CN116893644B - Remote intelligent control method and system for water conservancy gate - Google Patents

Remote intelligent control method and system for water conservancy gate Download PDF

Info

Publication number
CN116893644B
CN116893644B CN202311149130.7A CN202311149130A CN116893644B CN 116893644 B CN116893644 B CN 116893644B CN 202311149130 A CN202311149130 A CN 202311149130A CN 116893644 B CN116893644 B CN 116893644B
Authority
CN
China
Prior art keywords
garbage
water
gate
transmission
flow
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
CN202311149130.7A
Other languages
Chinese (zh)
Other versions
CN116893644A (en
Inventor
刘树军
孙晓明
郭东纪
刘振新
刘振潇
仇鹏
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.)
Dongping Xintong Floating Bridge Co ltd
Original Assignee
Dongping Xintong Floating Bridge Co ltd
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 Dongping Xintong Floating Bridge Co ltd filed Critical Dongping Xintong Floating Bridge Co ltd
Priority to CN202311149130.7A priority Critical patent/CN116893644B/en
Publication of CN116893644A publication Critical patent/CN116893644A/en
Application granted granted Critical
Publication of CN116893644B publication Critical patent/CN116893644B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00Programme-control systems
    • G05B19/02Programme-control systems electric
    • G05B19/04Programme control other than numerical control, i.e. in sequence controllers or logic controllers
    • G05B19/042Programme control other than numerical control, i.e. in sequence controllers or logic controllers using digital processors
    • G05B19/0423Input/output
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00Program-control systems
    • G05B2219/20Pc systems
    • G05B2219/23Pc programming
    • G05B2219/23051Remote control, enter program remote, detachable programmer

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Filtration Of Liquid (AREA)

Abstract

The application discloses a remote intelligent control method and a system for a water conservancy gate, which belong to the technical field of water conservancy gate control and comprise an acquisition module, wherein the acquisition module is used for acquiring water flow parameters, opening and closing state parameters and garbage state parameters of the downstream of the gate; the water flow collection module is used for collecting data from the inlet end of the gate and the water flow; the control module is used for controlling the opening and closing degree of the gate and the operation of the trash removal module; the communication module is used for data transmission; the central processing unit is used for calling data and instructions, carrying out gate state operation and carrying out garbage state detection operation. According to the invention, the gate opening, the actual flow data and the garbage state data are collected through the collecting module, the gate opening and the cleaning module are controlled, the dirt at the inlet end of the gate is cleaned, the influence on water flow is avoided, the garbage is judged through garbage identification, and the damage to the device is avoided.

