CN202693771U - Portable storage battery state detection system - Google Patents

Abstract

The utility model discloses a portable storage battery state detection system. The system comprises a data acquisition module, a multi-channel A/D (Analog/Digital) conversion module, an ARM9 microcontroller module and a data memory module, wherein the voltage, current and temperature signals of a storage battery are acquired and sent to the A/D conversion module by the data acquisition module; analog signals are converted by the multi-channel A/D conversion module, and the multi-channel A/D conversion module is then connected with the ARM9 microcontroller module by an acquisition conversion/control interface; the data acquisition module is controlled by the ARM9 microcontroller module by a Qt/Embedded visualization interface and the acquisition conversion/control interface; after the performance parameters of the storage battery, namely the data of voltage at the two ends of the storage battery, the data of current in a circuit and the data of temperature in charging and discharging processes, are acquired, information acquired in real time is processed by an RBF (Radial Basis Function) neural network algorithm, the health state of the storage battery is detected, and the data is displayed and stored in external USB (Universal Serial Bus) equipment; and the data memory module is a USB interface and connected with the external USB equipment. Data memory equipment is provided. Alarm and interface prompt are carried out on defective batteries to remind a worker to timely process the batteries.

Description

The portable storage battery condition detecting system

Technical field

The utility model relates to a kind of portable storage battery condition detecting system, can be applicable to accumulator and produces line or other battery condition Site Detection environment, finishes the flexible Measurement accuracy to the accumulator property parameter.

Background technology

Lead-acid accumulator is widely used back-up source in the present standby power system, and its reliability is related to the safety and stablization of computing machine, telecommunications, the numerous critical systems of field of petrochemical industry.In case battery malfunction or off-capacity just might cause a serious accident, so battery detecting and monitoring are the domestic and international hot issues of research always.Whether the use state that the detects battery how residual capacity of Main Basis battery satisfies job requirement.And the size of capacity not only operational factor such as operating ambient temperature, the terminal voltage etc. with battery are relevant, also the constructing variable with battery is relevant, such as density of electrolyte etc.Wherein in the accumulator charging/discharging process voltage at two ends, the electric current in the circuit and temperature as the most important parameter of battery, have closely with capacity and to contact, it can not only reflect the state-of-charge that battery is current, and the good and bad degree of reflection battery, but the ageing failure complicated mechanism of lead-acid accumulator, the ageing failure pattern is subjected to the impact of factors, be difficult to reflect with simple model, work and data acquisition by reality are found, rely on the simple data that discharge and recharge to estimate that accurately the performance state of battery is very difficult thing.Therefore, can use the soft computing method of neural network, carry out modeling for the accumulator detection system, thereby effectively solve the on-line monitoring of battery health situation (SOH).But the speed of convergence that exists of traditional neural network is absorbed in the shortcomings such as local extremum, modeling accuracy are not high slowly, easily, now adopt radial basis (RBF, being Radial Basis Function) neural network optimizes the low shortcoming of approximation accuracy of traditional neural network, and nonlinear system had better self-learning capability, approximation capability and generalization ability, to improve the local minimum problem of neural network, improve modeling accuracy, the performance state of correct detection accumulator.

Performance and the life-span of the fluctuation situation of two parameters in the charge and discharge process (for example representing the up and down undulating quantity of two parameters and normal data with Differential) reflection battery.

At present, domestic battery-detecting instrument device mainly contains the instruments such as battery tension logging, electric conductance test instrument for storage battery and internal resistance capacity tester, these instruments just can be realized detection battery condition in various degree, can not reflect the state in the accumulator charging/discharging process comprehensively.Produce on the line the flexible Measurement accuracy of accumulator property parameter for satisfying accumulator, need a kind of portable detector of detection accumulator running status of perfect performance.

The utility model content

Technology of the present utility model is dealt with problems and is: can overcome the complex operation of existing battery-detecting instrument device and carry inconvenient deficiency, a kind of portable storage battery condition detecting system is provided, this system satisfies accumulator and produces on the line demand to the flexible Measurement accuracy of accumulator property parameter, process by radial basis (RBF in data, be Radial Basis Function) the neural network battery model, the performance state of correct analysis accumulator, adopt simultaneously data storage device is provided, the problem battery is reported to the police and interface prompt, remind the staff in time to process accumulator.

