CN201255729Y - Wood material moisture percentage intelligent detecting device based on multi-sensor data fusion - Google Patents
Wood material moisture percentage intelligent detecting device based on multi-sensor data fusion Download PDFInfo
- Publication number
- CN201255729Y CN201255729Y CNU2008200909781U CN200820090978U CN201255729Y CN 201255729 Y CN201255729 Y CN 201255729Y CN U2008200909781 U CNU2008200909781 U CN U2008200909781U CN 200820090978 U CN200820090978 U CN 200820090978U CN 201255729 Y CN201255729 Y CN 201255729Y
- Authority
- CN
- China
- Prior art keywords
- moisture content
- temperature
- module
- data
- wood
- 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.)
- Expired - Fee Related
Links
Landscapes
- Investigating Or Analyzing Materials By The Use Of Electric Means (AREA)
Abstract
The utility model discloses a wood moisture content intelligent inspection device based on multi-sensor data fusion, which adopts multiple sensors to inspect the temperature, balance moisture content and wood moisture content and the like of a wood drying kiln, and adopts a signal adjusting module and a data acquisition module to input the equivalent electric signals of the temperature and the wood moisture content and the like as a data fusion model, and confirms the structure and algorism of neural network fusion module under training mode, to fuse and output wood moisture content in drying process, while a host computer module can real-timely display the temperature, balance moisture content and wood moisture content of the kiln. The utility model adopts multi-sensor data fusion technique, to eliminate the coupling between temperature and wood moisture content, thereby accurately, reliably and intelligently inspecting wood drying parametes based on the prior inspection instruments of wood drying plants, improving inspection accuracy and reliability, reducing wood consumption caused by electric consumption and drying faults, and improving labor productivity.
Description
Technical field
The utility model provides a kind of moisture content intelligent detection device that merges based on multi-sensor data, is mainly used in the detection of moisture content in the timber drying in the wood processing field.
Background technology
The world today is being faced with the forest reserves and is reducing environmental protection and the ecological problem that is brought day by day, and for protecting the earth of human mutual survival, most countries has all taked limit to cut down policy to the forest reserves (particularly wildwood) in the world.The national dependence on import material of some scarce materials that comprises China will be restricted day by day with the situation of covering the shortage.Effectively utilizing limited timber resources seems and becomes more and more important.In recent years, because timber has the function of many excellences and can satisfy the requirement that people go back to nature, it is widely used at building, interior decoration, furniture, car and boat, traffic and transport field, the requirement cumulative year after year of timber.In the face of the such Shaolin country of China, how to improve the timber usability better and improve its utilization factor, become pendulum one of problem demanding prompt solution in face of the timber scientific worker.
Timber will utilize, and at first will carry out dried.In timber drying, moisture content reaches the final purpose that the drying process requirement is the drying of wood, also is to guarantee that drying of wood final products satisfy important quality control means of customer requirements.And moisture content still is one of main feedback quantity of timber drying control, and the accuracy that moisture content detects will directly have influence on the quality of wood drying quality, the height of drying cost and the length of drying cycle.Therefore, the accuracy problem of its measuring accuracy has become one of forward position research topic of domestic and international wood-processing industry extensive concern.
Chinese patent NO.98243294.1 discloses a kind of moisture content measurement mechanism that is used for the timber dryer system, its circuit structure comprises measuring probe, logarithmic amplifier, removes memory circuit, V/F change-over circuit, photoelectric isolating circuit, interface and power circuit, this utility model is amplified, is eliminated the electricity memory and adopt V/F (voltage/frequency) conversion measure by adopting logarithm, improves resolution and precision that moisture content is measured.But this measurement mechanism does not have to consider the complicacy and the uncertainty of duty in lumber kiln, and from the conductive microstructure mechanism of timber, the moisture content value is main relevant with temperature and seeds; From the macro-test process, moisture content also with dry kiln in temperature, equilibrium moisture content, air-flow wind speed envirment factor exist and intercouple and correlationship.No matter from microcosmic or macroanalysis, temperature all is the leading factor that influences the moisture content measuring accuracy.Single moisture content measurement mechanism can not make full use of multi-sensor data resource in the dry kiln, can not optimal combination multiple sensors redundancy or complementary information, can not solve the influence of envirment factor, thereby make raising moisture content accuracy of detection be subjected to certain restriction the moisture content measuring accuracy.
Summary of the invention
The purpose of this utility model is to overcome the deficiency of prior art, a kind of moisture content intelligent detection device that merges based on multi-sensor data is provided, by gathering temperature, equilibrium moisture content and the moisture content electric signal of different measuring point in the dry run, utilize data fusion method, improve validity, reliability, accuracy, robustness that moisture content detects effectively.
In order to achieve the above object, the moisture content intelligent detection device that merges based on multi-sensor data that provides of the utility model comprises sensor assembly, signal condition module, data acquisition module, neural Network Data Fusion module and host computer display module.Temperature in the lumber kiln, equilibrium moisture content, moisture content parameter detect, export and each measured corresponding electric signal through sensor assembly; Electric signal is through the amplification of signal condition module, filtering input data acquisition module; Data acquisition module comprises dsp controller, serial communication interface, light-coupled isolation.Dsp controller is the simulating signal of signal condition module input that digital signal is delivered to master controller through analog to digital conversion as data processing core, and the master controller chip distributes the address to patrol survey automatically to temperature, humidity, moisture content parameter.Temperature, equilibrium moisture content need be calculated, tabling look-up obtains, and are sent to the real-time video data of host computer display module by the serial communication interface of primary controller.
The neural Network Data Fusion module with the multi-source sensor information in the dry kiln as input, select neural network model structure and learning algorithm according to system requirements and data characteristics, the nerve network system of being set up is carried out off-line learning, study is also adjusted weights, until training error to be reduced to the scope of permission, determine the connection weights and the syndeton of network, this Fusion Model is recorded in the database.Do in the process the actual drying firer, the temperature equivalent electric signal and the moisture content equivalent electric signal of data acquisition module output are input to the neural Network Data Fusion module, through normalized, the weights of gained behind the network training are written in the data fusion model, merge output moisture content value, deliver to the host computer display module and show the moisture content value in real time.
The host computer display module shows drying of wood kiln temperature, the equilibrium moisture content data by data acquisition module output in real time, and merge the moisture content value of output, and show that simultaneously drying regime, working time, drying schedule stage, data ultra-range in the kiln report to the police by the neural Network Data Fusion module.
The utility model has the advantages that this intelligent detection device has solved the environment main factor in the lumber kiln effectively---the relevant and coupled relation between temperature and the moisture content.Having overcome single-sensor can only provide the local message of the drying of wood, can not reflect the defective that timber drying and moisture content change comprehensively, because the information category of gathering is many, anti-jamming capacity is strong, also is not easy to cause because collecting error message the detection error of moisture content.The utility model adopts sensor Data Fusion, on the basis of the existing operation in timber drying plant measuring instrument, realization is to accurate, the reliable intellectual monitoring of drying of wood parameter and moisture content variation, having improved the sensing detection unit provides accuracy and the reliability of data, realization with any single-sensor can't realize to comprehensive, the high-quality intellectual monitoring of timber drying, reduced power consumption and, improved labour productivity because of seasoning defect drying defect causes lumber consumption.Simultaneously, also can provide for the Automatic Control of timber drying accurately, timely foundation.
Description of drawings
The moisture content intelligent detection device synoptic diagram that Fig. 1 merges based on multi-sensor data.
Fig. 2 multilayer perceptron structure and single neuron model synoptic diagram.
Embodiment
Below in conjunction with drawings and Examples embodiment of the present utility model is further described.
The moisture content intelligent detection device that merges based on multi-sensor data that the utility model provides comprises sensor assembly, signal condition module, data acquisition module, neural Network Data Fusion module and host computer display module.In dry run is used, the sets of temperature sensors in the sensor assembly and equilibrium moisture content group are settled the centre position of every block check plate in air intake, outlet air end and the material heap of dry kiln respectively, totally eight sensor groups; A block check plate is respectively settled in the material heap upper, middle and lower of the air intake in lumber kiln, outlet air end both sides, totally six moisture content sensor probes.In order to make the moisture content sensor fully contact test specimen, and assurance measuring accuracy, in utility model, select needle electrode for use---the steel nail probe is as the moisture content sensor, on moisture content breadboard in the steel nail insertion dry kiln, probe must be away from the plate end, away from defects in timber, insertion water percentage, thickness, the representational plank of seeds, the representational plank in the insertion position of plank in kiln.With plug the steel nail probe is linked to each other with cable on the dry kiln wall, cable lead-in wire kiln discharge is connected with the signal condition module.This utility model electrifying startup, temperature detection and wet and dry bulb temperature testing circuit are selected the three-wire system temperature measurement circuit for use, and the thermometric amplifier adopts the instrumentation amplifier CA3140 of high input impedance, high cmrr to eliminate channel difference; Analog switch selects for use the 74HC4051 (high-speed cmos simulant electronic switch) of TEXAX Instruments company to import multi way temperature signal and wet and dry bulb temperature signal successively.The moisture content testing circuit adopts the white steady nucleus amplifier ICL7650 of high precision copped wave, cooperates a slice CD4052 to constitute programmable amplifier (programmable amplifier), to the water percentage areal survey.Analog switch is selected the 74HC4053 sequential sampling moisture content signal of TEXAX Instruments company for use, and the equivalent resistance of moisture content is converted into voltage signal.Temperature, equilibrium moisture content, moisture content parameter detect, export and each measured corresponding voltage signal through sensor assembly; Voltage signal is through the amplification of signal condition module, filtering input data acquisition module; The serial communication interface of data acquisition module is input to the neural Network Data Fusion module to temperature equivalent voltage data and moisture content equivalent voltage data, merges the output moisture content.The host computer display module shows temperature, equilibrium moisture content and the moisture content parameter value that comes from data acquisition module and neural Network Data Fusion module in real time.
