CN105510693B - A kind of method based on Hall current and image change monitoring Large-scale Medical Equipment use state - Google Patents

A kind of method based on Hall current and image change monitoring Large-scale Medical Equipment use state Download PDF

Info

Publication number
CN105510693B
CN105510693B CN201510829345.2A CN201510829345A CN105510693B CN 105510693 B CN105510693 B CN 105510693B CN 201510829345 A CN201510829345 A CN 201510829345A CN 105510693 B CN105510693 B CN 105510693B
Authority
CN
China
Prior art keywords
mrow
image change
use state
medical equipment
change rate
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.)
Active
Application number
CN201510829345.2A
Other languages
Chinese (zh)
Other versions
CN105510693A (en
Inventor
王清波
虞成
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hangzhou First Peoples Hospital
Original Assignee
Hangzhou First Peoples Hospital
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hangzhou First Peoples Hospital filed Critical Hangzhou First Peoples Hospital
Priority to CN201510829345.2A priority Critical patent/CN105510693B/en
Publication of CN105510693A publication Critical patent/CN105510693A/en
Application granted granted Critical
Publication of CN105510693B publication Critical patent/CN105510693B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
    • G01R19/25—Arrangements for measuring currents or voltages or for indicating presence or sign thereof using digital measurement techniques
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R15/00—Details of measuring arrangements of the types provided for in groups G01R17/00 - G01R29/00, G01R33/00 - G01R33/26 or G01R35/00
    • G01R15/14—Adaptations providing voltage or current isolation, e.g. for high-voltage or high-current networks
    • G01R15/20—Adaptations providing voltage or current isolation, e.g. for high-voltage or high-current networks using galvano-magnetic devices, e.g. Hall-effect devices, i.e. measuring a magnetic field via the interaction between a current and a magnetic field, e.g. magneto resistive or Hall effect devices
    • G01R15/202—Adaptations providing voltage or current isolation, e.g. for high-voltage or high-current networks using galvano-magnetic devices, e.g. Hall-effect devices, i.e. measuring a magnetic field via the interaction between a current and a magnetic field, e.g. magneto resistive or Hall effect devices using Hall-effect devices

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

The invention discloses it is a kind of based on Hall current and image change monitoring Large-scale Medical Equipment use state method, including：Step 1, using the running current intensity of Hall current sensor detection Large-scale Medical Equipment, and the display image of Large-scale Medical Equipment is gathered simultaneously, after display image is converted into gray level image, calculate image change rate；Step 2, cluster analysis is carried out to running current intensity and image change rate, running current intensity and image change rate is made to correspond to different use states, use state includes at least shutdown, standby, operation；Step 3, the acquisition running current intensity of Large-scale Medical Equipment and image change rate, foundation cluster result obtain corresponding use state in real time.The present invention can carry out nonlinear state analysis automatically to the use state of Large-scale Medical Equipment, break away from manual intervention, obtain accurate use state information.

