CN117543803A - Dual-power on-line standby circuit and control method thereof - Google Patents
Dual-power on-line standby circuit and control method thereof Download PDFInfo
- Publication number
- CN117543803A CN117543803A CN202410032882.3A CN202410032882A CN117543803A CN 117543803 A CN117543803 A CN 117543803A CN 202410032882 A CN202410032882 A CN 202410032882A CN 117543803 A CN117543803 A CN 117543803A
- Authority
- CN
- China
- Prior art keywords
- power supply
- control module
- power
- running state
- data
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 230000005856 abnormality Effects 0.000 claims description 75
- 238000004458 analytical method Methods 0.000 claims description 69
- 239000013598 vector Substances 0.000 claims description 64
- 239000002245 particle Substances 0.000 claims description 52
- 238000012544 monitoring process Methods 0.000 claims description 37
- 238000004422 calculation algorithm Methods 0.000 claims description 19
- 238000012545 processing Methods 0.000 claims description 18
- 230000002159 abnormal effect Effects 0.000 claims description 17
- 230000009977 dual effect Effects 0.000 claims description 17
- 238000000605 extraction Methods 0.000 claims description 13
- 238000004140 cleaning Methods 0.000 claims description 11
- 230000006870 function Effects 0.000 claims description 11
- 238000006243 chemical reaction Methods 0.000 claims description 10
- 238000010606 normalization Methods 0.000 claims description 10
- 230000004927 fusion Effects 0.000 claims description 9
- 238000013507 mapping Methods 0.000 claims description 8
- 230000002776 aggregation Effects 0.000 claims description 4
- 238000004220 aggregation Methods 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 9
- 230000002265 prevention Effects 0.000 abstract 1
- 230000008569 process Effects 0.000 description 16
- 238000004364 calculation method Methods 0.000 description 7
- 230000008859 change Effects 0.000 description 5
- 238000001514 detection method Methods 0.000 description 5
- 238000011217 control strategy Methods 0.000 description 4
- 238000007726 management method Methods 0.000 description 4
- 238000005457 optimization Methods 0.000 description 4
- 230000006378 damage Effects 0.000 description 3
- 238000007405 data analysis Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000003993 interaction Effects 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 241000288105 Grus Species 0.000 description 2
- 230000004913 activation Effects 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000011084 recovery Methods 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 102100028680 Protein patched homolog 1 Human genes 0.000 description 1
- 101710161390 Protein patched homolog 1 Proteins 0.000 description 1
- 102100036894 Protein patched homolog 2 Human genes 0.000 description 1
- 101710161395 Protein patched homolog 2 Proteins 0.000 description 1
- 230000003213 activating effect Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 238000011022 operating instruction Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008380 pattern binding Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000010223 real-time analysis Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 230000011273 social behavior Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J9/00—Circuit arrangements for emergency or stand-by power supply, e.g. for emergency lighting
- H02J9/04—Circuit arrangements for emergency or stand-by power supply, e.g. for emergency lighting in which the distribution system is disconnected from the normal source and connected to a standby source
- H02J9/06—Circuit arrangements for emergency or stand-by power supply, e.g. for emergency lighting in which the distribution system is disconnected from the normal source and connected to a standby source with automatic change-over, e.g. UPS systems
- H02J9/061—Circuit arrangements for emergency or stand-by power supply, e.g. for emergency lighting in which the distribution system is disconnected from the normal source and connected to a standby source with automatic change-over, e.g. UPS systems for DC powered loads
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H9/00—Emergency protective circuit arrangements for limiting excess current or voltage without disconnection
- H02H9/02—Emergency protective circuit arrangements for limiting excess current or voltage without disconnection responsive to excess current
- H02H9/026—Current limitation using PTC resistors, i.e. resistors with a large positive temperature coefficient
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
- H02J7/0029—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries with safety or protection devices or circuits
- H02J7/00304—Overcurrent protection
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
- H02J7/0047—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries with monitoring or indicating devices or circuits
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
- H02J7/34—Parallel operation in networks using both storage and other dc sources, e.g. providing buffering
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
Abstract
The application relates to the field of power supply control and discloses a dual-power supply online standby circuit and a control method thereof. The circuit comprises: the power supply control system comprises a first power supply control module and a second power supply control module, wherein the first power supply control module is connected with the second power supply control module in parallel, the output end of a first power supply in the first power supply control module is connected with the input end of a first diode, the output end of the first diode is connected with the input end of a first thermistor, the output end of the first thermistor is connected with the VCC end of a preset target circuit, and the input end of the first power supply is connected with the GND end of the target circuit; the output of second power among the second power control module is connected with the input of second diode, and the output of second diode is connected with the input of second thermistor, and the output of second thermistor is connected with the VCC end of target circuit, and the input of second power is connected with the GND end of target circuit, and this application has improved power supply system prevention power failure effect.
Description
Technical Field
The application relates to the technical field of power supply control, in particular to a dual-power supply online standby circuit and a control method thereof.
Background
In a data storage center, a monitoring control system, an industrial control automatic control system, an automobile function safety/intelligent driving system and other application scenes, sudden power failure can generate great harm to the whole system, even generate application risks, the first power failure in the data storage center or the monitoring system can possibly generate data loss, in the industrial control system (the automatic control system), under the condition of power failure of the system, control system stagnation, procedure disorder and the like can possibly generate, so that unnecessary economic loss is caused; particularly in an intelligent safe driving system, even short power failure can cause driving risks and personal safety hazards. Therefore, strict requirements are applied to the system standby power in many application scenes, so that the control system can keep stable and safe operation under the condition that the first power supply is powered down or loses power. At present, the dual power mode of the first power supply and the auxiliary control power supply is the most application solutions, but the control modes of the dual power supplies are different, and the application effects of preventing power failure/power failure are different.
As shown in fig. 1, the prior art includes: 1. the first power supply is used as a first power supply, the switch 1 is used for access control, and the fuse 1 provides primary overcurrent/overload protection; 2. the second power supply is used as an auxiliary control power supply, when the first power supply works abnormally, the first power supply is connected into a main circuit through the control switch 2 to supply power, and the fuse 2 also provides primary overcurrent/overload protection; 3. in the whole control system, the first power supply needs to be monitored and monitored to a certain extent, and the switch 1 is controlled to be switched on and off according to different conditions, so that the first power supply is switched on and off; and the second power supply also performs switching control on the switch 2 according to the working operation condition of the system on the first power supply, so that the access control of the second power supply is realized under the condition that the first power supply exits from power supply. Based on the technical scheme, the method can be as follows: 1. the switch 1 and the switch 2 are used as active control switches, a central processing system is required to control the control switches, and the switching control of the second power supply is generally required to be completed before power failure or power failure, otherwise, the switching is possibly not timely, so that power failure is caused; or requires an additional control system (independent of the first/second power supply) to control switch 1/switch 2 separately; 2. in the system, the fuse 1/the fuse 2 are used as unrecoverable fuses, the self-recovery function is not provided, once the fuse fails, the first power supply/the second power supply permanently lose the opportunity of accessing the circuit to supply power, the system maintenance is needed, and the normal operation of the system can be kept through the power supply of the power supply module outside the equipment; in this system, it is generally ensured that one of the first power source and the second power source is operated, and the switch 1/2 must be maintained in a controlled state of being opened and closed. If the control failure of the switch 1/the switch 2 is connected into the power supply system, a power short circuit between the second power supplies is formed, the system heating and the power consumption are increased, and potential safety hazards can be generated.
