CN115509626A - Method and device for realizing pause state setting based on energy prediction in nonvolatile processor - Google Patents
Method and device for realizing pause state setting based on energy prediction in nonvolatile processor Download PDFInfo
- Publication number
- CN115509626A CN115509626A CN202211390764.7A CN202211390764A CN115509626A CN 115509626 A CN115509626 A CN 115509626A CN 202211390764 A CN202211390764 A CN 202211390764A CN 115509626 A CN115509626 A CN 115509626A
- Authority
- CN
- China
- Prior art keywords
- energy
- power
- period
- prediction model
- state
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000013528 artificial neural network Methods 0.000 claims abstract description 20
- 230000008859 change Effects 0.000 claims abstract description 19
- 230000007613 environmental effect Effects 0.000 claims abstract description 18
- 238000012360 testing method Methods 0.000 claims abstract description 10
- 238000012549 training Methods 0.000 claims abstract description 7
- 238000012545 processing Methods 0.000 claims abstract description 6
- 238000007781 pre-processing Methods 0.000 claims abstract description 5
- 238000005070 sampling Methods 0.000 claims description 26
- 238000003062 neural network model Methods 0.000 claims 1
- 239000002699 waste material Substances 0.000 abstract description 4
- 230000008569 process Effects 0.000 description 5
- 230000008901 benefit Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000011084 recovery Methods 0.000 description 3
- 230000036760 body temperature Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003990 capacitor Substances 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005059 dormancy Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007477 logistic regression Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000001208 nuclear magnetic resonance pulse sequence Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000036387 respiratory rate Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000002459 sustained effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/4401—Bootstrapping
- G06F9/4418—Suspend and resume; Hibernate and awake
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Computer Security & Cryptography (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Power Sources (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The method provided by the invention comprises the following steps: the method comprises the steps of firstly, collecting the change trend of environmental energy, and carrying out data preprocessing on the change trend to change the environmental energy into environmental energy data for use; secondly, establishing an energy prediction model based on a neural network, and training and testing the energy prediction model according to the preprocessed environmental energy data, so that the prediction result of the energy prediction model reaches a high-accuracy level; thirdly, creating a solution aiming at different prediction results of the energy prediction model; fourthly, when the processing system is powered off, predicting the future energy trend of the system after the power off according to the energy prediction model; and fifthly, according to the energy trend predicted by the energy prediction model, applying a corresponding solution to guide the selection of the pause state and the backup state of the system. The method and the device for realizing the pause state setting based on the energy prediction in the nonvolatile processor can avoid unnecessary waste of limited available energy after energy power failure in the system, make the best state selection for the system and further realize the maximization of the execution efficiency.
Description
Technical Field
The present invention relates to the field of electronic technologies, and in particular, to a method and an apparatus for implementing energy prediction based suspend state setting in a non-volatile processor.
Background
Along with the rapid development of the internet of things, wearable devices such as smart bracelets and smart watches are widely applied. Most wearable devices are capable of detecting physiological data such as heart rate, body temperature, and respiratory rate of a user in real time. Typically, wearable devices are powered by equipped rechargeable batteries. The rechargeable battery has the problem of frequent charging, and the rechargeable battery with larger volume is not suitable for the medical wearable device.
In order to solve the power supply problem of tiny internet of things equipment, people also utilize an energy collection system to convert the energy of the surrounding environment such as solar energy, wind energy, body temperature and the like into electric energy for use, so that the internet of things equipment realizes self-power supply, and further obtains the effect of working for a super long time. However, there is a problem that the energy of the surrounding environment is converted into electric energy, and the power supply is unstable, which may cause frequent power failure of the wearable device. When the wearable device is interrupted by energy, the conventional processor needs to perform multiple rollback operations on the executed process, so that the backup overhead is greatly increased.
Conventional non-volatile processors may solve the above problems. When the energy is lost, the energy stored in the capacitor can support the backup of volatile data into the nonvolatile memory, and after the energy is recovered, the data is copied back to the processor for continuous execution, so that the loss of a program process is avoided, and the normal operation of the program is ensured.