Description

It is a kind of that Large-scale Medical Equipment use state is monitored based on Hall current and image change Method
Technical field
The present invention relates to monitoring of equipment technical fields, and in particular to one kind is large-scale based on Hall current and image change monitoring The method of medical equipment use state.
Background technology
With the constantly improve of China's medical system structure, the valuable Medical Devices of major, Medium Sized Hospitals large size and instrument Quantity be also continuously increased, therefore, how to improve the service efficiency of Large-scale Medical Equipment, improve the comprehensive benefit of hospital, be doctor Treat the topic that should be inquired into deeply in appliance management.
In the daily management of hospital, such as radiation, ultrasound equipment carry out equipment substantially all by the way of labor management The daily statistics use states such as enable, stop, the mode subjectivity manually counted is stronger, and can not accomplish to supervise in real time Control.In this process, if labor management is slightly neglected, i.e., with the waste of equipment electric energy and the drop of equipment service efficiency It is low.
It is inquired about through available data, more approximately there is entitled " the Medical Devices use state monitoring system based on RFID of article System ", in June, 2015 is published in《Chinese medical equipment》, it is mainly used for recording the standby of Medical Devices, use state and uses RFID mark instrument numbers.In this article, monitoring system mainly includes client and server, wherein, client is by electricity Four parts such as detection module, RFID identification module, central control module, wireless transport module composition is flowed, is mainly responsible for identification doctor Identity information, the monitoring Medical Devices neutral line current of equipment are treated, and above-mentioned identity information and neutral line current data sending are extremely taken Business device, the Organization Chart of client are as shown in Figure 1.
Using the ac current measurement circuit based on current transformer in the document, to flow through the current value of equipment zero curve into Row detection, wherein ac current measurement use punch current transfomer, after measuring electric current, classify according to such as lower threshold value, per a kind of It corresponds to standby or use state.
Standby current threshold value=preliminary survey standby current × (1+10%)
Operating current threshold value=preliminary survey operating current × (1-10%)
New Medical Devices are started the test pattern of client by operating personnel, surveyed when first time putting into Clinical practice Under die trial formula, operating personnel successively make equipment be in holding state and working condition, client measurements record Medical Devices respectively Preliminary survey standby current and preliminary survey operating current, and the number information of the two preliminary survey electric currents and equipment is sent to service together Device, server set 2 grades of threshold values according to preliminary survey electric current, and then by microcontroller gathered data, wireless transport module passes through UART Interface neutral line current data and device RF ID number informations, are then sent to service by Wi-Fi via wireless router Device.
The method that the document is reported has the disadvantages that：
(1) using circuit block of the punch current transfomer as current measurement, if AC current transformer is opened a way, can produce Raw high voltage, can puncture line insulation when serious, generate the safety that electric shock threatens staff, simultaneously as tested electric current is abnormal The influence of change, multiple harmonic, non-sinusoidal waveform etc., the accuracy for making Measurement Result of Current Transformer are poor.
(2) foundation of disaggregated model is, it is necessary to which independent test pattern, the test pattern need the manpower intervention of staff Operation, if the number of devices in hospital is more, will appear from the substantial amounts of duplication of labour, it is more difficult to promote and apply.
(3) object-oriented is Medical Devices, but article main body is patient monitor, lung ventilator etc., which is within the hospital Middle-size and small-size equipment, equipment state are divided into standby mode and use pattern substantially, and standby mode is substantially at equipment out of use state, There is larger difference with the pattern of large scale equipment.
Separately there is article to be published in《Chinese medicine is equipped》, article is entitled " to concentrate the center of supervision ultrasonic device open state Monitoring system is researched and developed ", this article is completed to set ultrasound by carrying out ultrasonic image equipment information-based, centralization start detection Standby centralized management and supervision.Specific method is：B ultrasound current signal, comparison circuit conversion are gathered using Hall current sensor Into 0-1 status informations；Turn network module using MSP430F449 and serial ports, send data to acquisition terminal；Acquisition terminal number According to and be sent to central monitoring server after being arranged.
In the document, although object-oriented is Large-scale Medical Equipment, state only has 0-1 states, i.e. startup and shutdown shape State, cannot be distinguished under open state it is standby, the different conditions such as use.The judgement of different conditions is also sentenced using threshold value in the document Fixed mode is set by comparing the slide rheostat resistance value of circuit on hardware circuit, i.e., is manually set using the method for experience It is fixed.And in actual use, because machine performance changes, it is also possible to there is equipment use, but electric current compared to before instead under The situation of drop.