Disclosure of Invention
The application provides a dual-power online standby circuit and a control method thereof, which are used for improving the effect of preventing power failure of a power supply system by adopting dual-power online standby.
In a first aspect, the present application provides a dual power on-line standby circuit, the dual power on-line standby circuit comprising:
the power supply control system comprises a first power supply control module and a second power supply control module, wherein the first power supply control module is connected with the second power supply control module in parallel, the first power supply control module comprises a first power supply, a first diode and a first thermistor, the output end of the first power supply is connected with the input end of the first diode, the output end of the first diode is connected with the input end of the first thermistor, the output end of the first thermistor is connected with the VCC end of a preset target circuit, and the input end of the first power supply is connected with the GND end of the target circuit; the second power supply control module comprises a second power supply, a second diode and a second thermistor, wherein the output end of the second power supply is connected with the input end of the second diode, the output end of the second diode is connected with the input end of the second thermistor, the output end of the second thermistor is connected with the VCC end of the target circuit, and the input end of the second power supply is connected with the GND end of the target circuit;
Acquiring power difference information between the first power supply and the second power supply, and continuously supplying power to the target circuit without interruption according to the power difference information; the first diode and the second diode are used for preventing short circuit between the first power supply and the second power supply; the first thermistor and the second thermistor are used for dynamically overload or overcurrent protection of the target circuit; the first thermistor or the second thermistor is specifically configured to: when the current in the target circuit is excessive, the resistance of the first thermistor or the second thermistor is controlled to be increased so as to reduce the overload current in the target circuit.
In a second aspect, the present application provides a dual-power online standby electric control method, where the dual-power online standby electric control method includes:
the method comprises the steps of performing state monitoring on a first power supply control module to obtain first running state data of the first power supply control module, and performing state monitoring on a second power supply control module to obtain second running state data of the second power supply control module;
extracting the running state characteristics of the first running state data through a preset first self-encoder to obtain a first running state characteristic set, and extracting the running state characteristics of the second running state data through a preset second self-encoder to obtain a second running state characteristic set;
Inputting the first running state characteristic set into a preset power supply abnormality analysis model to perform power supply abnormality analysis to obtain a first abnormality analysis result of the first power supply control module, and inputting the second running state characteristic set into the power supply abnormality analysis model to perform power supply abnormality analysis to obtain a second abnormality analysis result of the second power supply control module;
comprehensively analyzing the first abnormal analysis result and the second abnormal analysis result to obtain a comprehensive abnormal analysis result;
according to the comprehensive anomaly analysis result, load demand data of the dual-power supply online standby circuit are obtained;
inputting the load demand data into a preset particle swarm algorithm for control parameter analysis, and generating a corresponding first control parameter strategy;
performing standby power switching control and real-time state monitoring on the dual-power online standby circuit according to the first control parameter strategy to obtain real-time state monitoring data;
and carrying out strategy adjustment on the first control parameter strategy according to the real-time state monitoring data to obtain a second control parameter strategy.
In the technical scheme provided by the application, the invention comprises the following steps: the power supply control system comprises a first power supply control module and a second power supply control module, wherein the first power supply control module is connected with the second power supply control module in parallel, the output end of a first power supply in the first power supply control module is connected with the input end of a first diode, the output end of the first diode is connected with the input end of a first thermistor, the output end of the first thermistor is connected with the VCC end of a preset target circuit, and the input end of the first power supply is connected with the GND end of the target circuit; the output end of the second power supply in the second power supply control module is connected with the input end of the second diode, the output end of the second diode is connected with the input end of the second thermistor, the output end of the second thermistor is connected with the VCC end of the target circuit, and the input end of the second power supply is connected with the GND end of the target circuit. The invention adopts the self-adaptive mode to automatically realize the application of mutual backup of the dual power supplies, does not need additional control and detection, and has simple application; compared with the prior art, the dual power supply is always in an on-line working state, the single power supply is low in configuration, the overall power supply capacity is high, the load bearing capacity is high, the power supply application efficiency is high, and the cost is low; the PTC application technology is adopted to realize the restorable overcurrent overload protection capability; the dual-power on-line standby power has obvious, reliable and efficient effect of preventing power failure of the system power supply system.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained based on these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a circuit diagram of a dual power backup in an embodiment of the present application;
FIG. 2 is a schematic diagram of an embodiment of a dual power on-line standby circuit according to an embodiment of the present application;
fig. 3 is a schematic diagram of an embodiment of a dual power on-line standby control method in an embodiment of the present application.
Detailed Description
The embodiment of the application provides a dual-power on-line standby circuit and a control method thereof. The terms "first," "second," "third," "fourth" and the like in the description and in the claims of this application and in the above-described figures, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, circuit, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, circuit, article, or apparatus.
For ease of understanding, a specific flow of an embodiment of the present application is described below, referring to fig. 2, where an embodiment of a dual power on-line standby circuit in an embodiment of the present application includes:
the power supply control system comprises a first power supply control module and a second power supply control module, wherein the first power supply control module is connected with the second power supply control module in parallel, the first power supply control module comprises a first power supply, a first diode and a first thermistor, the output end of the first power supply is connected with the input end of the first diode, the output end of the first diode is connected with the input end of the first thermistor, the output end of the first thermistor is connected with the VCC end of a preset target circuit, and the input end of the first power supply is connected with the GND end of the target circuit; the second power supply control module comprises a second power supply, a second diode and a second thermistor, wherein the output end of the second power supply is connected with the input end of the second diode, the output end of the second diode is connected with the input end of the second thermistor, the output end of the second thermistor is connected with the VCC end of the target circuit, and the input end of the second power supply is connected with the GND end of the target circuit;
Acquiring power difference information between the first power supply and the second power supply, and continuously supplying power to the target circuit without interruption according to the power difference information; the first diode and the second diode are used for preventing short circuit between the first power supply and the second power supply; the first thermistor and the second thermistor are used for dynamically overload or overcurrent protection of the target circuit; the first thermistor or the second thermistor is specifically configured to: when the current in the target circuit is excessive, the resistance of the first thermistor or the second thermistor is controlled to be increased so as to reduce the overload current in the target circuit.