Research shows that the self-powered system has a high probability of restoring power supply shortly after power failure. Thus, in such conventional non-volatile processors, the system may employ a suspend mode to optimize the operations of backup and restore. Namely, when the system is powered down, the data backup is not carried out at first, and the system enters a pause state. When the system enters the dormancy state in the suspended state and waits for the incoming call again, if the system recovers the incoming call in a short time, the program continues to run, and therefore the expenses of backup and recovery are reduced.
Although conventional non-volatile processors may utilize the suspend state to reduce the number of times of backup, the lack of knowledge of the future trend of energy changes makes the system too conservative in selecting the suspend state and the backup state, and thus the advantage of the suspend state cannot be maximized. The invention predicts the future change trend of energy through a prediction model established based on a neural network, and guides a system to make more reasonable selection between a pause state and a backup state based on a prediction result, thereby maximizing the energy efficiency.
Disclosure of Invention
To overcome, at least in part, the problems in the related art, the present application provides a method and apparatus for implementing energy prediction based suspend state setting in a non-volatile processor.
The method provided by the invention comprises the following steps: preprocessing the data according to the acquired energy to enable the data to become usable data; establishing an energy prediction model according to the neural network; training and testing a prediction model according to the preprocessed environmental energy data, so that the prediction result of the prediction model reaches a high-accuracy level; predicting the future energy trend of the system after power failure according to the application prediction model; and guiding the selection of the pause state and the backup state of the system according to the predicted energy trend.
That is, the prediction model is used to predict the future energy variation trend, and the predicted result is input into the state selection model, so that the state selection model can perform more reasonable and favorable state selection result according to the predicted environmental energy variation trend.
An embodiment of the present application provides a method for designing a halt state based on energy prediction in a non-volatile processor, which includes the following steps:
firstly, collecting environmental energy, preprocessing the collected data, and processing and analyzing the environmental energy;
designing an energy prediction algorithm based on a neural network, and continuously optimizing the prediction algorithm to achieve higher accuracy;
designing a pause state selection scheme;
when the processor system is powered off, a prediction algorithm is applied to predict the change trend of future energy, and the predicted result is divided into four types;
according to the predicted energy change trend category, different selection strategies are carried out according to different categories, so that the self-powered system can be selected more reasonably under different scenes, the program process is maximized, and the energy efficiency is improved.
Optionally, the acquiring environmental energy includes:
sampling environment energy within a preset time range to obtain an energy trace signal;
and filtering the energy trace signal into a pulse sequence under the action of a preset threshold, wherein the preset threshold represents that the environmental energy can meet the normal operation of the system and can be used as a signal.
Optionally, the energy prediction algorithm based on the neural network is implemented by a fully-connected lightweight neural network, the neural network has three input parameters and four output results, and the future energy change trend can be predicted according to the energy condition of the previous power sampling period.
Optionally, the continuous optimization of the prediction algorithm is to apply the test set to continuously debug and verify by adjusting the network structure, parameter indexes, and the like, so that the prediction model can achieve a prediction result with high accuracy.
Optionally, the predicted result is divided into four categories, namely, continuous power down, power down after power up, power up after power down, and continuous power up, and corresponding terms are represented by 00, 01, 10, and 11, where 0 represents power down and 1 represents power up.
Optionally, the selection strategy is as follows:
when the predicted trend is 00, comparing the current power-off period with the size of a waiting period which can be provided by the system for selection, if the power-off period is greater than the waiting period, the system enters a backup state, if the power-off period is less than the waiting period, the system enters a pause state, predicting again in the predicted last power sampling period, accumulating the power-off period and entering the next judgment cycle;
when the predicted trend is 01, comparing the current power-off period with the size of a waiting period which can be provided by the system for selection, if the power-off period is greater than the waiting period, the system enters a backup state, and if the power-off period is less than the waiting period, the system enters a pause state;
when the predicted trend is 10, comparing the power-off period with the size of a waiting period which can be provided by the system for selection, if the power-off period is greater than the waiting period, the system enters a backup state, if the power-off period is less than the waiting period, the system enters a pause state, predicting again in the predicted last power sampling period, and entering the next judgment cycle;
when the predicted trend is 11, comparing the current power-off period with the size of a waiting period which can be provided by the system for selection, if the power-off period is larger than the waiting period, the system enters a backup state, and if the power-off period is smaller than the waiting period, the system enters a pause state.