Technical solution of the present utility model is:

A kind of portable storage battery condition detecting system, it comprises ARM9 micro controller module, A/D modular converter, data acquisition module and data memory module;

Wherein, the voltage of data collecting module collected accumulator, electric current and temperature signal are sent into the A/D modular converter;

Multichannel A/D modular converter is changed simulating signal, then is connected with the ARM9 micro controller module through Collect conversion/control interface;

The ARM9 micro controller module is processed the information that gathers, and by Qt/Embedded visualization interface and Collect conversion/control interface the data acquisition module is controlled; After finishing the accumulator property parameter and be the data acquisition of voltage, the electric current in the circuit and temperature at accumulator two ends in the charge and discharge process, by the RBF neural network algorithm information of Real-time Collection is processed, detect quickly and easily the health status of accumulator, the health status of final accumulator is shown and store in the external USB device;

Data memory module is USB interface, is connected with the USB device of outside.

Described data acquisition module comprises rectification circuit, current collection circuit, temperature collection circuit, amplifying circuit and noise processing circuit; Wherein,

Noise processing circuit is removed the noise signal in the accumulator signal that gathers, and is connected with rectification circuit, current collection circuit, temperature collection circuit, amplifying circuit;

Rectification circuit is single phase bridge type rectifier circu, finishes the processing to the voltage signal at battery two ends in the accumulator charging/discharging process; Then the voltage signal after will processing directly is sent to the A/D modular converter;

Current collection circuit is current divider circuit, and the magnitude of voltage of shunt output is 0 ~ 75mV;

Temperature collection circuit is temperature sensor;

Current collection circuit, temperature collection circuit are connected with the A/D modular converter through amplifying circuit.

A kind of detection method that adopts the portable storage battery condition detecting system, its step is:

1) current signal, voltage signal, the temperature signal of collection accumulator after amplification, rectification, filtering processing, are sent into the AD converter interface;

2) Qt/Embedded under the ARM microprocessor employing Linux is as the interface development platform, the data of system acquisition and processing are shown and the control work of system, at Correct Analysis after the influence factor of accumulator health status SOH, proposed take electric current, time and temperature as the input, voltage as output the RBF neural network, in order to estimate the health status of accumulator;

3) with the various parameters in the accumulator charging/discharging process of different health status as study and training sample, gather voltage, electric current, temperature and time signal that many groups are in accumulator each stage in charge and discharge process of different health status, as the training level of neural network signal, the correct battery model of setting up;

4) performance parameter of the collection accumulator of real-time online according to the model of setting up in the step 3), analyzes the health status SOH of accumulator, and to the problem accumulator, system's processing of reporting to the police is preserved warning message simultaneously simultaneously and shown; To the problem accumulator, the later stage does further detection again and judges;

5) after data are handled, by the Config_write function with in the ini file of deposit data in the USB device, data comprise the health status SOH of data acquisition time Date, voltage Voltage, electric current Current, temperature T emperature and accumulator, check historical data by the historical data button, the Config_read function in USB device in the ini file sense data be shown among the tableWidget, the technician can check the performance state of the accumulator that certain time point gathers clearly, to data operation and analysis.

In the described step 3), concrete study and training process are as follows:

Input variable is

x _j=[x ₁, x ₂] ^T=[i _j, ψ _j] ^T(formula 1)

The hidden layer radial basis function adopts Gauss's activation function

$u_i=\exp [-\frac{{\left|\right|x-c_i\left|\right|}^2}{2{{\mathrm{\σ}}_i}^2}](i=\mathrm{1,2},\·\·\·,q)$ (formula 2)

In the formula, u _iBe the output of i hidden node; c _iIt is the generalized constant of i hidden node; Q is the hidden layer node number; X=(x ₁, x ₂) ^TBe the input sample; c _iBe the center vector of i hidden node Gaussian function, this vector is a column vector identical with the dimension of input sample x;

The magnitude of voltage at accumulator two ends is as the output of RBF neural network

