a monolithic reinforced concrete (RC) specimen and two PC specimens with unique connection details. To avoid reinforcement congestions and also to secure emulative connecting performances, the PC specimens were integrated by asymmetrically placed connecting bars anchored into the column panel zone, and their anchorage end details were chosen as a key testing variable. Detailed numerical models were developed to identify the failure mechanisms and key influencing factors of the cyclic responses of PC wide beam-column connections. It appeared that the precast wide beam-column connections with asymmetrical anchorage details can also provide the so-called emulative seismic performance as a monolithic connection. Significant effects of bond-slip behaviors of the anchorage connecting reinforcements in the joint and axial compression level of the column on the seismic performance of PC wide beam-column connections were also identified by the nonlinear numerical simulations.

End-to-End Image Classification and Compression With Variational Autoencoders

The past decade has witnessed the rising dominance of deep learning and artificial intelligence in a wide range of applications. In particular, the ocean of wireless smartphones and IoT devices continue to fuel the tremendous growth of edge/cloud-based machine learning (ML) systems, including image/speech recognition and classification. To overcome the infrastructural barrier of limited network bandwidth in cloud ML, existing solutions have mainly relied on traditional compression codecs such as JPEG that were historically engineered for human-end users instead of ML algorithms. Traditional codecs do not necessarily preserve features important to ML algorithms under limited bandwidth, leading to potentially inferior performance. This work investigates application-driven optimization of programmable commercial codec settings for networked learning tasks such as image classification. Based on the foundation of variational autoencoders (VAEs), we develop an end-to-end networked learning framework by jointly optimizing the codec and classifier without reconstructing images for a given data rate (bandwidth). Compared with the standard JPEG codec, the proposed VAE joint compression and classification framework achieves classification accuracy improvement by over 10% and 4%, respectively, for CIFAR-10 and ImageNet-1k data sets at data rate of 0.8 bpp. Our proposed VAE-based models show 65%–99% reductions in encoder size, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\times 1.5$ </tex-math></inline-formula> – <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\times 13.1$ </tex-math></inline-formula> improvements in inference speed, and 25%–99% savings in power compared to baseline models. We further show that a simple decoder can reconstruct images with sufficient quality without compromising classification accuracy.

Cyclic loading history alters the joint compression tolerance and regional indentation responses in the cartilaginous endplate

This study quantified the effect of subthreshold loading histories that differed by joint posture (neutral, flexed), peak loading variation (10%, 20%, 40%), and loading duration (1000, 3000, 5000 cycles) on the post-loading Ultimate Compressive Tolerance (UCT), yield force, and regional Cartilaginous End Plate (CEP) indentation responses (loading stiffness and creep displacement). One hundred and fourteen porcine spinal units were included. Following conditioning and cyclic compression exposures, spinal units were transected and one endplate from each vertebra underwent subsequent UCT or microindentation testing. UCT testing was conducted by compressing a single vertebra at a rate of 3 kN/s using an indenter fabricated to a representative intervertebral disc size and shape. Force and actuator position were sampled at 100 Hz. Non-destructive uniaxial CEP indentation was performed at five surface locations (central, anterior, posterior, right, left) using a Motoman robot and aluminum indenter (3 mm hemisphere). Force and end-effector position were sampled at 10 Hz. A significant three-way interaction was observed for UCT (p = 0.038). Compared to neutral, the UCT was, on average, 1.9 kN less following each flexed loading duration. No effect of variation was observed in flexion; however, 40% variation caused the UCT to decrease by an average of 2.13 kN and 2.06 kN following 3000 and 5000 cycles, respectively. The indentation stiffness in the central CEP mimicked the UCT response. These results demonstrate a profound effect of posture on post-loading UCT and CEP behaviour. Control of peak compression exposures became particularly relevant only when a neutral posture was maintained and beyond the midpoint of the predicated lifespan.