Description

Remote intelligent control method and system for water conservancy gate
Technical Field
The application relates to the technical field of water conservancy gate control, in particular to a remote intelligent control method and system for a water conservancy gate.
Background
In hydraulic engineering, a dam and the like are often adopted for water storage and water discharge, the dam is used for controlling the opening and closing of a water outlet through a gate, the purpose of water storage and water discharge is achieved, and the opening of the gate is required to be controlled when water is discharged.
The prior art publication No. CN104635571B provides a gate remote control system and a remote control method thereof, the device comprises a remote upper computer, a gate motor control module and a power module, the control system collects and processes data of a gate sensor and a water level sensor, the data are stored after being added with time labels at set time intervals, and the date and time are provided by a clock chip.
The prior art schemes described above, although achieving the relevant advantageous effects by the prior art structure, still have the following drawbacks.
The gate front end can be through trash rack with rubbish and spot interception usually, avoids rubbish to block up the gate, so after the trash accumulated to the trash rack, can lead to the actual flow that the gate aperture corresponds to be less than predetermined flow, then can lead to the gate to need open and surpass the settlement aperture, after the actual flow of maximum aperture also is less than predetermined flow, can exist the gate and continue to open and lead to the condition of extrusion damage.
In the related art, the inventor considers that the front end of the gate normally intercepts the garbage and the dirt through the trash rack, so that the trash can be prevented from blocking the gate, after the trash accumulates on the trash rack, the actual flow corresponding to the opening of the gate is smaller than the preset flow, the gate needs to be opened to exceed the set opening, and when the actual flow of the maximum opening is also smaller than the preset flow, the gate continues to be opened to cause extrusion damage.
The document of prior art publication No. CN110872831a provides a rubbish filtering and salvaging device for water conservancy gate department of being convenient for, including top, transmission pivot, fixed axle, second motor, second gyro wheel and slider, the centre of top is provided with first motor, the both sides of first motor are provided with fixed pivot, and the inside of fixed pivot is provided with first lifter, the transmission pivot sets up in the top of fixed pivot, and installs the driving rope in the centre of transmission pivot, the fixed axle is fixed in the left side of fixed pivot, and the below of fixed axle is provided with the second lifter, the below of second lifter is provided with the third lifter, and installs first gyro wheel in the below of third lifter, the centre of first gyro wheel is fixed with first pivot. This be convenient for water conservancy gate department is with rubbish filtration and fishing device, the transmission pivot is through rotatory, lets first lifter reciprocate to make the salvaging part of machine can reciprocate, can rise when not using, avoid the machine bubble in water. Although the related beneficial effects can be achieved by the structure of the prior art, the following drawbacks still exist: the salvaging device salvages the garbage at the gate by using a net-shaped garbage collection box, the size of the garbage collection box is fixed, and when large garbage is generated and the size of the garbage is far larger than that of the garbage collection box, the device does not know whether the target garbage can damage the garbage or not; or if the garbage articles at the gate are heavier, the garbage collection box cannot smoothly salvage the garbage articles; if salvaging is carried out according to a preset working flow, the current device can be blocked, so that the device cannot operate, or damage or abrasion are caused to the device.
In view of the above, we propose a remote intelligent control method and system for a water conservancy gate.
Disclosure of Invention
1. The technical problem to be solved.
An object of the present application is to provide a remote intelligent control method and system for a water conservancy gate, which solve the technical problem set forth in the background art, and implement cleaning of dirt at the inlet end of the gate, avoiding influencing water flow, and avoiding damaging the device by identifying and judging garbage.
2. The technical proposal is that.
The technical scheme of this application provides a remote intelligent control system of water conservancy gate, contains.
The collection module is used for collecting water flow parameters at the downstream of the gate, opening and closing state parameters of the gate and garbage state parameters.
And the trash removal module is used for removing trash from the inlet end and the water flow at the upstream of the gate according to the collected data.
The control module is used for controlling the opening and closing degree of the gate and the operation of the trash removal module.
And the communication module is used for data transmission.
And the central processing unit is used for calling data and instructions, carrying out operation processing on the gate state and carrying out operation processing on garbage state detection.
The central processing unit is provided with a picture processing unit, a garbage identification model and an early warning unit; the picture processing unit performs image graying, filtering denoising, gradient sharpening and image normalization processing on the acquired picture; the garbage identification model is used for comparing and identifying the pictures processed by the picture processing unit and judging whether garbage in the pictures contains garbage which can cause damage to the device or not; if the device contains garbage which can cause damage to the device, the device is stopped, the photo is sent to the monitoring terminal, and the early warning unit sends out an alarm through a wireless signal to remind workers of processing as soon as possible.
The acquisition module comprises an opening sensor, a flow sensor and a camera, wherein the opening sensor is arranged on the gate and is used for acquiring the opening and closing degree of the gate; the flow sensor is arranged at the downstream of the gate and is used for collecting water flow data; a plurality of cameras and illuminating lamps are installed above the dam body, and the cameras are used for collecting garbage states. A plurality of cameras are arranged above the side of the water outlet on the dam body. The camera is an infrared thermal imaging camera with a laser range finder and has a waterproof function. A brightness sensor is fixedly arranged above the dam body, and the brightness sensor can enable the illuminating lamp to be started at night or when the light is bad, so that the effect of enhancing illumination is achieved; the inner side wall of the dam body is fixedly provided with a water level sensor.
Through above-mentioned scheme, through collection module collection gate aperture and actual flow data to control gate aperture and the module of decontaminating, clear up the dirty of gate entry end through the module of decontaminating, avoid influencing rivers.
Optionally, the cleaning module comprises a mounting plate arranged on the dam body, a water outlet is formed in the bottom of the dam body, and a gate for controlling the water outlet to open and close is arranged in the water outlet and is an automatic control gate; the trash rack is fixedly arranged at the inlet end of the water outlet, two rotating shafts are rotatably sleeved on the mounting plate, chain wheels are fixedly sleeved at two ends of each rotating shaft, chains are sleeved between the chain wheels in a transmission manner on the same side, a driving mechanism is mounted on the mounting plate and is in transmission connection with one rotating shaft, and a cleaning mechanism is fixedly mounted between the two chains.
Through the scheme, when the difference value between the actual flow and the theoretical flow is larger than the threshold value, the driving mechanism drives the chain to reciprocate, so that the cleaning mechanism moves from the position of the top of one chain wheel to the water outlet to brush the trash rack, then moves to the top of the other chain wheel to reversely move, and when the cleaning mechanism moves back and forth, the trash rack is brushed reciprocally, so that attached dirt is brushed off, water flow is facilitated, the dirt suspended in the water is carried by the cleaning mechanism to be collected from the water surface to the top of the chain wheel, the aim of cleaning a water area is fulfilled, and the water flow is ensured.
Optionally, actuating mechanism includes the motor, motor fixed mounting has the slip frame in the mounting panel, slip joint has the slip frame in the mounting panel, the top of slip frame and the equal fixed mounting of bottom inner wall have first rack, the output shaft of motor stretches into in the slip frame and fixed the cup joint half gear, half gear can intermittent type with two first racks meshing, the bottom fixed mounting of slip frame has the second rack, the mounting panel rotation of slip frame bottom has cup jointed the transmission shaft, fixed cup joint driven gear on the transmission shaft, driven gear meshing second rack, driven gear installs matched with drive wheel and drive belt with one of them pivot between, half gear and driven gear's drive ratio is 1:2, drive wheel and drive belt between transmission shaft and pivot are synchronous belt structure.
Through the scheme, when the difference value between the actual flow and the theoretical flow is larger than the threshold value, the control module controls the trash rack to clean the trash rack and clean the water area around the trash rack, the motor drives the semi-gear to continuously rotate during cleaning, the semi-gear is alternately meshed with the two first racks, so that the sliding frame is driven to slide reciprocally, the sliding frame drives the second racks to move reciprocally, the second racks drive the driven gears to rotate reciprocally, the driven gears drive the transmission shaft to rotate reciprocally, the transmission shaft drives the rotating shaft to rotate reciprocally through the transmission wheel with a large transmission ratio, the chain wheel drives the chain to reciprocate every time when the rotating shaft rotates reciprocally, the cleaning mechanism moves to the water outlet from the position of the top of one chain wheel to brush the trash rack, then moves to the top of the other chain wheel to reversely move again, and brushes the trash rack reciprocally when the cleaning mechanism moves back and forth, so that attached dirt brushes down, water flow is facilitated, and the dirt suspended in the water is carried out of the water surface by the cleaning mechanism to the top of the chain wheel to collect, and water flow is guaranteed.
Preferably, the spray pipes are arranged at the two ends of the top of the mounting plate, the spray pipes are connected with water pumps, and water pumping pipes and water pumping ports of the water pumps are embedded in the dam body and aligned with the trash rack.
Optionally, the cleaning mechanism comprises upright rods, wherein an upright rod is fixedly arranged at the corresponding position of the outer walls of the two chains, and the two upright rods are fixedly connected; the outer wall of one upright rod close to the dam body is fixedly provided with bristles, a filter frame is rotatably clamped between the two upright rods, pin shafts are fixedly arranged on two sides of the filter frame, the pin shafts are rotatably sleeved with the upright rods, the outer wall of the upright rod far away from the dam body is rotatably provided with idler wheels, the wheel shafts of the idler wheels are fixedly sleeved with transmission gears on the pin shafts, and the two transmission gears are in meshed transmission connection; torsion springs are arranged between the pin shafts, the wheel shafts of the rollers and the vertical rods, supporting plates are fixedly arranged on the outer walls of the mounting plates, the supporting plates penetrate through the inner cavities of the chains, guide rails are fixedly arranged at the tops of the two ends, far away from the dam body, of the supporting plates, and collecting frames are clamped on the mounting plates at the tops of the supporting plates.