$y=U=\underset{i=1}{\overset{q}{\mathrm{\Σ}}}{w}_i{u}_i-U_i$ (formula 3)

In the formula, y is the output of RBF neural network, w _iBe the weighting coefficient of hidden layer to output layer; U _iThreshold value for hidden layer; For j sample, (x _j, y _j) specifically learning procedure is as follows:

(1) presses respectively the output u that (formula 2) and (formula 3) calculates each hidden layer unit of RBF neural network _i(x _i) and the output y of network _j

(2) error between the output of calculating actual acquisition accumulator both end voltage and RBF neural network || e _j||=|| U-y _j|| and the distance of sample and already present hidden layer unit

d _j＝||x _j-c _j|| (j＝1,2,…，q)

(3) make dmin=min (d _jIf) satisfy condition

$\left\{\begin{array}{c}\left|\right|e_j\left|\right|=\mathrm{\ϵ},d_{\min }>\mathrm{\λ}\left(j\right)\\ \mathrm{\λ}\left(j\right)=\max ({\mathrm{\λ}}_{\max }{\mathrm{\γ}}^j,{\mathrm{\λ}}_{\min })\end{array}\right.,$ Error between i.e. expectation output and the actual output then increases a hidden layer neuron greater than setting accuracy, and in the following formula, ε is the precision of network expectation; λ _jBe j when input network fit precision, along with the carrying out of study, λ _jFrom λ _MaxBe reduced to λ _Min; γ is decay factor, 0＜γ＜1;

(4) if do not satisfy the condition of step 3, press the least square method of recursion regulating networks and connect weights;

(5) if all satisfy for N sample of continuous input δ is that predefined constant (getting N=100) is then with i Hidden unit deletion in the formula;

(6) j+1 sample of input repeats said process;

Through behind a series of learning trainings, obtain the health status model of accumulator.

The utility model is in order to satisfy the demand of data processing speed, and system has adopted the ARM9 micro controller module, and model is the S3C2440 microcontroller, and a series of peripherals are provided, comprise multichannel A/D modular converter, have sampling and keep function, satisfy system requirements.

Data storages adopt have easy to use, speed fast, connect flexibly, the USB device of plug and play and the characteristics such as bus-powered, its small and exquisite being easy to carry, memory data output is large, dependable performance and can reduce system cost.

Employing is controlled the man-machine close friend in interface based on the GUI development environment of Qt/Embedded application software, has improved the display effect of human-computer interaction interface.

Detection system has improved the reliability that accumulator parameter is measured greatly take the Linux of aboundresources, cheap, stable performance as operating system.

The S3C2440 microcontroller is finished the data acquisition of accumulator property parameter (voltage at accumulator two ends, the electric current in the circuit and temperature in the charge and discharge process), then by the soft computing method of neural network, carry out modeling for the accumulator detection system, thereby effectively solve battery health situation (SOH(State of health)) on-line monitoring.The problem that the neural network model of SOH exists input selection and the input space to divide, the SOH of the flash-over characteristic prediction battery of dependence short time, selection that must be correct can reflect fully that the measurement data of accumulator SOH is as input and output.Select electric current, temperature in the charge-discharge circuit as the input of model by comparative analysis, judge the performance state of accumulator according to the magnitude of voltage of the input quantity magnitude of voltage of model and standard.

The learning process of RBF neural network is divided into two stages.Phase one, all input samples are carried out the K-mean cluster, calculate the Gaussian function central value C of each node of hidden layer _jWith generalized constant δ _jSubordinate phase after definite hidden layer parameter, is utilized least square method according to sample, obtains the weights Wi of output layer.Then, according to sample, again the parameter of hidden layer and output layer is proofreaied and correct, with the approximation accuracy of further raising network.

Accumulator is discharged and recharged experiment, gets T constantly at whole charge and discharge process, measure each constantly voltage at accumulator two ends, obtain the magnitude of voltage of each accumulator constantly:

Ui＝{Ui∈R,i＝1,2,…,T}

Top n by sequence is each value constantly of M behind the value prediction constantly, and then always to organize number be k=T-M-N+1 to sample, can be with the input as the RBF neural network of the top n value of each sample, and rear M value exported as target.

