CN112505576A - Power failure rapid detection method based on least square classifier - Google Patents
Power failure rapid detection method based on least square classifier Download PDFInfo
- Publication number
- CN112505576A CN112505576A CN202011411645.6A CN202011411645A CN112505576A CN 112505576 A CN112505576 A CN 112505576A CN 202011411645 A CN202011411645 A CN 202011411645A CN 112505576 A CN112505576 A CN 112505576A
- Authority
- CN
- China
- Prior art keywords
- classifier
- square
- output
- input voltage
- power failure
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 34
- 230000002159 abnormal effect Effects 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 3
- 238000000034 method Methods 0.000 claims description 2
- 230000007547 defect Effects 0.000 abstract description 4
- 238000012850 discrimination method Methods 0.000 abstract description 3
- 230000009977 dual effect Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000007796 conventional method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/40—Testing power supplies
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Evolutionary Biology (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Mathematics (AREA)
- Mathematical Optimization (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- Artificial Intelligence (AREA)
- Algebra (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Testing Electric Properties And Detecting Electric Faults (AREA)
Abstract
The invention discloses a power failure rapid detection method based on a least square classifier, which comprises the following steps: the power failure detection method has the advantages that the least square classifier embedded into the DSP controller is built in the DSP-controlled dual-power system, DSP hardware collects two paths of power supply voltages in real time and inputs the two paths of power supply voltages into the least square classifier, the least square classifier trained by collecting actual voltage data has the discrimination capability more suitable for actual conditions, the defects of false detection or missed detection or overlong detection time of a conventional discrimination method are avoided, the power failure detection speed is increased, according to a classification result, whether two paths of power supplies are normal or not can be monitored in real time, the power supplies can be safely, quickly and effectively switched, and the power utilization reliability of loads is guaranteed.
Description
Technical Field
The invention relates to the technical field of power failure rapid detection, in particular to a power failure rapid detection method based on a least square classifier.
Background
In order to improve the power utilization reliability of the load, a common power supply system adopts dual power supplies, for example, STS and ATS usually supply power to two mains supplies, one mains supply and one UPS, and UPS usually supplies power to one battery-input inverter and one mains supply. How to safely, quickly and effectively switch the dual power supplies directly affects the power utilization reliability of the load, and especially the acceptable power failure time of many devices is extremely harsh, such as: the power failure time is required to be not more than 10ms, and some power failure time even requires 5 ms.
The switching time of STS, ATS and UPS includes power down detection time and power switch operation time, and it is obvious that the use of a semiconductor device with faster switching speed can improve the switching speed, but increase the cost, and therefore, it is more valuable to study a faster power down detection method.
The conventional power failure detection method compares the deviation of the instantaneous value and the effective value of the actual input voltage with the deviation of the instantaneous value and the effective value of the reference input voltage, but the requirement of rapidity is difficult to meet, and in the case of inputting a special waveform voltage as shown in fig. 1, if the conventional method is adopted, false detection or missing detection is easy, or the detection time is too long. How to provide a technical scheme capable of meeting the requirement of rapidly detecting power failure under conventional and specific conditions is a problem to be solved by the technical personnel in the field at present.
Disclosure of Invention
The invention aims to overcome the technical defects and provides a power failure rapid detection method based on a least square classifier, so that a power supply is switched safely, rapidly and effectively, and the power utilization reliability of a load is ensured.