Specifically, as shown in fig. 2, 1. A first power supply and an auxiliary control power supply are directly connected into a system through a Diode1/Diode2 Diode, and continuous power supply is performed in a self-adaptive manner through respective power supply differences of a second power supply; the diode has unidirectional conduction capability, so that internal consumption between the second power supplies is prevented, and the possibility of short circuit between the double power supplies is avoided; 2. the dual power supply online standby power makes the whole system not need to carry out additional detection and switch control on the access control of the second power supply, and has the advantages of simple application, convenient control, low cost, high efficiency and good effect. 3. The PTC1/PTC2 thermistor is used for dynamic overload/overcurrent protection; when the current is overlarge, the resistance of the thermistor is increased, the overload current is automatically reduced, the thermistor is usually used as a recoverable fuse, and after overload or overcurrent is relieved, the thermistor can be automatically recovered to a normal working state, so that the thermistor is safe to use and easy to maintain; 4. because the dual power supplies of the system are simultaneously connected into the system, the capacity of coping with load fluctuation in the system is strong, the power supply capacity of the power supply is enhanced, the specification configuration of the dual power supplies can be reduced from the traditional application, and the cost is further reduced; the dual-power on-line standby circuit can ensure that the control system is powered normally as long as one of the two power supplies is powered normally, reduces the control risks in other aspects, further reduces the probability of power failure/power failure, has obvious standby effect and is safe and reliable to apply.
In an embodiment of the present application, the present invention includes: the power supply control system comprises a first power supply control module and a second power supply control module, wherein the first power supply control module is connected with the second power supply control module in parallel, the output end of a first power supply in the first power supply control module is connected with the input end of a first diode, the output end of the first diode is connected with the input end of a first thermistor, the output end of the first thermistor is connected with the VCC end of a preset target circuit, and the input end of the first power supply is connected with the GND end of the target circuit; the output end of the second power supply in the second power supply control module is connected with the input end of the second diode, the output end of the second diode is connected with the input end of the second thermistor, the output end of the second thermistor is connected with the VCC end of the target circuit, and the input end of the second power supply is connected with the GND end of the target circuit. The invention adopts the self-adaptive mode to automatically realize the application of mutual backup of the dual power supplies, does not need additional control and detection, and has simple application; compared with the prior art, the dual power supply is always in an on-line working state, the single power supply is low in configuration, the overall power supply capacity is high, the load bearing capacity is high, the power supply application efficiency is high, and the cost is low; the PTC application technology is adopted to realize the restorable overcurrent overload protection capability; the dual-power on-line standby power has obvious, reliable and efficient effect of preventing power failure of the system power supply system.
Referring to fig. 3, an embodiment of a dual-power online standby electric control method in an embodiment of the present application includes:
step S101, performing state monitoring on a first power supply control module to obtain first running state data of the first power supply control module, and performing state monitoring on a second power supply control module to obtain second running state data of the second power supply control module;
specifically, first, the first power control module is monitored in real time, and first multidimensional state data are collected through sensors installed on the first power supply, the first diode and the first thermistor, and the data reflect working states of various components, such as voltage, current and temperature of the first power supply, conduction state of the first diode and resistance change of the first thermistor. And meanwhile, the same monitoring operation is carried out on the second power supply control module, so that the working state data of the second power supply, the second diode and the second thermistor are obtained, and the overall monitoring of the whole standby power system is ensured. And then, performing data cleaning on the collected first multi-dimensional state data and second multi-dimensional state data, including operations of removing noise, correcting deviation, filling missing values and the like, so as to ensure the accuracy and reliability of the data. After data cleaning, data is subjected to standardized processing, and data of different sources and different dimensions are converted into a unified format which can be compared and analyzed, so that subsequent data analysis and decision making are facilitated. And after data cleaning and standardization processing, obtaining first running state data of the first power supply control module and second running state data of the second power supply control module.
Step S102, extracting the running state characteristics of the first running state data through a preset first self-encoder to obtain a first running state characteristic set, and extracting the running state characteristics of the second running state data through a preset second self-encoder to obtain a second running state characteristic set;
specifically, first, the running state data of the first power supply control module is processed through a preset first self-encoder, wherein the self-encoder comprises a forward long-short-time memory network and a backward long-short-time memory network. The forward LSTM focuses on extracting the time dependence of past and current data points, inputting first running state data in time sequence, extracting key forward time sequence features affecting the current state through calculation of a plurality of hidden layers, and encoding the features into a high-dimensional first forward time sequence state feature vector. Then, the backward LSTM is responsible for extracting the time sequence information from the future to the past, and inputs the same running state data into the network in reverse time sequence, so as to capture the potential influence of the future state on the current state. After the first forward and backward time sequence state feature vectors are obtained, the two vectors are combined through a feature fusion technology, and finally a first running state feature set integrating time information is obtained through a series of nonlinear mapping and transformation, wherein the first running state feature set contains key features extracted from current and historical data, and a comprehensive and deep view is provided for subsequent state monitoring and abnormality detection. And processing the running state data of the second power supply control module through a preset second self-encoder. The second self-encoder also includes forward and backward LSTM networks for respectively performing forward and backward time sequence feature extraction on the second operation state data to form a second forward time sequence state feature vector and a second backward time sequence state feature vector. And then, carrying out feature fusion and integrated mapping on the two vectors, and finally converting the two vectors into a second running state feature set, wherein the second running state feature set also contains rich time sequence feature information, and providing data support for ensuring stable running and timely responding to potential problems of the power supply system.
Step S103, inputting the first running state characteristic set into a preset power supply abnormality analysis model to perform power supply abnormality analysis to obtain a first abnormality analysis result of the first power supply control module, and inputting the second running state characteristic set into the power supply abnormality analysis model to perform power supply abnormality analysis to obtain a second abnormality analysis result of the second power supply control module;
specifically, first, feature encoding and feature normalization are performed on a first running state feature set and a second running state feature set to obtain a plurality of first normalized running state features and a plurality of second normalized running state features. Feature encoding is the conversion of original features into a format that can be understood by the model, while feature normalization is the conversion of data of different dimensions into the same dimensions, which helps to improve the accuracy and convergence speed of the model. After data preprocessing is completed, vector conversion is carried out on the normalized running state feature, and a plurality of features are combined into a high-dimensional feature vector to obtain a first running state feature vector and a second running state feature vector. Next, the feature vectors are respectively input to a preset power abnormality analysis model. The model is a deep learning structure comprising two layers of threshold loop networks (Gated Recurrent Unit, GRU) and fully connected layers. The two layers of GRUs are used for capturing time dependence and nonlinear relations in feature vectors, hidden state features are extracted through the structure of a circulation network, and the features can reflect deep operation states of a power supply system. The GRU network is good at processing time sequence data, can effectively solve the problem of long-term dependence, and is suitable for analysis of a power supply system which is a dynamic system changing with time. After each run state feature vector passes through the two layers of GRUs, a series of hidden features are obtained, which are then input to the fully connected layer. In the fully connected layer, the hidden features are subjected to nonlinear transformation by using a ReLU activation function, and key information for power supply abnormality prediction is further extracted and enhanced. And finally obtaining the abnormal probability predicted values of the first power supply and the second power supply through the processing of the full connection layer. Finally, these predictions are compared with a preset anomaly probability threshold. If the predicted value exceeds the threshold value, the power control module is indicated that an abnormality may exist, and a first abnormality analysis result or a second abnormality analysis result is generated.