By RBF network struction and the training to sample, realize from input sample R ^NTo output sample R ^MMapping, and finally realize the monitoring of accumulator health status.

Data after will processing at last and the health status of accumulator store in the USB device and carry inconvenience, measure inflexible deficiency to solve the traditional storage battery measuring equipment, the in addition friendly visualization interface based on Qt/Embedded, it is more flexible that accumulator parameter is measured, eliminated as the needs of system configuration additional devices, be applicable to portable equipment.

Multichannel AD modular converter, this module has extensibility, can scan hyperchannel to change or single channel is changed continuously, and reference voltage adopts the 3.3V canonical reference voltage of processor inside, and is stable and reliable for performance.

USB module among the S3C2440 that data memory module adopts has two USB mouths in the design, a mouth is USB A mouth, be mainly used in connecting the equipment such as USB flash disk, MP3, another one is USB B mouth, and the USB extended line by standard is connected with PC, mainly finishes data transfer.

It is operating system that host computer adopts aboundresources, Linux cheap, stable performance, and the driving collection of embedded chip commonly used is provided, and supports a large amount of peripheral hardware devices, drives and enriches, and has the kernel of simplifying, and has greatly improved the efficient of control system.By Qt/Embedded Software for Design graphical interfaces, finish the functions such as demonstration of control detection system and data.

The beneficial effects of the utility model are: the utility model is applied to high-performance 32-bit microprocessor S3C2440 and embedded Linux system, Qt/Embedded in the design of the portable detection system in the actual industrial production and the research and development of bottom layer driving equipment, has certain feasibility and perspective.

Description of drawings

Fig. 1 is system construction drawing of the present utility model;

Fig. 2 is that shunt is measured current circuit figure;

Fig. 3 is system flowchart;

Fig. 4 is the step that the ARM microprocessor utilizes the Processing with Neural Network data.

Wherein, 1.ARM9 micro controller module, 2.A/D modular converter, 3. data acquisition module, 4. data memory module, 5. accumulator, 6. Collect conversion/control interface, 7.Qt/Embedded visualization interface, 8. rectification circuit, 9. current collection circuit, 10. temperature collection circuit, 11. amplifying circuit, 12. noise processing circuits.

Embodiment

Below in conjunction with accompanying drawing and embodiment the utility model is described further.

Among Fig. 1, the utility model hardware mainly comprises ARM9 micro controller module 1, A/D modular converter 2, data acquisition module 3 and data memory module 4;

Wherein, data acquisition module 3 gathers voltage, electric current and the temperature signal of accumulator 5, and it can be that the ARM9 micro controller module is inner integrated to send into A/D modular converter 2(, also can be external);

Multichannel A/D modular converter 2 is changed simulating signal, then is connected with ARM9 micro controller module 1 through Collect conversion/control interface 6;

ARM9 micro controller module 1 is processed the information that gathers, and is controlled by 6 pairs of data acquisition modules 3 of Collect conversion/control interface by Qt/Embedded visualization interface 7; After finishing the data acquisition and data processing of voltage, the electric current in the circuit and temperature that accumulator 5 performance parameters are accumulator 5 two ends in the charge and discharge process, store in the external USB device;

Data memory module 4 is USB interface, is connected with the USB device of outside.

Described data acquisition module 3 comprises rectification circuit 8, current collection circuit 9, temperature collection circuit 10, amplifying circuit 11 and noise processing circuit 12; Wherein,

Noise processing circuit 12 is removed the noise signal in accumulator 5 signals that gather, and is connected with rectification circuit 8, current collection circuit 9, temperature collection circuit 10, amplifying circuit 11;

Rectification circuit 8 is single phase bridge type rectifier circu, finishes the processing to the voltage signal at accumulator 5 two ends in accumulator 5 charge and discharge process; Then the voltage signal after will processing directly is sent to A/D modular converter 2;

Current collection circuit 9 is shunt, and the magnitude of voltage of shunt output is 0 ~ 75mV;

Temperature collection circuit 10 is temperature sensor;

Current collection circuit 9, temperature collection circuit 10 are connected with A/D modular converter 2 through amplifying circuit.

