In the manuscript, we have investigated the effective use of sensors in a bearing ring grinder for failure classification in the condition-based maintenance context. The proposed methodology combines domain knowledge of process monitoring and condition monitoring to successfully achieve failure mode prediction with high accuracy using only a few key sensors. This enables manufacturing equipment to take advantage of advanced data processing and machine learning techniques.

The grinding machine is of type SGB55 from Lidköping Machine Tools and is used to produce functional raceway surface of inner rings of type SKF-6210 deep groove ball bearing. Additional sensors like vibration, acoustic emission, force, and temperature sensors are installed to monitor machine condition while producing bearing components under different operating conditions. Data is sampled from sensors as well as the machine's numerical controller during operation. Selected parts are measured for the produced quality.

Technical failures in machines are major sources of unplanned downtime in any production and result in reduced efficiency and system reliability. Despite the well-established potential of Machine Learning techniques in condition-based maintenance (CBM), the lack of access to failure data in production machines has limited the development of a holistic approach to address machine-level CBM. This paper presents a practical approach for failure mode prediction using multiple sensors installed in a bearing ring grinder for process control as well as condition monitoring. Bearing rings are produced in a set of 7 experimental runs, including 5 frequently occurring production failures in the critical subsystems. An advanced data acquisition setup, implemented for CBM in the grinder, is used to capture information about each individual grinding cycle. The dataset is pre-processed and segmented into grinding cycle stages before time and frequency domain feature extraction. A sensor ranking algorithm is proposed to optimize feature selection for failure classification and the installation cost. Random forest models, benchmarked as best performing classifiers, are trained in a two-step classification framework. The presence of failure mode is predicted in the first step and the failure mode type is identified in the second step using the same feature set. Defining the feature set in the failure detection step improves the predictor generalization with the classifiers’ performance accuracy of 99%99% on the test dataset. The presented approach demonstrates an efficient failure mode classification by selecting crucial sensors resulting in a cost-effective CBM implementation in a bearing ring grinder.

Grinding processes’ stochastic nature poses a challenge in predicting the quality of the resulting surfaces. Post-production measurements for form, surface roughness, and circumferential waviness are commonly performed due to infeasibility in measuring all quality parameters during the grinding operation. Therefore, it is challenging to diagnose the root cause of quality deviations in real-time resulting from variations in the machine’s operating condition. This paper introduces a novel approach to predict the overall quality of the individual parts. The grinder is equipped with sensors to implement condition-based maintenance and is induced with five frequently occurring failure conditions for the experimental test runs. The crucial quality parameters are measured for the produced parts. Fuzzy c-means (FCM) and Hotelling’s T-squared (T2) have been evaluated to generate quality labels from the multi-variate quality data. Benchmarked random forest regression models are trained using fault diagnosis feature set and quality labels. Quality labels from the T2 statistic of quality parameters are preferred over FCM approach for their repeatability. The model, trained from T2 labels achieves more than 94% accuracy when compared to the measured ring disposition. The predicted overall quality using the sensors’ feature set is compared against the threshold to reach a trustworthy maintenance decision.

This paper discusses a lean instrumentation framework for guiding the introduction of the lean concept in condition monitoring in order to enhance the organizational capability (i.e. human, technical and management trichotomy) and reduce the complexity in the maintenance management systems of industrial companies. Additionally, decision-making, based on severity diagnosis and prognosis in condition monitoring, is a complex maintenance function which is based on large data-set of sensors measurements. Yet, the entirety of such decision-making is not dependent on only the sensors measurements, but also on other important indices, such as the human factors, organizational aspects and knowledge management. This is because, the ability to identify significant features from large amount of measured data is a major challenge for automated defect diagnosis, a situation that necessitate the need to identify signal transformations and features in new domains. The need for the lean instrumentation framework is justified by the desire to have a modern condition monitoring system with the capability of pruning to the optimal level the number of sensors required for efficient and effective serviceability of the maintenance process. It is concluded that there are methodologies that can be developed to enable more efficient condition monitoring systems, with benefits for many processes along the value chain.

The phase diagram of three-flavor quark matter under compact star constraints is investigated within a Nambu-Jona-Lasinio model. Global color and electric charge neutrality is imposed for beta-equilibrated superconducting quark matter. The constituent quark masses and the diquark condensates are determined self-consistently in the plane of temperature and quark chemical potential. Both strong and intermediate diquark coupling strengths are considered. We show that in both cases, gapless superconducting phases do not occur at temperatures relevant for compact star evolution, i.e., below T~50 MeV. The stability and structure of isothermal quark star configurations are evaluated. For intermediate coupling, quark stars are composed of a mixed phase of normal (NQ) and two-flavor superconducting (2SC) quark matter up to a maximum mass of 1.21 M[sun]. At higher central densities, a phase transition to the three-flavor color flavor locked (CFL) phase occurs and the configurations become unstable. For the strong diquark coupling we find stable stars in the 2SC phase, with masses up to 1.33 M[sun]. A second family of more compact configurations (twins) with a CFL quark matter core and a 2SC shell is also found to be stable. The twins have masses in the range 1.30...1.33 M[sun]. We consider also hot isothermal configurations at temperature T=40 MeV. When the hot maximum mass configuration cools down, due to emission of photons and neutrinos, a mass defect of 0.1 M[sun] occurs and two final state configurations are possible

We discuss a Nambu-Jona-Lasinio (NJL)-type quantum field theoretical approach to the quark matter equation of state with color superconductivity and construct hybrid star models on this basis. It has recently been demonstrated that with increasing baryon density, the different quark flavors may occur sequentially, starting with down-quarks only, before the second light quark flavor and at highest densities the strange quark flavor also appears. We find that color superconducting phases are favorable over non-superconducting ones, which entails consequences for thermodynamic and transport properties of hybrid star matter. In particular, for NJL-type models no strange quark matter phases can occur in compact star interiors due to mechanical instability against gravitational collapse, unless a sufficiently strong flavor mixing as provided by the Kobayashi-Maskawa-'t Hooft determinant interaction is present in the model. We discuss observational data on mass-radius relationships of compact stars which can put constraints on the properties of the dense matter equation of state.

Recently, observations of compact stars have provided new data of high accuracy which put strong constraints on the high-density behaviour of the equation of state of strongly interacting matter otherwise not accessible in terrestrial laboratories. The evidence for neutron stars with high mass (M = 2.1 +/- 0.2 M-circle dot for PSR J0751 + 1807) and large radii (R > 12 km for RX J1856-3754) rules out soft equations of state and has provoked a debate whether the occurrence of quark matter in compact stars can be excluded as well. In this contribution, it is shown that modern quantum field theoretical approaches to quark matter including colour superconductivity, and a vector meanfield allow a microscopic description of hybrid stars which fulfil the new, strong constraints. The deconfinement transition in the resulting stiff hybrid equation of state is weakly first order so that its signals have to be expected due to specific changes in transport properties governing the rotational and cooling evolution caused by the colour superconductivity of quark matter. A similar conclusion holds for the investigation of quark deconfinement in future generations of nucleus-nucleus collision experiments at low temperatures and high baryon densities such as CBM @ FAIR.

We suggest a scenario where the three light quark flavors are sequentially deconfined under increasing pressure in cold asymmetric nuclear matter as, e.g., in neutron stars. The basis for our analysis is a chiral quark matter model of Nambu–Jona-Lasinio (NJL) type with diquark pairing in the single flavor color-spin-locking (CSL), 2-flavor (2SC) and 3-flavor color-flavor locking (CFL) channels, and a Dirac–Brueckner–Hartree–Fock (DBHF) approach in the nuclear matter sector. We find that nucleon dissociation sets in at about the saturation density, n0, when the down-quark Fermi sea is populated (d-quark dripline) due to the flavor asymmetry imposed by β-equilibrium and charge neutrality. At about 3n0 u-quarks appear forming a 2-flavor color superconducting (2SC) phase, while the s-quark Fermi sea is populated only at still higher baryon density. The hybrid star sequence has a maximum mass of 2.1 M. Two- and 3-flavor quark matter phases are found only in gravitationally unstable hybrid star solutions.

A scenario is suggested in which the three light quark flavors are sequentially deconfined under increasing pressure in cold asymmetric nuclear matter as found, for example, in neutron stars. The basis for this analysis is a chiral quark matter model of Nambu–Jona-Lasinio (NJL) type with diquark pairing in the spin-1 single-flavor, spin-0 two-flavor, and three-flavor channels. Nucleon dissociation sets in at about the saturation density, n0, when the down-quark Fermi sea is populated (d-quark drip line) because of the flavor asymmetry induced by β equilibrium and charge neutrality. At about 3n0, u-quarks appear and a two-flavor color superconducting (2SC) phase is formed. The s-quark Fermi sea is populated only at still higher baryon density, when the quark chemical potential is of the order of the dynamically generated strange quark mass. Two different hybrid equations of state (EOSs) are constructed using the Dirac-Brueckner Hartree-Fock (DBHF) approach and the EOS of Shen et al. [H. Shen, H. Toki, K. Oyamatsu, and K. Sumiyoshi, Nucl. Phys. A637, 435 (1998)] in the nuclear matter sector. The corresponding hybrid star sequences have maximum masses of 2.1 and 2.0 M respectively. Two- and three-flavor quark-matter phases exist only in gravitationally unstable hybrid star solutions in the DBHF case, whereas the Shen-based EOSs produce stable configurations with a 2SC phase component in the core of massive stars. Nucleon dissociation via d-quark drip could act as a deep crustal heating process, which apparently is required to explain superbursts and cooling of x-ray transients.

We suggest a scenario where the three light quark flavors are sequentially deconfined under increasing pressure in cold asymmetric nuclear matter as, e.g., in neutron stars. The basis for our analysis is a chiral quark matter model of Nambu-Jona-Lasinio (NJL) type with diquark pairing in the spin-1 single flavor (CSL), spin-0 two flavor (2SC) and three flavor (CFL) channels. We find that nucleon dissociation sets in at about the saturation density, n0, when the down-quark Fermi sea is populated (d-quark dripline) due to the flavor asymmetry induced by β-equilibrium and charge neutrality. At about 3n0 u-quarks appear and a two-flavor color superconducting (2SC) phase is formed. The s-quark Fermi sea is populated only at still higher baryon density, when the quark chemical potential is of the order of the dynamically generated strange quark mass. We construct two different hybrid equations of state (EoS) using the Dirac-Brueckner Hartree-Fock (DBHF) approach and the EoS by Shen et al. in the nuclear matter sector. The corresponding hybrid star sequences have maximum masses of, respectively, 2.1 and 2.0 M⊙. Two- and three-flavor quark-matter phases exist only in gravitationally unstable hybrid star solutions in the DBHF case, while the Shen-based EoS produce stable configurations with a 2SC phase component in the core of massive stars. Nucleon dissociation due to d-quark drip at the crust-core boundary fulfills basic criteria for a deep crustal heating process which is required to explain superbusts as well as cooling of X-ray transients.