Step S104, comprehensively analyzing the first abnormal analysis result and the second abnormal analysis result to obtain a comprehensive abnormal analysis result;
specifically, first, key abnormality indexes and parameters such as the frequency, duration, influence range, severity and the like of occurrence of abnormality are extracted. Then, weight distribution is performed on different abnormality indexes according to preset rules, so that more important or dangerous abnormalities can be ensured to be sufficiently focused. For example, the system may be given a higher weight for those serious anomalies that may cause a system shutdown or safety accident. Then, based on these weighted abnormality indexes, a comprehensive logic judgment is made to analyze the correlation and interaction between the abnormality results of the first power supply and the second power supply, and judge whether they belong to independent events or whether there is some inherent correlation or interaction. And then, integrating all information and judgment by using a comprehensive evaluation algorithm to obtain a comprehensive abnormality analysis result. This result reflects not only the abnormal condition of a single power supply, but also the interaction between the two power supplies and the overall health of the overall system. The composite anomaly analysis results may include an overall assessment of the current state of the system, a prediction of future potential risk, and suggested response measures.
Step 105, obtaining load demand data of the dual-power online standby circuit according to the comprehensive abnormal analysis result;
specifically, first, the actual influence of the abnormal analysis result on the dual-power on-line standby circuit is analyzed, including the influence on the power output stability, the influence on the standby switching logic and the influence on the overall power management strategy. For example, if the analysis results show that an overload anomaly is frequently occurring in a certain power module, the system may predict that the load demand of that module will increase in a certain period of time in the future, thus requiring additional power to handle the overload condition and maintain normal operation. Then, corresponding data acquisition strategies are formulated based on these analyses and predictions. This may include increasing the frequency of monitoring critical components, adjusting data acquisition parameters to better capture load changes, or initiating more advanced data analysis modes to track and predict load demand in real time. These strategies may be implemented by automated algorithms that can dynamically adjust the manner and frequency of data acquisition based on changes in system state and external environment. Meanwhile, the system needs to be closely cooperated with other management and control modules of the standby circuit, so that all relevant information and decisions can be timely transmitted and executed. For example, if a large load fluctuation is predicted for a short period of time, the system informs the power scheduling module in advance so that it can adjust the power distribution strategy to ensure that there is sufficient backup capacity to cope with the possible load increase.
Step S106, inputting load demand data into a preset particle swarm algorithm to perform control parameter analysis, and generating a corresponding first control parameter strategy;
specifically, first, a plurality of control parameter adjustment ranges of the first power supply control module and the second power supply control module are defined, and the ranges reflect possible maximum values and minimum values of the respective control parameters and possible change steps or change rules. Therefore, the adjustment of the control parameters is ensured to be carried out in a safe and effective interval, and unstable or low efficiency of the system caused by incorrect parameter setting is prevented. And then, carrying out control parameter random initialization on the first power supply control module or the second power supply control module according to the load demand data, namely randomly selecting an initial value in the adjustment range of each parameter, thereby obtaining a corresponding initialization control parameter set. This set represents one random sample of possible control parameter configurations, which is the starting point for the particle swarm algorithm to initially explore the solution space. Then, particle population construction is carried out on the initialized control parameter set through a preset particle swarm algorithm, each control parameter configuration is regarded as a particle, and all particles form a population. In the particle swarm algorithm, each particle represents a potential solution that will move and search in the solution space, guiding the search direction together through individual experience and population experience to find the optimal solution. Next, particle fitness calculation is performed on the particle population, i.e., the performance or quality of each particle (control parameter configuration) is evaluated. The fitness function is usually designed according to the actual requirement of the system, and may include multiple indexes such as stability, response speed, energy efficiency and the like of the power supply system. The fitness calculation of each particle will determine its direction and distance of movement in subsequent iterations. Subsequently, an iterative calculation is performed on the particle population, and each iteration updates the position (control parameter configuration) of the particle according to the individual experience of the particle and the common experience of the population. This process continues until a preset stop condition is met, such as a maximum number of iterations is reached, the quality of the solution reaches a predetermined threshold, or the improvement amplitude is below a certain limit. Finally, the particle swarm algorithm generates a corresponding optimal solution, namely, the control parameter configuration with optimal system performance under the given control parameter range and load demand conditions. According to the optimal solution, a corresponding first control parameter strategy is generated, and the strategy guides the actual operation of the power supply control module, so that the system operates in an optimal state while meeting the load demand, and the performance and reliability of the whole dual-power supply online standby circuit are improved.
Step S107, performing standby power switching control and real-time state monitoring on the dual-power online standby circuit according to a first control parameter strategy to obtain real-time state monitoring data;
first, the first control parameter strategy is parsed in detail and converted into a specific control command. These control parameters may include the triggering condition of the power supply switching, the priority order of the switching, the power supply output adjustment during the switching, the control of the standby time, and other relevant protection measures, etc. After analyzing these parameters, a detailed set of operating instructions is obtained, which are input into the power management control system for directing the specific execution of the standby power switch. Then, standby power switching control is executed according to the instruction. For example, monitoring the status of the primary power source, evaluating the availability of the backup power source, dynamically adjusting the power output, activating a backup power switching mechanism if necessary, and the like. The core goal of the standby power switching control is to ensure that when the primary power source fails or is insufficient to support the current load, a smooth and rapid switch to the standby power source is enabled, while ensuring that no interference or damage is caused to the system during the switching process. And when the standby power is switched and controlled, real-time state monitoring is performed to ensure that the control strategy is correctly executed and timely responds to any emergency. Real-time status monitoring typically involves continuous tracking of critical parameters of the circuit, including voltage, current, power, temperature, and other factors that may affect system operation. Real-time data of these parameters are collected by various sensors and monitoring devices installed in the power supply system and then transmitted to a data processing center for real-time analysis and processing. By analyzing the real-time monitoring data, the standby power switching control can be ensured to be executed according to a preset strategy, and any abnormal situation or unexpected situation can be timely found. For example, if a sudden drop in the output of the backup power source or a failure recovery of the primary power source is monitored, the system reacts quickly to these real-time data, adjusts the control strategy, or triggers additional protection measures.
And S108, performing strategy adjustment on the first control parameter strategy according to the real-time state monitoring data to obtain a second control parameter strategy.