The S3C2440(ARM920T kernel that micro controller module adopts Samsung company to release).This microcontroller has the characteristics such as low-power consumption, high-performance, is specially adapted to portable equipment.The inner integrated A/D converter of 8 passages does not namely need to add the Real-time Collection that the A/D chip just can complete paired data.Meanwhile, S3C2440 provides a series of complete system peripherals equipment, supports the operating systems such as WinCE, Linux, has eliminated to be the needs of system configuration additional devices, has reduced the cost of whole system; Have touch screen interface, provide good resource for the virtual frames of data shows; Has the usb host interface and the USB device interface that make things convenient for data transmission and communication between devices.

Data acquisition is mainly finished the collection of voltage, electric current and temperature signal and analog to digital conversion etc.In the data acquisition, for preventing the confusion of data, the A/D modular converter is elected hyperchannel as, and each only has one road signal to be in on-state constantly, has reduced the phase mutual interference between the signal, improves the reliability of signals collecting.Behind the EOC directly with data reading, obtain corresponding digital quantity, this digital quantity is sent to ARM9 micro controller module 1 by Collect conversion/control interface 6, then ARM9 micro controller module 1 is by Collect conversion/control interface 6 control frequency acquisitions, the mode that adopts multi collect to average in the system, the error of reduction system.In order to improve the dirigibility of system.

The noise processed module 12 main noises of removing guarantee distortionless each passage that is sent to A/D converter of signal.

Current rectifying and wave filtering circuit is selected single phase bridge type rectifier circu, in the accumulator charging/discharging process, though there has been rectification circuit in the circuit, still there is small fluctuation in voltage, so need to add current rectifying and wave filtering circuit in the circuit.This main circuit will be finished the processing to the voltage signal at battery two ends in the accumulator charging/discharging process, to the single-unit accumulator, the voltage signal at two ends is 2V ~ 2.4V, for improving the stability of voltage signal, needing the one-level feedback circuit in the circuit is that voltage follower is cushioned voltage, while noise isolation signal, the respective channel that the voltage signal after then will processing directly is sent to A/D converter gets final product;

In the accumulator charging/discharging process, current ratio is larger in the circuit, therefore adopts the electric current in the lower-cost shunt metering circuit, and circuit as shown in Figure 2.The magnitude of voltage of shunt output is 0 ~ 75mV, needs signal is amplified processing for improving acquisition precision, and amplification chip adopts AD620, and according to actual needs, enlargement factor is set to 25, and the signal after then will amplifying is sent to the respective channel of A/D converter;

To temperature signal, temperature sensor is selected PT100, and the voltage signal after the three-wire system circuit conversion is fainter, need to amplify processing, and enlargement factor elects 100 as, and the signal after then will amplifying is sent to the respective channel of A/D converter.

Adopt wave filter that the noise rate in the signal is removed in the signal processing circuit, the wave filter that adopts here is phase locked loop filter.

Data memory module 4 adopts the USB module.For the easily passage between S3C2440 and USB device and the main frame is provided, system USB Device interface circuit.System can identify USB device automatically, has avoided the leakage of data to deposit.

Among Fig. 3, the concrete steps of the utility model method of work are:

(a) collection of the utility model performance parameter is obtained mainly and is finished by 3 among Fig. 1,8,9 parts, then the data that collect being delivered to data processing circuit processes, mainly finishing amplification, the filtering of current signal, temperature signal processes, the functions such as the rectification of voltage signal, filtering guarantee the distortionless AD converter interface that is sent to of signal.

(b) transplanting of (SuSE) Linux OS is installed.(SuSE) Linux OS has good user interface, abundant network function, reliably security of system, good advantages such as portable, so that application program has good expandability, for further exploitation is provided convenience with research.Consider operating position and compatibility, what select in the utility model is Fedora10 operating system, and the disk space that this operating system needs is smaller, and the multiple driving that needs can be installed simultaneously.First (SuSE) Linux OS is transplanted after the host computer initialization in the TQ2440 development board, for the exploitation of driver and installation and signals collecting provide required interface and environment.