We discuss a hybrid equation of state (EoS) that fulfills constraints for mass-radius relationships and cooling of compact stars. The quark matter EoS is obtained from a Polyakov-loop Nambu-Jona-Lasinio (PNJL) model with color superconductivity, and the hadronic one from a relativistic mean-field (RMF) model with density-dependent couplings (DD-RMF). For the construction of the phase transition regions we employ here for simplicity a Maxwell construction. We present the phase diagram for symmetric matter which exhibits two remarkable features: (1) a "nose"-like structure of the hadronic-to-quark matter phase border with an increase of the critical density at temperatures below T \sim 150 MeV and (2) a high critical temperature for the border of the two-flavor color superconducting (2SC) phase, Tc >160 MeV. We show the trajectories of heavy-ion collisions in the plane of excitation energy versus baryon density calculated using the UrQMD code and conjecture that for incident energies of 4 . . . 8 A GeV as provided,  e.g., by the Nuclotron-M at JINR Dubna or by lowest energies at the future heavy-ion collision experiments CBM at FAIR and NICA at JINR, the color superconducting quark matter phase becomes accessible.

The short-loading cycle is a construction task where a wheel loader scoops material from a nearby pile in order to move that material to the tipping body of a dump truck. The short-loading cycle is a vital task performed in high quantities and is often part of a more extensive never-ending process to move material for further refinement. This, together with the highly repetitive nature of the short-loading cycle, makes it a suitable candidate for automation. However, the short-loading cycle is a complex task where the mechanics of the wheel loader together with the interaction between the wheel loader and the environment needs to be considered. This must be achieved while maintaining some productivity goal and, concurrently, minimizing the used energy. The main objective of this work is to analyze the short-loading cycle, assess the current state of research in this field, and discuss the steps required to progress towards a minimal viable product consisting of individual automation solutions that can perform the short-loading cycle well enough to be used by early adopters. This is achieved through a comprehensive literature study and consequent analysis of the review results. From this analysis, the requirements of an MVP are defined and some gaps which are currently hindering the realization of the MVP are presented.

Machine vision is required by autonomous heavy construction equipment to navigate and interact with the environment. Wheel loaders need the ability to identify different objects and other equipment to perform the task of automatically loading and dumping material on dump trucks, which can be achieved using deep neural networks. Training such networks from scratch requires the iterative collection of potentially large amounts of video data, which is challenging at construction sites because of the complexity of safely operating heavy equipment in realistic environments. Transfer learning, for which pretrained neural networks can be retrained for use at construction sites, is thus attractive, especially if data can be acquired without full-scale experiments. We investigate the possibility of using scalemodel data for training and validating two different pretrained networks and use real-world test data to examine their generalization capability. A dataset containing 268 images of a 1:16 scale model of a Volvo A60H dump truck is provided, as well as 64 test images of a full-size Volvo A25G dump truck. The code and dataset are publicly available 1 . The networks, both pretrained on the MS-COCO dataset, were fine-tuned to the created dataset, and the results indicate that both networks can learn the features of the scale-model dump truck (validation mAP of 0.82 for YOLOv3 and 0.95 for RetinaNet). Both networks can transfer these learned features to detect objects on a full-size dump truck with no additional training (test mAP of 0.70 for YOLOv3 and 0.79 for RetinaNet).

The short-loading cycle is a repetitive task performed in high quantities, making it a great alternative for automation. In the short-loading cycle, an expert operator navigates towards a pile, fills the bucket with material, navigates to a dump truck, and dumps the material into the tipping body. The operator has to balance the productivity goal while minimising the fuel usage, to maximise the overall efficiency of the cycle. In addition, difficult interactions, such as the tyre-to-surface interaction further complicate the cycle. These types of hard-to-model interactions that can be difficult to address with rule-based systems, together with the efficiency requirements, motivate us to examine the potential of data-driven approaches. In this paper, the possibility of teaching an agent through reinforcement learning to approach a dump truck's tipping body and get in position to dump material in the tipping body is examined. The agent is trained in a 3D simulated environment to perform a simplified navigation task. The trained agent is directly transferred to a real vehicle, to perform the same task, with no additional training. The results indicate that the agent can successfully learn to navigate towards the dump truck with a limited amount of control signals in simulation and when transferred to a real vehicle, exhibits the correct behaviour. 

Data collection and annotation is a time consuming and costly process, yet necessary for machine vision. Automation of construction equipment relies on seeing and detecting different objects in the vehicle’s surroundings. Construction equipment is commonly used to perform frequent repetitive tasks, which are interesting to automate. An example of such a task is the short-loading cycle, where the material is moved from a pile into the tipping body of a dump truck for transport. To complete this task, the wheel loader needs to have the capability to locate the tipping body of the dump truck. The machine vision system also allows the vehicle to detect unforeseen dangers such as other vehicles and more importantly human workers. In this work, we investigate the viability to perform semi-automatic annotation of video data using linear interpolation. The data is collected using scale-models mimicking a wheel-loaders approach towards a dump truck during the short-loading cycle. To measure the viability of this type of solution, the workload is compared to the accuracy of the model, YOLOv3. The results indicate that it is possible to maintain the performance while decreasing the annotation workload by about 95%. This is an interesting result for this application domain, as safety is critical and retaining the vision system performance is more important than decreasing the annotation workload. The fact that the performance seems to retain with a large workload decrease is an encouraging sign.