Multi-source sensor information in the dry kiln is at first carried out pre-service through normalization as the input of neural Network Data Fusion model, makes inhomogeneous data have same or analogous form to realize data fusion.The foundation of neural Network Data Fusion model at first will be selected the BP neural network model according to the requirement and the data characteristic of system, and typical B P network structure is shown in Fig. 2 (A).Network also has one or more layers hidden layer node except that the input and output node, with in the node layer without any coupling.2 (B) have provided a neuronic model, and neuronic input/output relation determines that by basis function u () and excitation function f () neuronic excitement levels can represent that the connection weights between the neuron are represented by w with θ.Merge knowledge according to existing multi-source data and system and adopt certain learning method, the nerve network system of being set up is carried out off-line learning, determine the connection weights and the syndeton of network, at last the network that obtains is used in the middle of the data fusion of real system.For the training of multilayer perceptron, select error backpropagation algorithm usually for use.Based on the neural Network Data Fusion modeling scheme two kinds of working methods are arranged: training mode and mode of operation.Therefore, corresponding data set also comprises two parts: training set and test set.In training mode, neural network input comprises the temperature voltage value after moisture content magnitude of voltage, normalized equilibrium moisture content magnitude of voltage and the normalization after the normalization, training process is selected the BP algorithm for use, study is also adjusted weights, until training error to be reduced to the scope of permission, if this extended mode can not meet the demands, then change the function item number of times, and constantly adjust training set in error is reduced to desired scope.In case after this mathematical model was determined, the sensor of same kind can be used this model so, and this mathematical model is recorded in the database.In mode of operation, during promptly actual application, the weights of gained behind the network training are written in the neural network model, calculate the moisture content value after the fusion.
Choose ashtree and be the test seeds, the whole water percentage stage 18% is a test sample to 10%, the exact value of moisture content is by national standard (GB1931-91), the employing oven drying method obtains, the variation of ambient temperature scope is 25.2 ℃ to 80 ℃ in the kiln, based on the dry run of reality, measure the output voltage values of a moisture content sensor, the output voltage values and the kiln temperature (magnitude of voltage) of equilibrium moisture content sensor every half an hour, import as sample; By the quality (after the dry run end,, obtain the over dry quality of inspection panel, extrapolate water percentage at that time) of weight method records tests plate, export simultaneously as sample by oven dry.Through arrangement, every group of data are made up of the measured data of the examination material of the difference same kiln temperature under, obtain 138 groups of inputoutput datas altogether, and wherein 100 groups are used for the training of model, other 38 groups of tests that are used for model.For removing amount of redundancy, improving speed of convergence, make easier study of neural network and training, all sample inputoutput datas are carried out normalization, make its value interior.It is 16,31,5,12 to train that the utility model is in use successively got the number of hidden nodes, determines that at last the number of hidden nodes is 12, η=0.7, and α=0.8, e-learning 1000 times, the error mean square root is less than 7 * 10
-2, at this moment network training finishes.After specified data Fusion Model parameter, the equivalent voltage of the temperature of dry run, equilibrium moisture content, moisture content parameter is input to the neural Network Data Fusion module, merge output moisture content value, on the host computer interface, show.Simultaneously, the equivalent electric signal of temperature, equilibrium moisture content parameter is converted to temperature and balance water cut value through tabling look-up, and shows on the host computer interface.
Claims (6)
1, a kind of moisture content intelligent detection device that merges based on multi-sensor data, it is characterized in that this pick-up unit comprises sensor assembly, the signal condition module, data acquisition module, the neural Network Data Fusion module, the host computer display module, temperature in the lumber kiln, equilibrium moisture content, the moisture content parameter detects through sensor assembly, output and each measured corresponding electric signal, voltage signal amplifies through the signal condition module, filtering input data acquisition module, the serial communication interface of data acquisition module is input to the neural Network Data Fusion module to temperature and moisture content equivalent electric signal, merge the output moisture content, the host computer display module shows the temperature that comes from data acquisition module and neural Network Data Fusion module in real time, equilibrium moisture content and moisture content value.
2, according to the described moisture content intelligent detection device that merges based on multi-sensor data of claim 1, it is characterized in that described sensor assembly comprises sets of temperature sensors, equilibrium moisture content sensor groups, moisture content sensor groups, sensor assembly is converted to the equivalent electric signal to temperature, equilibrium moisture content, moisture content and inputs to the signal condition module.
3, according to the described moisture content intelligent detection device that merges based on multi-sensor data of claim 1, it is characterized in that described signal condition module is the temperature that comes from sensor assembly, equilibrium moisture content, moisture content equivalent electric signal amplifies, input to data acquisition module after the filtering, wherein temperature signal adopts high input impedance, the instrumentation amplifier of high cmrr amplifies, analog switch is selected for use the high-speed cmos simulant electronic switch to patrol successively and is surveyed the multi-point temp signal, the moisture content testing circuit adopts operational amplifier and analog switch to constitute programmable amplifier to the water percentage areal survey, and the equivalent resistance of moisture content is converted into electric signal.
4, according to the described moisture content intelligent detection device that merges based on multi-sensor data of claim 1, it is characterized in that described data acquisition module comprises light-coupled isolation, dsp controller, serial communication interface, data acquisition module is changed the data that come from the signal condition module through AD, intercom mutually with the serial ports of host computer by serial communication interface, temperature, moisture content equivalent electric signal input neural network data fusion module, simultaneously temperature, equilibrium moisture content data input host computer display module.
5, according to the described moisture content intelligent detection device that merges based on multi-sensor data of claim 1, it is characterized in that described neural Network Data Fusion module is input with temperature equivalent electric signal and the moisture content equivalent electric signal that comes from data acquisition module, merge the output moisture content through neural network, show the moisture content value that distributes in the lumber kiln by the host computer display module in real time.
6, according to the described moisture content intelligent detection device that merges based on multi-sensor data of claim 1, it is characterized in that described host computer display module shows drying of wood kiln temperature, the equilibrium moisture content data by data acquisition module output in real time, and merge the moisture content value of output, and show that simultaneously drying regime, working time, drying schedule stage, data ultra-range in the kiln report to the police by the neural Network Data Fusion module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNU2008200909781U CN201255729Y (en) | 2008-09-26 | 2008-09-26 | Wood material moisture percentage intelligent detecting device based on multi-sensor data fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNU2008200909781U CN201255729Y (en) | 2008-09-26 | 2008-09-26 | Wood material moisture percentage intelligent detecting device based on multi-sensor data fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN201255729Y true CN201255729Y (en) | 2009-06-10 |
Family
ID=40739053
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNU2008200909781U Expired - Fee Related CN201255729Y (en) | 2008-09-26 | 2008-09-26 | Wood material moisture percentage intelligent detecting device based on multi-sensor data fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN201255729Y (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102033043A (en) * | 2010-10-19 | 2011-04-27 | 浙江大学 | Grain moisture content detecting method based on hyperspectral image technology |
| RU2444725C2 (en) * | 2010-03-11 | 2012-03-10 | Государственное образовательное учреждение высшего профессионального образования "Тамбовский государственный технический университет" ГОУ ВПО ТГТУ | Method of estimating moisture content of paste-like material when drying in roll-band drier |