(c) exploitation of driver and installation.(SuSE) Linux OS has the operate of three types, that is: character type, block device and the network equipment.AD equipment belongs to character type equipment, in device driver, mainly finish " open " of AD equipment, " close ", " read ", respective function when the operation such as " write " produces, in " read " function with the data COPY among the ADDATA in the user storer, be convenient to reading out data in the application program; " write " mainly finishes the setting to the control word of ADCON register, comprises setting of acquisition channel and frequency acquisition etc.The entry address of these functions is positioned in " AD_file_operations " structure in the program.

After writing linux system A/D and USB driver, define implementor name, compile appearance as file zImage.bin by the make order, programming is in the kernel of built-in Linux, namely finished the dynamic load of driver, in application program, just can carry out read-write operation to equipment by functions such as open, read.The support that layer with the end drives, the user can write application program according to the needs of oneself, does not need the hardware of S3C2440 is configured again.

(d) design at data demonstration and interface.The design of the visual friendly interface of the utility model adopts Qt/Embedded under the Linux as the interface development platform, and the data of system acquisition and processing are shown and the control work of system.Qt/Embedded adopts C Plus Plus, can be directly and the Linux-C hybrid programming, and convenient and swift.In (c), the driver of equipment is loaded in the kernel, provides interface function for the combination of Qt/Embedded and Linux-C and to the read-write operation of special device file.Therefore just can open equipment in the design of Qt/Embedded upper computer software, the operation such as read and write.

Human Machine Interface is based on main by the demonstration of various data in the assembly completion systems such as button, tableWidget and edit box and the control operation of each equipment in the Graphic Design of Qt/Embedded.In order to make graphical interfaces more friendly visual, adopted the mode of Chinese and English combination, for staff's operation and use provides convenience.Carry out in the demonstration of data at tableWidget, for improving the clear readability of data, the data with up-to-date collection are presented in the screen all the time.By program design, the position of real-time control tableWidget.

By gathering and VECTOR pushbutton enable detection system, gather battery condition parameter (voltage, electric current and temperature) collection control/translation interface and deliver to the ARM microprocessor, data are carried out obtaining actual value after the conversion process, process for late time data and prepare.

(e) utilize the RBF neural network to carry out step that data process as shown in Figure 4 in the ARM microprocessor.

Step 1 is determined the input/output argument of neural network model.The utility model at Correct Analysis after the influence factor of accumulator health status SOH, proposed with electric current, time and temperature are input, voltage is the RBF neural network of output, in order to estimate the health status of accumulator, concrete evaluation criterion: for charging process, adopt certain electric current to give charge in batteries, according to the charge value Q that is filled with accumulator after the Δ t time and current temperature value T, obtain the magnitude of voltage of current accumulator, then with neural network model in the magnitude of voltage Uo that exports relatively, if the difference of magnitude of voltage surpasses 0.5V, think that then this accumulator out of condition, need to do further processing, otherwise normal; Same discharge process also is like this.

Step 2, with the various parameters in the accumulator charging/discharging process of different health status as study and training sample, gather voltage, electric current, temperature and time signal that many groups are in accumulator each stage in charge and discharge process of different health status, as the training level of neural network signal, the correct battery model of setting up.

Concrete study and training process are as follows:

Input variable is

x _j=[x ₁, x ₂] ^T=[i _j, ψ _j] ^T(formula 1)

The hidden layer radial basis function adopts Gauss's activation function

$u_i=\exp [-\frac{{\left|\right|x-c_i\left|\right|}^2}{2{{\mathrm{\σ}}_i}^2}](i=\mathrm{1,2},\·\·\·,q)$ (formula 2)

In the formula, u _iBe the output of i hidden node; c _iIt is the generalized constant of i hidden node; Q is the hidden layer node number; X=(x ₁, x ₂) ^TBe the input sample; c _iBe the center vector of i hidden node Gaussian function, this vector is a column vector identical with the dimension of input sample x.