It is not two usually using state for Large-scale Medical Equipment, but there are many state, for example, shut down, Standby, use state etc. for Large-scale Medical Equipment, because power consumption is larger, if stopping using on the day of machine, should be at shutting down State saves energy consumption.Under open state, standby non-working condition and use state are also classified into, use state is for example：Such as B Super camera shooting, CT exposures etc..
Large-scale Medical Equipment is under the special states such as patient information operation, with using current conditions under holding state without apparent Difference can not accurately distinguish merely in the prior art by current strength.Further, in the prior art, if to large size The use state of medical equipment be judged as it is standby, and for a long time be in holding state, it is still necessary to manually Large-scale Medical Equipment is closed Machine.
The content of the invention
The present invention provides it is a kind of based on Hall current and image change monitoring Large-scale Medical Equipment use state method, Nonlinear state analysis can be carried out automatically to the use state of Large-scale Medical Equipment, break away from manual intervention, accurately made Use status information.
A kind of method based on Hall current and image change monitoring Large-scale Medical Equipment use state, including：
Step 1, using the running current intensity of Hall current sensor detection Large-scale Medical Equipment, and large size is gathered simultaneously The display image of medical equipment after display image is converted into gray level image, calculates image change rate；
Step 2, cluster analysis is carried out to running current intensity and image change rate, makes running current intensity and image change Rate corresponds to different use states, and use state includes at least shutdown, standby, operation；
Step 3, the acquisition running current intensity of Large-scale Medical Equipment and image change rate, foundation cluster result obtain in real time To corresponding use state.
Hall current sensor is using electromagnetic induction principle, alternating current is measured in real time, using accurate constant current Technology and linear temperature compensation technology, by the DC signal output that its isolated variable is standard, the present invention is passed using Hall current Sensor measures the running current of Large-scale Medical Equipment, and security is higher, and the linearity is good, and without to medical equipment on hardware Destructive improvement is carried out, it is easy to implement.
In the present invention using image change etc. with using relevant factor, in combination with the electric current of operation intensity, to doctor The analysis of operating status is carried out with equipment, the different state such as shutdown, standby, operation can be distinguished, manual intervention is broken away from, realize Automatic classification.
Preferably, step 1 is fallen into a trap nomogram as the method for change rate it is as follows：
1-a, each pixel gray value of current time gray level image are denoted as ymn, the total quantity of pixel is S, when current The pixel gray value for carving the average gray image in retrospect predetermined amount of time forward is denoted as yavg mn；
1-b, formula z is utilizedmn=| ymn-yavg mn| the pixel gray value z of calculating difference gray level imagemn；
Pixel gray value is more than threshold value y in 1-c, calculating difference gray level imagecPixel quantity Rt；
1-d, x (t)=R is utilizedt/ S calculates image change rate x (t).
Due to considering the data of two dimensions of running current intensity and image change rate simultaneously, make in different With under state, aspherical distribution is presented in data, and traditional cluster mode cannot use, and needs to consider using aspherical distribution Clustering algorithm, the present invention obtains classification center using density peaks quick clustering algorithm, for real-time grading sample storehouse, specifically Ground, in step 2, the step of cluster analysis, is as follows：
2-a, image change rate and current signal are normalized using following equation：
In formula, x (t) and i (t) are respectively image change rate and running current intensity；
xnorm(t) and inorm(t) image change rate and running current intensity after respectively normalizing；
xmax(t) and imax(t) be respectively image change rate and running current intensity maximum；
2-b, in two-dimensional space, respectively using the image change rate after normalization and running current intensity as each moment Sample coordinate point, calculate the local density ρ of each sample pointi, to the distance δ of high local density's pointiAnd high local density Point mark ki；
2-c, J is utilizedi=δi×ρiCalculate characteristic value Ji, and the characteristic value of all sample points is ranked up, take maximum For several sample points as classification center, each classification center corresponds to a kind of use state of Large-scale Medical Equipment；
The order of 2-d, remaining sample point according to local density from high to low determines that affiliated medical equipment uses shape successively State.
The performance of Large-scale Medical Equipment changes with usage time, in order to accurately know the use of medical equipment State, it is necessary to adjust cluster analysis as a result, more accurately to tell the use state of medical equipment, the present invention using point Class center more new algorithm, it is preferable that in step 2, after being spaced predetermined amount of time, be updated to cluster result, update method is such as Under：