Optionally, the energy efficiency improvement is performed by applying a corresponding experimental platform and a simulation mode to verify the advantages of the design compared with other schemes before, and the advantages are verified by using indexes such as throughput rate, energy efficiency and energy utilization rate.
The technical scheme provided by the application can comprise the following beneficial effects:
the method and the device for realizing the pause state setting based on the energy prediction in the nonvolatile processor can avoid unnecessary waste of limited available energy in the system after energy power failure, make the best state selection for the system and further realize the maximization of execution efficiency.
Drawings
The drawings illustrate various embodiments, by way of example and not by way of limitation, and together with the description and claims, serve to explain the inventive embodiments. The same reference numbers will be used throughout the drawings to refer to the same or like parts, where appropriate. Such embodiments are illustrative, and are not intended to be exhaustive or exclusive embodiments of the present apparatus or method.
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application.
FIG. 1 is a flowchart of a method for implementing energy-prediction-based pause state setting in a non-volatile processor according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a trace signal of environmental energy provided by an embodiment of the present application;
FIG. 3 is a schematic illustration of FIG. 2 after data processing;
FIG. 4 is a schematic diagram of a neural network for energy prediction provided by an embodiment of the present application;
fig. 5 is a schematic diagram of an optimal state selection policy provided in an embodiment of the present application.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of methods and apparatus consistent with certain aspects of the present application, as detailed in the appended claims.
Fig. 1 is a flowchart of a method for implementing energy prediction based suspend state setting in a non-volatile processor according to an embodiment of the present application, and referring to fig. 1, the method includes the following steps:
s1, collecting environmental energy through an energy collecting device. Energy data collected referring to fig. 2, we collected three typical environmental energies, where the start threshold was 3456 microwatts, we considered the system powered on at the time when the energy intensity was greater than the start threshold, and conversely, the system powered off at the time when the energy intensity was less than the start threshold.
After the environmental energy is collected, the collected data is processed. Referring to fig. 3, for the convenience of analyzing data, data larger than a threshold value is denoted as 1, data smaller than the threshold value is denoted as 0, and fig. 3 shows a case after partial data processing.
S2, firstly, a prediction model based on a neural network is established, the neural network structure refers to FIG. 4, and the neural network has three inputs, four outputs and a plurality of hidden layers. The input data are respectively the energy intensity at the moment of prediction, the energy intensity mean value of the first three power sampling periods and the energy intensity logistic regression value of the first three power sampling periods. Wherein the output result has four values, which are 00, 01, 10 and 11 respectively, corresponding to different energy variation trends, wherein 00 represents that the future energy level is in continuous power failure; 01 represents that the future energy level is that the power is cut off first and then the power is supplied; 10 represents future energy levels with power on first and then off; 11 represents a future energy level present with a sustained call.
And dividing the data collected in the step S1 into a training set and a test set, wherein the training set is used for training the neural network prediction model established in the step S2, and the test set is used for testing the accuracy of the trained neural network-based prediction model. Through continuous testing and adjustment of parameters of the neural network, the testing result reaches a level with higher accuracy.
And S3, designing a corresponding pause state selection strategy according to the predicted result.
S4, after the prediction model based on the neural network is trained to be mature, the prediction model based on the neural network can be applied to self-powered equipment, when a processor system is powered down, the data condition before the power down is brought into the prediction model based on the neural network, namely, three input data are input into the neural network, and therefore four basic future energy change trends can be predicted through the prediction model.
And S5, after the future energy change trend is known, a more reasonable state selection strategy can be carried out according to the future energy change trend. Referring to fig. 5, the predicted result is divided into a pre-power sampling period and a post-power sampling period. The two results are obtained before and after the power sampling period, so that four results are formed, namely 00, 01, 10 and 11, wherein the time of the power sampling period interval is recorded as T, and the longest waiting incoming call period which can be provided by the system is recorded as Tw.