Specifically, first, the performance of the current control strategy is evaluated and the possible direction of adjustment is explored. This may include modeling different parameter tuning schemes, predicting their possible impact on system performance, and assessing the feasibility and safety of these tuning schemes. In this process, various factors are considered, such as the load trend of the power supply, historical fault data, expected operating environment changes, etc., to ensure that policy adjustments can solve the current problem without introducing new risks. On the basis of evaluation and exploration, a new set of control parameter strategies, namely a second control parameter strategy, is formulated. The set of strategies will make necessary adjustments and optimizations to the first control parameter strategy based on the results of the real-time data analysis. These adjustments may include changing the triggering conditions of the backup power switch, adjusting the range of regulation of the power supply output, optimizing the logic sequence of power supply management, or introducing new protection mechanisms, etc. And then, verifying and testing the second control parameter strategy to ensure that the new strategy can solve the existing problems and can stably and reliably operate. This may include testing the effect of the new strategy in a simulated environment, evaluating its impact on the overall performance of the system, and monitoring whether its performance in each case meets expectations.
In a specific embodiment, the process of executing step S101 may specifically include the following steps:
(1) The method comprises the steps of monitoring the state of a first power supply control module to obtain first multidimensional state data of the first power supply control module, wherein the first multidimensional state data comprise: first power state data, first diode state data, and first thermistor state data;
(2) The state monitoring is carried out on the second power supply control module to obtain second multidimensional state data of the second power supply control module, wherein the second multidimensional state data comprises: second power state data, second diode state data, and second thermistor state data;
(3) And performing data cleaning and data standardization processing on the first multi-dimensional state data to obtain first running state data of the first power supply control module, and performing data cleaning and data standardization processing on the second multi-dimensional state data to obtain second running state data of the second power supply control module.
Specifically, first, the first power control module is subjected to state monitoring, which includes monitoring of key parameters such as output voltage, current, power, temperature and the like of the first power supply, and tracking of the working states of the first diode and the first thermistor. Operational data of these components is collected in real time by sensors and detection devices mounted on them, which data constitute first multidimensional status data. Similarly, the second power supply control module is subjected to state monitoring, state data of the second power supply, state data of the second diode and state data of the second thermistor are collected to form second multidimensional state data, and the data also cover various parameters of power supply output and working states of key components. And then, carrying out data cleaning and data standardization processing on the multidimensional state data. The data cleaning is to remove noise and errors generated during the collection process, such as correcting deviations caused by sensor errors, removing invalid or incomplete records, filling in missing values, and the like. This helps to improve data quality and analysis accuracy. After data cleaning is completed, the cleaned data is subjected to standardized processing, and data from different sources and with different dimensions are converted into a consistent format and scale. By data normalization, different state data can be compared and analyzed under the same criteria, which facilitates subsequent data mining and pattern recognition activities. And obtaining the first running state data of the first power supply control module and the second running state data of the second power supply control module through data cleaning and standardization processing.
In a specific embodiment, the process of executing step S102 may specifically include the following steps:
(1) Performing forward time sequence state feature extraction on the first running state data through a forward long-short time memory network in the first self-encoder to obtain a first forward time sequence state feature vector, and performing backward time sequence state feature extraction on the first running state data through a backward long-short time memory network in the first self-encoder to obtain a first backward time sequence state feature vector;
(2) Performing feature fusion and aggregation mapping on the first forward time sequence state feature vector and the first backward time sequence state feature vector to obtain a first running state feature set;
(3) Performing forward time sequence state feature extraction on the second running state data through a forward long-short time memory network in the second self-encoder to obtain a second forward time sequence state feature vector, and performing backward time sequence state feature extraction on the second running state data through a backward long-short time memory network in the second self-encoder to obtain a second backward time sequence state feature vector;
(4) And performing feature fusion and aggregation mapping on the second forward time sequence state feature vector and the second backward time sequence state feature vector to obtain a second running state feature set.
Specifically, first, a self-encoder model is built for each of the first and second power control modules, each model including a forward LSTM network and a backward LSTM network. These networks are specifically designed to process time series data and are capable of capturing and extracting long-term dependencies and complex patterns in time series. In a first self-encoder model, first operating state data is input to a forward LSTM network. Forward LSTM networks process data in a time-sequential manner, gradually learning and extracting features of the time series from past to present. As data propagates forward in the network, the LSTM unit updates its internal state and generates a feature representation at each time step, which successive feature representations together form a first forward timing state feature vector. This feature vector captures the dynamic changes and time dependence from history to current time, helping to understand and predict future behavior of the system. At the same time, the first operating state data is input to the backward LSTM network in the same self-encoder model. The backward LSTM network, as opposed to the forward network, never comes to past processing data and learning the impact on the current state from the subsequent states. Similar to the forward network, the backward network also generates feature representations at each time step, which together constitute a first backward timing state feature vector. This vector captures the time series characteristics from another perspective. And after the first forward and backward time sequence state feature vectors are obtained, feature fusion is carried out on the two vectors. Feature fusion generally involves combining two vector elements or transformed elements to form a more comprehensive, richer feature representation. This fused representation of features integrates past to future information and more fully reflects the operational state of the system. This fused feature representation is then converted to a first set of operating state features by set mapping. And extracting second forward and backward time sequence state feature vectors through the forward and backward LSTM networks in the second self-encoder, and carrying out feature fusion and integration mapping to finally obtain a second running state feature set.
In a specific embodiment, the process of executing step S103 may specifically include the following steps:
(1) Performing feature coding and feature normalization on the first running state feature set to obtain a plurality of first normalized running state features, and performing feature coding and feature normalization on the second running state feature set to obtain a plurality of second normalized running state features;
(2) Performing feature vector conversion on the first normalized running state features to obtain first running state feature vectors, and performing feature vector conversion on the second normalized running state features to obtain second running state feature vectors;
(3) Inputting a first running state feature vector into a preset power supply abnormality analysis model, wherein the power supply abnormality analysis model comprises a two-layer threshold circulation network and a full-connection layer, extracting hidden features of the first running state feature vector through the two-layer threshold circulation network to obtain first running state hidden features, predicting power supply abnormality probability of the first running state hidden features through a ReLU function in the full-connection layer to obtain a first power supply abnormality probability predicted value, and comparing the first power supply abnormality probability predicted value with a preset first power supply abnormality probability threshold to obtain a first abnormality analysis result of a first power supply control module;
(4) Inputting the second running state feature vector into a preset power supply abnormality analysis model, extracting hidden features of the second running state feature vector through a two-layer threshold circulation network to obtain second running state hidden features, predicting power supply abnormality probability of the second running state hidden features through a ReLU function in a full-connection layer to obtain a second power supply abnormality probability predicted value, and comparing the second power supply abnormality probability predicted value with a preset second power supply abnormality probability threshold to obtain a second abnormality analysis result of a second power supply control module.