It is assumed that the sample point that need to be currently substituted is Xm, for substituting the data point of the sample point as Xs；
Calculate XmThe distance between other sample points dmj, j ∈ sample point sets, record dmj<dcSample point set be Ψ, dcTo block distance, dmjFor sample point m to the Euclidean distance of sample point j；
Calculate XsThe distance between other sample points dsj, j ∈ sample point sets, record dsj<dcSample point set be Π, dsjFor sample point s to the Euclidean distance of sample point j；
It obtains other sample points Zhong Gao local densities point and is denoted as the sample set of m as Ρ；
Replace sample point XmFor Xs, and do following update：
Update ρaFor ρa'：ρa'=ρa- 1, a ∈ Ψ；
Update ρbFor ρb'：ρb'=ρb+ 1, b ∈ Π；
Update ρmFor ρ 'm：Wherein, if dmj-dc<0, thenOtherwise
Update δmFor δm'：
Update δvFor δv'：
Preferably, further including step 4, the use state of gained is subjected to remote data transmission.
By the use state remote data transmission of gained to server, convenient for the reality concentrated to each Large-scale Medical Equipment When monitor.
Preferably, further include step 5, if use state is determined as standby, and standby time is more than threshold value, then automatic to close Disconnected Medical Devices.The threshold value is set as needed, by the way that power-off control module is set to turn off Medical Devices.
Beneficial effects of the present invention are as follows：
(1) real-time operating current intensity is detected by Hall current sensor, reduces the difficulty of scrap build, hall sensing Device current detecting can realize isolation detection, not influence existing equipment performance.
(2) using density peaks quick clustering algorithm, compared to traditional clustering method, aspherical classification shape can be handled State, classification center more new algorithm can avoid clustering again completely for sample, borrow legacy data, quick to realize sample storehouse more Newly.
(3) use state is transmitted to central monitoring unit for monitoring in real time and recording using network communications technology.
(4) network communication can be used for the remote monitoring that Large-scale Medical Equipment uses.Power-off control can automatically shut down doctor Institute's equipment is placed carelessness and is used, energy saving.
Description of the drawings
Fig. 1 is the client framework map in " Medical Devices use state monitor system " based on RFID, in June, 2015 It is published in《Chinese medical equipment》；
Fig. 2 is the flow chart of method provided by the invention；
Fig. 3 is the system schematic for implementing the method for the present invention；
Fig. 4 is power detecting module diagram in the system for implement the method for the present invention；
Fig. 5 is the schematic diagram of image detection module in the system for implement the method for the present invention；
Fig. 6 is the schematic diagram of real-time three condition detection module in the system for implement the method for the present invention；
Fig. 7 is the schematic diagram of communication transmission module in the system for implement the method for the present invention；
Fig. 8 is the schematic diagram of control module in the system for implement the method for the present invention.
Specific embodiment
Below in conjunction with the accompanying drawings, Large-scale Medical Equipment use state is monitored to the present invention is based on Hall currents and image change Method is described in detail.
As shown in figure 3, realizing the monitoring system structure of method provided by the invention includes：Current detection module, image become Change detection module, real-time three condition detection module, communication transmission module and power-off control module；Wherein current detection module, figure As change detection module and the communication connection of real-time three condition detection module, real-time three condition detection module and communication transmission module phase Even, power-off control module is connected with communication transmission module.
The magnetic field that alternating current generates is converted to linear DC voltage by current detection module by Hall effect, and is converted For accessible numerical data；Image Change Detection module obtains the image data of Medical Devices by wired connection, is converted to Gray level image passes through image change rate detection algorithm, real-time image acquisition change rate；Real-time three condition detection module respectively with electricity Detection module wired connection is flowed, with Image Change Detection module wired connection；Communication transmission module is wiredly connected to by serial ports.
Accessible numerical data is obtained from current detection module, is converted to and represents shutdown, standby, operation 3 classification shape State information, is provided in communication transmission module；Communication transmission module by receiving status information of classifying, turns network mould using serial ports Block is sent to computer and is received, while the teledata transmitting element on computer sends data to teledata and receives list Member completes remote data monitoring.
As shown in figure 4, current detection module include power supply unit, Hall current sensor, signal modulation circuit and Analog/digital conversion circuit.Power supply unit includes Switching Power Supply, analog voltage circuit and reference voltage circuit.Wherein switch electricity Source provides 12V DC power supply and 9V direct current supplys, analog voltage circuit connects its 9V using the extra large Switching Power Supply JMD30-1209 of letter Output is exported using 7805 chip voltages to 5V, and reference voltage circuit uses AD780, and AD780 is a superhigh precision band gap base Reference voltage source provides 2.5V outputs, inputs and export offer by analog voltage 5V.12V DC is for electrical connection Hall current sensor Driving voltage end, analog voltage 5V outputs connection signal modulation circuit and analog/digital conversion circuit, wherein 2.5V reference voltage outputs It is inputted as analog/digital conversion circuit reference voltage.
Hall current sensor is using Nanjing very A-CT2010LT Hall current sensors suddenly.Signal modulation circuit is using general Logical integrated transporting discharging follows circuit, and chip uses rail-to-rail amplifier OPA343.Analog/digital conversion circuit uses AD7705 preferred circuits, Reference voltage is provided by AD780, and data transmission is attached with real-time three condition detection module.
As shown in figure 5, Image Change Detection module includes image acquisition unit, image conversion unit, image change rate meter Calculate unit.Image acquisition unit and image conversion unit wired connection, image conversion unit can with image change rate computing unit On one piece of microprocessor, internal processes connection.Image acquisition unit is responsible for image and is converted to unified output, technical solution choosing Composite video BNC outputs are selected as, conventional large scale equipment output format includes DVI, VGA, composite video (BNC) output substantially.Such as Fruit interface exports for DVI, and usable DVI turns HDMI wiring connection HDMI modular converters, and HDMI modular converters can select PlayvisionLC101HDMI-BNC converters；If interface exports for VGA, HDMI conversions can be turned using green VGA Device is converted to HDMI outputs, PlayvisionLC101HDMI-BNC converters is recycled to be converted to SD video, after progress Reason；BNC outputs turn HDMI converter because output frequency is higher, first with bright strong BNC and are converted to high definition HDMI signals, then profit SD video is converted to PlayvisionLC101HDMI-BNC converters, is post-processed.Using day falcon image FI1206VCABoard development boards are responsible for carrying image converting unit, image change rate computing unit relative program, use TVP5151 digit chips and TMS320DM6437 chips.
DM6437 configures TVP5146 chips by I2C buses, such as input clock frequency, video formats and regards Frequency input format etc..TVP5151 chips under control of the synchronization signal, by 8 position datawires by the image data of acquisition, with The format transmission of YCbCr video data streams is to VPFE.
As shown in fig. 6, three condition detection module includes state classification unit, the real-time training unit of grader, cluster in real time Data storage cell.State classification unit is examined using the SPI interface of dsp chip TMS320F2812, TMS320F2812 with electric current AD7705 wired connections in module analog/digital conversion circuit are surveyed, obtain 16Bit data.The real-time training unit of grader also uses TMS320F2812 passes through serial ports wired connection with state classification unit chip.Storage unit is AT24C256 memories, AT24C256 stores training sample point, storage current data, image change data, sample point local density, to high local density Distance, the high local density point mark of point.
As shown in fig. 7, communication transmission module includes microprocessor, Date Conversion Unit, teledata transmitting element and remote Journey data receipt unit.Microprocessor uses circuit of the STC12C5A60S2 chips for core, and serial ports 1 receives real-time three condition inspection Survey the data mode that module is come, serial ports 2 with Date Conversion Unit is wired is connected, responsible data are transmitted to medical equipment network And reception.Microprocessor I/O port carries out wired connection with power-off modular simultaneously, and transmission equipment cut-off signals are 0-1 data.Data Converting unit can be used serial ports and turn network module, use commercially available USR-TCP232-2 modules.VB6.0 progress can be used in software Programming is programmed using Winsock controls, using Transmission Control Protocol, needs to establish connection before communication.Server unit is set It puts oracle listener and is responsible for data decryptor, and carry out monitoring data and show.
As shown in figure 8, power-off control module includes low-voltage control high voltage unit and solid-state relay.Low-voltage controls High voltage unit is using the T73SRD-5VDC-SL-C relays that voltage is 5V are controlled, and input terminal concatenates triode, and triode is defeated Enter the equipment cut-off signals for communication transmission module.1 end of output termination is connected with 12V Switching Power Supplies, the other end and solid-state relay Device wired connection.Low-voltage controls high voltage unit to isolate for circuit.Mage that model MGR- can be used in solid-state relay The three-phase solid relay of 30323840Z controls voltage to control the output terminal wired connection of high voltage unit, solid-state for low-voltage Relay input terminal accesses power supply interface, and output terminal connection power supply leading-out terminal can be matched with device line and is connected.
As shown in Fig. 2, a kind of method based on Hall current and image change monitoring Large-scale Medical Equipment use state, bag It includes：
Step 1, using the running current intensity of Hall current sensor detection Large-scale Medical Equipment, and large size is gathered simultaneously The display image of medical equipment after display image is converted into gray level image, calculates image change rate.
It is obtained with the form of YCbCr video data streams and shows image, and be converted to Y, Cr, Cb component using equation below RGB image：
R=1.164* (Y-16)+1.596* (Cr-128)；
G=1.164* (Y-16) -0.392* (Cb-128) -0.813* (Cr-128)；
B=1.164* (Y-16)+2.017* (Cb-128).
RGB component is converted into gray level image using following publicity：
Gray=(R*299+G*587+B*114+500)/1000.
The method for calculating image change rate is as follows：