| CN102564106A (en) * | 2011-12-05 | 2012-07-11 | 东北林业大学 | Wood drying remote monitoring system based on DSP (digital signal processor) |
| CN102072922B (en) * | 2009-11-25 | 2013-04-03 | 东北林业大学 | Particle swarm optimization neural network model-based method for detecting moisture content of wood |
| CN104391468A (en) * | 2014-12-05 | 2015-03-04 | 芜湖中艺企业管理咨询有限公司 | Internet of Things based intelligent control system for drying kiln |
| CN104713910A (en) * | 2015-02-17 | 2015-06-17 | 合肥艾瑞德电气有限公司 | Resistive grain moisture measurement system |
| CN109724398A (en) * | 2019-02-02 | 2019-05-07 | 北京木业邦科技有限公司 | A kind of drying of wood control method and device based on artificial intelligence |
| CN110044977A (en) * | 2019-05-07 | 2019-07-23 | 中山市武汉理工大学先进工程技术研究院 | A kind of sandstone aggregate detecting device for moisture content and method |
| CN112686303A (en) * | 2020-12-29 | 2021-04-20 | 辽宁工程技术大学 | Intelligent wood warehousing system |
-
2008
- 2008-09-26 CN CNU2008200909781U patent/CN201255729Y/en not_active Expired - Fee Related
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102072922B (en) * | 2009-11-25 | 2013-04-03 | 东北林业大学 | Particle swarm optimization neural network model-based method for detecting moisture content of wood |
| RU2444725C2 (en) * | 2010-03-11 | 2012-03-10 | Государственное образовательное учреждение высшего профессионального образования "Тамбовский государственный технический университет" ГОУ ВПО ТГТУ | Method of estimating moisture content of paste-like material when drying in roll-band drier |
| CN102033043A (en) * | 2010-10-19 | 2011-04-27 | 浙江大学 | Grain moisture content detecting method based on hyperspectral image technology |
| CN102564106A (en) * | 2011-12-05 | 2012-07-11 | 东北林业大学 | Wood drying remote monitoring system based on DSP (digital signal processor) |
| CN104391468A (en) * | 2014-12-05 | 2015-03-04 | 芜湖中艺企业管理咨询有限公司 | Internet of Things based intelligent control system for drying kiln |
| CN104713910A (en) * | 2015-02-17 | 2015-06-17 | 合肥艾瑞德电气有限公司 | Resistive grain moisture measurement system |
| CN109724398A (en) * | 2019-02-02 | 2019-05-07 | 北京木业邦科技有限公司 | A kind of drying of wood control method and device based on artificial intelligence |
| CN110044977A (en) * | 2019-05-07 | 2019-07-23 | 中山市武汉理工大学先进工程技术研究院 | A kind of sandstone aggregate detecting device for moisture content and method |
| WO2020223998A1 (en) * | 2019-05-07 | 2020-11-12 | 中山市武汉理工大学先进工程技术研究院 | Device and method for detecting water content of sand and gravel aggregate |
| CN112686303A (en) * | 2020-12-29 | 2021-04-20 | 辽宁工程技术大学 | Intelligent wood warehousing system |
| CN112686303B (en) * | 2020-12-29 | 2024-04-02 | 辽宁工程技术大学 | Intelligent wood warehousing system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN201255729Y (en) | Wood material moisture percentage intelligent detecting device based on multi-sensor data fusion | |
| CN107389746B (en) | Grain pile moisture content on-line detection method and system | |
| CN102095658B (en) | The detection method of sheet cigarette quality and system | |
| CN110231111A (en) | A kind of contactless stress method of real-time of buried pipeline | |
| CN101576593B (en) | Array Type Inferior Insulator Local Electric Field Detection Device and Reverse Diagnosis Method | |
| CN105067125B (en) | Cannon high speed colorimetric infrared temperature measurement apparatus and method | |
| CN113188984B (en) | Intelligent monitoring system and method for corrosion state of steel bar in concrete | |
| CN101101265A (en) | Power transmission and transformation equipment outer insulation equivalent ash deposit density measuring method and device | |
| CN201576052U (en) | Insulation testing device with humidity compensation function | |
| CN208459884U (en) | A kind of accelerated degradation test system of electric energy meter | |
| CN201749100U (en) | Device for testing moisture content of wood | |
| CN105277111A (en) | Star arrow locking device strain monitoring system | |
| CN101949825A (en) | Leaf water near infrared non-destructive testing device and method in light open environment | |
| CN201653954U (en) | Lumber drying process parameter automatic detection device | |
| CN101799463A (en) | Intelligent fault diagnosis device for gas sensor | |
| CN201069467Y (en) | Testing platform for battery manager | |
| CN103558177B (en) | Raw meat organizes moisture near-infrared detection apparatus and detection method | |
| CN205373755U (en) | Wireless monitoring system of chicken coop environment based on zigBee network | |
| CN2872383Y (en) | Multi-station tester according to power-supplier standard and its multiplexer | |
| CN105445331A (en) | Concrete material apparent crack monitoring and detecting system, and monitoring and detecting method thereof | |
| CN106225842A (en) | A kind of dust pollution online monitoring system | |
| CN111796063A (en) | Method for automatically measuring response time of mining gas sensor | |
| CN213689329U (en) | Self-powered environmental corrosion monitoring system for wind power equipment | |
| CN109458306A (en) | A kind of power characteristic of wind driven generator set test macro based on DSP | |
| CN205718833U (en) | A kind of Structural Engineering safety automation monitoring system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090610 Termination date: 20130926 |