Earth-moving machines are heavy-duty vehicles designed for construction operations involving earthworks. The tasks performed by such machines typically involve navigation and interaction with materials such as soil, gravel, and blasted rock. Skilled operators use a combination of visual, sound, tactile and possibly motion feedback to perform tasks efficiently. We survey the literature in this research area and analyse the relative importance of different sensor system modalities focusing on deep-learning-based vision and automation for the short-cycle loading task. This is a common and repetitive task that is attractive to automate. The analysis indicates that computer vision, in combination with onboard sensors, is more critical than coordinate-based positioning. Furthermore, we find that data-driven approaches, in general, have high potential in terms of productivity, adaptability, versatility and wear and tear with respect to automation system solutions. The main knowledge gaps identified relate to loading non-fine heterogeneous material and navigation during loading and unloading.

This study investigates the use of transfer learning and modular design for adapting a pretrained model to optimize energy efficiency and heat reuse in edge data centers while meeting local conditions, such as alternative heat management and hardware configurations. A Physics-Informed Data-Driven Recurrent Neural Network (PIDD RNN) is trained on a small scale-model experiment of a six-server data center to control cooling fans and maintain the exhaust chamber temperature within safe limits. The model features a hierarchical regularizing structure that reduces the degrees of freedom by connecting parameters for related modules in the system. With a RMSE value of 1.69, the PIDD RNN outperforms both a conventional RNN (RMSE: 3.18), and a State Space Model (RMSE: 2.66). We investigate how this design facilitates transfer learning when the model is fine-tuned over a few epochs to small dataset from a second set-up with a server located in a wind tunnel. The transferred model outperforms a model trained from scratch over hundreds of epochs.

Recent observational results for the masses and radii of some neutron stars are in contrast with typical observations and theoretical predictions for "normal" neutron stars. We propose that their unusual properties can be interpreted as the signature of a dark matter core inside them. This interpretation requires that the dark matter is made of some form of stable, long-living or in general non-annihilating particles, that can accumulate in the star (Sandin and Ciarcelluti (2009). In the proposed scenario all mass-radius measurements can be explained with one nuclear matter equation of state and a dark core of varying relative size. This hypothesis will be challenged by forthcoming observations and could eventually be a useful tool for the determination of dark matter.

This paper presents experimental results with tele-remote operation of a wheel-loader and proposes a method to semi-automate the process. The different components of the tele-remote setup are described in the paper. We focus on the short loading cycle, which is commonly used at quarry and construction sites for moving gravel from piles onto trucks. We present results from short-loading-cycle experiments with three operators, comparing productivity between tele-remote operation and manual operation. A productivity loss of 42% with tele-remote operation motivates the case for more automation. We propose a method to automate the bucket-filling process, which is one of the key operations performed by a wheel-loader.

The automation of bucket loading for repetitive tasks of earth-moving operations is desired in several applications at mining sites, quarries and construction sites where larger amounts of gravel and fragmented rock are to be moved. In load and carry cycles the average bucket weight is the dominating performance parameter, while fuel efficiency and loading time also come into play with short loading cycles. This paper presents the analysis of data recorded during loading of different types of gravel piles with a Volvo L110G wheel loader. Regression models of lift and tilt actions are fitted to the behavior of an expert driver for a gravel pile. We present linear regression models for lift and tilt action that explain most of the variance in the recorded data and outline a learning approach for solving the automatic bucket loading problem. A general solution should provide good performance in terms of average bucket weight, cycle time of loading and fuel efficiency for different types of material and pile geometries. We propose that a reinforcement learning approach can be used to further refine models fitted to the behavior of expert drivers, and we briefly discuss the scooping problem in terms of a Markov decision process and possible value functions and policy iteration schemes.

Automatic bucket filling is an open problem since three decades. In this paper, we address this problem with supervised machine learning using data collected from manual operation. The range-normalized actuations of lift joystick, tilt joystick and throttle pedal are predicted using information from sensors on the machine and the prediction errors are quantified. We apply linear regression, k-nearest neighbors, neural networks, regression trees and ensemble methods and find that an ensemble of neural networks results in the most accurate predictions. The prediction root-mean-square-error (RMSE) of the lift action exceeds that of the tilt and throttle actions, and we obtain an RMSE below 0.2 for complete bucket fillings after training with as little as 135 bucket filling examples

Bucket-filling is a repetitive task in earth-moving operations with wheel-loaders, which needs to be automated to enable efficient remote control and autonomous operation. Ideally, an automated bucket-filling solution should work for different machine-pile environments, with a minimum of manual retraining. It has been shown that for a given machine-pile environment, a time-delay neural network can efficiently fill the bucket after imitation-based learning from 100 examples by one expert operator. Can such a bucket-filling network be automatically adapted to different machine-pile environments without further imitation learning by optimization of a utility or reward function? This paper investigates the use of a deterministic actor-critic reinforcement learning algorithm for automatic adaptation of a neural network in a new pile environment. The algorithm is used to automatically adapt a bucket-filling network for medium coarse gravel to a cobble-gravel pile environment. The experiments presented are performed with a Volvo L180H wheel-loader in a real-world setting. We found that the bucket-weights in the novel pile environment can improve by five to ten percent within one hour of reinforcement learning with less than 40 bucket-filling trials. This result was obtained after investigating two different reward functions motivated by domain knowledge.

Automation of earth-moving industries (construction, mining and quarry) require automatic bucket-filling algorithms for efficient operation of front-end loaders. Autonomous bucket-filling is an open problem since three decades due to difficulties in developing useful earth models (soil, gravel and rock) for automatic control. Operators make use of vision, sound and vestibular feedback to perform the bucket-filling operation with high productivity and fuel efficiency. In this paper, field experiments with a small time-delayed neural network (TDNN) implemented in the bucket control-loop of a Volvo L180H front-end loader filling medium coarse gravel are presented. The total delay time parameter of the TDNN is found to be an important hyperparameter due to the variable delay present in the hydraulics of the wheel-loader. The TDNN network successfully performs the bucket-filling operation after an initial period (100 examples) of imitation learning from an expert operator. The demonstrated solution show only 26% longer bucket-filling time, an improvement over manual tele-operation performance.

Sparse signal models with learned dictionaries of morphological features provide efficient codes in a variety of applications. Such models can be useful to reduce sensor data rates and simplify the communication, processing and analysis of information, provided that the algorithm can be realized in an efficient way and that the signal allows for sparse coding. In this paper we outline an FPGA prototype of a general purpose "analog-to-feature converter", which learns an overcomplete dictionary of features from the input signal using matching pursuit and a form of Hebbian learning. The resulting code is sparse, event-based and suitable for analysis with parallel and neuromorphic processors. We present results of two case studies. The first case is a blind source separation problem where features are learned from an artificial signal with known features. We demonstrate that the learned features are qualitatively consistent with the true features. In the second case, features are learned from ball-bearing vibration data. We find that vibration signals from bearings with faults have characteristic features and codes, and that the event-based code enable a reduction of the data rate by at least one order of magnitude.

This paper presents the technology concepts for a “thumb”-sized self-contained ultrasonic IoT measurement sys- tem. An overall architecture is proposed, and key elements are discussed with solutions using existing technology, thus arguing that realization is possible with the current technology.

Such an ultrasonic IoT measurement system is constrained by its size and available energy, although it requires at least decent computational and communication resources. Because streaming data from such a device is not advisable from an energy viewpoint, there is a need for resource efficient (energy, memory and computational power) data analysis.

An architecture with the following parts as well as some implementation details and performance data are proposed here:

• Energy supply, battery and super capacitor

• Transducer excitation achieving almost zero electrical losses

• Event detection sensor interface

• Data aggregation using sparse approximation and learned

  feature dictionaries, adapted to resource constrained em-

  bedded systems

• IoT communication protocols and implementations enabling

  event -based communication and System of Systems integra- tion capabilities

  The optimization of system level performance requires each subsystem to be optimized for the specific measurement situation taking into account the subsystem interdependencies. This can be performed using a combined electrical and acoustical model of the system. Here, the model allowing electronic and acoustic co-simulation using SPICE is an important tool bridging the electronic and acoustic domains. 

Analogy-making is a key function of human cognition. Therefore, the development of computational models of analogy that automatically learn from examples can lead to significant advances in cognitive systems. Analogies require complex, relational representations of learned structures, which is challenging for both symbolic and neurally inspired models. Vector symbolic architectures (VSAs) are a class of connectionist models for the representation and manipulation of compositional structures, which can be used to model analogy. We study a novel VSA network for the analogical mapping of compositional structures, which integrates an associative memory known as sparse distributed memory (SDM). The SDM enables non-commutative binding of compositional structures, which makes it possible to predict novel patterns in sequences. To demonstrate this property we apply the network to a commonly used intelligence test called Raven’s Progressive Matrices. We present results of simulation experiments for the Raven’s task and calculate the probability of prediction error at 95% confidence level. We find that non-commutative binding requires sparse activation of the SDM and that 10–20% concept-specific activation of neurons is optimal. The optimal dimensionality of the binary distributed representations of the VSA is of the order 10^4, which is comparable with former results and the average synapse count of neurons in the cerebral cortex.

We present a computational model for the analogical mapping of compositional structures that com- bines two existing ideas known as holistic mapping vec- tors and sparse distributed memory. The model enables integration of structural and semantic constraints when learning mappings of the type x_i → y_i and computing analogies x_j → y_j for novel inputs x_j. The model has a one-shot learning process, is randomly initialized and has three exogenous parameters: the dimensionality D of representations, the memory size S and the prob- ability χ for activation of the memory. After learning three examples the model generalizes correctly to novel examples. We find minima in the probability of generalization error for certain values of χ, S and the number of different mapping examples learned. These results indicate that the optimal size of the memory scales with the number of different mapping examples learned and that the sparseness of the memory is important. The optimal dimensionality of binary representations is of the order 10^4, which is consistent with a known analytical estimate and the synapse count for most cortical neurons. We demonstrate that the model can learn analogical mappings of generic two-place relationships and we calculate the error probabilities for recall and generalization.

The rapid integration of physical systems with cyberspace infrastructure, the so-called Internet of Things, is likely to have a significant effect on how people interact with the physical environment and design information and communication systems. Internet-connected systems are expected to vastly outnumber people on the planet in the near future, leading to grand challenges in software engineering and automation in application domains involving complex and evolving systems. Several decades of artificial intelligence research suggests that conventional approaches to making such systems interoperable using handcrafted "semantic" descriptions of services and information are difficult to apply. In this paper we outline a bioinspired learning approach to creating interoperable systems, which does not require handcrafted semantic descriptions and rules. Instead, the idea is that a functioning system (of systems) can emerge from an initial pseudorandom state through learning from examples, provided that each component conforms to a set of information coding rules. We combine a binary vector symbolic architecture (VSA) with an associative memory known as sparse distributed memory (SDM) to model context-dependent prediction by learning from examples. We present simulation results demonstrating that the proposed architecture can enable system interoperability by learning, for example by human demonstration.

The rapid integration of physical systems with cyberspace infrastructure, the so-called Internet of Things, is likely to have a significant effect on how people interact with the physical environment and design information and communication systems. Internet-connected systems are expected to vastly outnumber people on the planet in the near future, leading to grand challenges in software engineering and automation in application domains involving complex and evolving systems. Several decades of artificial intelligence research suggests that conventional approaches to making such systems interoperable using handcrafted "semantic" descriptions of services and information are difficult to apply. In this paper we outline a bioinspired learning approach to creating interoperable systems, which does not require handcrafted semantic descriptions and rules. Instead, the idea is that a functioning system (of systems) can emerge from an initial pseudorandom state through learning from examples, provided that each component conforms to a set of information coding rules. We combine a binary vector symbolic architecture (VSA) with an associative memory known as sparse distributed memory (SDM) to model context-dependent prediction by learning from examples. We present simulation results demonstrating that the proposed architecture can enable system interoperability by learning, for example by human demonstration.

We compare two classes of hybrid equations of state with a hadron-to-quark matter phase transition in their application to core collapse supernova simulations. The first one uses the quark bag model and describes the transition to three-flavor quark matter at low critical densities. The second one employs a Polyakov-loop extended Nambu-Jona-Lasinio (PNJL) model with parameters describing a phase transition to two-flavor quark matter at higher critical densities. These models possess a distinctly different temperature dependence of their transition densities which turns out to be crucial for the possible appearance of quark matter in supernova cores. During the early post-bounce accretion phase quark matter is found only if the phase transition takes place at sufficiently low densities as in the study based on the bag model. The increase critical density with increasing temperature, as obtained for our PNJL parametrization, prevents the formation of quark matter. The further evolution of the core collapse supernova as obtained applying the quark bag model leads to a structural reconfiguration of the central protoneutron star where, in addition to a massive pure quark matter core, a strong hydrodynamic shock wave forms and a second neutrino burst is released during the shock propagation across the neutrinospheres. We discuss the severe constraints in the freedom of choice of quark matter models and their parametrization due to the recently observed 2M ⊙ pulsar and their implications for further studies of core collapse supernovae in the QCD phase diagram.

Autoencoders are commonly trained using element-wise loss. However, element-wise loss disregards high-level structures in the image which can lead to embeddings that disregard them as well. A recent improvement to autoencoders that helps alleviate this problem is the use of perceptual loss. This work investigates perceptual loss from the perspective of encoder embeddings themselves. Autoencoders are trained to embed images from three different computer vision datasets using perceptual loss based on a pretrained model as well as pixel-wise loss. A host of different predictors are trained to perform object positioning and classification on the datasets given the embedded images as input. The two kinds of losses are evaluated by comparing how the predictors performed with embeddings from the differently trained autoencoders. The results show that, in the image domain, the embeddings generated by autoencoders trained with perceptual loss enable more accurate predictions than those trained with element-wise loss. Furthermore, the results show that, on the task of object positioning of a small-scale feature, perceptual loss can improve the results by a factor 10. The experimental setup is available online: https://github.com/guspih/Perceptual-Autoencoders

This work investigates three methods for calculating loss for autoencoder-based pretraining of image encoders: The commonly used reconstruction loss, the more recently introduced deep perceptual similarity loss, and a feature prediction loss proposed here; the latter turning out to be the most efficient choice. Standard auto-encoder pretraining for deep learning tasks is done by comparing the input image and the reconstructed image. Recent work shows that predictions based on embeddings generated by image autoencoders can be improved by training with perceptual loss, i.e., by adding a loss network after the decoding step. So far the autoencoders trained with loss networks implemented an explicit comparison of the original and reconstructed images using the loss network. However, given such a loss network we show that there is no need for the time-consuming task of decoding the entire image. Instead, we propose to decode the features of the loss network, hence the name “feature prediction loss”. To evaluate this method we perform experiments on three standard publicly available datasets (LunarLander-v2, STL-10, and SVHN) and compare six different procedures for training image encoders (pixel-wise, perceptual similarity, and feature prediction losses; combined with two variations of image and feature encoding/decoding). The embedding-based prediction results show that encoders trained with feature prediction loss is as good or better than those trained with the other two losses. Additionally, the encoder is significantly faster to train using feature prediction loss in comparison to the other losses. The method implementation used in this work is available online. https://github.com/guspih/Perceptual-Autoencoders

This chapter describes how district heating substations can be controlled in order to improve customer comfort and system performance, and how energy measurement data can be used to monitor the functionality of substations. We primarily consider indirectly connected district heating substations, where the primary energy supply system is separated from internal building heating and tap water systems. Different approaches to achieve higher efficacy of the district heating substation and the district heating system are introduced, as are methods for the detection of typical faults in district heating substations. These methods enable identification of faults that result in abnormal energy metering data, such as incorrectly configured or installed substations, or faults in the energy meter instrumentation, including the communication system and sensors.

In the context of the standard model of particle physics, there is a definite upper limit to the density of stable compact stars. However, if a more fundamental level of elementary particles exists, in the form of preons, stability may be re-established beyond this limiting density. We show that a degenerate gas of interacting fermionic preons does allow for stable compact stars, with densities far beyond that in neutron stars and quark stars. In keeping with tradition, we call these objects "preon stars", even though they are small and light compared to white dwarfs and neutron stars. We briefly note the potential importance of preon stars in astrophysics, e.g., as a candidate for cold dark matter and sources of ultra-high energy cosmic rays, and a means for observing them.

The use of machine learning to accelerate computer simulations is on the rise. In atomistic simulations, the use of machine learning interatomic potentials (ML-IAPs) can significantly reduce computational costs while maintaining accuracy close to that of ab initio methods. To achieve this, ML-IAPs are trained on large datasets of images, which are atomistic configurations labeled with data from ab initio calculations. Focusing on carbon, we use deep learning to train neural network potentials (NNPs), a form of ML-IAP, based on the state-of-the-art end-to-end NNP architecture SchNet and investigate how the choice of training and validation data affects the performance of the NNPs. Training is performed on the CA-9 dataset, a 9-carbon allotrope dataset constructed using data obtained via ab initio molecular dynamics (AIMD). Our results show that image generation with AIMD causes a high degree of similarity between the generated images, which has a detrimental effect on the performance of the NNPs. But by carefully choosing which images from the dataset are included in the training and validation data, this effect can be mitigated. We conclude by benchmarking our trained NNPs in applications such as relaxation and phonon calculation, where we can reproduce ab initio results with high accuracy.

Modeling Industrial Internet of Things (IIoT) architectures for the automation of wind turbines and farms(WT/F), as well as their condition monitoring (CM) is a growing concept among researchers. Several end-to-end automated cloud-based solutions that digitize CM operations intelligently to reduce manual efforts and costs are being developed. However, establishing robust and secure communication across WT/F is still difficult for the wind energy industry. We propose a fully automated cloud-based collaborative learning (CCL) architecture using the Eclipse Arrowhead Framework and an unsupervised dictionary learning (USDL) CM approach. The scalability of the framework enabled digitization and collaboration across the WT/Fs. Collaborative learning is a novel approach that allows all WT/Fs to learn from each other in real-time. Each turbine has CCL based CM using USDL as micro-services that autonomously perform feature selection and failure prediction to optimize cost, computation, and resources. The fundamental essence of the USDA approach is to enhance the WT/F’s learning and accuracy. We use dictionary distances as a metric for analyzing the CM of WT in our proposed USDL approach. A dictionary indicates an anomaly if its distances increased from the dictionary computed at a healthy state of that WT. Using CCL, a WT/F learns all types of failures that could occur in a similar WT/F, predicts any machinery failure, and sends alerts to the technicians to ensure guaranteed proactive maintenance. The results of our research support the notion that when testing a turbine with dictionaries of all the other turbines, every dictionary converges to similar behavior and captures the fault that occurs in that turbine.

“He who knows no foreign languages knows nothing of his own.” Johann Wolfgang emphasized the worth of languages for expanding ones learning horizons. This work instills the same notion into the industrial internet of things (IIoT) sensory devices paradigm. We study the interoperability problem setting with a new perspective of envisioning knowledge graphs (KGs) modeling for the device to device ontology alignment. Ontology alignment is structured as entity alignment in which similar entities are linked from two heterogeneous knowledge graphs. The novelty is conceiving the IIoT ontology graph as a language of the sensory device and then addressing it through the natural language processing (NLP) language translation approach. The IIoT ontology graph nodes have unique URIs so they act as words (sentences) for the NLP model and the schema of the graph is depicted as the language structure. Existing methods give less attention to the importance of structural information which ignores the fact of even when a node pair has similar entity labels it may not refer to a similar context and vice versa. To deal with these issues, we propose a novel solution using a modified BERT_INT model on graph Triplets for ontology alignment among heterogeneous IIoT devices. Moreover, an iterative framework is designed to leverage the alignments within nodes as well as among relations. As the first attempt at this problem, the proposed model is tested on a contemporary language dataset of DBP15K and compared with the best state-of-the-art results. The proposed model outperforms the target baseline BERT_INT model by 2.1% in terms of HR@1, HR@10, and MRR. Next, a dataset on ontology instances is constructed on smart building sensors using two W3C standardized IIoT ontologies i.e. SSN and SOSA. Comprehensive experiments and analysis with ablation study on language and structural encoders demonstrate the effectiveness of our model.

The technical capabilities of modern Industry 4.0 and Industry 5.0 are vast and growing exponentially daily. The present-day Industrial Internet of Things (IIoT) combines manifold underlying technologies that require real-time interconnection and communication among heterogeneous devices. Smart cities are established with sophisticated designs and control of seamless machine-to-machine (M2M) communication, to optimize resources, costs, performance, and energy distributions. All the sensory devices within a building interact to maintain a sustainable climate for residents and intuitively optimize the energy distribution to optimize energy production. However, this encompasses quite a few challenges for devices that lack a compatible and interoperable design. The conventional solutions are restricted to limited domains or rely on engineers designing and deploying translators for each pair of ontologies. This is a costly process in terms of engineering effort and computational resources. An issue persists that a new device with a different ontology must be integrated into an existing IoT network. We propose a self-learning model that can determine the taxonomy of devices given their ontological meta-data and structural information. The model finds matches between two distinct ontologies using a natural language processing (NLP) approach to learn linguistic contexts. Then, by visualizing the ontological network as a knowledge graph, it is possible to learn the structure of the meta-data and understand the device's message formulation. Finally, the model can align entities of ontological graphs that are similar in context and structure.Furthermore, the model performs dynamic M2M translation without requiring extra engineering or hardware resources.