1-a, each pixel gray value of current time gray level image are denoted as ymn, the total quantity of pixel is S, when current Carve the pixel gray level for tracing the average gray image of (predetermined amount of time is set as needed) in predetermined amount of time forward Value is denoted as yavg mn；
1-b, formula z is utilizedmn=| ymn-yavg mn| the pixel gray value z of calculating difference gray level imagemn；
Pixel gray value is more than threshold value y in 1-c, calculating difference gray level imagecPixel quantity Rt；
1-d, x (t)=R is utilizedt/ S calculates image change rate x (t).
Step 2, cluster analysis is carried out to running current intensity and image change rate, makes running current intensity and image change Rate corresponds to different use states, and use state includes at least shutdown, standby, operation.
If memory head byte is 0, show to there is no cluster result, carry out clustering learning, the step of cluster analysis is as follows：
2-a, image change rate and current signal are normalized using following equation：
In formula, x (t) and i (t) are respectively image change rate and running current intensity；
xnorm(t) and inorm(t) image change rate and running current intensity after respectively normalizing；
xmax(t) and imax(t) be respectively image change rate and running current intensity maximum；
2-b, in two-dimensional space, respectively using the image change rate after normalization and running current intensity as each moment Sample coordinate point, calculate the local density ρ of each sample pointi, to the distance δ of high local density's pointiAnd high local density Point mark ki。
The local density ρ of each sample point i is calculated using following formulai：
If dij-dc<0, thenOtherwise
dijFor the Euclidean distance between other sample points j and sample point i；dcTo block distance, when other sample points j and sample Euclidean distance between this i less than this block apart from when, sample point i local densities add 1, block the desirable normalization of distance acquiescence The 1% of data maximums can modify according to actual needs.
Sample point i is calculated to the distance δ of high local density's point using following formulai：
Wherein dijFor the Euclidean distance between sample point j and sample point i；
High local density's point indicates kiMeet：
dki=δi
dkiFor the Euclidean distance between sample point k and sample point i；
Sample point is to the distance beta of Ci Gao local densities pointi：
2-c, J is utilizedi=δi×ρiCalculate characteristic value Ji, and the characteristic value of all sample points is ranked up, take maximum For several sample points as classification center, each classification center corresponds to a kind of use state of Large-scale Medical Equipment.
In present embodiment, use state is divided into shutdown, standby, three kinds of operation, each corresponds to a classification center.
The order of 2-d, remaining sample point according to local density from high to low determines that affiliated medical equipment uses shape successively State.
After determining classification center, classification center is numbered by with center zero apart from size, is respectively 0,1,2, represents Shutdown, standby, three states of operation.Order of other sample points according to local density from high to low determines generic successively, The classification of each non-categorical central sample point is the classification corresponding to classification center nearest in neighborhood.Memory head byte status It is changed to data storage area first address.
If memory head byte is not 0, shows have cluster result, clustering learning is not required to, into real-time grading pattern. During start, if the real-time training unit of grader, into after real-time grading pattern, state classification unit receives the electric current number of AD7705 According to and image change rate data, and pass through the real-time training unit of grader and compareed with sample storehouse, it is nearest according to Euclidean distance Principle, determine most like training sample, determine class categories.Serial data is sent to communication transfer after determining class categories Module.
Data and classification results deposit caching to real-time grading, carry out the data under current class every 1 hour It is average, and the storage data under the category are updated, pointer cyclic shift.It is locally close to update corresponding sample point simultaneously Degree, the distance to high local density's point, high local density point beacon information.
After being spaced predetermined amount of time, cluster result is updated, update method is as follows：
It is assumed that the sample point that need to be currently substituted is Xm, for substituting the data point of the sample point as Xs；
Calculate XmThe distance between other sample points dmj, j ∈ sample point sets, record dmj<dcSample point set be Ψ, dcTo block distance, dmjFor sample point m to the Euclidean distance of sample point j；
Calculate XsThe distance between other sample points dsj, j ∈ sample point sets, record dsj<dcSample point set be Π, dsjFor sample point s to the Euclidean distance of sample point j；
It obtains other sample points Zhong Gao local densities point and is denoted as the sample set of m as Ρ；
Replace sample point XmFor Xs, and do following update：
Update ρaFor ρa'：ρa'=ρa- 1, a ∈ Ψ；
Update ρbFor ρb'：ρb'=ρb+ 1, b ∈ Π；
Update ρmFor ρ 'm：Wherein, if dmj-dc<0, thenOtherwise
Update δmFor δm'：
Update δvFor δv'：
Step 3, the acquisition running current intensity of Large-scale Medical Equipment and image change rate, foundation cluster result obtain in real time To corresponding use state.
Step 4, the use state of gained is subjected to remote data transmission.
Include step 5, if use state be determined as it is standby, and standby time be more than threshold value, then automatically shut down Medical Devices.