Specifically, first, feature encoding and feature normalization are performed on the first running state feature set. Feature encoding is the conversion of non-numeric features in a feature set into numeric form so that algorithms can better process these data. For example, if a feature is of a class type, such as a power state "normal" or "abnormal," the system will convert it to a numeric type, such as 0 and 1. Feature normalization is then performed to scale all features to a common numerical range, typically 0 to 1, to eliminate the effects of different dimensions and ensure that the model is able to treat all features fairly. And performing feature coding and normalization on the second running state feature set to obtain a plurality of second normalized running state features. These normalized features are more suitable for processing of machine learning models, and can improve the efficiency of model training and the accuracy of prediction. Then, feature vector conversion is performed on the first normalized operating state features, and the normalized features are combined into a single high-dimensional feature vector. This high-dimensional feature vector contains comprehensive information of the operating state of the first power control module. And similarly, performing feature vector conversion on the second normalized running state feature to obtain a feature vector representing the running state of the second power supply control module. After the first and second operation state feature vectors are obtained, these vectors are input to a preset power supply abnormality analysis model, respectively. This model is a complex structure based on deep learning, comprising a two-layer threshold loop network (Gated Recurrent Unit, GRU) and a fully connected layer. Two-layer GRU networks are designed to process time series data, and can capture time dependence and complex patterns in feature vectors. The first layer of GRU network receives the input feature vector and generates a series of hidden states that are passed to the second layer of GRU network to further extract deep timing features. The hidden features extracted by the GRU network reflect the deep dynamic change of the running state of the power supply system, and provide key information for predicting power supply abnormality. The hidden feature is then sent to the fully connected layer for further processing. The fully-connected layer typically contains one or more linear transforms, and a nonlinear activation function, such as a ReLU function, that is capable of converting the linearly transformed data into nonlinear data, increasing the expressive power of the model. In this layer, the hidden feature of the first operating state is converted into a power anomaly probability prediction value that indicates the likelihood of an anomaly in the first power control module. The system compares the predicted value with a preset first power supply abnormality probability threshold value, and if the predicted value exceeds the threshold value, the system considers that the first power supply control module possibly has abnormality, so as to generate a first abnormality analysis result. And sending the second running state feature vector into the same power supply abnormality analysis model, obtaining a second power supply abnormality probability prediction value through the processing of the two layers of GRU networks and the full-connection layer, and comparing the second power supply abnormality probability prediction value with a preset second power supply abnormality probability threshold value to obtain a second abnormality analysis result. This result reflects the health and potential risk of the second power control module.
In a specific embodiment, the process of executing step S106 may specifically include the following steps:
(1) Defining a plurality of control parameter adjustment ranges of a first power supply control module or a second power supply control module, randomly initializing control parameters of the first power supply control module or the second power supply control module according to load demand data through the plurality of control parameter adjustment ranges to obtain corresponding initialization control parameter sets, and constructing particle populations of the initialization control parameter sets through a preset particle swarm algorithm to obtain particle populations;
(2) Carrying out particle fitness calculation on the particle population to obtain a particle fitness set corresponding to the particle population;
(3) And carrying out iterative computation on the particle fitness set until a preset condition is met, generating an optimal solution corresponding to the particle population, and generating a corresponding first control parameter strategy according to the optimal solution.
Specifically, first, a plurality of control parameter adjustment ranges of the first power control module or the second power control module are defined. These control parameters may include the voltage of the power supply output, current limits, temperature control thresholds, time parameters of the switching logic, etc., and the adjustment range of each parameter defines the minimum and maximum values that the parameter may vary. These parameters are then randomly initialized based on the current load demand data. Each control parameter is set to a random value within its adjustment range to generate a preliminary, random configuration of control parameters. This configuration represents one control strategy that the power control module may take under certain conditions. And respectively carrying out random initialization for the first power supply control module and the second power supply control module to obtain two groups of initialization control parameter sets. Next, the initialization control parameter set is input into a preset particle swarm algorithm. In the algorithm, each set of control parameter configurations is considered a "particle" and all particles together constitute a search population. The particle swarm algorithm is an optimization algorithm based on swarm intelligence, simulates social behaviors of shoals or shoals, and searches for an optimal solution in a solution space. Each particle has a position and a velocity, the position representing a potential solution, i.e. a control parameter configuration, and the velocity determining the direction and distance of movement of the particle. And after the population construction is completed, carrying out fitness calculation on each particle. The fitness function is usually designed according to the performance index of the power control module, such as stability, response time, energy efficiency, and the like of the power supply. The fitness value of each particle reflects the performance of the control parameter configuration represented by the particle, and a high fitness value means that the parameter configuration is better. Subsequently, an iterative calculation is performed on the population of particles. In each iteration, the particles update their velocity and position based on their own experience and the common experience of the population. Specifically, each particle remembers its historically best location (the individual best solution) and the historically best location (the global best solution) of the entire population, and adjusts its flight direction and distance based on this information. In this way, the particles gradually approach the optimal solution region. This iterative process continues until a preset stop condition is met, such as a maximum number of iterations is reached or the quality of the solution reaches a predetermined threshold. After the stopping condition is met, the algorithm generates an optimal solution corresponding to the particle population, namely, the optimal solution in all control parameter configurations under the current load demand and the power performance index. The system generates a corresponding first control parameter strategy according to the first control parameter strategy to guide the operation of the first power supply control module. The same optimization process is also applicable to the second power control module, generating a second control parameter strategy.
According to the technical scheme, the first power control module and the second power control module are subjected to state monitoring, so that running state data can be obtained in real time. This enables a rapid perception of the state change of the power supply and a corresponding control adjustment to be made in time. The feature extraction is carried out on the running state data through the preset self-encoder, so that the key features in the power supply system can be captured. Such feature extraction may reduce data dimensionality, improve understanding of the power state, and provide valuable input for subsequent anomaly analysis. And carrying out abnormality analysis on the running state characteristic set by using a power supply abnormality analysis model, thereby being beneficial to identifying abnormal conditions in a power supply system. This provides predictive capability for power failures, enabling early discovery and addressing of potential problems, thereby improving reliability and stability of the system. Load demand data is obtained through comprehensive abnormal analysis results, and control parameter analysis is carried out by utilizing a particle swarm algorithm, so that the system operation parameters are optimized. The optimization can ensure that the system can provide stable and reliable power supply under different load conditions, and the adaptability and performance of the system are improved. And carrying out real-time control and standby switching on the dual-power supply online standby circuit according to the generated control parameter strategy. By adjusting the control parameter strategy according to the real-time state monitoring data, different working conditions can be dynamically adapted in the running process, and the flexibility and the robustness of the system are improved. A thermistor is used for dynamic overload or overcurrent protection, and when the current in a target circuit is excessive, the overload current is limited by adjusting the resistance value. This adaptive fault handling mechanism helps to prevent system damage caused by power supply overload.