For the 00 situation, since 00 represents that the future energy variation trend is a continuous power-off level, we first determine whether the power-off period is greater than the longest waiting power-on period Tw, and since the two future power sampling periods are both power-off levels, the power-off period at this time is 3T, that is, determine the magnitude relationship between 3T and Tw. When the power-off period is longer than the longest waiting period, the system should select the backup state, because the longest waiting period which can be provided by the system is shorter than the power-off period, that is, the existing energy of the system is not enough to insist on coming of the incoming call moment, the system selects the backup state, backs up data as early as possible, and reduces energy loss; when the power down period is less than the longest waiting period, that is, the longest waiting period that the system can provide is greater than 3T, the system supports the time of waiting for 3T, at this time, the system should enter a pause state, and then the predicted future energy change trend is predicted again at the moment of the predicted post-power sampling period, that is, the post-power sampling period enters the judgment cycle again. However, at this time, the power-down period should be accumulated, and when the cycle is repeated, the accumulated power-down period should be used for comparison with the longest waiting period.
For the case 01, since 01 represents the future energy change trend that power is first turned off and then turned on, we first determine whether the power-off period is greater than the longest waiting incoming call period Tw, and since the previous power sampling period in the future is the power-off condition and the next power sampling period is the incoming call condition, the power-off period at this time is 2T, that is, the relationship between 2T and Tw is determined. When the power-off period is longer than the longest waiting period, the system should select the backup state, because the longest waiting period which can be provided by the system is shorter than the power-off period, namely the existing energy of the system is not enough to insist on the coming of the incoming call moment, the system selects the backup state, backs up data as early as possible, and reduces the energy loss; when the power down period is less than the longest waiting period, that is, the longest waiting period that the system can provide is greater than 2T, the system supports the time of waiting for 2T, at this time, the system should enter a pause state, and the system can resume the incoming call after waiting for 2T, thus avoiding the performance waste of one backup and recovery data.
For the 10 cases, since 10 represents the future energy variation trend is the case of powering on first and then powering off, we first determine whether the power-off period is greater than the longest waiting power-on period Tw, and since the future previous power sampling period is the case of powering on and the next power sampling period is the case of powering off, the power-off period at this time is T, that is, determine the magnitude relationship between T and Tw. When the power-off period is longer than the longest waiting period, the system should select the backup state, because the longest waiting period which can be provided by the system is shorter than the power-off period, that is, the existing energy of the system is not enough to insist on coming of the incoming call moment, the system selects the backup state, backs up data as early as possible, and reduces energy loss; when the power-down period is less than the longest waiting period, that is, the longest waiting period that the system can provide is greater than T, the system supports the time for waiting T, at this time, the system should enter a pause state first, and then predict the future energy change trend again at the predicted post-power sampling period, that is, the post-power sampling period enters the judgment cycle again. However, the power down period is not accumulated at this time.
For the case of 11, since 11 represents that the future energy variation trend is the continuous incoming call level, we first determine whether the power down period is greater than the longest waiting incoming call period Tw, and since the future power sampling period before and after is the incoming call condition, the power down period at this time is T, that is, determine the magnitude relationship between T and Tw. When the power-off period is longer than the longest waiting period, the system should select the backup state, because the longest waiting period which can be provided by the system is shorter than the power-off period, that is, the existing energy of the system is not enough to insist on coming of the incoming call moment, the system selects the backup state, backs up data as early as possible, and reduces energy loss; when the power failure cycle is less than the longest waiting cycle, that is, the longest waiting cycle that the system can provide is greater than T, the system supports the time of waiting T, and at this time, the system should enter a suspend state first, and then wait for T time, the system can resume the incoming call, thus avoiding the performance waste of a backup and recovery data.
It is to be understood that the same or similar components in the above embodiments may be referred to each other, and the same or similar components in other embodiments may be referred to in some embodiments without detailed description, and the same or similar components in method embodiments may be referred to in other embodiments without detailed description.
It should be noted that, in the description of the present application, the terms "first", "second", etc. are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. Further, in the description of the present application, the meaning of "a plurality" means at least two unless otherwise specified.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and the scope of the preferred embodiments of the present application includes other implementations in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present application.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried out in the method of implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and the program, when executed, includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present application may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc.
In the description herein, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Although embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application, and that variations, modifications, substitutions and alterations may be made to the above embodiments by those of ordinary skill in the art within the scope of the present application.
Claims (7)
1. A method for implementing energy prediction based suspend state setting in a non-volatile processor, comprising the steps of:
the method comprises the steps of firstly, collecting the change trend of environmental energy, and carrying out data preprocessing on the change trend to change the environmental energy into environmental energy data for use;
secondly, establishing an energy prediction model based on a neural network, and training and testing the energy prediction model according to the preprocessed environmental energy data, so that the prediction result of the energy prediction model reaches a high-accuracy level;
thirdly, creating a solution aiming at different prediction results of the energy prediction model;
fourthly, when the processing system is powered off, predicting the future energy trend of the system after the power off according to the energy prediction model;
and fifthly, according to the energy trend predicted by the energy prediction model, applying a corresponding solution to guide the selection of the pause state and the backup state of the system.
2. The method of claim 1, wherein the data preprocessing in the first step comprises:
the size of the data is scaled, and the data is labeled.
3. The method of claim 1, wherein the establishing of the neural network-based energy prediction model in the second step comprises:
the energy prediction result of the selection and setting of the input value is the energy trend after the system is powered off and is not the single energy intensity.
4. The method of claim 1, wherein the training and testing of the energy prediction model in the second step comprises:
the prediction precision is improved by continuously debugging relevant parameters and input values in the neural network model.
5. Method according to claim 1, characterized in that the solution in the third step comprises:
and setting strategies correspondingly according to different results predicted by the energy prediction model.
6. The method of claim 5,
the solution divides the energy prediction trend into a front power sampling period and a rear power sampling period, and provides a corresponding scheme according to the change condition of the front power sampling period and the rear power sampling period; the key judgment conditions for deciding the state selection are as follows: the power down period is related to the size of the longest waiting period.
7. The method of claim 6,
the corresponding scheme is provided according to the change conditions of the front and rear power sampling periods, wherein 0 represents the power-off condition, 1 represents the power-on condition, and the change conditions of the front and rear power sampling periods are 00, 01, 10 and 11; for each case, a comparison is made of the own power-down period with the longest waiting period that the system can provide; for the 00 condition, if the power-off period is larger than the longest waiting period, entering a backup state, otherwise entering a pause state, and accumulating the power-off period before entering the next judgment; if the power-off period is longer than the longest waiting period under the condition of 01, entering a backup state, otherwise, entering a pause state; for 10 cases, if the power-off period is greater than the longest waiting period, entering a backup state, otherwise, entering a pause state, predicting in a later period, and then entering next judgment; for the 11 case, the power-off period is greater than the longest waiting period, then the backup state is entered, otherwise the suspend state is entered.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211390764.7A CN115509626B (en) | 2022-11-07 | 2022-11-07 | Method and device for realizing energy prediction-based pause state setting in nonvolatile processor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211390764.7A CN115509626B (en) | 2022-11-07 | 2022-11-07 | Method and device for realizing energy prediction-based pause state setting in nonvolatile processor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115509626A true CN115509626A (en) | 2022-12-23 |
| CN115509626B CN115509626B (en) | 2024-02-02 |
Family
ID=84513686
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211390764.7A Active CN115509626B (en) | 2022-11-07 | 2022-11-07 | Method and device for realizing energy prediction-based pause state setting in nonvolatile processor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115509626B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109116968A (en) * | 2018-08-06 | 2019-01-01 | 清华大学 | Write policy control method, system and its applicable electronic equipment |
| CN109116967A (en) * | 2018-08-06 | 2019-01-01 | 清华大学 | Based on time-multiplexed neural network prediction device and its applicable electronic equipment |
| CN109144214A (en) * | 2018-08-06 | 2019-01-04 | 清华大学 | Energy Management System, method, electronic equipment, device and non-volatile processor |
| US20210150383A1 (en) * | 2019-11-19 | 2021-05-20 | International Business Machines Corporation | Prediction of a data protection activity time for a backup environment |
| CN113077172A (en) * | 2021-04-19 | 2021-07-06 | 瑞湖智科数据(苏州)有限公司 | Equipment state trend analysis and fault diagnosis method |
| CN115115030A (en) * | 2022-06-24 | 2022-09-27 | 平安科技（深圳）有限公司 | System monitoring method and device, electronic equipment and storage medium |
-
2022
- 2022-11-07 CN CN202211390764.7A patent/CN115509626B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109116968A (en) * | 2018-08-06 | 2019-01-01 | 清华大学 | Write policy control method, system and its applicable electronic equipment |
| CN109116967A (en) * | 2018-08-06 | 2019-01-01 | 清华大学 | Based on time-multiplexed neural network prediction device and its applicable electronic equipment |
| CN109144214A (en) * | 2018-08-06 | 2019-01-04 | 清华大学 | Energy Management System, method, electronic equipment, device and non-volatile processor |
| US20210150383A1 (en) * | 2019-11-19 | 2021-05-20 | International Business Machines Corporation | Prediction of a data protection activity time for a backup environment |
| CN113077172A (en) * | 2021-04-19 | 2021-07-06 | 瑞湖智科数据(苏州)有限公司 | Equipment state trend analysis and fault diagnosis method |
| CN115115030A (en) * | 2022-06-24 | 2022-09-27 | 平安科技（深圳）有限公司 | System monitoring method and device, electronic equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| ZEJUN SHI等: "leveraging energy cycle regularity to predict adaptive mode for non-volatile processors", 2019 IEEE 30TH INTERNATIONAL CONFERENCE ON APPLICATION-SPECIFIC SYSTEMS, ARCHITECTURES AND PROCESSORS (ASAP), pages 189 - 196 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115509626B (en) | 2024-02-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7716500B2 (en) | Power source dependent program execution | |
| US20170091878A1 (en) | Method and system for controlling charge and discharge of high powered energy storage systems | |
| CN114047452B (en) | Method and device for determining cycle life of battery | |
| CN105893141A (en) | Regulation control method and apparatus for multi-core processor and mobile terminal using method | |
| US11734103B2 (en) | Behavior-driven die management on solid-state drives | |
| CN115509626A (en) | Method and device for realizing pause state setting based on energy prediction in nonvolatile processor | |
| CN111737053B (en) | Nonvolatile processor backup method and system based on instruction analysis | |
| Lee et al. | Energy-efficient control of mobile processors based on long short-term memory | |
| CN103593220B (en) | OPENCL compilation | |
| CN111082487B (en) | Charging control method and device and electronic equipment | |
| WO2023207062A1 (en) | Maximum power tracking control method, and photovoltaic system and energy storage device | |
| Kiarie et al. | Raspberry Pi based recording system for acoustic monitoring of bird species | |
| CN110991519A (en) | Intelligent switch state analysis and adjustment method and system | |
| US20220146996A1 (en) | Reward generation method to reduce peak load of electric power and action control apparatus performing the same method | |
| CN109301820A (en) | A kind of enterprise's electrical control method and system | |
| CN111007406B (en) | Internet of things node power consumption statistical learning method | |
| US11714564B2 (en) | Systems and methods of power management | |
| CN106560761A (en) | Computer system and method for providing both main power and auxiliary power on a single power bus | |
| Henares et al. | Cyber-physical systems design methodology for the prediction of symptomatic events in chronic diseases | |
| CN115389947B (en) | Lithium battery health state prediction method and device, electronic equipment and storage medium | |
| CN111509810A (en) | Lithium battery charging state detection method and device and storage medium | |
| CN117519759B (en) | Remote upgrading method and system for household energy storage system equipment | |
| KR102650678B1 (en) | Device control apparatus and method for controlling an inactive mode of a device | |
| CN115566762B (en) | Self-balancing electric vehicle intelligent charge and discharge management method for local power distribution network | |
| KR102259096B1 (en) | Solar cell power plant and control system, and its method of operating thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |