Power-on and power-off self-reset control circuit and control method thereof

Abstract

The present invention discloses a control circuit for power-on and power-off self-reset and a control method thereof, which obtains a power Good signal and a reset signal within the predetermined time period; performs feature analysis on the power Good signal and the reset signal to obtain the power Good signal waveform feature and the reset signal waveform feature; performs transfer association coding on the power Good signal waveform feature and the reset signal waveform feature to obtain the power Good-reset signal transfer association feature; and controls the generation of a power-on signal based on the power Good-reset signal transfer association feature. In this way, the system can be automatically powered on and turned on after an abnormal power failure, thereby improving the reliability and stability of the electric energy meter, ensuring that the electric energy meter can resume normal operation and protect the integrity of the data in the event of an abnormal power failure.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to a control circuit for self-reset when power is on or off and a control method thereof. Background Art

[0002] With the continuous development and maturity of electronic energy meter technology, the functions of energy meters are also increasing and improving. In addition to the basic energy metering function, modern energy meters also have a variety of communication methods, such as RS485 communication, infrared communication, carrier communication, wireless public network 4G communication and micro-power wireless communication. In addition, there has been further development and maturity in LCD display, prepayment and payment rate functions. Energy meters with multiple rate functions have been widely installed and used throughout the country.

[0003] In order to ensure the normal operation of various functions of the energy meter, the normal operation of the MCU (microcontroller unit) is crucial. Especially in the case of a power outage, it is particularly important to ensure that the MCU can complete data processing and storage work normally to avoid data errors or confusion. In order to improve the reliability of the MCU, some existing solutions use a battery backup module to cope with the power supply needs after the mains power outage.

[0004] In addition, as data centers have higher and higher requirements for hardware intelligence, they also need to ensure that the MCU can automatically power on and work normally when power is restored after a power outage. However, in the case of abnormal power failure of the system, such as accidental power failure of the MCU or chip overheating causing the system to power off, it becomes relatively difficult to achieve automatic detection and automatic power on. Currently, there is no solution that can achieve automatic power on after an abnormal power failure of the system.

[0005] Therefore, an optimized power-on / power-off self-reset control circuit is desired. Summary of the invention

[0006] The embodiment of the present invention provides a control circuit for power-on and power-off self-reset and a control method thereof, which obtains a power Good signal and a reset signal within the predetermined time period; performs feature analysis on the power Good signal and the reset signal to obtain a power Good signal waveform feature and a reset signal waveform feature; performs transfer association coding on the power Good signal waveform feature and the reset signal waveform feature to obtain a power Good-reset signal transfer association feature; and controls the generation of a power-on signal based on the power Good-reset signal transfer association feature. In this way, the system can be automatically powered on and turned on after an abnormal power failure, thereby improving the reliability and stability of the electric energy meter, ensuring that the electric energy meter can resume normal operation and protect the integrity of the data in the event of an abnormal power failure.

[0007] The embodiment of the present invention further provides a power-on and power-off self-reset control circuit, which comprises:

[0008] A signal acquisition module, used to obtain a power Good signal and a reset signal within the predetermined time period;

[0009] A signal characteristic analysis module, used for performing characteristic analysis on the power Good signal and the reset signal to obtain a power Good signal waveform characteristic and a reset signal waveform characteristic;

[0010] A signal feature transfer correlation coding module, used for performing transfer correlation coding on the power Good signal waveform feature and the reset signal waveform feature to obtain a power Good-reset signal transfer correlation feature;

[0011] The power-on signal control module is used to control the generation of a power-on signal based on the power Good-reset signal transfer association feature.

[0012] The embodiment of the present invention also provides a control method for power-on and power-off self-reset, which includes:

[0013] Obtaining a power Good signal and a reset signal within the predetermined time period;

[0014] Performing feature analysis on the power Good signal and the reset signal to obtain power Good signal waveform features and reset signal waveform features;

[0015] Performing transfer association coding on the power Good signal waveform feature and the reset signal waveform feature to obtain a power Good-reset signal transfer association feature;

[0016] Based on the power Good-reset signal transfer correlation feature, control is performed to generate a power-on signal. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0018] Figure 1 A simple power-off and power-on voltage detection circuit diagram provided in an embodiment of the present invention.

[0019] Figure 2 The present invention provides a block diagram of a power-on and power-off self-reset control circuit in an embodiment of the present invention.

[0020] Figure 3 Schematic diagram of a voltage detection chip provided in an embodiment of the present invention.

[0021] Figure 4 A voltage detection circuit diagram for powering on and off an electric energy meter provided in an embodiment of the present invention.

[0022] Figure 5 The present invention provides a flow chart of a method for controlling power-on and power-off self-reset in an embodiment of the present invention.

[0023] Figure 6 A schematic diagram of the system architecture of a power-on / power-off self-reset control method provided in an embodiment of the present invention.

[0024] Figure 7 This is a diagram of an application scenario of a power-on / power-off self-reset control circuit provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0025] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0026] Unless otherwise specified, all technical and scientific terms used in the embodiments of the present application have the same meaning as those commonly understood by those skilled in the art of the present application. The terms used in this application are only for the purpose of describing specific embodiments and are not intended to limit the scope of this application.

[0027] In the embodiments of the present application, it should be noted that, unless otherwise specified and limited, the term "connection" should be understood in a broad sense. For example, it can be an electrical connection or a connection between two components. It can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to the specific circumstances.

[0028] It should be noted that the terms "first\second\third" involved in the embodiments of the present application are only used to distinguish similar objects, and do not represent a specific order for the objects. It is understandable that the "first\second\third" can be interchanged with a specific order or sequence where permitted. It should be understood that the objects distinguished by "first\second\third" can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein.

[0029] With the continuous development and maturity of electronic energy meter technology, the functions of energy meters are also increasing and maturing, such as communication methods RS485 communication, infrared communication, carrier communication, wireless public network 4G communication, micro-power wireless communication, etc. For example, the functions of LCD display, prepayment, and multiple rates are also developing and maturing. Energy meters with multiple rates have been fully installed and used throughout the country. In order to ensure the normal operation of the energy meter in various functions, it is necessary to ensure the normal operation of the energy meter MCU. When the mains power is cut off under normal working conditions (power supply voltage is about 5V), ensure that the MCU can complete the data processing and storage work normally in the case of power off, so that the data will not produce errors or confusion. At the same time, it is also necessary to ensure that the MCU works normally in the case of power on, and the saved data processing and other normal operations continue to be processed. Therefore, the single-chip MCU has a special I / O port for connecting the circuit to judge the power off, that is, the voltage detection circuit for power off and power on.

[0030] Existing voltage detection circuits for powering off and on are as follows Figure 1 As shown. This circuit is the simplest voltage detection circuit for power-off and power-on, and can also be said to be the circuit for judging power-off of the I / O port of the MCU. When the mains power supply is normal (the power supply voltage VIN is about 5V), the detected voltage VOUT is about 5V.

[0031] This technical solution has the following defects:

[0032] When performing a quick power-on and power-off test on a product, if the MCU's power-on and power-off requirements are not met, the MCU will often fail to start or even lock up.

[0033] MCU has strict requirements for power on and off. In product design, attention should be paid to the power on and off requirements of core components, including power on and off timing, slope, etc. Unreasonable design often causes abnormal situations such as MCU power failure.

[0034] When the MCU is powered off and the load cannot release energy quickly, the MCU and other digital devices will power off slowly. If the power is turned on again at this time, since the chip cannot be "reset" in time, it is an uncertain state for digital devices such as the MCU. If the system is powered on again at this time, it is easy to cause MCU logic confusion, resulting in device latching and the MCU cannot be started.

[0035] The problem of slow power-off is easily overlooked during the design process. It is often too late to discover the problem during the product debugging stage. Adding a fast discharge circuit to the system is both time-consuming and labor-intensive.

[0036] Therefore, in the present application, an optimized power-on and power-off self-reset control circuit is provided.

[0037] In one embodiment of the present invention, Figure 2 FIG. 1 is a block diagram of a power-on / power-off self-reset control circuit provided in an embodiment of the present invention. Figure 2 As shown, the power-on and power-off self-reset control circuit 100 according to an embodiment of the present invention includes: a signal acquisition module 110, used to obtain a power Good signal and a reset signal within the predetermined time period; a signal feature analysis module 120, used to perform feature analysis on the power Good signal and the reset signal to obtain a power Good signal waveform feature and a reset signal waveform feature; a signal feature transfer association coding module 130, used to perform transfer association coding on the power Good signal waveform feature and the reset signal waveform feature to obtain a power Good-reset signal transfer association feature; a power-on signal control module 140, used to control the generation of a power-on signal based on the power Good-reset signal transfer association feature.

[0038] In the signal acquisition module 110, the power Good signal and the reset signal within a predetermined time period are obtained. During implementation, it is ensured that the signal acquisition module can accurately acquire and collect these two signals for subsequent feature analysis and processing. In the signal feature analysis module 120, the power Good signal and the reset signal are feature analyzed to obtain their waveform features. When performing feature analysis, it is noted that appropriate algorithms and methods are selected to accurately extract the key features of the signal, such as falling edge, duration, frequency, etc., so as to provide reliable input data for subsequent transfer association coding. In the signal feature inter-transfer association coding module 130, the power Good signal waveform feature and the reset signal waveform feature are transferred and associated coded to obtain the power Good-reset signal transfer association feature. During implementation, appropriate coding algorithms and methods are selected to ensure that the transfer association feature can accurately reflect the relationship between the power Good signal and the reset signal, so as to provide an effective judgment basis for the subsequent power-on signal control. In the power-on signal control module 140, based on the power Good-reset signal transfer association feature, the power-on signal is controlled to be generated. During implementation, the triggering conditions and generation methods of the power-on signal are determined according to specific needs and system design. By accurately determining the correlation characteristics between the power Good signal and the reset signal, it is possible to ensure automatic power-on after an abnormal power failure, thereby improving the reliability and stability of the system.

[0039] In this way, through feature analysis and transfer-related coding, the abnormal power-off situation can be accurately judged, so as to realize automatic power-on and boot, and improve the reliability and stability of the system. Automatic power-on and boot after abnormal power-off can ensure that the electric energy meter can resume normal operation and protect the integrity of the data. Through the analysis and control of the power Good signal and the reset signal, more accurate power-on control can be achieved, and the functional performance of the electric energy meter can be improved. Through signal feature analysis and transfer-related coding, intelligent judgment and control of the power Good signal and the reset signal can be realized, and the intelligence level of the system can be improved.

[0040] In electronic devices, power failure refers to a situation where the power supply is interrupted or abnormal, causing the device to fail to work properly. In order to ensure the stability and reliability of the device, it is usually necessary to monitor the power status and take corresponding measures. Power failure self-reset is a common control circuit that can automatically reset or restart the device to resume normal operation when a power anomaly is detected.

[0041] When designing a control circuit for power-on and power-off self-reset, it is necessary to analyze the power Good signal and the reset signal to determine whether an abnormal power failure has occurred. The power Good signal is usually used to indicate whether the power supply is working properly, while the reset signal is used to restore the device to its initial state. By monitoring and analyzing the changes in these two signals, it is possible to determine whether an abnormal power failure has occurred.

[0042] Specifically, the technical concept of the present application is to perform correlation analysis on the power Good signal and the reset signal through signal processing and analysis algorithms, so as to determine whether there is a 0 bit in the falling edge change data of the power Good signal. If so, it is determined whether the reset signal data is 0, so as to determine whether it is an abnormal power outage. In this way, the system can automatically power on after an abnormal power outage, thereby improving the reliability and stability of the electric energy meter, ensuring that the electric energy meter can resume normal operation and protect the integrity of the data in the event of an abnormal power outage.

[0043] Specifically, in the technical solution of the present application, first, a power Good signal and a reset signal within the predetermined time period are obtained. The power Good signal is a signal indicating the power status, usually used to indicate whether the power supply is working properly. When the power supply is stable and provides sufficient voltage and current, the power Good signal will be set to a high level or logic "1". If the power supply is abnormal, such as unstable voltage, insufficient current or power failure, the power Good signal will be set to a low level or logic "0". By detecting the state of the power Good signal, the system can determine whether the power supply is working properly, and then take corresponding control or protection measures.

[0044] A reset signal is a signal used to restore an electronic device or circuit to its initial state. When the reset signal is triggered, the device or circuit will perform a series of reset operations to restore the internal registers, logic circuits, etc. to a predetermined initial state. Reset signals are usually generated by external triggers, reset circuits, or specific reset chips, and can be used to initialize the system, clear error states, restore the device to an operational state, etc.

[0045] Acquiring the power Good signal and the reset signal within the predetermined time period means collecting and monitoring the two signals within a specific time period, so that the working status of the power supply and the reset status of the system can be understood, and accurate information can be provided for subsequent control and processing. By acquiring and analyzing these signals, the functions of reliable startup, abnormality handling and status monitoring of electronic equipment can be realized.

[0046] By obtaining the status of the power Good signal, it can be determined whether the power supply is working properly. If the power Good signal is high or logical "1", it means that the power supply is stable and provides sufficient voltage and current, and the power-on operation can be performed safely. If the power Good signal is low or logical "0", it means that the power supply is abnormal, such as unstable voltage, insufficient current or power failure. In this case, the system can determine that it is an abnormal power failure and take corresponding control measures, such as delaying the startup or performing fault processing.

[0047] By obtaining the trigger state of the reset signal, it can be determined whether the system is in the reset state. When the reset signal is triggered, the system is in the initial state, the internal registers, logic circuits, etc. are cleared or initialized, and the system is in a controllable state. Based on the trigger state of the reset signal, the power-on signal can be controlled. When the reset signal is released or the reset is completed, the system can perform the power-on operation according to the predetermined process and control logic, start the device and resume normal operation.

[0048] Acquiring the power Good signal and reset signal within a predetermined time period can provide key information for controlling the generation of the power-on signal. The power Good signal is used to determine the power status and eliminate abnormal power failures, while the reset signal is used to determine the system status and control the power-on operation, which ensures that the power-on signal is generated at the appropriate time and under the appropriate conditions, so that the system can start safely and reliably and operate normally.

[0049] In one embodiment of the present application, the signal feature analysis module 120 includes: a signal waveform feature extraction unit, used to perform feature extraction on the power Good signal and the reset signal respectively through a signal waveform feature extractor based on a deep neural network model to obtain a power Good signal waveform feature matrix and a reset signal waveform feature matrix; a waveform feature spatial manifestation unit, used to pass the power Good signal waveform feature matrix and the reset signal waveform feature matrix through a spatial attention module respectively to obtain a spatially manifested power Good signal waveform feature matrix as the power Good signal waveform feature and a spatially manifested reset signal waveform feature matrix as the reset signal waveform feature.

[0050] Among them, the signal waveform feature extractor based on the deep neural network model is a signal waveform feature extractor based on a two-dimensional convolutional layer.

[0051] Next, considering that the power Good signal and the reset signal are both waveforms in the time domain, a convolutional neural network model with excellent performance in implicit feature extraction of images is used to perform feature mining of the waveforms of the two signals. Specifically, the power Good signal and the reset signal are respectively subjected to feature mining in a signal waveform feature extractor based on a two-dimensional convolutional layer to extract the implicit feature distribution information of the waveform of the power Good signal and the waveform of the reset signal, respectively, thereby obtaining a power Good signal waveform feature matrix and a reset signal waveform feature matrix. These feature matrices can contain important information about the signal, such as peak value, duration, spectrum distribution, etc., which is helpful for subsequent analysis and abnormal drop point judgment.

[0052] Then, considering that when performing the interactive correlation analysis between the waveform features of the power Good signal and the reset signal, more attention should be paid to the falling edge change area and the 0-bit area of ​​the power Good signal, and the 0-bit area of ​​the reset signal should also be paid attention to. In view of the fact that the attention mechanism can select the focus position, produce a more distinguishable feature representation, and the features after adding the attention module will produce adaptive changes as the network deepens. Therefore, in the technical solution of the present application, in order to improve the accuracy of abnormal point drop detection, it is necessary to further enhance the spatial features of the power Good signal waveform feature matrix and the reset signal waveform feature matrix through the spatial attention module, so as to obtain the spatially manifested power Good signal waveform feature matrix and the spatially manifested reset signal waveform feature matrix. It is worth mentioning that here, the waveform features extracted by the spatial attention module reflect the weight of the spatial dimension feature difference, which is used to suppress or strengthen the features of different spatial positions, so as to focus on the spatially falling edge area features and 0-bit area features of the power Good signal and the 0-bit area feature information of the reset signal.

[0053] In one embodiment of the present application, the signal feature transfer association coding module 130 includes: a signal feature transfer association coding unit, which is used to calculate the transfer matrix of the spatially manifested power Good signal waveform feature matrix relative to the spatially manifested reset signal waveform feature matrix to obtain a power Good-reset signal transfer association feature matrix; a feature distribution optimization unit, which is used to perform soft matching of the rank arrangement distribution of the power Good-reset signal transfer association feature matrix using the feature scale as an imitation mask to obtain an optimized power Good-reset signal transfer association feature matrix; a transfer association feature full perception unit, which is used to pass the optimized power Good-reset signal transfer association feature matrix through a global perception module to obtain a power Good-reset signal global perception association feature vector as the power Good-reset signal transfer association feature.

[0054] Furthermore, the transfer matrix of the spatially manifested power Good signal waveform feature matrix relative to the spatially manifested reset signal waveform feature matrix is ​​calculated to obtain the power Good-reset signal transfer correlation feature matrix, so as to represent the transfer correlation feature information between the power Good signal waveform feature and the reset signal waveform feature focused on the spatial position. It should be understood that the transfer matrix can be used to describe the evolution and transfer process of a signal in time. In this case, by calculating the transfer matrix of the spatially manifested power Good signal waveform feature matrix relative to the spatially manifested reset signal waveform feature matrix, the transfer relationship between the power Good signal and the reset signal, such as the timing relationship, transfer mode and correlation information, can be obtained, so as to facilitate more accurate abnormal power failure detection in the subsequent period.

[0055] Next, in order to globally analyze and synthesize the entire power Good-reset signal transfer correlation feature matrix, so as to more accurately judge the falling edge change of the power Good signal and the 0-bit regional feature information of both signals, so as to determine whether there is an abnormal power failure, in the technical solution of the present application, it is also necessary to pass the power Good-reset signal transfer correlation feature matrix through a global perception module to obtain a power Good-reset signal global perception correlation feature vector. In this process, the global perception module will globally process and analyze the power Good-reset signal transfer correlation feature matrix to obtain a more comprehensive and advanced feature representation. In other words, by inputting the power Good-reset signal transfer correlation feature matrix into the global perception module, the power Good-reset signal global perception correlation feature vector containing the global information and advanced features in the transfer correlation feature matrix can be obtained, so that the correlation and importance feature information between the power Good signal and the reset signal can be better represented, which is conducive to more accurate detection and judgment of abnormal power failure.

[0056] In particular, in the technical solution of the present application, the spatially explicit power Good signal waveform feature matrix expresses the image semantic features of the local spatial distribution enhancement of the signal waveform graphs of the power Good signal and the reset signal relative to the spatially explicit reset signal waveform feature matrix. Therefore, when calculating the transfer matrix of the spatially explicit power Good signal waveform feature matrix relative to the spatially explicit reset signal waveform feature matrix, if the spatially explicit power Good signal waveform feature matrix relative to the spatially explicit reset signal waveform feature matrix expresses the image semantic features of the spatial attention as the foreground object features, then the transfer matrix will introduce background distribution noise while introducing the image semantic space domain transfer feature representation. Moreover, when calculating the high-rank distribution representation of the image semantic features of the transfer matrix, the image semantic spatial heterogeneous distribution of the high-dimensional features of the spatially manifested power Good signal waveform feature matrix relative to the spatially manifested reset signal waveform feature matrix will also cause an image semantic probability density mapping error of the power Good-reset signal transfer association feature matrix as the transfer matrix relative to the spatially manifested power Good signal waveform feature matrix relative to the spatially manifested reset signal waveform feature matrix, affecting the accuracy of the classification result obtained by the classifier of the power Good-reset signal global perception association feature vector obtained by the global perception module through the power Good-reset signal transfer association feature matrix.

[0057] Based on this, the applicant of the present application performs soft matching of the rank arrangement distribution of the power Good-reset signal transfer correlation feature matrix, for example, denoted as M, with the feature scale as the imitation mask, specifically expressed as: the power Good-reset signal transfer correlation feature matrix is ​​soft matched with the feature scale as the imitation mask rank arrangement distribution using the following optimization formula to obtain the optimized power Good-reset signal transfer correlation feature matrix; wherein the optimization formula is:

[0058]

[0059] Wherein, M is the power Good-reset signal transfer correlation characteristic matrix, m i，j is the eigenvalue of the (i, j)th position of the power Good-reset signal transfer correlation feature matrix, S is the scale of the power Good-reset signal transfer correlation feature matrix, represents the square of the Frobenius norm of the power Good-reset signal transfer correlation feature matrix, ||M||2 represents the bi-norm of the power Good-reset signal transfer correlation feature matrix, and α is a weighted hyperparameter, exp(·) represents an exponential operation, and m′ i，jis the eigenvalue of the (i, j)th position of the optimized power Good-reset signal transfer correlation characteristic matrix.

[0060] Here, the soft matching of the rank arrangement distribution of the feature scale as an imitation mask can focus on the foreground object features and ignore the background distribution noise when mapping the high-dimensional features to be regressed into the probability density space, and the distribution soft matching of the pyramid rank arrangement distribution is performed by different norms of the power Good-reset signal transfer associated feature matrix M, so as to effectively capture the correlation between the central area and the tail area of ​​the probability density distribution, and avoid the probability density mapping deviation caused by the heterogeneous distribution of the image semantic space of the high-dimensional features of the power Good-reset signal transfer associated feature matrix M, thereby improving the accuracy of the classification result obtained by the classifier of the global perception associated feature vector of the power Good-reset signal. In this way, it is possible to realize automatic power-on and startup after abnormal power failure of the system, so as to improve the reliability and stability of the electric energy meter, thereby ensuring that the electric energy meter can resume normal operation and protect the integrity of the data in the case of abnormal power failure.

[0061] In one embodiment of the present application, the power-on signal control module 140 includes: an abnormal power-off detection unit, used to pass the power Good-reset signal global perception association feature vector through a classifier to obtain a classification result, and the classification result is used to indicate whether it is an abnormal power-off; a power-on signal generation unit, used to control the generation of a power-on signal based on the classification result.

[0062] Among them, the abnormal power-off detection unit, the fully-connected encoding subunit, is used to use multiple fully-connected layers of the classifier to fully-connect the power Good-reset signal global perception association feature vector to obtain a coded classification feature vector; and the classification subunit is used to pass the coded classification feature vector through the Softmax classification function of the classifier to obtain the classification result.

[0063] Then, the power Good-reset signal global perception association feature vector is passed through a classifier to obtain a classification result, and the classification result is used to indicate whether it is an abnormal power failure. In other words, the transfer association full perception feature information between the power Good signal waveform feature and the reset signal waveform feature is used for classification processing, so as to effectively detect whether it is an abnormal power failure, and based on the classification result, control the generation of a power-on signal. In this way, the system can be automatically powered on after an abnormal power failure, which improves the reliability and stability of the electric energy meter, and ensures that the electric energy meter can resume normal operation and protect the integrity of the data in the event of an abnormal power failure.

[0064] By inputting the global perception correlation feature vector of the power Good-reset signal into the classifier, the abnormal power failure detection unit can accurately determine whether an abnormal power failure has occurred. This can help the system identify power failure events in a timely manner and provide an accurate judgment basis for subsequent processing. The existence of the abnormal power failure detection unit can improve the reliability of the system. It can detect abnormal power failures in a timely manner and trigger corresponding control measures, such as automatic power-on or data preservation, to protect the integrity of the device and data.

[0065] The power-on signal generation unit controls the generation of a power-on signal based on the classification result of the abnormal power-off detection unit and whether it is an abnormal power-off situation. This means that in the case of a normal power-off, the system can automatically perform a power-on operation, reducing the need for manual intervention and improving the automation of the equipment. By automatically generating a power-on signal, the power-on signal generation unit can speed up the recovery of the system. The system can restart quickly after an abnormal power-off, reducing downtime, improving production efficiency and user experience. The power-on signal generation unit can ensure that the system can perform a power-on operation according to a predetermined process after an abnormal power-off, thereby ensuring the stability of the system, and can control the power-on timing and related initialization operations to ensure that the device starts in a reliable state.

[0066] In summary, the power-on and power-off self-reset control circuit 100 based on the embodiment of the present invention is explained, and the power Good signal and the reset signal are correlated and analyzed through signal processing and analysis algorithms to determine whether there is a 0 bit in the falling edge change data of the power Good signal. If so, it is determined whether the reset signal data is 0, so as to determine whether it is an abnormal power off. In this way, the system can automatically power on after an abnormal power off, thereby improving the reliability and stability of the electric energy meter, ensuring that the electric energy meter can resume normal operation and protect the integrity of the data in the event of an abnormal power off.

[0067] In one embodiment of the present application, a detection mechanism (method) for powering on and off an electric energy meter is provided, and its design technology and solution are as follows: Figure 3 The voltage detection circuit diagram of the electric energy meter designed for power on and off is shown in FIG. Figure 4 shown.

[0068] Meeting the MCU's power-on timing is one of the most basic and important requirements for system design. Therefore, it is very necessary to carefully study the chip's power-on and power-off timing.

[0069] Power-on part:

[0070] The main requirements of MCU for power-on are:

[0071] 1) The power-on time generally cannot exceed 500ms;

[0072] 2) The voltage before power-on generally needs to be lower than 200mV for at least 12us.

[0073] The above requirements can be found in the MCU's User Manual.

[0074] This requires that the power-on time be shortened as much as possible, especially when there are large capacitors or supercapacitors in the circuit. Too long power-on time can easily lead to system failure to start or device latch-up problems.

[0075] Shorten the power-on time and control the CE pin of the power chip (voltage detection chip). Cleverly using the voltage divider resistor of the CE pin can effectively shorten the power-on time of the system. When using the power chip, it is generally enabled by default with an external pull-up. If the enable threshold is reached too early, the output will follow the input, that is, the slower the input, the slower the output, and the jitter at the input when powered on will also be transmitted to the output.

[0076] By using voltage divider resistors properly, when VIN rises to 70% to 80%, the voltage of CE reaches the enable threshold. At this time, the output rising edge is steep and the output is stable, eliminating the unstable stage of the input power supply and reducing the impact of input voltage fluctuations. At the same time, a 20% to 30% margin is reserved to avoid output shutdown due to power supply fluctuations.

[0077] Power off part:

[0078] When the system fails to start, you should first use an oscilloscope to check whether the device's power supply pins are powered on slowly or not completely. When the power is turned on and off quickly, whether the system is powered off "cleanly" is also closely related to the system startup.

[0079] Therefore, after the power is turned off, the MCU power supply is quickly reduced to nearly 0V, so that the system reaches a certain state in a short time. This is also a key factor for the system to start normally when the power is quickly restarted.

[0080] When the input voltage is powered off, if the CE voltage is lower than the enable threshold, the fast discharge circuit inside the voltage detection chip will be started, and a 240Ω discharge resistor R1 will be loaded at the output end to quickly power off the output voltage. At this time, the output voltage of the voltage detection chip, that is, the input power supply of the MCU, can quickly "return to zero" to avoid system startup failure when the power is quickly turned on again.

[0081] In this application, the beneficial effects produced include:

[0082] 1. Effectively avoid the phenomenon of MCU crash of electric energy meter caused by frequent power failure, power on or power on during the test of closing and closing the switch;

[0083] 2. Ensure the normal operation of the MCU of the energy meter; ensure the correct and reliable storage of the energy meter measurement and its data;

[0084] 3. Effectively improve the stability, reliability and operation quality level of the electric energy meter, thereby ensuring the quality of the electric energy meter or improving and ensuring the service life requirements.

[0085] Figure 5 The present invention provides a flow chart of a method for controlling power-on and power-off self-reset in an embodiment of the present invention. Figure 6 Schematic diagram of the system architecture of a power-on / power-off self-reset control method provided in an embodiment of the present invention. Figure 5 and Figure 6 As shown, a control method for power-on and power-off self-reset includes: 210, obtaining a power Good signal and a reset signal within the predetermined time period; 220, performing feature analysis on the power Good signal and the reset signal to obtain a power Good signal waveform feature and a reset signal waveform feature; 230, performing transfer association coding on the power Good signal waveform feature and the reset signal waveform feature to obtain a power Good-reset signal transfer association feature; 240, controlling the generation of a power-on signal based on the power Good-reset signal transfer association feature.

[0086] In the power-on and power-off self-reset control method, feature analysis is performed on the power Good signal and the reset signal to obtain power Good signal waveform features and reset signal waveform features, including: feature extraction is performed on the power Good signal and the reset signal respectively through a signal waveform feature extractor based on a deep neural network model to obtain a power Good signal waveform feature matrix and a reset signal waveform feature matrix; the power Good signal waveform feature matrix and the reset signal waveform feature matrix are respectively passed through a spatial attention module to obtain a spatially explicit power Good signal waveform feature matrix as the power Good signal waveform feature and a spatially explicit reset signal waveform feature matrix as the reset signal waveform feature.

[0087] In the power-on / power-off self-reset control method, the signal waveform feature extractor based on the deep neural network model is a signal waveform feature extractor based on a two-dimensional convolutional layer.

[0088] Those skilled in the art will appreciate that the specific operations of each step in the above power-on / power-off self-reset control method have been described in detail above. Figures 1 to 6 The power-on / power-off self-reset control circuit has been described in detail, and therefore, its repeated description will be omitted.

[0089] Figure 7FIG. 1 is an application scenario diagram of a power-on / power-off self-reset control circuit provided in an embodiment of the present invention. Figure 7 As shown, in this application scenario, first, a power Good signal (for example, Figure 7 C1) and a reset signal (e.g., Figure 7 Then, the obtained power Good signal and reset signal are input to a server (for example, Figure 7 In the S) shown in , the server can process the power Good signal and the reset signal based on the control circuit algorithm of power-on and power-off self-reset to control the generation of a power-on signal.

[0090] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A power-on / power-off self-reset control circuit, characterized in that: include: A signal acquisition module is used to obtain a power Good signal and a reset signal within a predetermined time period; A signal characteristic analysis module, used for performing characteristic analysis on the power Good signal and the reset signal to obtain a power Good signal waveform characteristic and a reset signal waveform characteristic; A signal feature transfer correlation coding module, used for performing transfer correlation coding on the power Good signal waveform feature and the reset signal waveform feature to obtain a power Good-reset signal transfer correlation feature; A power-on signal control module, used for controlling the generation of a power-on signal based on the power Good-reset signal transfer correlation feature; Wherein, the signal feature analysis module includes: A signal waveform feature extraction unit, used to extract features of the power Good signal and the reset signal respectively through a signal waveform feature extractor based on a deep neural network model to obtain a power Good signal waveform feature matrix and a reset signal waveform feature matrix; A waveform feature spatial manifestation unit, used for respectively passing the power Good signal waveform feature matrix and the reset signal waveform feature matrix through a spatial attention module to obtain a spatially manifested power Good signal waveform feature matrix as the power Good signal waveform feature and a spatially manifested reset signal waveform feature matrix as the reset signal waveform feature; The signal feature transfer association coding module includes: A signal feature transfer correlation coding unit, used to calculate the transfer matrix of the spatially manifested power Good signal waveform feature matrix relative to the spatially manifested reset signal waveform feature matrix to obtain a power Good-reset signal transfer correlation feature matrix; A feature distribution optimization unit, used for performing soft matching of the rank arrangement distribution of the power Good-reset signal transfer correlation feature matrix with the feature scale as the imitation mask to obtain an optimized power Good-reset signal transfer correlation feature matrix; A transfer correlation feature full perception unit, used for passing the optimized power Good-reset signal transfer correlation feature matrix through a global perception module to obtain a power Good-reset signal global perception correlation feature vector as the power Good-reset signal transfer correlation feature; The feature distribution optimization unit is used to: perform soft matching of the rank arrangement distribution of the power Good-reset signal transfer correlation feature matrix with the feature scale as the imitation mask using the following optimization formula to obtain the optimized power Good-reset signal transfer correlation feature matrix; Wherein, the optimization formula is: Wherein, M is the power Good-reset signal transfer correlation characteristic matrix, m i,j is the eigenvalue of the (i, j)th position of the power Good-reset signal transfer correlation feature matrix, S is the scale of the power Good-reset signal transfer correlation feature matrix, represents the square of the Frobenius norm of the power Good-reset signal transfer correlation feature matrix, ||M||2 represents the bi-norm of the power Good-reset signal transfer correlation feature matrix, and α is a weighted hyperparameter, exp(·) represents an exponential operation, and m' i,j is the eigenvalue of the (i, j)th position of the optimized power Good-reset signal transfer correlation characteristic matrix.

2. The power-on / power-off self-reset control circuit according to claim 1, characterized in that: The signal waveform feature extractor based on the deep neural network model is a signal waveform feature extractor based on a two-dimensional convolutional layer.

3. The power-on / power-off self-reset control circuit according to claim 2, characterized in that: The power-on signal control module comprises: An abnormal power failure detection unit, used for passing the power Good-reset signal global perception association feature vector through a classifier to obtain a classification result, wherein the classification result is used to indicate whether it is an abnormal power failure; A power-on signal generating unit is used to control the generation of a power-on signal based on the classification result.

4. The power-on / power-off self-reset control circuit according to claim 3, characterized in that: The abnormal power failure detection unit, A fully connected encoding subunit, used for performing fully connected encoding on the power Good-reset signal global perception association feature vector using multiple fully connected layers of the classifier to obtain an encoded classification feature vector; as well as The classification subunit is used to pass the encoded classification feature vector through the Softmax classification function of the classifier to obtain the classification result.

5. A control method for power-on and power-off self-reset, characterized in that: include: Obtain power Good signal and reset signal within a predetermined time period; Performing feature analysis on the power Good signal and the reset signal to obtain a power Good signal waveform feature and a reset signal waveform feature; Performing transfer association coding on the power Good signal waveform feature and the reset signal waveform feature to obtain a power Good-reset signal transfer association feature; Based on the power Good-reset signal transfer correlation feature, controlling the generation of a power-on signal; The feature analysis of the power Good signal and the reset signal to obtain the power Good signal waveform feature and the reset signal waveform feature includes: The power Good signal and the reset signal are respectively subjected to feature extraction by a signal waveform feature extractor based on a deep neural network model to obtain a power Good signal waveform feature matrix and a reset signal waveform feature matrix; The power Good signal waveform feature matrix and the reset signal waveform feature matrix are respectively passed through a spatial attention module to obtain a spatially manifested power Good signal waveform feature matrix as the power Good signal waveform feature and a spatially manifested reset signal waveform feature matrix as the reset signal waveform feature; The power Good signal waveform feature and the reset signal waveform feature are transferred and associated encoded to obtain a power Good-reset signal transfer association feature, including: Calculating the transfer matrix of the spatially manifested power Good signal waveform characteristic matrix relative to the spatially manifested reset signal waveform characteristic matrix to obtain a power Good-reset signal transfer correlation characteristic matrix; Performing soft matching of the power Good-reset signal transfer correlation feature matrix with the feature scale as the rank arrangement distribution of the imitation mask to obtain an optimized power Good-reset signal transfer correlation feature matrix; The optimized power Good-reset signal transfer correlation feature matrix is ​​passed through a global perception module to obtain a power Good-reset signal global perception correlation feature vector as the power Good-reset signal transfer correlation feature; Among them, the power Good-reset signal transfer correlation feature matrix is ​​softly matched with the rank arrangement distribution of the feature scale as the imitation mask to obtain the optimized power Good-reset signal transfer correlation feature matrix, including: using the following optimization formula to softly match the power Good-reset signal transfer correlation feature matrix with the feature scale as the rank arrangement distribution of the imitation mask to obtain the optimized power Good-reset signal transfer correlation feature matrix; Wherein, the optimization formula is: Wherein, M is the power Good-reset signal transfer correlation characteristic matrix, m i,j is the eigenvalue of the (i, j)th position of the power Good-reset signal transfer correlation feature matrix, S is the scale of the power Good-reset signal transfer correlation feature matrix, represents the square of the Frobenius norm of the power Good-reset signal transfer correlation feature matrix, ||M||2 represents the bi-norm of the power Good-reset signal transfer correlation feature matrix, and α is a weighted hyperparameter, exp(·) represents an exponential operation, and m' i,j is the eigenvalue of the (i, j)th position of the optimized power Good-reset signal transfer correlation characteristic matrix.

6. The power-on / power-off self-reset control method according to claim 5, characterized in that: The signal waveform feature extractor based on the deep neural network model is a signal waveform feature extractor based on a two-dimensional convolutional layer.

Images