It will be clear to those skilled in the art that, for convenience and brevity of description, the specific operation of the above-described system, system and unit may refer to the corresponding procedure in the foregoing circuit embodiment, which is not repeated here.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, including several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the circuits described in the various embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (random acceS memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The above embodiments are merely for illustrating the technical solution of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the corresponding technical solutions.
Claims (6)
1. The utility model provides a dual supply on-line spare circuit which characterized in that, dual supply on-line spare circuit includes:
the power supply control system comprises a first power supply control module and a second power supply control module, wherein the first power supply control module is connected with the second power supply control module in parallel, the first power supply control module comprises a first power supply, a first diode and a first thermistor, the output end of the first power supply is connected with the input end of the first diode, the output end of the first diode is connected with the input end of the first thermistor, the output end of the first thermistor is connected with the VCC end of a preset target circuit, and the input end of the first power supply is connected with the GND end of the target circuit; the second power supply control module comprises a second power supply, a second diode and a second thermistor, wherein the output end of the second power supply is connected with the input end of the second diode, the output end of the second diode is connected with the input end of the second thermistor, the output end of the second thermistor is connected with the VCC end of the target circuit, and the input end of the second power supply is connected with the GND end of the target circuit;
Acquiring power difference information between the first power supply and the second power supply, and continuously supplying power to the target circuit without interruption according to the power difference information; the first diode and the second diode are used for preventing short circuit between the first power supply and the second power supply; the first thermistor and the second thermistor are used for dynamically overload or overcurrent protection of the target circuit; the first thermistor or the second thermistor is specifically configured to: when the current in the target circuit is excessive, the resistance of the first thermistor or the second thermistor is controlled to be increased so as to reduce the overload current in the target circuit.
2. The dual-power online standby electric control method is characterized by comprising the following steps of:
the method comprises the steps of performing state monitoring on a first power supply control module to obtain first running state data of the first power supply control module, and performing state monitoring on a second power supply control module to obtain second running state data of the second power supply control module;
extracting the running state characteristics of the first running state data through a preset first self-encoder to obtain a first running state characteristic set, and extracting the running state characteristics of the second running state data through a preset second self-encoder to obtain a second running state characteristic set;
Inputting the first running state characteristic set into a preset power supply abnormality analysis model to perform power supply abnormality analysis to obtain a first abnormality analysis result of the first power supply control module, and inputting the second running state characteristic set into the power supply abnormality analysis model to perform power supply abnormality analysis to obtain a second abnormality analysis result of the second power supply control module;
comprehensively analyzing the first abnormal analysis result and the second abnormal analysis result to obtain a comprehensive abnormal analysis result;
according to the comprehensive anomaly analysis result, load demand data of the dual-power supply online standby circuit are obtained;
inputting the load demand data into a preset particle swarm algorithm for control parameter analysis, and generating a corresponding first control parameter strategy;
performing standby power switching control and real-time state monitoring on the dual-power online standby circuit according to the first control parameter strategy to obtain real-time state monitoring data;
and carrying out strategy adjustment on the first control parameter strategy according to the real-time state monitoring data to obtain a second control parameter strategy.
3. The dual-power on-line standby control method according to claim 2, wherein the performing state monitoring on the first power control module to obtain first operation state data of the first power control module, and performing state monitoring on the second power control module to obtain second operation state data of the second power control module comprises:
The state monitoring is performed on the first power supply control module to obtain first multidimensional state data of the first power supply control module, wherein the first multidimensional state data comprises: first power state data, first diode state data, and first thermistor state data;
performing state monitoring on the second power supply control module to obtain second multidimensional state data of the second power supply control module, wherein the second multidimensional state data comprises: second power state data, second diode state data, and second thermistor state data;
and performing data cleaning and data standardization processing on the first multi-dimensional state data to obtain first running state data of the first power supply control module, and performing data cleaning and data standardization processing on the second multi-dimensional state data to obtain second running state data of the second power supply control module.
4. The dual-power on-line standby control method according to claim 3, wherein the performing, by a preset first self-encoder, the operation state feature extraction on the first operation state data to obtain a first operation state feature set, and performing, by a preset second self-encoder, the operation state feature extraction on the second operation state data to obtain a second operation state feature set, includes:
The first operation state data is subjected to forward time sequence state feature extraction through a forward long-short time memory network in the first self-encoder to obtain a first forward time sequence state feature vector, and the first operation state data is subjected to backward time sequence state feature extraction through a backward long-short time memory network in the first self-encoder to obtain a first backward time sequence state feature vector;
performing feature fusion and aggregation mapping on the first forward time sequence state feature vector and the first backward time sequence state feature vector to obtain a first running state feature set;
the forward time sequence state characteristic extraction is carried out on the second running state data through a forward long-short time memory network in the second self-encoder to obtain a second forward time sequence state characteristic vector, and the backward time sequence state characteristic extraction is carried out on the second running state data through a backward long-short time memory network in the second self-encoder to obtain a second backward time sequence state characteristic vector;
and performing feature fusion and aggregation mapping on the second forward time sequence state feature vector and the second backward time sequence state feature vector to obtain a second running state feature set.
5. The dual-power on-line standby control method according to claim 4, wherein the inputting the first running state feature set into a preset power abnormality analysis model for power abnormality analysis to obtain a first abnormality analysis result of the first power control module, and inputting the second running state feature set into the power abnormality analysis model for power abnormality analysis to obtain a second abnormality analysis result of the second power control module comprises:
performing feature coding and feature normalization on the first running state feature set to obtain a plurality of first normalized running state features, and performing feature coding and feature normalization on the second running state feature set to obtain a plurality of second normalized running state features;
performing feature vector conversion on the plurality of first normalized running state features to obtain a first running state feature vector, and performing feature vector conversion on the plurality of second normalized running state features to obtain a second running state feature vector;
inputting the first running state feature vector into a preset power supply abnormality analysis model, wherein the power supply abnormality analysis model comprises a two-layer threshold circulation network and a full-connection layer, extracting hidden features of the first running state feature vector through the two-layer threshold circulation network to obtain a first running state hidden feature, predicting the power supply abnormality probability of the first running state hidden feature through a ReLU function in the full-connection layer to obtain a first power supply abnormality probability predicted value, and comparing the first power supply abnormality probability predicted value with a preset first power supply abnormality probability threshold to obtain a first abnormality analysis result of the first power supply control module;
Inputting the second running state feature vector into a preset power supply abnormality analysis model, extracting hidden features of the second running state feature vector through the two-layer threshold circulation network to obtain second running state hidden features, predicting power supply abnormality probability of the second running state hidden features through a ReLU function in the full-connection layer to obtain a second power supply abnormality probability predicted value, and comparing the second power supply abnormality probability predicted value with a preset second power supply abnormality probability threshold to obtain a second abnormality analysis result of the second power supply control module.