Claims (5)

  1. A kind of 1. method based on Hall current and image change monitoring Large-scale Medical Equipment use state, which is characterized in that bag It includes：
    Step 1, using the running current intensity of Hall current sensor detection Large-scale Medical Equipment, and gather simultaneously large-scale medical The display image of equipment after display image is converted into gray level image, calculates image change rate；
    The method of the calculating image change rate is as follows：
    1-a, each pixel gray value of current time gray level image are denoted as ymn, the total quantity of pixel is S, current time to The pixel gray value of average gray image in preceding retrospect predetermined amount of time is denoted as yavg mn；
    1-b, formula z is utilizedmn=| ymn-yavg mn| the pixel gray value z of calculating difference gray level imagemn；
    Pixel gray value is more than threshold value y in 1-c, calculating difference gray level imagecPixel quantity Rt；
    1-d, x (t)=R is utilizedt/ S calculates image change rate x (t)；
    Step 2, cluster analysis is carried out to running current intensity and image change rate, makes running current intensity and image change rate pair Should be in different use states, use state includes at least shutdown, standby, operation；
    Step 3, the acquisition running current intensity of Large-scale Medical Equipment and image change rate in real time according to cluster result, obtains pair The use state answered.
  2. 2. the method based on Hall current and image change monitoring Large-scale Medical Equipment use state as described in claim 1, It is characterized in that, in step 2, the step of cluster analysis, is as follows：
    2-a, image change rate and current signal are normalized using following equation：
    <mrow> <msub> <mi>x</mi> <mrow> <mi>n</mi> <mi>o</mi> <mi>r</mi> <mi>m</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <mi>x</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> <mrow> <msub> <mi>x</mi> <mrow> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>,</mo> <msub> <mi>i</mi> <mrow> <mi>n</mi> <mi>o</mi> <mi>r</mi> <mi>m</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <mi>i</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> <mrow> <msub> <mi>i</mi> <mrow> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </mfrac> </mrow>
    In formula, x (t) and i (t) are respectively image change rate and running current intensity；
    xnorm(t) and inorm(t) image change rate and running current intensity after respectively normalizing；
    xmax(t) and imax(t) be respectively image change rate and running current intensity maximum；
    2-b, in two-dimensional space, respectively using the image change rate after normalization and running current intensity as the sample at each moment This coordinate points calculates the local density ρ of each sample point ii, to the distance δ of high local density's pointiAnd high local density's point mark Show ki；
    2-c, J is utilizedi=δi×ρiCalculate characteristic value Ji, and the characteristic value of all sample points is ranked up, take the several of maximum For sample point as classification center, each classification center corresponds to a kind of use state of Large-scale Medical Equipment；
    The order of 2-d, remaining sample point according to local density from high to low determines affiliated medical equipment use state successively.
  3. 3. the method based on Hall current and image change monitoring Large-scale Medical Equipment use state as claimed in claim 2, It is characterized in that, in step 2, after being spaced predetermined amount of time, cluster result is updated, update method is as follows：
    It is assumed that the sample point that need to be currently substituted is Xm, for substituting the data point of the sample point as Xs；
    Calculate XmThe distance between other sample points dmj, j ∈ sample point sets, record dmj<dcSample point set be Ψ, dc To block distance, dmjFor sample point m to the Euclidean distance of sample point j；
    Calculate XsThe distance between other sample points dsj, j ∈ sample point sets, record dsj<dcSample point set be Π, dsj For sample point s to the Euclidean distance of sample point j；
    It obtains other sample points Zhong Gao local densities point and is denoted as the sample set of m as P；
    Replace sample point XmFor Xs, and do following update：
    Update ρaFor ρa'：ρa'=ρa- 1, a ∈ Ψ；
    Update ρbFor ρb'：ρb'=ρb+ 1, b ∈ Π；
    Update ρmFor ρ 'm：Wherein, if dmj-dc<0, thenOtherwise
    Update δmFor δm'：
    Update δvFor δv'：
  4. 4. the method based on Hall current and image change monitoring Large-scale Medical Equipment use state as claimed in claim 3, It is characterized in that, further including step 4, the use state of gained is subjected to remote data transmission.
  5. 5. the method based on Hall current and image change monitoring Large-scale Medical Equipment use state as claimed in claim 4, It is characterized in that, further include step 5, if use state is determined as standby, and standby time is more than threshold value, then automatically shuts down medical treatment Equipment.
CN201510829345.2A 2015-11-24 2015-11-24 A kind of method based on Hall current and image change monitoring Large-scale Medical Equipment use state Active CN105510693B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510829345.2A CN105510693B (en) 2015-11-24 2015-11-24 A kind of method based on Hall current and image change monitoring Large-scale Medical Equipment use state