The magnitude of voltage at accumulator two ends is as the output of RBF neural network

$y=U=\underset{i=1}{\overset{q}{\mathrm{\Σ}}}{w}_i{u}_i-U_i$ (formula 3)

In the formula, y is the output of RBF neural network, w _iBe the weighting coefficient of hidden layer to output layer; U _iThreshold value for hidden layer.For j sample, (x _j, y _j) specifically learning procedure is as follows:

(1) presses respectively the output u that (formula 2) and (formula 3) calculates each hidden layer unit of RBF neural network _i(x _i) and the output y of network _j

(2) error between the output of calculating actual acquisition accumulator both end voltage and RBF neural network || e _j||=|| U-y _j|| and the distance of sample and already present hidden layer unit

d _j＝||x _j-c _j|| (j＝1,2,…，q)

(3) make dmin=min (d _jIf) satisfy condition

$\left\{\begin{array}{c}\left|\right|e_j\left|\right|=\mathrm{\ϵ},d_{\min }>\mathrm{\λ}\left(j\right)\\ \mathrm{\λ}\left(j\right)=\max ({\mathrm{\λ}}_{\max }{\mathrm{\γ}}^j,{\mathrm{\λ}}_{\min })\end{array}\right.,$ Error between i.e. expectation output and the actual output then increases a hidden layer neuron greater than setting accuracy, and in the following formula, ε is the precision of network expectation; λ _jBe j when input network fit precision, along with the carrying out of study, λ _jFrom λ _MaxBe reduced to λ _Min; γ is decay factor, 0＜γ＜1.

(4) if do not satisfy the condition of step 3, press the least square method of recursion regulating networks and connect weights.

(5) if all satisfy for N sample of continuous input

δ is that predefined constant (getting N=100) is then with i Hidden unit deletion in the formula.

(6) j+1 sample of input repeats said process.

Through behind a series of learning trainings, obtain the health status model of accumulator.

Step 3, the performance parameter of the collection accumulator of real-time online according to the model of setting up in the step 2, analyzes the health status SOH of accumulator, and to the problem accumulator, system's processing of reporting to the police is preserved warning message simultaneously simultaneously and is shown.To the problem accumulator, the later stage does further detection again and judges.

(f) data storage.After data were handled, with in the ini file of deposit data in the USB device, data comprised the health status SOH of data acquisition time Date, voltage Voltage, electric current Current, temperature T emperature and accumulator by the Config_write function.

Check historical data by the historical data button, the Config_read function in USB device in the ini file sense data be shown among the tableWidget, the technician can check the performance state of the accumulator that certain time point gathers clearly, to data operation and analysis.

Claims (2)

1. a portable storage battery condition detecting system is characterized in that, it comprises ARM9 micro controller module, A/D modular converter, data acquisition module and data memory module;

Wherein, the voltage of data collecting module collected accumulator, electric current and temperature signal are sent into the A/D modular converter;

Multichannel A/D modular converter is changed simulating signal, then is connected with the ARM9 micro controller module through Collect conversion/control interface;

The ARM9 micro controller module is processed the information that gathers, and by Qt/Embedded visualization interface and Collect conversion/control interface the data acquisition module is controlled; After finishing the accumulator property parameter and be the data acquisition of voltage, the electric current in the circuit and temperature at accumulator two ends in the charge and discharge process, the information that gathers is processed the health status that detects accumulator, the health status of final accumulator is shown and store in the external USB device;

Data memory module is USB interface, is connected with the USB device of outside.

2. portable storage battery condition detecting system as claimed in claim 1 is characterized in that, described data acquisition module comprises rectification circuit, current collection circuit, temperature collection circuit, amplifying circuit and noise processing circuit; Wherein,

Noise processing circuit is removed the noise signal in the accumulator signal that gathers, and is connected with rectification circuit, current collection circuit, temperature collection circuit, amplifying circuit;

Rectification circuit is single phase bridge type rectifier circu, finishes the processing to the voltage signal at battery two ends in the accumulator charging/discharging process; Then the voltage signal after will processing directly is sent to the A/D modular converter;

Current collection circuit is current divider circuit, and the magnitude of voltage of shunt output is 0 ~ 75mV;

Temperature collection circuit is temperature sensor;

Current collection circuit, temperature collection circuit are connected with the A/D modular converter through amplifying circuit.

Images