6. The dual-power on-line standby control method according to claim 5, wherein the inputting the load demand data into a preset particle swarm algorithm for control parameter analysis, generating a corresponding first control parameter strategy, comprises:
defining a plurality of control parameter adjustment ranges of the first power supply control module or the second power supply control module, carrying out control parameter random initialization on the first power supply control module or the second power supply control module according to the load demand data through the plurality of control parameter adjustment ranges to obtain a corresponding initialization control parameter set, and carrying out particle population construction on the initialization control parameter set through a preset particle swarm algorithm to obtain a particle population;
Calculating the particle fitness of the particle population to obtain a particle fitness set corresponding to the particle population;
and carrying out iterative computation on the particle fitness set until a preset condition is met, generating an optimal solution corresponding to the particle population, and generating a corresponding first control parameter strategy according to the optimal solution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410032882.3A CN117543803A (en) | 2024-01-10 | 2024-01-10 | Dual-power on-line standby circuit and control method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410032882.3A CN117543803A (en) | 2024-01-10 | 2024-01-10 | Dual-power on-line standby circuit and control method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117543803A true CN117543803A (en) | 2024-02-09 |
Family
ID=89792336
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410032882.3A Pending CN117543803A (en) | 2024-01-10 | 2024-01-10 | Dual-power on-line standby circuit and control method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117543803A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103984316A (en) * | 2014-05-16 | 2014-08-13 | 刘玮 | Energy management device and system |
| CN104600827A (en) * | 2014-12-05 | 2015-05-06 | 超威电源有限公司 | Composite power supply and control method thereof and electric vehicle with same |
| CN205725129U (en) * | 2016-04-27 | 2016-11-23 | 南京因泰莱配电自动化设备有限公司 | A kind of FTU power supply possessing double loop power supply input function |
| CN109638832A (en) * | 2019-02-12 | 2019-04-16 | 深圳市风云实业有限公司 | Dual power supply redundancy power supply is realized and monitoring system and equipment |
| CN213521394U (en) * | 2020-09-29 | 2021-06-22 | 北京东土科技股份有限公司 | Clock circuit and equipment with power down holding function |
| US20220085646A1 (en) * | 2020-09-14 | 2022-03-17 | Shenzhen Carku Technology Co., Limited | Intelligent control system, emergency starting power supply, and intelligent battery clip |
| CN114781552A (en) * | 2022-06-17 | 2022-07-22 | 深圳硅山技术有限公司 | Motor performance testing method, device, equipment and storage medium |
| CN116643178A (en) * | 2023-07-27 | 2023-08-25 | 深圳凌奈智控有限公司 | SOC estimation method and related device of battery management system |
| CN220156279U (en) * | 2023-05-04 | 2023-12-08 | 立讯精密科技(西安)有限公司 | Dual-power supply circuit and device |
-
2024
- 2024-01-10 CN CN202410032882.3A patent/CN117543803A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103984316A (en) * | 2014-05-16 | 2014-08-13 | 刘玮 | Energy management device and system |
| CN104600827A (en) * | 2014-12-05 | 2015-05-06 | 超威电源有限公司 | Composite power supply and control method thereof and electric vehicle with same |
| CN205725129U (en) * | 2016-04-27 | 2016-11-23 | 南京因泰莱配电自动化设备有限公司 | A kind of FTU power supply possessing double loop power supply input function |
| CN109638832A (en) * | 2019-02-12 | 2019-04-16 | 深圳市风云实业有限公司 | Dual power supply redundancy power supply is realized and monitoring system and equipment |
| US20220085646A1 (en) * | 2020-09-14 | 2022-03-17 | Shenzhen Carku Technology Co., Limited | Intelligent control system, emergency starting power supply, and intelligent battery clip |
| CN213521394U (en) * | 2020-09-29 | 2021-06-22 | 北京东土科技股份有限公司 | Clock circuit and equipment with power down holding function |
| CN114781552A (en) * | 2022-06-17 | 2022-07-22 | 深圳硅山技术有限公司 | Motor performance testing method, device, equipment and storage medium |
| CN220156279U (en) * | 2023-05-04 | 2023-12-08 | 立讯精密科技(西安)有限公司 | Dual-power supply circuit and device |
| CN116643178A (en) * | 2023-07-27 | 2023-08-25 | 深圳凌奈智控有限公司 | SOC estimation method and related device of battery management system |
Non-Patent Citations (1)
| Title |
|---|
| 冯成安;: "双电源自动转换装置设计", 机电工程技术, vol. 38, no. 6, 15 June 2009 (2009-06-15), pages 69 - 73 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101984730B1 (en) | Automatic predicting system for server failure and automatic predicting method for server failure | |
| KR102118670B1 (en) | System and method for management of ict infra | |
| Fink | Data-driven intelligent predictive maintenance of industrial assets | |
| CN116683588B (en) | Lithium ion battery charge and discharge control method and system | |
| CN117713221B (en) | Micro-inversion photovoltaic grid-connected optimization system | |
| CN117436004B (en) | Motor performance real-time monitoring system and method | |
| CN116700193A (en) | Factory workshop intelligent monitoring management system and method thereof | |
| CN112488541A (en) | Multi-mode fault early warning method and system for electric equipment in construction site | |
| CN117394529A (en) | SCADA-based auxiliary decision method and system for main distribution network loop-closing reverse power supply control conditions | |
| Di Maio et al. | Transient identification by clustering based on Integrated Deterministic and Probabilistic Safety Analysis outcomes | |
| CN116880151A (en) | Redundancy computer control system | |
| CN117639602A (en) | Self-adaptive motor running state adjusting method and system | |
| Basher et al. | Autonomous control of nuclear power plants | |
| CN117114454A (en) | DC sleeve state evaluation method and system based on Apriori algorithm | |
| CN117543803A (en) | Dual-power on-line standby circuit and control method thereof | |
| Xu et al. | A multivariate anomaly detector for satellite telemetry data using temporal attention-based lstm autoencoder | |
| Borissova et al. | A concept of intelligent e-maintenance decision making system | |
| De Santis et al. | Fault recognition in smart grids by a one-class classification approach | |
| Mol et al. | Neural and genetic-based approaches to nuclear transient identification including ‘don't know’response | |
| CN113076217A (en) | Disk fault prediction method based on domestic platform | |
| Xu et al. | Probabilistic Prognosis of Wind Turbine Faults with Feature Selection and Confidence Calibration | |
| CN112115128A (en) | 5G real-time dynamic fault processing system based on multi-dimensional sensor | |
| Papataxiarhis et al. | Event correlation and forecasting over high-dimensional streaming sensor data | |
| CN117687307B (en) | Multi-rocker control method, device, equipment and storage medium | |
| Papataxiarhis et al. | Event correlation and forecasting over multivariate streaming sensor data |
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 |