We present a hybrid equation of state (EoS) for dense matter that satisfies phenomenological constraints from modern compact star (CS) observations which indicate high maximum masses (M = 2 Msun) and large radii (R> 12 km). The corresponding isospin symmetric EoS is consistent with flow data analyses of heavy-ion collisions and a deconfinement transition at approx. 0.55 fm-3. The quark matter phase is described by a 3-flavor Nambu--Jona-Lasinio model that accounts for scalar diquark condensation and vector meson interactions while the nuclear matter phase is obtained within the Dirac-Brueckner-Hartree-Fock (DBHF) approach using the Bonn-A potential. We demonstrate that both pure neutron stars and neutron stars with quark matter cores (QCSs) are consistent with modern CS observations. Hybrid star configurations with a CFL quark core are unstable.

One of the key ingredients to understand the properties of neutrons stars1 (NS) is the equation of state at finite densities far beyond nuclear saturation. Investigating the phase structure of quark matter that might be realized in the core of NS inspires theory and observation. We discuss recent results of our work to point out our view on challenges and possibilities in this evolving field by means of a few examples.

Nowadays, lots of data are generated in industries using vibration sensors to evaluate the equipment’s working condition and identify faults. A significant challenge is that only a small fraction of data can be transmitted for intelligent fault diagnosis and storage. The edge processing capacity is often insufficient for advanced analysis due to time and resource constraints. The neuromorphic signal encoding scheme efficiently reduces the data rate by encoding relevant signal changes into spike trains while discarding redundant information and noise, enabling energy-efficient neuromorphic processing. Due to the presence of dominant operational features and noise in the original measurements, signal pre-processing is required to extract the relevant features before spike coding and processing. The work investigates the effects of different filter banks (pre-processing methods) on the spike encodings for vibration measurements from bearings. This also includes bearing fault features diagnosis based on statistical analysis of generated spikes. The comparative analysis is made for benchmarking different signal pre-processing methods (e.g., envelope, empirical mode decomposition (EMD), and gammatone filter) on bearing vibration datasets. An event-triggered scheme, i.e., Level-crossing analog-to-digital converters (LC-ADCs) is applied to encode the vibration measurement to spikes. Inter-spike intervals (ISIs) statistics are analysed for fault diagnosis of bearings. The results obtained for CWRU bearing databases indicate a possible fault detection and diagnosis with significant data rate reduction and an opportunity for improved computational efficiency. With the developed approach, the envelope filter is found to be the most efficient of all. This work enables a new approach to improve the energy efficiency of condition monitoring systems and further sets a new course of research development in this area using neuromorphic technologies. 

Labelled industry datasets are one of the most valuable assets in prognostics and health management (PHM) research. However, creating labelled industry datasets is both difficult and expensive, making publicly available industry datasets rare at best, in particular labelled datasets.Recent studies have showcased that industry annotations can be used to train artificial intelligence models directly on industry data ( https://doi.org/10.36001/ijphm.2022.v13i2.3137 , https://doi.org/10.36001/phmconf.2023.v15i1.3507 ), but while many industry datasets also contain text descriptions or logbooks in the form of annotations and maintenance work orders, few, if any, are publicly available.Therefore, we release a dataset consisting with annotated signal data from two large (80mx10mx10m) paper machines, from a Kraftliner production company in northern Sweden. The data consists of 21 090 pairs of signals and annotations from one year of production. The annotations are written in Swedish, by on-site Swedish experts, and the signals consist primarily of accelerometer vibration measurements from the two machines.The dataset is structured as a Pandas dataframe and serialized as a pickle (.pkl) file and a JSON (.json) file. The first column (‘id’) is the ID of the samples; the second column (‘Spectra’) are the fast Fourier transform and envelope-transformed vibration signals; the third column (‘Notes’) are the associated annotations, mapped so that each annotation is associated with all signals from ten days before the annotation date, up to the annotation date; and finally the fourth column (‘Embeddings’) are pre-computed embeddings using Swedish SentenceBERT. Each row corresponds to a vibration measurement sample, though there is no distinction in this data between which sensor or machine part each measurement is from.

Annotations in condition monitoring systems contain information regarding asset history and fault characteristics in the form of unstructured text that could, if unlocked, be used for intelligent fault diagnosis. However, processing these annotations with pre-trained natural language models such as BERT is problematic due to out-of-vocabulary (OOV) technical terms, resulting in inaccurate language embeddings. Here we investigate the effect of OOV technical terms on BERT and SentenceBERT embeddings by substituting technical terms with natural language descriptions. The embeddings were computed for each annotation in a pre-processed corpus, with and without substitution. The K-Means clustering score was calculated on sentence embeddings, and a Long Short-Term Memory (LSTM) network was trained on word embeddings with the objective to recreate the output from a keyword-based annotation classifier. The K-Means score for SentenceBERT annotation embeddings improved by 40% at seven clusters by technical language substitution, and the labelling capacityof the BERT-LSTM model was improved from 88.3 to 94.2%. These results indicate that the substitution of OOV technical terms can improve the representation accuracy of the embeddings of the pre-trained BERT and SentenceBERT models, and that pre-trained language models can be used to process technical language.