Through the scheme, when the cleaning mechanism moves back and forth, brush the trash rack with the brush hair on the upright rod close to the dam body, the upright rod moves along with the chain, when the upright rod is located at the bottom of the chain wheel, the pin shaft and the torsion spring on the wheel shaft of the roller enable the filter frame to be parallel to the upright rod, the camera shoots the suspended matters in water and the trash picture on the water surface in real time, the image is processed by the central processing unit, the central processing unit detects the trash state in the trash picture after being processed at present, whether damage is caused to the device is judged, if damage is caused, the operation of the device is stopped, if damage is not caused, the suspended matters in water and the trash filtration on the water surface are carried out of the water surface when the filter frame moves, the roller gradually moves to the top of the chain wheel, and is contacted with the guide rail, at the moment, the upright rod moves continuously, so that the roller rotates, the transmission gear enables the pin shaft to slightly rotate, the filter frame is slightly inclined, the water pump starts to send the water on the inner side of the dam to the position of the water pipe when the inclined filter frame passes through the water spraying pipe, the water pump sprays out of the water flow into the trash collection frame, the trash collection frame falls into the trash collection frame, the trash collection frame is convenient to flow in the water collection frame, and the trash collection is convenient to clean the trash flow after the water flow from the dam for a long time, and the water flow is collected after the water flow is pulled out of the trash collection frame.
Optionally, the one end that the filter frame is close to the chain is the slope structure, the fixed surface of gyro wheel has cup jointed the rubber friction circle, the filter frame is the network structure with collecting the frame, the top of collecting the frame is less than the filter frame to the distance of chain, the lateral wall fixed mounting of collecting the frame has the couple of a plurality of L type structures, the link with couple looks adaptation is installed to the outer wall of mounting panel, the spray pipe is all installed at the top both ends of mounting panel.
Through above-mentioned scheme, thereby the automatic vertical filtration clearance aquatic rubbish of filter frame of being convenient for through the torsional spring, the mouth of drawing water sets up the filter screen and filters dirty to the brush is wiped away the filter screen brush of mouth of drawing water and is cleaned after the trash rack is wiped away to the brush, thereby then spray pipe water spray washes the filter frame.
The application discloses a control method of a remote intelligent control system of a water conservancy gate, which comprises the following steps.
S1, controlling the gate to be opened according to water conservancy requirements so as to drain water through the water outlet, collecting opening of the gate and water flow data at an inner cavity at the downstream of the water outlet through the collecting module, comparing actual water flow with theoretical water flow corresponding to the opening of the gate at the moment through the central processing unit, comparing a difference value with a preset threshold value, and when the difference value between the actual flow and the theoretical flow is larger than the threshold value, indicating that the trash rack at the inlet of the water outlet is seriously blocked, so that water drainage is affected, otherwise, the trash rack is not greatly affected.
S2, when the difference value between the actual flow and the theoretical flow is larger than a threshold value, the control module controls the trash rack to clean the water area around the trash rack, the motor drives the semi-gear to continuously rotate, the semi-gear alternately engages with the two first racks to drive the sliding frame to reciprocally slide, the sliding frame drives the second rack to reciprocally move, the second rack drives the driven gear to reciprocally rotate, the driven gear drives the transmission shaft to reciprocally rotate, the transmission shaft drives the rotating shaft to reciprocally rotate through the transmission wheel with a large transmission ratio, the chain wheel drives the chain to reciprocally move each time when the rotating shaft reciprocally rotates, the cleaning mechanism moves to the water outlet from the position of the top of one chain wheel to brush the trash rack, then moves to the top of the other chain wheel to reciprocally move, and when the cleaning mechanism reciprocally moves, the trash rack is reciprocally brushed, so that attached dirt is brushed down, water flow is convenient, and the dirt suspended in the water is carried out of the water surface to the collection frame of the top of the chain wheel by the cleaning mechanism, and the water flow is guaranteed.
S3, the camera shoots pictures of suspended matters in water and garbage on the water surface and transmits the pictures to the central processing unit.
S4, the central processing unit processes the input pictures.
S41, image graying, converting the color image into a gray image, obtaining each pixel value of the image by adopting maximum value, respectively obtaining RGB component values of the image, and then taking the largest one of the three components as the component value of the pixel.
S42, performing geometric transformation, namely processing the acquired image through geometric transformation such as translation, transposition, mirror image, rotation, scaling and the like, and correcting the systematic error and the random error of the instrument position (imaging angle, perspective relation and even the self reason of the lens) when the camera acquires the garbage image.
S43, filtering and denoising, wherein Gaussian filtering is adopted for the image, the Gaussian filtering is a process of carrying out weighted average on the whole image, and the value of each pixel point is obtained by carrying out weighted average on the value of each pixel point and other pixel values in the neighborhood; the gaussian function formula used in the filtering process is: g (x, y) =1/(2ζσ2) e- (x x+y)/2σ, wherein x, y are values of pixel points in the image; sigma is a constant.
S44, utilizing gradient sharpening to make the image more prominent, facilitating analysis, calculating the absolute value of the value difference between the current pixel point and the next pixel point, adding the absolute value of the value difference between the current pixel point and the next line of current pixel point, and if the absolute value is larger than the threshold value, setting the value of the absolute value addition as the value of the current pixel point.
S45, dividing the garbage image, wherein one image possibly contains multiple garbage, and dividing the multiple garbage; the method for dividing the image by using the edge detection comprises the following main steps: edge detection, edge connection, edge refinement, and region filling.
S46, normalizing the image to convert the image into a fixed standard form; by adopting a Min-Max normalization method, setting Max and Min values by traversing each pixel point in an image matrix, and carrying out data normalization processing, wherein the formula is as follows: x '= (x-min (x))/(max (x) -min (x)), the calculated result x' is the normalized pixel value, and x is the original pixel value.
S5, the central processing unit performs garbage identification on the processed picture, and whether the current garbage picture contains garbage which can damage the device or not is judged.
S51, collecting different garbage pictures in advance, and inputting densities corresponding to different garbage; carrying out image preprocessing, manually judging whether garbage in the picture damages a device or not, and marking the picture; taking the processed garbage picture as a characteristic image, and taking whether damage is caused to the device or not as a label; the features + labels form a sample set and are divided into a training set and a test set.
S52, constructing a garbage identification model, wherein the garbage identification model based on machine learning adopts an AlexNet neural network algorithm; the AlexNet comprises eight layers of transformation, and comprises five layers of rolling and two layers of fully-connected hidden layers and one layer of fully-connected output layer; the convolution operation formula is: f=lowerbound ((i+2p-k)/s) +1, i is the size of the input feature image, p is the filling, k is the convolution kernel size, s is the step size; the Relu function formula is: f (x) =max (0, x), x being the input vector.
S521, a first convolution layer, ninety six convolution kernels with the size of 11 x 11, the step length of four, and the activation function of Relu.
S522, a first maximum pooling layer, wherein the window size is 3*3, and the step size is two steps.
S523, a second convolution layer, two hundred fifty-six convolution kernels with the size of 5*5, wherein the step size is one, and the activation function is a Relu function.
S524, a second maximum pooling layer, wherein the window size is 3*3, and the step size is two steps.
S525, the third convolution layer and the fourth convolution layer are three hundred eighty four convolution kernels with the size of 3*3, the step size is one, and the activation function is a Relu function.
S526, a fifth convolution layer, two hundred fifty-six convolution kernels with the size of 3*3, the step size of one, and the activation function of Relu.
S527, a third maximum pooling layer, wherein the window size is 3*3, and the step size is two steps.
S528, two full-connection layer hidden layers, namely four thousand zero ninety six neurons, wherein an activation function is a Relu function, a Dropout method is introduced to randomly inactivate, some hidden neurons in a network are randomly deleted, and the input and output neurons are kept unchanged.
S529, fully connecting the output layers, wherein the activation function is a sigmoid function; the sigmoid activation function expression is: σ (z) =1/(1+e-z), z being the output result of the full link layer.
S53, training the constructed garbage identification model by using a training set.
S531, defining a loss function, and adopting a cross entropy loss function, wherein the formula is as follows: loss=1/batch_size Σj=1 batch_size Σ2i=1-yjilogyji ' - (1-yji) log (1-yji '), batch_size is the number of samples per model entered, yji is the actual label of the sample, yji ' is the predictive label of the sample.
S532, defining an optimizer, adopting an Adam optimizer, integrating an adaptive gradient mechanism and a momentum gradient mechanism in RMSprop, and adaptively adjusting the learning rate of each parameter according to the historical gradient and the updating condition of each parameter in the training process; the update rule is as follows.
mt=β1mt-1+(1-β1)gt。
vt=β2vt-1+(1-β2)gt2。
mt’=mt/(1-β1t)。
vt’=vt/(1-β2t)。
θt+1=θt- η/(a+ε) ×mt ', a is the under-root vt'.
Where gt is the gradient of the parameter, β1 and β2 are the attenuation coefficients of the two exponentially weighted averages, mt 'and vt' are the moving average after correction of the deviation of the gradient, θt+1 is the updated parameter, η is the learning rate, ε is a small constant for avoiding division by zero.
S54, evaluating the trained garbage recognition model by using the test set; evaluating by calculating accuracy, precision, recall, F1 value, etc.; the actual sample prediction result of damage is TP, the actual sample prediction result of damage is FN, the actual sample prediction result of damage is TN, and the actual sample prediction result of damage is FP.
S541, counting how many prediction pairs are in all prediction results according to the accuracy; the calculation formula of the accuracy rate: accuracy= (tp+tn)/(tp+fp+tn+fn).
S542, counting how many samples are actually damaged in samples which are damaged according to all prediction results; the calculation formula is as follows: precision = TP/(tp+fp).
S543, the recall rate is used for counting the number of samples which are actually damaged in all samples, and the prediction result of the number of samples is damage; the calculation formula is as follows: recovery=tp/(tp+fn).
The S544 and F1 values are the harmonic mean of the accuracy rate and the recall rate; and (5) calculating a formula.
F1=((recall-1+precision-1)/2)-1=2*precision*recall/(precision+recall)。
S55, inputting the processed pictures into an estimated garbage classification model, classifying garbage in the pictures by the garbage classification model, shooting pictures with size scales by a plurality of cameras, marking the sizes of different types of garbage, obtaining the volume of the garbage through operation, and multiplying the volume by the density corresponding to the garbage to obtain the weight; judging whether the garbage which can damage the device is contained.
S551, calculating to obtain that the size of the garbage is larger than a set threshold value, and then the garbage is contained and damaged; the device stops running and an alarm is raised.
S552, calculating to obtain that the weight of the garbage is larger than a set threshold value, and then the garbage with damage is contained; the device stops running and an alarm is raised.
S553, if the step S551 and the step S552 judge that the garbage causing damage is not contained, the device continues to work according to the established working flow.
S6, when the cleaning mechanism moves back and forth, brush hair on the upright rod close to the dam body brushes the trash rack, the upright rod moves along with the chain, when the upright rod is positioned at the bottom of the chain wheel, the pin shaft and the torsion spring on the wheel shaft of the roller enable the filter frame to be parallel to the upright rod, when the filter frame moves to carry suspended matters in water and garbage on the water surface out of the water surface, the roller gradually contacts with the guide rail when the upright rod gradually moves to the top of the chain wheel, at the moment, the upright rod continuously moves, so that the roller rotates, the transmission gear enables the pin shaft to slightly rotate, the filter frame slightly tilts, when the tilted filter frame passes through the water spraying pipe, the water flow sprayed by the water spraying pipe washes the garbage on the filter frame into the collecting frame, then the garbage is collected in the collecting frame, the water flow is convenient, and after long-term use, the collecting frame is pulled out of the dam body.
3. Has the beneficial effects of.
One or more of the technical solutions provided in the technical solutions of the present application have at least the following technical effects or advantages.
1. This application is greater than the threshold value when actual flow and theoretical flow's difference, actuating mechanism drive chain reciprocating motion for cleaning mechanism removes the water outlet department from the position at a sprocket top and brushes the trash rack and wipe, then remove another sprocket top and reverse the removal again, then cleaning mechanism round trip movement time, brush the trash rack with reciprocating, make the dirt that adheres to brush off, the rivers flow of being convenient for, and brush the dirt that falls the suspension in water and carry out the surface of water by cleaning mechanism and collect to the sprocket top, reach the purpose of clearing up the waters, guarantee rivers flow, make actual flow be close to the settlement flow, avoid the gate to open excessively and lead to the damage.
2. Shooting suspended matters in water and garbage on the water surface through a camera, transmitting the shot pictures into a central processing unit, preprocessing the pictures at the central processing unit, transmitting the processed pictures into a garbage recognition model for recognition, judging whether the pictures contain garbage which can damage the device or not, stopping the device to continue to operate if the pictures contain garbage which can damage the device, and continuing to operate according to a set working flow if the pictures do not contain garbage which can damage the device. Avoiding the damage to the device caused by suspended matters in water and garbage on the water surface.
3. When the filter frame moves, suspended matters in water and garbage on the water surface are filtered and carried out of the water surface, the idler wheels are gradually contacted with the guide rail when the vertical rods gradually move to the top of the chain wheel, at the moment, the vertical rods continuously move, so that the idler wheels rotate, the transmission gear enables the pin shafts to slightly rotate, the filter frame is slightly inclined, when the inclined filter frame passes through the spray pipe, garbage on the filter frame is flushed and falls into the collecting frame by the water flow sprayed by the spray pipe, then the garbage is collected in the collecting frame, the water flow is facilitated, the garbage is prevented from remaining in the water, and the blocking of the trash rack is continued.
Drawings
Fig. 1 is a schematic diagram of the overall structure of a remote intelligent control system for a water conservancy gate according to a preferred embodiment of the present application.
Fig. 2 is a schematic view of a structure of the decontamination module.
FIG. 3 is a schematic cross-sectional view of the mounting plate.
Fig. 4 is an enlarged schematic view of the structure at a in fig. 2.
Fig. 5 is an enlarged schematic view of the structure at B in fig. 3.
Fig. 6 is an enlarged schematic view of the structure at C in fig. 4.
Fig. 7 is a schematic view of the structure of the collecting frame.
The reference numerals in the figures illustrate: 100. a dam body; 2. a water outlet; 3. a trash rack; 4. a mounting plate; 5. a rotating shaft; 6. a sprocket; 7. a chain; 8. a cleaning mechanism; 81. a vertical rod; 82. brushing; 83. a filter frame; 831. a pin shaft; 84. a roller; 85. a transmission gear; 86 a support plate; 87. a collection frame; 88. a guide rail; 9. a driving mechanism; 91. a motor; 92. a sliding frame; 93. a first rack; 94. a half gear; 95. a second rack; 96. a transmission shaft; 97. a driven gear; 10. a camera; 11. an illuminating lamp.
Detailed Description
The present application is described in further detail below in conjunction with the drawings attached to the specification.
Referring to fig. 1 and 2, an embodiment of the present application provides a remote intelligent control system for a water conservancy gate, including.
The collection module is used for collecting water flow parameters at the downstream of the gate, opening and closing state parameters of the gate and garbage state parameters.
And the sewage disposal module is used for cleaning the inlet end at the upstream of the gate and the water flow according to the collected data.
The control module is used for controlling the opening and closing degree of the gate and the operation of the trash removal module.
And the communication module is used for data transmission.
The central processing unit is used for calling data and instructions, carrying out gate state operation and carrying out garbage state detection operation; the central processing unit is provided with a picture processing unit, a garbage identification model and an early warning unit; the picture processing unit performs image graying, filtering denoising, gradient sharpening and image normalization processing on the acquired picture; the garbage identification model is used for comparing and identifying the pictures processed by the picture processing unit and judging whether garbage in the pictures contains garbage which can cause damage to the device or not; if the device contains garbage which can cause damage to the device, the device is stopped, the photo is sent to the monitoring terminal, and the early warning unit sends out an alarm through a wireless signal to remind workers of processing as soon as possible.
The gate opening, the actual flow data and the garbage state data are collected through the collecting module, so that the gate opening and the cleaning module are controlled, dirt at the inlet end of the gate is cleaned through the cleaning module, water flow is prevented from being influenced, garbage is judged and identified through the garbage identification model, and damage to the device is avoided.
Further, the acquisition module comprises an opening sensor, a flow sensor and a camera 10, wherein the opening sensor is arranged on the gate and is used for acquiring the opening and closing degree of the gate; the flow sensor is arranged at the downstream of the gate and is used for collecting water flow data; a plurality of cameras 10 are installed above the dam body, and the cameras 10 are used for collecting garbage states. The camera 10 is an infrared thermal imaging camera with a laser range finder and has a waterproof function.
Referring to fig. 3 and 4, the cleaning module includes a mounting plate 4 mounted on the dam 100, a water outlet 2 is provided at the bottom of the dam 100, and a gate for controlling the opening and closing of the water outlet 2 is mounted in the water outlet 2, and the gate is an automatic control gate, which is not described in detail herein.
The inlet end of the water outlet 2 is fixedly provided with a trash rack 3, the mounting plate 4 is rotatably sleeved with two rotating shafts 5, the two ends of each rotating shaft 5 are fixedly sleeved with chain wheels 6, a chain 7 is in transmission sleeve connection between the two chain wheels 6 on the same side, the mounting plate 4 is provided with a driving mechanism 9, the driving mechanism 9 is in transmission connection with one of the rotating shafts 5, and a cleaning mechanism 8 is arranged between the two chain wheels 7.
The trash rack 3 filters water flow at the water outlet 2, when the difference value between the actual flow and the theoretical flow is larger than a threshold value, the driving mechanism 9 starts the driving chain 7 to reciprocate, so that the trash rack 3 is brushed by the cleaning mechanism 8 from the position of the top of one chain wheel 6 to the water outlet 2, then the trash rack 3 is brushed by the cleaning mechanism 8 after the trash rack 3 moves to the top of the other chain wheel 6 and then reversely moves, and when the cleaning mechanism 8 moves back and forth, the attached dirt is brushed off, and the water flow is facilitated; the dam body 100 top is fixed to be provided with a plurality of cameras 10 and light 11, and the dam body 100 top is fixed to be provided with the luminance sensor, and the luminance sensor can make light 11 start at night or when light is bad, plays the effect of reinforcing illumination.
A water level sensor is fixedly arranged on the inner side wall of the dam body 100.
A camera 10 is arranged above the side of the water outlet 2 on the dam body 100; the camera 10 is a waterproof camera.
The camera 10 on the dam body 100 shoots pictures of suspended matters in water and garbage on the water surface in real time, two adjacent sides of the pictures are provided with size scale marks, the pictures are transmitted into the central processing unit for picture processing, the pictures are transmitted into the garbage recognition model of the central processing unit after picture processing, garbage in the pictures is recognized, whether the pictures contain garbage which can damage the device is judged, if the garbage which can damage the device is contained, the device is stopped to operate and give an alarm, a worker is reminded to timely process, if the garbage which can damage the device is not contained, the device can continue to work according to a set working flow, and the dirt which is brushed and suspended in the water is carried out of the water surface to the top of the chain wheel 6 by the cleaning mechanism 8 for collection, so that the purpose of cleaning the water area is achieved, and the normal flow of the water flow is ensured.
The water spraying pipes are arranged at the two ends of the top of the mounting plate 4 and connected with the water pump, and the water pumping pipe water pumping mouth of the water pump is embedded in the dam body 100 and aligned with the trash rack 3.
Referring to fig. 2 and 7, the driving mechanism 9 includes a motor 91, the motor 91 is fixedly installed in the installation plate 4, a sliding frame 92 is slidably clamped in the installation plate 4, first racks 93 are fixedly installed on the inner walls of the top and bottom of the sliding frame 92, an output shaft of the motor 91 extends into the sliding frame 92 and is fixedly sleeved with a half gear 94, and the half gear 94 can be intermittently meshed and transmitted with the two first racks 93.
The bottom of the sliding frame 92 is fixedly provided with a second rack 95, the mounting plate 4 is rotatably sleeved with a transmission shaft 96, the transmission shaft 96 is fixedly sleeved with a driven gear 97, the driven gear 97 is meshed with the second rack 95, a matched transmission wheel and a transmission belt are arranged between the driven gear 97 and one of the rotating shafts 5, the transmission ratio of the half gear 94 to the driven gear 97 is 1:2, the transmission ratio of the transmission wheel between the transmission shaft 96 and the rotating shaft 5 is 1:4, and the transmission wheel and the transmission belt between the transmission shaft 96 and the rotating shaft 5 are of synchronous belt structures.
When the difference value between the actual flow and the theoretical flow is larger than a threshold value, the control module controls the trash cleaning module to clean the trash rack 3 and clean the water area around the trash rack 3; during cleaning, the motor 91 drives the half gears 94 to continuously rotate, the half gears 94 are alternately meshed with the two first racks 93, so that the sliding frame 92 is driven to reciprocate, the sliding frame 92 drives the second racks 95 to reciprocate, then the second racks 95 drive the driven gears 97 to reciprocate, the driven gears 97 drive the transmission shafts 96 to reciprocate, the transmission shafts 96 drive the rotation shafts 5 to reciprocate through the transmission wheels with large transmission ratios, the chain wheels 6 drive the chains 7 to reciprocate each time when the rotation shafts 5 reciprocate, the cleaning mechanism 8 moves from the position at the top of one chain wheel 6 to the water outlet 2 to brush the trash rack 3, then moves to the top of the other chain wheel 6 to reversely rotate, and then the trash rack 3 is brushed reciprocally when the cleaning mechanism 8 moves back and forth, so that attached dirt is brushed off, water flow is facilitated, the dirt brushed off and suspended in water is collected above the chain wheels 6 by the cleaning mechanism 8, and the purpose of cleaning water area is achieved.
Referring to fig. 3 to 6, the cleaning mechanism 8 comprises upright posts 81, wherein an upright post 81 is fixedly arranged at the corresponding position of the outer walls of two chains 7, the two upright posts 81 are fixedly connected with each other, bristles 82 are fixedly arranged on the outer wall of one upright post 81 close to the dam body 100, a filter frame 83 is rotatably clamped between the two upright posts 81, pin shafts 831 are fixedly arranged on two sides of the filter frame 83, the pin shafts 831 are rotatably sleeved with the upright posts 81, rollers 84 are rotatably arranged on the outer walls of the upright posts 81 far away from the dam body 100, transmission gears 85 are fixedly sleeved on the wheel shafts of the rollers 84 and the pin shafts 831, and the two transmission gears 85 are in meshed transmission connection; torsion springs are arranged between the axle shafts of the pin shafts 831 and the rollers 84 and the upright rods 81.
The outer wall fixed mounting of mounting panel 4 has backup pad 86, and backup pad 86 runs through the inner chamber of chain 7, and backup pad 86 keeps away from the symmetrical fixed mounting in both ends top of dam body 100 and has guide rail 88, and the joint has collection frame 87 on the mounting panel 4 at backup pad 86 top.
When the cleaning mechanism 8 moves back and forth, the brush bristles 82 on the vertical rods 81 close to the dam body 100 brush the trash rack 3, the vertical rods 81 move along with the chain 7, when the vertical rods 81 are positioned at the bottom of the chain wheel 6, the pin shafts 831 and the torsion springs on the wheel shafts of the rollers 84 enable the filter frames 83 to be parallel to the vertical rods 81, when the filter frames 83 move, suspended matters in water and trash on the water surface are filtered and carried out of the water surface, when the vertical rods 81 gradually move to the top of the chain wheel 6, the rollers 84 gradually contact with the guide rails 88, the vertical rods 81 continue to move, so that the rollers 84 rotate, the transmission gears 85 enable the pin shafts 831 to slightly rotate, the filter frames 83 slightly incline, at the moment, the water pump starts pumping water on the inner side of the dam to the water spraying pipe, when the inclined filter frames pass through the water spraying pipe, the water flow sprayed out of the water spraying pipe washes the trash on the filter frames 83 into the collection frames 87, the trash is collected in the collection frames 87, and the trash can be cleaned after the water flows are convenient for a long time, and the collection frames 87 are pulled out of the dam body 100.
Referring to fig. 4 and 6, the end that filter frame 83 is close to chain 7 is the slope structure, the fixed surface of gyro wheel 84 has cup jointed the rubber friction circle, filter frame 83 and collection frame 87 are the network structure, the distance of collection frame 87's top to chain 7 is less than the distance of filter frame 83 to chain 7, the couple of a plurality of L type structures is installed to the lateral wall fixed mounting of collection frame 87, the link with couple looks adaptation is installed to the outer wall of mounting panel 4, thereby be convenient for filter frame 83 automatic vertical filtration clearance aquatic rubbish through the torsional spring, the mouth of taking water sets up the filter screen and filters dirty, and brush 82 brush and wipe the clearance to the filter screen brush of water sucking mouth after also brushing the trash rack 3, then the spray pipe water spray is washed filter frame 83.
The embodiment of the application provides a control method of a remote intelligent control system of a water conservancy gate, which comprises the following steps.
S1, controlling the gate to be opened according to water conservancy requirements so as to discharge water through the water outlet 2, collecting opening of the gate and water flow data at the inner cavity of the downstream of the water outlet 2 through the collecting module, comparing actual water flow with theoretical water flow corresponding to the opening of the gate at the moment through the central processing unit, comparing a difference value with a preset threshold value, and when the difference value between the actual flow and the theoretical flow is larger than the threshold value, indicating that the trash rack 3 at the inlet of the water outlet 2 is seriously blocked, so that the water discharge is influenced, otherwise, the influence is not large.
S2, when the difference value between the actual flow and the theoretical flow is larger than a threshold value, the control module controls the action of the trash cleaning module to clean the trash rack 3 and clean the water area around the trash rack 3; during cleaning, the motor 91 drives the half gears 94 to continuously rotate, the half gears 94 are alternately meshed with the two first racks 93, so that the sliding frame 92 is driven to reciprocate, the sliding frame 92 drives the second racks 95 to reciprocate, then the second racks 95 drive the driven gears 97 to reciprocate, the driven gears 97 drive the transmission shafts 96 to reciprocate, the transmission shafts 96 drive the rotation shafts 5 to reciprocate through the transmission wheels with large transmission ratio, the chain wheels 6 drive the chains 7 to reciprocate each time the rotation shafts 5 reciprocate, and when the chains 7 reciprocate, the cleaning mechanism 8 moves from the position at the top of one chain wheel 6 to the water outlet 2 to brush the trash rack 3, then moves to the top of the other chain wheel 6 to reversely rotate, and when the cleaning mechanism 8 reciprocates, the trash rack 3 is brushed, so that attached dirt is brushed down, and water flow is convenient; and the dirt suspended in the water by brushing is carried out the water surface by the cleaning mechanism 8 and discharged into the collecting frame 87 at the top of the chain wheel 6 for collection, so that the aim of cleaning the water area is fulfilled, and the flow of water is ensured.
S3, the camera shoots pictures of suspended matters in water and garbage on the water surface and transmits the pictures to the central processing unit.
S4, the central processing unit processes the input pictures.
S41, image graying, converting the color image into a gray image, obtaining each pixel value of the image by adopting maximum value, respectively obtaining RGB component values of the image, and then taking the largest one of the three components as the component value of the pixel.
S42, performing geometric transformation, namely processing the acquired image through geometric transformation such as translation, transposition, mirror image, rotation, scaling and the like, and correcting the systematic error and the random error of the instrument position (imaging angle, perspective relation and even the self reason of the lens) when the camera acquires the garbage image.
S43, filtering and denoising, wherein Gaussian filtering is adopted for the image, the Gaussian filtering is a process of carrying out weighted average on the whole image, and the value of each pixel point is obtained by carrying out weighted average on the value of each pixel point and other pixel values in the neighborhood; the gaussian function formula used in the filtering process is: g (x, y) =1/(2ζσ2) e- (x x+y)/2σ, wherein x, y are values of pixel points in the image; sigma is a constant.
S44, utilizing gradient sharpening to make the image more prominent, so that the analysis is convenient; and calculating the absolute value of the value difference between the current pixel point and the next pixel point, adding the absolute value of the value difference between the current pixel point and the next line of current pixel point, and setting the value added by the absolute values as the value of the current pixel point if the absolute value is larger than the threshold value.
S45, dividing the garbage image, wherein one image possibly contains multiple garbage, and dividing the multiple garbage; the method for dividing the image by using the edge detection comprises the following main steps: edge detection, edge connection, edge refinement, and region filling.
S46, normalizing the image to convert the image into a fixed standard form; by adopting a Min-Max normalization method, setting Max and Min values by traversing each pixel point in an image matrix, and carrying out data normalization processing, wherein the formula is as follows: x '= (x-min (x))/(max (x) -min (x)), the calculated result x' is the normalized pixel value, and x is the original pixel value.
S5, the central processing unit performs garbage identification on the processed picture, and whether the current garbage picture contains garbage which can damage the device or not is judged.
S51, collecting different garbage pictures in advance, inputting densities corresponding to different garbage, preprocessing images, manually judging whether the garbage in the pictures damages a device or not, and marking the pictures; taking the processed garbage picture as a characteristic image, and taking whether damage is caused to the device or not as a label, wherein the damage is not caused and is represented by 0 and the damage is represented by 1; feature + tag constitutes the sample set and is according to 3: the scale of 1 is divided into training and test sets.
S52, constructing a garbage identification model; the garbage identification model based on machine learning adopts Alex Net neural network algorithm; the AlexNet comprises eight layers of transformation, and comprises five layers of rolling and two layers of fully-connected hidden layers and one layer of fully-connected output layer; the convolution operation formula is: f=lowerbound ((i+2p-k)/s) +1, i is the size of the input feature image, p is the filling, k is the convolution kernel size, s is the step size; the Relu function formula is: f (x) =max (0, x), x being the input vector.
S521, a first convolution layer, ninety six convolution kernels with the size of 11 x 11, the step length of four, and the activation function of Relu.
S522, a first maximum pooling layer, wherein the window size is 3*3, and the step size is two steps.
S523, a second convolution layer, two hundred and fifty-six convolution kernels with the size of 5*5, wherein the step length is one step, and the activation function is a Relu function.
S524, a second maximum pooling layer, wherein the window size is 3*3, and the step size is two steps.
S525, the third convolution layer and the fourth convolution layer are three hundred eighty four convolution kernels with the size of 3*3, the step length is one step, and the activation function is a Relu function.
S526, a fifth convolution layer, two hundred and fifty-six convolution kernels with the size of 3*3, wherein the step length is one step, and the activation function is a Relu function.
S527, a third maximum pooling layer, wherein the window size is 3*3, and the step size is two steps.
S528, two fully-connected hidden layers, namely four thousand zero ninety six neurons, and an activation function is a Relu function; the Dropout method is introduced to inactivate randomly, then some hidden neurons are deleted, and the input and output neurons are kept unchanged.
S529, fully connecting the output layers, wherein the activation function is a sigmoid function; the sigmoid activation function expression is: σ (z) =1/(1+e-z), z being the output result of the full link layer.
S53, training the constructed garbage identification model by using a training set.
S531, defining a loss function; the cross entropy loss function is adopted, and the formula is as follows: loss=1/batch_size Σj=1 batch_size Σ2i=1-yjilogyji ' - (1-yji) log (1-yji '), batch_size is the number of samples per model entered, yji is the actual label of the sample, yji ' is the predictive label of the sample.
S532, defining an optimizer; an Adam optimizer is adopted, which integrates an adaptive gradient mechanism and a momentum gradient mechanism in RMSprop; the learning rate of each parameter can be adaptively adjusted according to the historical gradient and the updating condition of each parameter in the training process; the update rule is as follows.
mt=β1mt-1+(1-β1)gt。
vt=β2vt-1+(1-β2)gt2。
mt’=mt/(1-β1t)。
vt’=vt/(1-β2t)。
θt+1=θt- η/(a+ε) ×mt ', a is the under-root vt'.
Where gt is the gradient of the parameter, β1 and β2 are the attenuation coefficients of the two exponentially weighted averages, mt 'and vt' are the moving average after correction of the deviation of the gradient, θt+1 is the updated parameter, η is the learning rate, ε is a small constant for avoiding division by zero.
S54, evaluating the trained garbage recognition model by using the test set; evaluating by calculating accuracy, precision, recall, F1 value, etc.; the actual sample prediction result of damage is TP, the actual sample prediction result of damage is FN, the actual sample prediction result of damage is TN, and the actual sample prediction result of damage is FP.
S541, counting how many prediction pairs are in all prediction results according to the accuracy; the calculation formula of the accuracy rate: accuracy= (tp+tn)/(tp+fp+tn+fn).
S542, counting how many samples are actually damaged in samples which are damaged according to all prediction results; the calculation formula is as follows: precision = TP/(tp+fp).
S543, the recall rate is used for counting the number of samples which are actually damaged in all samples, and the prediction result of the number of samples is damage; the calculation formula is as follows: recovery=tp/(tp+fn).
The S544 and F1 values are the harmonic mean of the precision rate and the recall rate; the calculation formula is as follows: f1 = ((accuracy-1+accuracy-1)/2) -1=2 x accuracy/(accuracy+accuracy).
S55, inputting the processed pictures into an estimated garbage classification model, classifying garbage in the pictures by the garbage classification model, shooting pictures with size scales by a plurality of cameras, marking the sizes of different types of garbage, obtaining the volume of the garbage through operation, and multiplying the volume by the density corresponding to the garbage to obtain the weight; judging whether the garbage which can damage the device is contained.
S551, calculating to obtain that the size of the garbage is larger than a set threshold value, and then the garbage is contained and damaged; the device stops running and an alarm is raised.
S552, calculating to obtain that the weight of the garbage is larger than a set threshold value, and then the garbage with damage is contained; the device stops running and an alarm is raised.
S553, if the step S551 and the step S552 judge that the garbage causing damage is not contained, the device continues to work according to the established working flow.
S6, when the cleaning mechanism 8 moves back and forth, brush bristles 82 on the vertical rods 81 close to the dam body 100 brush the trash rack 3, the vertical rods 81 move along with the chain 7, when the vertical rods 81 are positioned at the bottom of the chain wheel 6, the filter frames 83 tend to be parallel to the vertical rods 81 due to the torsion springs on the axle of the pin shafts 831 and the roller wheels 84, suspended matters in water and trash on the water surface are filtered and carried out of the water surface when the filter frames 83 move, when the vertical rods 81 gradually move to the top of the chain wheel 6, the roller wheels 84 gradually contact with the guide rails 88, the vertical rods 81 move continuously, so that the roller wheels 84 rotate, the transmission gear 85 slightly rotates the pin shafts 831, the inclined filter frames slightly incline, when the water sprayed by the water spraying pipes washes the trash on the filter frames 83 into the collecting frames 87, the trash is collected in the collecting frames 87, the trash can flow conveniently, and after long-term use, the collecting frames 87 can be pulled out of the dam body 100.
The invention relates to a working principle of a remote intelligent control system of a water conservancy gate, which comprises the following steps: the gate is controlled to be opened according to water conservancy requirements so as to discharge water through the water outlet 2, the opening degree of the gate and water flow data at the inner cavity at the downstream of the water outlet 2 are collected through the collecting module, the actual water flow is compared with the theoretical water flow corresponding to the opening degree of the gate at the moment through the central processing unit, the difference value is compared with a preset threshold value, when the difference value between the actual flow and the theoretical flow is larger than the threshold value, the situation that the drain grating 3 at the inlet of the water outlet 2 is seriously blocked is indicated, and the water discharge is influenced, otherwise, the influence is not great; when the difference value between the actual flow and the theoretical flow is larger than a threshold value, the control module controls the trash cleaning module to clean the trash rack 3 and clean the water area around the trash rack 3; during cleaning, the motor 91 drives the half gears 94 to continuously rotate, the half gears 94 are alternately meshed with the two first racks 93, so that the sliding frame 92 is driven to reciprocate, the sliding frame 92 drives the second racks 95 to reciprocate, then the second racks 95 drive the driven gears 97 to reciprocate, the driven gears 97 drive the transmission shafts 96 to reciprocate, the transmission shafts 96 drive the rotation shafts 5 to reciprocate through the transmission wheels with large transmission ratio, the chain wheel 6 drives the chain 7 to reciprocate each time the rotation shafts 5 reciprocate, the cleaning mechanism 8 moves from the position at the top of one chain wheel 6 to the water outlet 2 to brush the trash rack 3 when the chain 7 reciprocates, then moves to the top of the other chain wheel 6 to reversely rotate, and bristles 82 on the upright rods 81 close to the dam body 100 brush the trash rack 3 when the cleaning mechanism 8 reciprocates, so that attached dirt is brushed down, and water flow is convenient; the camera 10 above the dam body 100 shoots pictures of suspended matters in water and garbage on the water surface in real time, and transmits the pictures to the central processing unit for picture processing, and then transmits the pictures to the garbage identification model of the central processing unit for identifying the garbage in the pictures, judging whether the pictures contain garbage which can damage the device, and stopping the operation of the device and giving an alarm if the pictures contain the garbage which can damage the device; if the device does not contain garbage which can cause damage to the device, the device can continue to work according to a set working flow; the dirt brushed and suspended in the water is carried out of the water surface by the cleaning mechanism 8 to the top of the chain wheel 6 for collection, so that the aim of cleaning the water area is fulfilled, and the flow of water is ensured; when the cleaning mechanism 8 moves back and forth, the upright rod 81 moves along with the chain 7, when the upright rod 81 is positioned at the bottom of the chain wheel 6, the pin shaft 831 and the torsion spring on the wheel shaft of the roller 84 enable the filter frame 83 to be parallel to the upright rod 81, when the filter frame 83 moves, suspended matters in water and garbage on the water are filtered and carried out of the water, when the upright rod 81 gradually moves to the top of the chain wheel 6, the roller 84 gradually contacts with the guide rail 88, at the moment, the upright rod 81 continues to move, so that the roller 84 rotates, the transmission gear 85 enables the pin shaft 831 to slightly rotate, and the filter frame 83 slightly tilts; at this time, the water pump starts, and the water pump pumps water to spray pipe department blowout, and when the filter frame of slope passed the spray pipe, the rubbish on the filter frame 83 was washed to spray pipe spun rivers and falls into in the collection frame 87, then the garbage collection is in the collection frame 87, and the rivers flow of being convenient for, after long-term use, pull out from the dam body 100 and collect frame 87, clear up rubbish can.
The foregoing has shown and described the basic principles, principal features and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the above-described embodiments, and that the above-described embodiments and descriptions are only preferred embodiments of the present invention, and are not intended to limit the invention, and that various changes and modifications may be made therein without departing from the spirit and scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.

Claims (7)

1. A remote intelligent control system for a water conservancy gate, comprising: acquisition module, module of decontaminating, control module, communication module and central processing unit, its characterized in that:
the acquisition module is used for acquiring water flow parameters at the downstream of the gate and opening and closing state parameters of the gate;
the water flow cleaning module is used for cleaning the inlet end at the upstream of the gate according to the collected data;
the control module is used for controlling the opening and closing degree of the gate and the operation of the trash removal module;
the communication module is used for data transmission;
the central processing unit is used for calling data and instructions and carrying out operation processing on the gate state;
The central processing unit is provided with a picture processing unit, a garbage identification model and an early warning unit; the picture processing unit performs image graying, filtering denoising, gradient sharpening and image normalization processing on the acquired picture; the garbage identification model is used for comparing and identifying the pictures processed by the picture processing unit and judging whether garbage in the pictures contains garbage which can cause damage to the device or not; if the device contains garbage which can cause damage to the device, stopping the operation of the device, sending the photo to a monitoring terminal, and sending an alarm by an early warning unit through a wireless signal;
the sewage disposal module comprises a mounting plate arranged on a dam body, a water outlet is formed in the bottom of the dam body, a gate for controlling the water outlet to open and close is arranged in the water outlet, a sewage blocking grid is fixedly arranged at the inlet end of the water outlet, two rotating shafts are rotatably sleeved on the mounting plate, chain wheels are fixedly sleeved at two ends of each rotating shaft, chains are in transmission sleeve connection between the two chain wheels on the same side, a driving mechanism is arranged on the mounting plate, a cleaning mechanism is arranged between the two chains, and the driving mechanism is in transmission connection with one rotating shaft;
the cleaning mechanism comprises upright rods, wherein an upright rod is fixedly arranged at the corresponding position of the outer walls of the chains, bristles are fixedly arranged on the outer wall of the upright rod close to the dam body, a filter frame is rotatably clamped between the two upright rods, pin shafts are fixedly arranged on two sides of the filter frame, the pin shafts are rotatably sleeved with the upright rods, idler wheels are rotatably arranged on the outer walls of the upright rods far away from the dam body, transmission gears are fixedly arranged on wheel shafts and the pin shafts of the idler wheels, and the two transmission gears are in meshed transmission connection;
The outer wall fixed mounting of mounting panel has the backup pad, the backup pad is kept away from the equal fixed mounting in both ends top of dam body and is had the guide rail, and the detachable is provided with on the mounting panel and collects the frame, the one end that the filter frame is close to the chain is the slope structure.
2. The remote intelligent control system of a water conservancy gate according to claim 1, wherein: the acquisition module comprises an opening sensor, a flow sensor and a camera, wherein the opening sensor is arranged on the gate and is used for acquiring the opening and closing degree of the gate; the flow sensor is arranged at the downstream of the gate and used for collecting water flow data, the plurality of cameras and the illuminating lamps are arranged above the dam body, the brightness sensor is fixedly arranged above the dam body, the plurality of cameras are arranged above the side of the water outlet on the dam body, and the cameras are used for collecting garbage states.
3. The remote intelligent control system of a water conservancy gate according to claim 1, wherein: the driving mechanism comprises a motor, the motor is fixedly arranged on a mounting plate, a sliding frame is fixedly clamped in the mounting plate, first racks are fixedly arranged on the inner walls of the top and the bottom of the sliding frame, a half gear is fixedly sleeved on an output shaft of the motor and can be intermittently connected with the two first racks in a meshed transmission manner, a second rack is fixedly arranged at the bottom of the sliding frame, a transmission shaft is rotatably sleeved on the mounting plate, a driven gear is fixedly sleeved on the transmission shaft and is meshed with the second rack, and a matched transmission wheel and a transmission belt are arranged between the driven gear and one of the shafts.
4. A remote intelligent control system for a water conservancy gate according to claim 3 and wherein: the transmission ratio of the half gear to the driven gear is 1:2, the transmission ratio of the transmission wheel between the transmission shaft and the rotating shaft is 1:4, and the transmission wheel and the transmission belt between the transmission shaft and the rotating shaft are of synchronous belt structures.
5. The remote intelligent control system of a water conservancy gate according to claim 1, wherein: the surface fixing of gyro wheel has cup jointed the rubber friction circle, all install the torsional spring between round pin axle and the shaft and the pole setting of gyro wheel, filter frame and collection frame are the network structure.
6. The remote intelligent control system of a water conservancy gate according to claim 1, wherein: the distance from the top of collecting the frame to the chain is less than the distance from the filter frame to the chain, the lateral wall fixed mounting of collecting the frame has the couple of a plurality of L type structures, the link with couple looks adaptation is installed to the outer wall of mounting panel, a plurality of spray pipes are all installed at the top both ends of mounting panel, fixedly on the dam body be provided with the water pump, the spray pipe is linked together with the outlet pipe of water pump.
7. A control method of a remote intelligent control system of a water conservancy gate according to any one of claims 1-6, comprising the steps of:
S1, controlling the gate to be opened according to water conservancy requirements so as to discharge water through a water outlet, collecting opening of the gate and water flow data at an inner cavity at the downstream of the water outlet through an acquisition module, comparing actual water flow with theoretical water flow corresponding to the opening of the gate at the moment through a central processing unit, and comparing a difference value with a preset threshold value, wherein when the difference value between the actual flow and the theoretical flow is larger than the threshold value, the situation that the drain grating at the inlet of the water outlet is seriously blocked is indicated, and the water discharge is influenced, otherwise, the influence is not great;
s2, when the difference value between the actual flow and the theoretical flow is larger than a threshold value, the control module controls the driving mechanism to drive the trash cleaning module to clean the trash rack, and the water area around the trash rack is cleaned; when the cleaning mechanism moves back and forth, the trash rack is brushed back and forth, so that attached dirt is brushed off, and the flow of water is ensured;
s3, the camera shoots pictures of suspended matters in water and garbage on the water surface and transmits the pictures to the central processing unit;
s4, the central processing unit processes the input pictures;
s5, the central processing unit performs garbage identification on the processed picture, and judges whether the current garbage picture contains garbage which can damage the device or not;
s51, collecting different garbage pictures in advance, inputting densities corresponding to different garbage, preprocessing images, manually judging whether the garbage in the pictures damages a device or not, and marking the pictures; taking the processed garbage picture as a characteristic image, taking whether damage is caused to the device as a label, and according to 3: the proportion of 1 is divided into a training set and a testing set;
S52, constructing a garbage identification model; the garbage identification model based on machine learning adopts Alex Net neural network algorithm;
s53, training the constructed garbage identification model by using a training set;
s54, evaluating the trained garbage recognition model by using the test set;
s55, inputting the processed pictures into an estimated garbage classification model, classifying garbage in the pictures by the garbage classification model, shooting pictures with size scales by a plurality of cameras, marking the sizes of different types of garbage, obtaining the volume of the garbage through operation, and multiplying the volume by the density corresponding to the garbage to obtain the weight; judging whether the garbage which can damage the device is contained or not;
s551, calculating to obtain that the size of the garbage is larger than a set threshold value, and then the garbage is contained and damaged; the device stops running and gives an alarm;
s552, calculating to obtain that the weight of the garbage is larger than a set threshold value, and then the garbage with damage is contained; the device stops running and gives an alarm;
s553, if the step S551 and the step S552 judge that the garbage causing damage is not contained, the device continues to work according to the established working flow;
s6, when the cleaning mechanism moves back and forth, brush the trash rack by brush hair on the vertical rod close to the dam body, the vertical rod moves along with the chain, when the vertical rod is positioned at the bottom of the chain wheel, the filter frame tends to be parallel to the vertical rod due to the torsion spring on the axle of the pin shaft and the roller, suspended matters in water and garbage on the water surface are filtered and carried out of the water surface when the filter frame moves, the roller gradually contacts with the guide rail when the vertical rod gradually moves to the top of the chain wheel, at the moment, the vertical rod continuously moves, so that the roller rotates, the driving gear enables the pin shaft to slightly rotate, the filter frame is slightly inclined, when the inclined filter frame passes through the water spraying pipe, the garbage on the filter frame is flushed by the water flow sprayed out of the water spraying pipe and falls into the collecting frame, the garbage is collected in the collecting frame, the water flow is convenient, and after long-term use, the collecting frame is pulled out of the dam body for garbage cleaning.
CN202311149130.7A 2023-09-07 2023-09-07 Remote intelligent control method and system for water conservancy gate Active CN116893644B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311149130.7A CN116893644B (en) 2023-09-07 2023-09-07 Remote intelligent control method and system for water conservancy gate

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311149130.7A CN116893644B (en) 2023-09-07 2023-09-07 Remote intelligent control method and system for water conservancy gate

Publications (2)

Publication Number Publication Date
CN116893644A CN116893644A (en) 2023-10-17
CN116893644B true CN116893644B (en) 2023-12-19

Family

ID=88311023

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311149130.7A Active CN116893644B (en) 2023-09-07 2023-09-07 Remote intelligent control method and system for water conservancy gate

Country Status (1)

Country Link
CN (1) CN116893644B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN209277109U (en) * 2018-10-31 2019-08-20 四川恒诚达科技有限公司 A kind of power station trash rack dirt crushing cleaning plant
CN110747823A (en) * 2019-10-30 2020-02-04 湖北省水利水电规划勘测设计院 Sluice control system and control method for long-distance water diversion and transfer project
CN111945690A (en) * 2020-08-11 2020-11-17 徐壮 Environment-friendly sewage suction device based on water treatment
CN115331166A (en) * 2022-08-12 2022-11-11 广州市净水有限公司 Intelligent grid monitoring and early warning method and system
CN116212514A (en) * 2023-02-17 2023-06-06 王大波 Gate filtering device and gate flow monitoring method
CN116430722A (en) * 2023-03-17 2023-07-14 中国长江三峡集团有限公司 Water supply network system, control method and device thereof and storage medium

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113585188A (en) * 2021-09-02 2021-11-02 鲁东大学 River dam with function of collecting floating garbage on water surface

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN209277109U (en) * 2018-10-31 2019-08-20 四川恒诚达科技有限公司 A kind of power station trash rack dirt crushing cleaning plant
CN110747823A (en) * 2019-10-30 2020-02-04 湖北省水利水电规划勘测设计院 Sluice control system and control method for long-distance water diversion and transfer project
CN111945690A (en) * 2020-08-11 2020-11-17 徐壮 Environment-friendly sewage suction device based on water treatment
CN115331166A (en) * 2022-08-12 2022-11-11 广州市净水有限公司 Intelligent grid monitoring and early warning method and system
CN116212514A (en) * 2023-02-17 2023-06-06 王大波 Gate filtering device and gate flow monitoring method
CN116430722A (en) * 2023-03-17 2023-07-14 中国长江三峡集团有限公司 Water supply network system, control method and device thereof and storage medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
闸门对河道污染物影响的模拟研究;阮燕云;张翔;夏军;张永勇;张洪斌;;武汉大学学报(工学版)(05);全文 *

Also Published As

Publication number Publication date
CN116893644A (en) 2023-10-17

Similar Documents

Publication Publication Date Title
CN116893644B (en) Remote intelligent control method and system for water conservancy gate
CN116159824A (en) Automatic cleaning device and method for boiler pipeline dirt
CN108268028A (en) The small watersheds and method of photovoltaic battery panel cleaning
CN213986450U (en) Miniature visual water quality monitoring device
CN209837256U (en) A cut dirty device for municipal drainage
CN114554065A (en) Intelligent home security device
CN111218911A (en) Energy-saving system and energy-saving method applied to road sweeping cleaning vehicle
CN113663985A (en) Full-automatic dust removal equipment and dust removal method for photovoltaic panel
CN116580828A (en) Visual monitoring method for full-automatic induction identification of cat health
CN109740566A (en) A kind of computer digital image processing equipment
CN115450310A (en) Intelligent dredging robot for pipeline
CN214959804U (en) Thing networking image collector
CN115592687A (en) Fault alarming and removing system and method applied to swimming pool robot
CN216366835U (en) Unattended environment detection system
CN115334223B (en) Binocular camera and light filling lamp mounting structure for detecting violation of vehicle high beam
CN216696066U (en) Signal acquisition device
CN117674720B (en) Be applied to abluent walking robot of photovoltaic
CN116059715B (en) Water treatment filtration equipment for environmental monitoring laboratory
CN218722548U (en) A glass defogging device for freezer
CN117626904B (en) Sewage filtering device for environmental treatment
CN213429915U (en) Ash removal robot convenient for collecting adhesive substances
CN115000879B (en) Power transmission line inspection robot based on image learning and inspection method
CN221056143U (en) Environment water sample detection device
CN210256226U (en) Visual identification device of security robot
CN118186970A (en) Dry-type intelligent cleaning dust collection vehicle

Legal Events

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