In order to solve the technical problems, the technical scheme provided by the invention is as follows: a power failure rapid detection method based on a least square classifier comprises the following steps:
step S1: the DSP controller detects and retains an input voltage U set { U (k) } …, U (k-n +1) } of n beats on line;
step S2: inputting the input voltages { u (k), …, u (k-n +1) } into a least square classifier, wherein a characteristic vector x (k) at the k-th moment is a formula (1):
x(k)＝P(y(k-1),...,y(k-n),u(k),...,u(k-n)) (1)
the discriminant function of the least squares classifier is formula (2):
g(x(k))＝ω·x(k) (2)
the Boolean value output of the least squares classifier is formula (3)
If g (x (k)) > (0), then y (k)) > (1; if g (x (k)) is <0, then y (k) is 0; (3)
step S3: sampling and retaining p beats of input voltage signals { u (k-1), …, u (k-p) }, dividing the input voltage into a normal class and a fault class, and expressing the input voltage signals by using Boolean values; recording inputs as { u (k), …, u (k-n +1) }, recording voltage normal output as y (k) -1 and voltage abnormal output as y (k) -0, taking an input voltage signal { u (k), …, u (k-p +1) } of p beats and an output signal { y (k), … and y (k-p +1) } of p beats as training samples, and obtaining a weight omega of the least square classifier through training;
step S4: the DSP controller detects and retains n-beat input voltage signals { u (k), …, u (k-n +1) } and n-beat least square classifier output signals { y (k-1), …, y (k-n) }, calculates x (k), then calculates output y (k) of a discrimination function of the least square classifier, and according to the value of output y (k), if y (k) is 1, the input voltage is normal, and if y (k) is 0, the input voltage is abnormal, the DSP controller sends an SCR control signal to switch to an auxiliary power supply.
Preferably, the least squares classifier of step S2 is a discrimination algorithm inside the DSP.
Preferably, in formula (1) of step S2, u (k), …, u (k-n), and y (k), …, y (k-n) are input voltage and classifier output at the k-th time of the least square classifier, n is the beat number of the input and output, in formula (2), ω is the weight of the least square classifier, and in formula (3), y (k) is the classifier output, where y (k) is a boolean value indicating whether the input power is normal or not.
Compared with the prior art, the invention has the advantages that: the power failure detection method has the advantages that the least square classifier embedded into the DSP controller is built in the DSP-controlled dual-power system, DSP hardware collects two paths of power supply voltages in real time and inputs the two paths of power supply voltages into the least square classifier, the least square classifier trained by collecting actual voltage data has the discrimination capability more suitable for actual conditions, the defects of false detection or missed detection or overlong detection time of a conventional discrimination method are avoided, the power failure detection speed is increased, according to a classification result, whether two paths of power supplies are normal or not can be monitored in real time, the power supplies can be safely, quickly and effectively switched, and the power utilization reliability of loads is guaranteed.
Drawings
Fig. 1 is a schematic diagram of a mains voltage waveform aimed at by the power failure rapid detection method based on the least square classifier of the present invention.
FIG. 2 is a schematic diagram of a dual power supply system to which the power failure rapid detection method based on the least square classifier of the present invention is applied.
FIG. 3 is a least square classification flow diagram of the power failure rapid detection method based on the least square classifier.
Detailed Description
The technical solutions in the embodiments of the present invention are clearly and completely described below, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention discloses a power failure rapid detection method based on a least square classifier, which comprises the following steps:
step S1: the DSP controller detects and retains an input voltage U set { U (k) } …, U (k-n +1) } of n beats on line;
step S2: inputting the input voltages { u (k), …, u (k-n +1) } into a least square classifier, wherein a characteristic vector x (k) at the k-th moment is a formula (1):
x(k)＝P(y(k-1),...,y(k-n),u(k),...,u(k-n)) (1)
the discriminant function of the least squares classifier is formula (2):
g(x(k))＝ω·x(k) (2)
the Boolean value output of the least squares classifier is formula (3)
If g (x (k)) > (0), then y (k)) > (1; if g (x (k)) is <0, then y (k) is 0; (3)
step S3: sampling and retaining p beats of input voltage signals { u (k-1), …, u (k-p) }, dividing the input voltage into a normal class and a fault class, and expressing the input voltage signals by using Boolean values; recording inputs as { u (k), …, u (k-n +1) }, recording voltage normal output as y (k) -1 and voltage abnormal output as y (k) -0, taking an input voltage signal { u (k), …, u (k-p +1) } of p beats and an output signal { y (k), … and y (k-p +1) } of p beats as training samples, and obtaining a weight omega of the least square classifier through training;
step S4: the DSP controller detects and retains n-beat input voltage signals { u (k), …, u (k-n +1) } and n-beat least square classifier output signals { y (k-1), …, y (k-n) }, calculates x (k), then calculates output y (k) of a discrimination function of the least square classifier, and according to the value of output y (k), if y (k) is 1, the input voltage is normal, and if y (k) is 0, the input voltage is abnormal, the DSP controller sends an SCR control signal to switch to an auxiliary power supply.
The least squares classifier of step S2 is a discrimination algorithm inside the DSP.
In the step S2, in formula (1), u (k), …, u (k-n) and y (k), …, y (k-n) are the input voltage and the classifier output at the k-th time of the least square classifier, n is the beat number of the input and the output, in formula (2), ω is the weight of the least square classifier, and in formula (3), y (k) is the classifier output, where y (k) is the boolean value, which indicates whether the input power is normal.
The method is suitable for dual power systems such as STS, ATS and UPS, a least square classifier embedded in a DSP controller is built in the DSP-controlled dual power system, DSP hardware acquires two paths of power supply voltages in real time and inputs the two paths of power supply voltages into the least square classifier, the least square classifier trained by acquiring actual voltage data has the discrimination capability more suitable for actual conditions, the defects of false detection or missed detection or overlong detection time of a conventional discrimination method are avoided, the power failure detection speed is accelerated, whether two paths of power supplies are normal or not can be monitored in real time according to a classification result, the power supply can be switched safely, quickly and effectively, and the power utilization reliability of a load is guaranteed.
The preferred embodiments of the invention disclosed above are intended to be illustrative only. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise embodiments disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best utilize the invention. The invention is limited only by the claims and their full scope and equivalents.
Claims (3)
1. A power failure rapid detection method based on a least square classifier is characterized in that: the method comprises the following steps:
step S1: the DSP controller detects and retains an input voltage U set { U (k) } …, U (k-n +1) } of n beats on line;
step S2: inputting the input voltages { u (k), …, u (k-n +1) } into a least square classifier, wherein a characteristic vector x (k) at the k-th moment is a formula (1):
x(k)＝P(y(k-1),...,y(k-n),u(k),...,u(k-n)) (1)
the discriminant function of the least squares classifier is formula (2):
g(x(k))＝ω·x(k) (2)
the Boolean value output of the least squares classifier is formula (3)
If g (x (k)) > (0), then y (k)) > (1; if g (x (k)) is <0, then y (k) is 0; (3)
step S3: sampling and retaining p beats of input voltage signals { u (k-1), …, u (k-p) }, dividing the input voltage into a normal class and a fault class, and expressing the input voltage signals by using Boolean values; recording inputs as { u (k), …, u (k-n +1) }, recording voltage normal output as y (k) -1 and voltage abnormal output as y (k) -0, taking an input voltage signal { u (k), …, u (k-p +1) } of p beats and an output signal { y (k), … and y (k-p +1) } of p beats as training samples, and obtaining a weight omega of the least square classifier through training;
step S4: the DSP controller detects and retains n-beat input voltage signals { u (k), …, u (k-n +1) } and n-beat least square classifier output signals { y (k-1), …, y (k-n) }, calculates x (k), then calculates output y (k) of a discrimination function of the least square classifier, and according to the value of output y (k), if y (k) is 1, the input voltage is normal, and if y (k) is 0, the input voltage is abnormal, the DSP controller sends an SCR control signal to switch to an auxiliary power supply.
2. The power failure rapid detection method based on the least square classifier as claimed in claim 1, characterized in that: the least squares classifier of step S2 is a discrimination algorithm inside the DSP.