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510829345.2A CN105510693B (en) 2015-11-24 2015-11-24 A kind of method based on Hall current and image change monitoring Large-scale Medical Equipment use state

Publications (2)

Publication Number Publication Date
CN105510693A CN105510693A (en) 2016-04-20
CN105510693B true CN105510693B (en) 2018-05-25

Family

ID=55718814

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510829345.2A Active CN105510693B (en) 2015-11-24 2015-11-24 A kind of method based on Hall current and image change monitoring Large-scale Medical Equipment use state

Country Status (1)

Country Link
CN (1) CN105510693B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106777984B (en) * 2016-12-19 2019-02-22 福州大学 A method of photovoltaic array Working state analysis and fault diagnosis are realized based on density clustering algorithm
CN108693804A (en) * 2018-05-11 2018-10-23 广东冉盛科学仪器有限公司 A kind of contactless monitoring device and processing method
CN111257805B (en) * 2019-11-29 2022-09-06 广东中科慈航信息科技有限公司 Method and device for monitoring running state of equipment based on Hall sensor
CN111256757A (en) * 2020-02-25 2020-06-09 深圳哈维生物医疗科技有限公司 Medical equipment monitoring system and method based on cloud computing
CN112307671A (en) * 2020-10-27 2021-02-02 杭州电子科技大学 Method for self-adapting to different large-scale equipment instrument state threshold values
CN114061659A (en) * 2021-11-15 2022-02-18 江苏中天互联科技有限公司 Device monitoring method, device, electronic device, storage medium, and program product

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101893655A (en) * 2009-05-21 2010-11-24 徐启峰 Digital optical current sensor
CN102565524A (en) * 2012-03-15 2012-07-11 江苏省电力公司常州供电公司 Real-time monitoring method for short-circuit current of near zone of transformer

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005003541A (en) * 2003-06-12 2005-01-06 Auto Network Gijutsu Kenkyusho:Kk Sensor circuit, magnetic sensor circuit, current sensor circuit, and method for detecting level of magnetism or current
EP1884787A1 (en) * 2006-07-10 2008-02-06 S. THIIM ApS A current sensor for measuring electric current in a conductor and a short circuit indicator system comprising such a sensor
KR20130088284A (en) * 2012-01-31 2013-08-08 에스케이이노베이션 주식회사 Power relay assembly

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101893655A (en) * 2009-05-21 2010-11-24 徐启峰 Digital optical current sensor
CN102565524A (en) * 2012-03-15 2012-07-11 江苏省电力公司常州供电公司 Real-time monitoring method for short-circuit current of near zone of transformer

Also Published As

Publication number Publication date
CN105510693A (en) 2016-04-20

Similar Documents

Publication Publication Date Title
CN105510693B (en) A kind of method based on Hall current and image change monitoring Large-scale Medical Equipment use state
CN109444566A (en) A kind of electric car electrically-charging equipment detection device and method
CN109816033A (en) A method of the supervised learning based on optimization carries out area user identification zone
CN107622342B (en) MVC (model view controller) architecture-based distribution network area data analysis system
CN108508396A (en) Multiple-in-one collecting device detecting system and method
CN106872793A (en) A kind of direct current formate field intensity based on wireless telecommunications, ion current density synchronized measurement system
CN103777169A (en) Electric energy metering transformer simulation verification system
CN103778827A (en) Electric energy metering transformer detection simulation training and evaluation system
CN106646038A (en) Low-voltage diagnosis system on basis of data acquired by intelligent electric meter in real time
CN107222272A (en) Short-wave receiver intelligent monitor system based on collection of letters functional completeness
CN206193128U (en) Electricity theft monitoring device is prevented to multidimension degree
CN108732524A (en) A kind of the automation Calibration Circuit and system of forming and capacity dividing test power supply
CN103964312A (en) Energy efficiency testing device and method for electric hoist
CN102095911A (en) Single-phase smart meter based on impedance identification technology
CN205940863U (en) Pointer instrument of transformer substation reading automatic identification equipment
CN106814333A (en) A kind of underground coal mine D.C. regulated power supply automatic detection device
CN108398150A (en) A kind of fermentation automatically records analysis system
CN104931907A (en) Digital display electrical measurement instrument quality group-detection system based on robot visual sense
CN206331009U (en) A kind of electric power metering device with long-distance positioning function
CN205720433U (en) A kind of Distribution Network Harmonics on-Line Monitor Device based on radio sensing network
CN209013419U (en) Portable air-conditioning system energy efficiency monitor
CN207909325U (en) A kind of power grid taiwan area identifying system
CN204008827U (en) Quality of power supply on-line measurement instrument
CN106970263A (en) The efficiency metering system and metering method of a kind of electric vehicle alternating-current charging pile
CN109613354A (en) A kind of power quality wireless monitor system and its monitoring method

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant