Abstract: Methods, systems, and apparatus, including medium-encoded computer program products, for generating synthetic images, include: obtaining a captured image of a symbol that encodes data; producing a synthetic image from the captured image using a trained machine learning model, wherein the trained machine learning model has been trained using first images and second images of examples of symbols, wherein the second images have a second image quality that is different than a first image quality of first images, and wherein parameters of the machine learning model have been adjusted responsive to image features to tradeoff content loss versus style loss using measurements of information content correlation between third images produced during training and each of the first images and the second images, the measurements being from a comparison metric; and providing the synthetic image for use by a program configured to identify information in images of symbols.

Abstract: Data characterizing a fuel storage facility can be received from one or more of a plurality of sensors disposed in the fuel storage facility. A fuel leak prediction for the fuel storage facility can be determined by a server, based on the received data, and further based on at least one predictive model that predicts whether a fuel leak exists in the fuel storage facility. The fuel leak prediction can be provided by the server. Related apparatus, systems, methods, techniques, and articles are also described.

Abstract: Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a user equipment may establish a communication connection with a serving cell associated with a first discontinuous reception (DRX) cycle configuration, determine, based at least in part on at least one of a first on duration of a first DRX cycle indicated by the first DRX cycle configuration or a type of traffic of one or more communications received during the first DRX cycle, an autonomous measurement gap associated with a neighboring cell, and perform a measurement of the neighboring cell during the autonomous measurement gap. Numerous other aspects are provided.

Abstract: Methods, systems, and devices for wireless communications are described. A user equipment (UE) may receive, from a network entity, one or more control signals indicating information associated with a set of occasions for a transmission time interval (TTI) corresponding to a low power operation mode of the UE. For example, the set of occasions may include one or more of a wake up signal (WUS) monitoring occasion, a synchronization signal block (SSB) occasion, a channel state information (CSI) reporting occasion, and a discontinuous reception (DRX) on duration occasion. The UE may determine whether to skip the WUS monitoring occasion based on a respective duration and a respective start time for each occasion of the set of occasions associated with the information indicated by the one or more control signals.

Abstract: Methods, systems, and apparatus, including medium-encoded computer program products, for generating synthetic images, include: obtaining a captured image of a symbol that encodes data; producing a synthetic image from the captured image using a trained machine learning model, wherein the trained machine learning model has been trained using first images and second images of examples of symbols, wherein the second images have a second image quality that is different than a first image quality of first images, and wherein parameters of the machine learning model have been adjusted responsive to image features to tradeoff content loss versus style loss using measurements of information content correlation between third images produced during training and each of the first images and the second images, the measurements being from a comparison metric; and providing the synthetic image for use by a program configured to identify information in images of symbols.

Abstract: Among other things, the techniques described herein include a method for receiving a plurality of images of one or more containers while the one or more containers are being emptied, the plurality of images comprising a training set of images and a validation set of images; labeling each image of the plurality of images as including either an overfilled container or a not-overfilled container; processing each image of the plurality of images to reduce bias of a machine learning model; training, and based on the labeling, the machine learning model using the plurality of images; and optimizing the machine learning model by performing learning against the validation set, the optimized machine learning model being used to generate a prediction for a new image of a container, the prediction indicating whether the container in the new image was overfilled prior to the new container being emptied.

Abstract: Methods and systems for ripple suppression in multi-phase buck converters may comprise a buck converter for providing an output voltage with controlled ripple current. The buck converter may include one or more main buck converter stages and one or more suppression buck converter stages coupled with the one or more main buck converter stages. The one or more suppression buck converter stages may provide suppression currents to reduce ripple currents generated in the one or main buck converter stages.

Abstract: A process includes obtaining a target thermal load profile and a target location; determining weather conditions associated with the target location; simulating regulation of the target thermal load profile by different temperature regulation systems having different corresponding sets of components, subject to the weather conditions, to obtain energy consumption data for each of the different temperature regulation systems; and providing at least one of the different temperature regulation systems based on the energy consumption data, to cause use of the at least one of the different temperature regulation systems having the corresponding set of components.

Abstract: Among other things, the techniques described herein include a method for receiving a plurality of images of one or more containers while the one or more containers are being emptied, the plurality of images comprising a training set of images and a validation set of images; labeling each image of the plurality of images as including either an overfilled container or a not-overfilled container; processing each image of the plurality of images to reduce bias of a machine learning model; training, and based on the labeling, the machine learning model using the plurality of images; and optimizing the machine learning model by performing learning against the validation set, the optimized machine learning model being used to generate a prediction for a new image of a container, the prediction indicating whether the container in the new image was overfilled prior to the new container being emptied.

Abstract: Methods, devices, apparatus, systems and computer-readable storage media for assessing damages on vehicles are provided. In one aspect, a method includes: accessing an image of a vehicle showing shapes each indicating at least one damage area on the vehicle; providing the image as input to a first model to identify the shapes in the image and obtaining shape data describing a position of each shape identified in the image; providing the image as input to a second model to identify one or more panels of the vehicle and obtaining panel data describing a position of each panel identified in the image; automatically correlating the one or more shapes and the one or more panels based on the shape data and the panel data to determine a number of shapes present on each panel; and generating a damage assessment report describing the number of shapes present on each panel.

Abstract: Methods and systems for ripple suppression in multi-phase buck converters may comprise a buck converter for providing an output voltage with controlled ripple current. The buck converter may include one or more main buck converter stages and one or more suppression buck converter stages coupled with the one or more main buck converter stages. The one or more suppression buck converter stages may provide suppression currents to reduce ripple currents generated in the one or main buck converter stages.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer for an industrial machine-learning operation model monitoring system, that include the actions of receiving monitoring data for an industrial machine-learning operations model, determining, from the monitoring data, to retrain the industrial machine-learning operations model, where the determining includes computing drift parameters, each of the drift parameters being indicative of a type of observable drift of the industrial machine-learning operations model, where the drift parameters include (i) a usage drift, (ii) a performance drift, (iii) a data drift, and (iv) a prediction drift, and where each drift parameter includes a respective retraining criteria, and confirming, from the drift parameters, the respective retraining criteria is met by at least one of the drift parameters, and triggering, in response to the determining to retrain the industrial machine-learning operations model, an update of the industrial machine-learn

Abstract: Among other things, systems and techniques are described for a predictive model for determining overall equipment effectiveness (OEE) in industrial equipment. Data including spectral features is obtained. A probability of survival is determined by fitting at least one degradation function to degradation data associated with the industrial equipment. An overall equipment effectiveness metric is predicted as a product of predicted planned production time, predicted performance, and predicted quality output by trained machine learning models.

Abstract: Among other things, techniques are described for an after-market service process digitization. Service data is obtained that is associated with at least one asset and comprises at least historical warranty data and current IoT data. Predictive analysis to generate an asset survival prediction is performed based on current data associated with a first asset and the service data. Troubleshooting data associated with the first asset from at least one knowledge data source is received. A warranty coverage metric is determined based on the asset survival prediction and the troubleshooting data, wherein the warranty coverage metric is calculated in real time according to the asset survival prediction and the troubleshooting data. The warranty coverage metric is transformed at a device into human-readable form.

Abstract: Among other things, systems and techniques are described for predictive maintenance of industrial equipment. Sensor data is obtained, e.g., using sensor hubs that are configured to capture sensor data associated with one or more operating conditions of the industrial equipment. The sensor data is input to a trained machine learning model. The trained machine learning model includes a physics based feature extraction model and a deep learning based automatic feature extraction model. Operating conditions associated with operation of the industrial equipment are predicted using the trained machine learning models.

Abstract: Data characterizing a fuel storage facility can be received from one or more of a plurality of sensors disposed in the fuel storage facility. A fuel leak prediction for the fuel storage facility can be determined by a server, based on the received data, and further based on at least one predictive model that predicts whether a fuel leak exists in the fuel storage facility. The fuel leak prediction can be provided by the server. Related apparatus, systems, methods, techniques, and articles are also described.

Abstract: An automated wetstock management system can include a plurality of sensors disposed in a fuel storage facility, the plurality of sensors configured to sense fuel data characterizing one or more aspects of the fuel storage facility, and a wetstock management server communicatively coupled to the plurality of sensors. The wetstock management server can process the fuel data to detect whether the fuel data satisfies an exception indicative of an operational issue of the fuel storage facility based on one or more predefined rules or models stored in the wetstock management server. In some embodiments, the wetstock management server can generate a workflow for assisting a user of the fuel storage facility to resolve the operational issue. In some embodiments, the wetstock management server can assign a risk category to the exception and electronically transmit an alert characterizing the operational issue to the user.

Abstract: In one aspect, fuel inventory data characterizing a level of fuel stored at a fuel storage facility can be received from a sensor in operable communication with the fuel storage facility. Flow meter data characterizing an amount of fuel dispensed from the fuel storage facility by a fuel dispenser nozzle can be received from a flow meter in fluid communication with the fuel dispenser nozzle and the fuel storage facility. Fuel transaction data characterizing the dispensing of fuel by the fuel dispenser nozzle and the dispensing of the fuel by one or more additional fuel dispenser nozzles in fluid communication with the fuel storage facility can be received. An estimate of meter drift characterizing a change in calibration of the flow meter can be determined based on the fuel inventory data, the flow meter data, and the fuel transaction data. The estimate of meter drift can be provided.

Abstract: An automated wetstock management system can include a plurality of sensors disposed in a fuel storage facility, the plurality of sensors configured to sense fuel data characterizing one or more aspects of the fuel storage facility, and a wetstock management server communicatively coupled to the plurality of sensors. The wetstock management server can process the fuel data to detect whether the fuel data satisfies an exception indicative of an operational issue of the fuel storage facility based on one or more predefined rules or models stored in the wetstock management server. In some embodiments, the wetstock management server can generate a workflow for assisting a user of the fuel storage facility to resolve the operational issue. In some embodiments, the wetstock management server can assign a risk category to the exception and electronically transmit an alert characterizing the operational issue to the user.

Abstract: Data characterizing a fuel storage facility can be received from one or more of a plurality of sensors disposed in the fuel storage facility. A fuel leak prediction for the fuel storage facility can be determined by a server, based on the received data, and further based on at least one predictive model that predicts whether a fuel leak exists in the fuel storage facility. The fuel leak prediction can be provided by the server. Related apparatus, systems, methods, techniques, and articles are also described.