3. The power failure rapid detection method based on the least square classifier as claimed in claim 1, characterized in that: in the step S2, in formula (1), u (k), …, u (k-n) and y (k), …, y (k-n) are the input voltage and the classifier output at the k-th time of the least square classifier, n is the beat number of the input and the output, in formula (2), ω is the weight of the least square classifier, and in formula (3), y (k) is the classifier output, where y (k) is the boolean value, which indicates whether the input power is normal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011411645.6A CN112505576A (en) | 2020-12-04 | 2020-12-04 | Power failure rapid detection method based on least square classifier |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011411645.6A CN112505576A (en) | 2020-12-04 | 2020-12-04 | Power failure rapid detection method based on least square classifier |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112505576A true CN112505576A (en) | 2021-03-16 |
Family
ID=74971843
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011411645.6A Pending CN112505576A (en) | 2020-12-04 | 2020-12-04 | Power failure rapid detection method based on least square classifier |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112505576A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116243234A (en) * | 2023-05-11 | 2023-06-09 | 石家庄科林电气股份有限公司 | Power failure detection method and system of multimode assembled electric energy meter and electric energy meter |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106953561A (en) * | 2017-04-24 | 2017-07-14 | 福州大学 | A Speed Regulation Method of Brushed DC Motor Based on Least Squares Classification Speed Measurement |
| US9874923B1 (en) * | 2005-05-30 | 2018-01-23 | Invent.Ly, Llc | Power management for a self-powered device scheduling a dynamic process |
| CN108183639A (en) * | 2018-01-15 | 2018-06-19 | 福州大学 | A kind of brshless DC motor least squared classified speed regulating method |
-
2020
- 2020-12-04 CN CN202011411645.6A patent/CN112505576A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9874923B1 (en) * | 2005-05-30 | 2018-01-23 | Invent.Ly, Llc | Power management for a self-powered device scheduling a dynamic process |
| CN106953561A (en) * | 2017-04-24 | 2017-07-14 | 福州大学 | A Speed Regulation Method of Brushed DC Motor Based on Least Squares Classification Speed Measurement |
| CN108183639A (en) * | 2018-01-15 | 2018-06-19 | 福州大学 | A kind of brshless DC motor least squared classified speed regulating method |
Non-Patent Citations (1)
| Title |
|---|
| 王俊伟;俞冬冬;吴秋轩;: "基于DSP的新型掉电快速检测系统" * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116243234A (en) * | 2023-05-11 | 2023-06-09 | 石家庄科林电气股份有限公司 | Power failure detection method and system of multimode assembled electric energy meter and electric energy meter |
| CN116243234B (en) * | 2023-05-11 | 2023-08-11 | 石家庄科林电气股份有限公司 | Power failure detection method and system of multimode assembled electric energy meter and electric energy meter |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10503126B2 (en) | Access control method for parallel direct current power supplies and device thereof | |
| Huang et al. | Multiple open-circuit fault diagnosis based on multistate data processing and subsection fluctuation analysis for photovoltaic inverter | |
| Geng et al. | Detection and localization strategy for switch open-circuit fault in modular multilevel converters | |
| CN101895155B (en) | Energy storage device with fault detection and automatic reorganization functions and recombination method | |
| CN109655703B (en) | Self-checking method, device and equipment for parallel power modules and storage medium | |
| CN103675591B (en) | A kind of electrokinetic cell output bus multipoint insulation fault detection circuit | |
| CN110703124B (en) | Method for detecting and protecting three-phase power supply and load phase loss | |
| US10024928B2 (en) | Inverter and detection method for an inverter for detecting a network fault | |
| CN112165071B (en) | Leakage current early warning protection circuit with self-test function | |
| EP4586431A1 (en) | Photovoltaic system, and circuit protection method for photovoltaic system | |
| CN106711953A (en) | Protection device and protection method for short-circuiting at direct current side of photovoltaic inverter | |
| KR101770987B1 (en) | Inverter fault diagnosis apparatus and the method thereof | |
| CN112505576A (en) | Power failure rapid detection method based on least square classifier | |
| Arsad et al. | Rule-based fuzzy controller for solid state transfer switch towards fast sensitive loads transfer | |
| JP2013085364A (en) | Grid connection power conversion equipment control device and grid connection power conversion equipment | |
| CN108037403A (en) | The inverter open fault diagnostic method and device of a kind of current hysteresis-band control | |
| JPH1139047A (en) | Static var compensator | |
| Caseiro et al. | Real-time multiple IGBT open-circuit fault diagnosis in three-level neutral-point-clamped converters with no additional sensors | |
| CN110687443A (en) | Primary equipment state discrimination method and system | |
| JP2012100494A (en) | Distributed power supply device | |
| CN213122151U (en) | Self-detection device of current output interface | |
| JPS6213119A (en) | Non-contact proximity switch | |
| Tumay et al. | Sequence reference frame-based new sag/swell detection method for static transfer switches | |
| CN112526338B (en) | Novel synchronous time window excitation power unit online fault diagnosis method | |
| Caseiro et al. | Open-circuit fault diagnosis in neutral-point-clamped UPS inverters with no additional sensors |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210316 |