Traditional Power Management Research Articles

Wind-driven suspended triboelectric-electromagnetic hybrid generator with vibration elimination for environmental monitoring in the high-voltage power transmission line

The sustainable development of the high-voltage power transmission system puts forward more significant requirements IoT sensor network. Currently, a large number of the IoT sensors are installed on the high-voltage transmission lines, and environmental wind energy harvesting around the outdoor power transmission lines enables a promising and feasible solution to supply distributed sensors. However, the traditional floor-mounted wind energy nanogenerator with high requirements for placement and fixation is no longer applicable to the suspension scenario of the high-voltage power transmission lines. In this work, a wind-driven suspended triboelectric-electromagnetic hybrid generator (WS-TEHG) is designed as a suspended wind cup with a three-layered integrated structure, which can be integrated with the damping structure that hanged and installed on the power transmission line, demonstrating the simultaneously achieving high efficiency wind energy harvesting and the operation stability of the device. After the structural and material optimization, WS-TEHG can achieve efficient energy collection under wide-range of the wind speed from 2.6 to 17.3 m s−1, whose TENG and EMG produced instantaneous peak power from 0.2 to 7.9 mW and 0.6–85.1 mW, respectively. Moreover, a standardized power management based on LTC 3588-based double-channel circuit was established to co-manage the outputs of the two generating modules in the hybrid generator and enhance the continuous stability of the standardized voltage output, demonstrating a shorter undervoltage charge accumulation, a faster start-up process, a lower power consumption compared with the traditional power management. Finally, a self-powered environmental monitoring system by a WS-TEHG equipped with a standardized power management circuit was developed to provide the weather and climate information around the power transmission lines, thereby demonstrating a potential and feasible for large-scale self-powered application in the field of the high-voltage power transmission lines.

The Use of Big Data Technology in Distributed Photovoltaic Power Generation Operation and Management--Take China as an Example

Given the lack of distributed PV power generation operation and management capability, this paper profoundly analyzes the current situation of the application of big data technology in the process and management of PV power generation in China and proves the necessity and importance of promoting new equipment, new technology, and new mode based on traditional power system operation and management, using the extensive data information formed in the process and management of power plants, meteorology, environment, and power grid, and launching ample data analysis research. In addition, the typical application scenarios of big data technology in the operation and management of distributed PV power generation are explored, and a new mode of operation and management of distributed PV power generation at the power plant, grid, and government levels is explored and constructed. The study shows that with the dramatic increase in the number of distributed PV power generation, the use of big data technology in scenarios such as the power generation side and grid side has excellent potential to provide adequate support for PV power generation operation, grid scheduling, market-based transaction settlement, and government decision-making.

Self-aware power management for multi-core microprocessors

Power management is one of the significant challenges to be addressed in multi-/many-core microprocessors. Furthermore, the multi-core microprocessors experience unforeseen scenarios such as performance degradation over time, manufacturing defects, power, and thermal impacts with time. Traditional power management techniques, though efficient, is not designed to handle such unseen scenarios. Furthermore, the variation in performance requirements is one of the challenges faced in the era of machine learning. We propose a self-aware power management scheme for multi-core microprocessors in this work to address the above-mentioned issues. We perform application-level power management in this work to overcome the overheads imposed by core-level power management and system-level power management inefficiency. The power management unit employs a linear predictor for workload prediction to perform DVFS. On top of the power manager, the self-aware controller is hierarchically placed to monitor the system components’ health and adapt the power manager's decision to meet the performance requirements and handle changes in system components’ health. We evaluate the proposed self-aware power manager under externally provided high performance goals, and resource contention. A power saving of up to 16% compared to existing power management techniques, and 2.4× speedup with 25% additional power to satisfy high performance compared to power management without self-awareness for a microprocessor with up to 32-cores is achieved.

A comparison of rule-based and model predictive controller-based power management strategies for fuel cell/battery hybrid vehicles considering degradation

Traditional power management systems for hybrid vehicles often focus on the optimization of one particular cost factor, such as fuel consumption, under specific driving scenarios. The cost factor is usually based on the beginning-of-life performance of system components. Typically, such strategies do not account for the degradation of the different components of the system over their lifetimes. This study incorporates the effect of fuel cell and battery degradation within their cost factors and investigates the impact of different power management strategies on fuel cell/battery loads and thus on the operating cost over the vehicle's lifetime. A simple rule-based power management system was compared with a model predictive controller (MPC) based system under a connected vehicle scenario (where the future vehicle speed is known a priori within a short time horizon). The combined cost factor consists of hydrogen consumption and the degradation of both the fuel cell stack and the battery. The results show that the rule-based power management system actually performs better and achieves lower lifetime cost compared to the MPC system even though the latter contains more information about the drive cycle. This result is explained by examining the changing dynamics of the three cost factors over the vehicle's lifetime. These findings reveal that a limited knowledge of traffic information might not be as useful for the power management of certain fuel cell/battery hybrid vehicles when degradation is taken into consideration, and a simple tuned rule-based controller is adequate to minimize the lifetime cost.

Chip-Specific Power Delivery and Consumption Co-Management for Process-Variation-Aware Manycore Systems Using Reinforcement Learning

Energy efficiency has become a critical design metric for high-performance systems. Various power management techniques have been proposed for the processor cores such as dynamic voltage and frequency scaling (DVFS), whereas few solutions consider the power losses suffered on the power delivery system (PDS), despite the fact that they have a significant impact on the overall energy efficiency of the system. With the explosive growth of system complexity and highly dynamic workloads variations, it is also challenging to find the optimal power management policies which can effectively match the power delivery with the power consumption. In addition, process variations (PVs) add heterogeneity to systems and make traditional power management methods less effective. To tackle the above problems, we propose a reinforcement-learning-based Chip-Specific Power co-Management (CSPM) scheme for PV-aware manycore systems. Both PDS and processor cores are jointly adjusted by distributed agents with modular Q-learning to improve the overall energy efficiency of the system. System characteristics are naturally included in the learning process to obtain chip-specific policies. Experimental results show that when applied to PV-aware manycore systems with a hybrid PDS constructed by both on- and off-chip voltage regulators, the proposed method achieves a 60.1% reduction of the overall energy delay product (EDP) of the system, on average, compared to a traditional DVFS approach.

Machine Learning for Power, Energy, and Thermal Management on Multicore Processors: A Survey

Due to the high integration density and roadblock of voltage scaling, modern multicore processors experience higher power densities than previous technology scaling nodes. When unattended, this issue might lead to temperature hot spots, that in turn may cause nonuniform aging, accelerate chip failure, impair reliability, and reduce the performance of the system. This paper presents an overview of several research efforts that propose to use machine learning (ML) techniques for power and thermal management on single-core and multicore processors. Traditional power and thermal management techniques rely on a certain a-priori knowledge of the chip’s thermal model, as well as information of the workloads/applications to be executed (e.g., transient and average power consumption). Nevertheless, these a-priori information is not always available, and even if it is, it cannot reflect the spatial and temporal uncertainties and variations that come from the environment, the hardware, or from the workloads/applications. Contrarily, techniques based on ML can potentially adapt to varying system conditions and workloads, learning from past events in order to improve themselves as the environment changes, resulting in improved management decisions.

Research on Intelligent Power Automation Technology Based on Edge Computing of Power Internet of Things

With the rapid development of modern economy and the rapid promotion of science and technology, electricity is the realistic need for further transformation of various industries in contemporary society. Because the traditional power management mode is relatively backward and the automation level of different industries is uneven, these will limit the development of intelligent power automation. Therefore, based on the edge computing of power Internet of Things, this paper studies intelligent power automation technology, which is of great value for realizing power automation and improving the management level of power equipment. In this paper, aiming at the research of intelligent power automation technology, based on the edge computing of power Internet of Things, the Internet of Things system based on RFID technology is studied. This system constructs the basic architecture of Internet of Things for smart power applications. The architecture is composed of sensing layer, network layer and application layer, which is used to realize information collection, identification and transmission, and its application in the production management of power Internet of Things. This paper analyzes the functional requirements, factors to be considered, main features and main functions of power equipment management system, gives the design ideas of power equipment management system, and designs an intelligent power equipment system based on edge computing of power Internet of Things. The research results show that RFID technology based on edge computing of power Internet of Things is feasible to realize intelligent power automation, can solve some problems in current power technology, and is of great significance to realize intelligent power automation.

Android-Based Intelligent Farmer Power Management System

Intelligent farmer power management system software is available for intelligent, small, single-user rural farmer systems. It is mainly used for real-time monitoring, data analysis and comprehensive coordinated control of solar energy, farm household consumption and energy storage units. The traditional farmer power management system is based on the Internet B/S structure and is developed using a standard J2EE platform. It requires an independent computer, software server, and database server to support operation and access. Based on the Android operating system, middleware and key applications, the intelligent farmer power management system software platform is based on the basic business of intelligent farmer power system consumption management, and solves the needs of farmers for real-time and historical data consumption. The easy-to-carry mobile phone smart terminal device overcomes the drawbacks that must be accessed by a computer, and can be conveniently and flexibly used for login, data access and remote control at any time, and is an important means for intelligent agriculture and intelligent farmer business promotion.

Utilization Aware Power Management in Reliable and Aggressive Chip Multi Processors

With increasing transistor density on a single chip, processor design in the nanoscale era is hitting power and frequency walls. Due to these challenges, processors not only need to run fast, but remain cool and consume less energy. At this juncture where no further improvement in clock frequency is possible, data dependent latching through timing speculation provides a silver lining. In this paper, (a) we present a novel power level switching mechanism using reliable and aggressive designs that support overclocking. Using the proposed framework, we achieve 40 percent speed-up and also observe 75 percent energy-delay squared product (ED $^2$ ) savings relative to the base architecture. (b) showcase the loss of efficiency in current chip multiprocessor systems due to excess power supplied. We propose a utilization-aware task scheduling (UTS)—a power management scheme that increases energy efficiency of chip multiprocessors. (c) demonstrate that UTS along with aggressive timing speculation extracts ample performance from the system without loss of efficiency, and also without breaching power and thermal constraints. We demonstrate that UTS improves performance by 12 percent due to aggressive power level switching and over 50 percent in ED $^2$ savings in comparing to traditional power management techniques.

Accurate energy modeling for many-core static schedules with streaming applications

Many-core systems provide a great performance potential with the massively parallel hardware structure. Yet, these systems are facing increasing challenges such as high operating temperatures, high electrical bills, unpleasant noise levels due to active cooling and high battery drainage in mobile devices; factors caused directly by poor energy efficiency. Furthermore by pushing the power beyond the limits of the power envelope, parts of the chip cannot be used simultaneously – a phenomenon referred to as “dark silicon”. Power management is therefore needed to distribute the resources to the applications on demand. Traditional power management systems have usually been agnostic to the underlying hardware, and voltage and frequency control is mostly driven by the workload. Static schedules, on the other hand, can be a preferable alternative for applications with timing requirements and predictable behavior since the processing resources can be more precisely allocated for the given workload. In order to efficiently implement power management in such systems, an accurate model is important in order to make the appropriate power management decisions at the right time. For making correct decisions, practical issues such as latency for controlling the power saving techniques should be considered when deriving the system model, especially for fine timing granularity. In this paper we present an accurate energy model for many-core systems which includes switching latency of modern power saving techniques. The model is used when calculating an optimal static schedule for many-core task execution on systems with dynamic frequency levels and sleep state mechanisms. We derive the model parameters for an embedded processor with the help of benchmarks, and we validate the model on real hardware with synthetic applications that model streaming applications. We demonstrate that the model accurately forecasts the behavior on an ARM multicore platform, and we also demonstrate that the model is not significantly influenced by variances in common type workloads.

A control-based methodology for power-performance optimization in NoCs exploiting DVFS

Abstract Networks-on-Chip (NoCs) are considered a viable solution to fully exploit the computational power of multi- and many-cores, but their non negligible power consumption requires ad hoc power-performance design methodologies. In this perspective, several proposals exploited the possibility to dynamically tune voltage and frequency for the interconnect, taking steps from traditional CPU-based power management solutions. However, the impact of the actuators, i.e. the limited range of frequencies for a PLL (Phase Locked Loop) or the time to increase voltage and frequency for a Dynamic Voltage and Frequency Scaling (DVFS) modules, are often not carefully accounted for, thus overestimating the benefits. This paper presents a control-based methodology for the NoC power-performance optimization exploiting the Dynamic Frequency Scaling (DFS). Both timing and power overheads of the actuators are considered, thanks to an ad hoc simulation framework. Moreover the proposed methodology eventually allows for user and/or OS interactions to change between different high level power-performance modes, i.e. to trigger performance oriented or power saving system behaviors. Experimental validation considered a 16-core architecture comparing our proposal with different settings of threshold-based policies. We achieved a speedup up to 3 for the timing and a reduction up to 33.17% of the power ∗ time product against the best threshold-based policy. Moreover, our best control-based scheme provides an averaged power-performance product improvement of 16.50% and 34.79% against the best and the second considered threshold-based policy setting.

Intelligent Agents under Collaborative Control in Emerging Power Systems

In the world of liberalized power markets traditional power management concepts have come to their limits. Optimal pricing can no longer be achieved, e.g. for very short-time needs across grids. Power line overload and grid stability, increasingly resulting in regional or even global black-outs, are at stake. With the highly desirable expansion of renewable energy production these challenges are experienced in quite an amplified way: We argue that for this emergent technology the traditional top-down and long-term power management is obsolete, due to the wide dispersion and high unpredictability of wind and solar-based power facilities. In the DECENT0F 1 R&D initiative we developed a multi-level, bottom-up solution where autonomous collaborative software agents negotiate available energy quantities and needs on behalf of consumer and producer groups (the DEZENT algorithm). We operate within very short time intervals of assumedly constant demand and supply, in our case periods of 0.5sec (switching delay for a light bulb). The solution has proven to be secure against a relevant variety of malicious attacks. Within this time interval we are also able to manage the coordinated power distribution, and achieve grid stability. In this paper the main contribution is to make the negotiation strategies themselves adaptive across periods: We derive the dynamic distributed learning algorithm DECOLEARN from Reinforcement Learning principles for providing the agents with collaborative intelligence and at the same time proving substantially superior to conventional (static) procedures. We report briefly on our extensive comparative simulation experiments. Keywords: Distributed Energy Management, Reinforcement Learning, Multi-agent Systems, Smart Grids

An Adaptive Interleaving Technique for Memory Performance-per-Watt Management

With the increased complexity of platforms coupled with data centers' servers sprawl, power consumption is reaching unsustainable limits. Researchers have addressed data centers' performance-per-watt management at different hierarchies going from server clusters to servers to individual components within the server platform. This paper addresses performance-per-watt maximization of memory subsystems in a data center. Traditional memory power management techniques rely on profiling the utilization of memory modules and transitioning them to some low-power mode when they are sufficiently idle. However, fully interleaved memory presents an interesting research challenge because data striping across memory modules reduces the idleness of individual modules to warrant transitions to low-power states. In this paper, we present a novel technique for performance-per-watt maximization of interleaved memory by dynamically reconfiguring (expanding or contracting) the degree of interleaving to adapt to incoming workload. The reconfigured memory hosts the application's working set on a smaller set of modules in a manner that exploits the platform's memory hierarchy architecture. This creates the opportunity for the remaining memory modules to transition to low-power states and remain in those states for as long as the performance remains within given acceptable thresholds. The memory power expenditure is minimized subject to application memory requirements and end-to-end memory access delay constraints. This is formulated as a performance-per-watt maximization problem and solved using an analytical memory power and performance model. Our technique has been validated on a real server using SPECjbb benchmark and on a trace-driven memory simulator using SPECjbb and gcc memory traces. On the server, our techniques are shown to give about 48.8 percent (26.7 kJ) energy savings compared to traditional techniques measured at 4.5 percent. The maximum improvement in performance-per-watt was measured at 88.48 percent. The simulator showed 89.7 percent improvement in performance-per-watt compared to the best performing traditional technique.

A Practical Dynamic Frequency Scaling Solution to DPM in Embedded Linux Systems

Traditional power management enables CPU to switch among different power consumption modes such as running, idle and standby to save energy to a certain degree. In battery-powered embedded Linux systems, power management plays an even more important role than in desktops. This paper proposes a practical solution to dynamic power management (DPM) based on dynamic frequency scaling (DFS). According to the system workload, it switches the working frequency among several predefined settings. This paper presents a design and implementation of the solution in an embedded Linux system. Experimental results on an Intel PXA250-based system have demonstrated that the system running a DPM-enabled Linux can save up to 44% of execution time with less than 8% extra power consumption in an email application.

Integrated CPU and l2 cache voltage scaling using machine learning

Embedded systems serve an emerging and diverse set of applications. As a result, more computational and storage capabilities are added to accommodate ever more demanding applications. Unfortunately, adding more resources typically comes on the expense of higher energy costs. New chip design with Multiple Clock Domains (MCD) opens the opportunity for fine-grain power management within theprocessor chip. When used with dynamic voltage scaling (DVS), we can control the voltage and power of each domain independently. A significant power and energy improvement has been shown when using MCD design in comparison to managing a single voltage domain for the whole chip, as in traditional chips with global DVS. In this paper, we propose PACSL a Power-Aware Compiler-based approach using Supervised Learning. PACSL automatically derives an integrated CPU-core and on-chip L2 cache DVS policy tailored to a specific system and workload. Our approach uses supervised machine learning to discover a policy, which relies on monitoring a few performance counters. We present our approach detailing the role of a compiler in constructing a custom power management policy. We also discuss some implementation issues associated with our technique. We show that PACSL improves on traditional power management techniques that are used in general MCD chips. Our technique saves 22% on average (up to 46%) in energy-delay product over a DVS technique that applies independent DVS decisions in each domain. Compared to no-power management, our technique improves energy-delay product by 26% on average (up to 64%).

Managing power consumption in networks on chips

In this paper, we present a new methodology for managing power consumption of networks-on-chips (NOCs). A power management problem is formulated for the first time using closed-loop control concepts. We introduce an estimator and a controller that implement our power management methodology. The estimator is capable of very fast and accurate tracking of changes in the system parameters. Parameters estimated are used to form the system model. Our system model combines node and network centric power management decisions. Node centric power management assumes no a priori knowledge of requests coming in from outside the core. Thus, it implements a more traditional dynamic voltage scaling and power management control algorithms. Network-centric power management utilizes interaction with the other system cores regarding the power and the quality of service (QoS) needs. The overall system model is based on Renewal theory and, thus, guarantees globally optimal results. We introduce a fast optimization method that runs multiple orders of magnitude faster than the previous optimization approaches while still having the same accuracy in obtaining the power management control. Finally, our controller implements the results of optimization in either hardware or software. The new methodology for power management of NOCs is tested on a system consisting of four satellite units, each implementing an estimator and a controller capable of both node and network centric power management. Our results show large savings in power with good QoS.