CN113277388A - Data acquisition control method for electric hanging basket - Google Patents
Data acquisition control method for electric hanging basket Download PDFInfo
- Publication number
- CN113277388A CN113277388A CN202110362590.2A CN202110362590A CN113277388A CN 113277388 A CN113277388 A CN 113277388A CN 202110362590 A CN202110362590 A CN 202110362590A CN 113277388 A CN113277388 A CN 113277388A
- Authority
- CN
- China
- Prior art keywords
- hanging basket
- data
- motor
- weight
- network
- 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.)
- Granted
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B1/00—Control systems of elevators in general
- B66B1/24—Control systems with regulation, i.e. with retroactive action, for influencing travelling speed, acceleration, or deceleration
- B66B1/28—Control systems with regulation, i.e. with retroactive action, for influencing travelling speed, acceleration, or deceleration electrical
- B66B1/30—Control systems with regulation, i.e. with retroactive action, for influencing travelling speed, acceleration, or deceleration electrical effective on driving gear, e.g. acting on power electronics, on inverter or rectifier controlled motor
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B1/00—Control systems of elevators in general
- B66B1/24—Control systems with regulation, i.e. with retroactive action, for influencing travelling speed, acceleration, or deceleration
- B66B1/28—Control systems with regulation, i.e. with retroactive action, for influencing travelling speed, acceleration, or deceleration electrical
- B66B1/30—Control systems with regulation, i.e. with retroactive action, for influencing travelling speed, acceleration, or deceleration electrical effective on driving gear, e.g. acting on power electronics, on inverter or rectifier controlled motor
- B66B1/304—Control systems with regulation, i.e. with retroactive action, for influencing travelling speed, acceleration, or deceleration electrical effective on driving gear, e.g. acting on power electronics, on inverter or rectifier controlled motor with starting torque control
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B1/00—Control systems of elevators in general
- B66B1/34—Details, e.g. call counting devices, data transmission from car to control system, devices giving information to the control system
- B66B1/3415—Control system configuration and the data transmission or communication within the control system
- B66B1/3446—Data transmission or communication within the control system
- B66B1/3453—Procedure or protocol for the data transmission or communication
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B1/00—Control systems of elevators in general
- B66B1/34—Details, e.g. call counting devices, data transmission from car to control system, devices giving information to the control system
- B66B1/3476—Load weighing or car passenger counting devices
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B5/00—Applications of checking, fault-correcting, or safety devices in elevators
- B66B5/0006—Monitoring devices or performance analysers
- B66B5/0012—Devices monitoring the users of the elevator system
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B5/00—Applications of checking, fault-correcting, or safety devices in elevators
- B66B5/0006—Monitoring devices or performance analysers
- B66B5/0018—Devices monitoring the operating condition of the elevator system
- B66B5/0031—Devices monitoring the operating condition of the elevator system for safety reasons
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B5/00—Applications of checking, fault-correcting, or safety devices in elevators
- B66B5/02—Applications of checking, fault-correcting, or safety devices in elevators responsive to abnormal operating conditions
- B66B5/14—Applications of checking, fault-correcting, or safety devices in elevators responsive to abnormal operating conditions in case of excessive loads
- B66B5/145—Applications of checking, fault-correcting, or safety devices in elevators responsive to abnormal operating conditions in case of excessive loads electrical
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention relates to a data acquisition control method for an electric hanging basket, which comprises the following steps: step 1: video acquisition and picture analysis and processing are specifically as follows: step 1.1: video acquisition and storage and data preprocessing; step 1.2: face recognition analysis processing; step 1.3: predicting the weight of the employee by the neural network; step 1.4: self-adaptive repairing of network packet loss; step 2: the real-time data analysis and processing of the parameters of the hanging basket are as follows: step 2.1: estimating the total load of the motor; step 2.2: self-adaptive starting strategy of the motor; step 2.3: the scheme mainly adopts an LSTM visual neural network to estimate the weight according to the length and width of the body contour of a person, so as to realize self-adaptive estimation of the complex weight from the length and width of the person in a picture; the effect of load evaluation is achieved, the motor starting strategy is further adjusted in a self-adaptive mode, the use of an expensive gravity sensor is avoided, and the cost in industrial production is reduced.
Description
Technical Field
The invention relates to a control method, in particular to a data acquisition control method for an electric hanging basket, and belongs to the technical field of data acquisition control of electric hanging baskets.
Background
The electric hanging basket is a manned carrying and lifting device which is commonly used on a construction site, and can be used for assisting high-altitude operation personnel to complete the work of installing a glass curtain wall and the like outside a building. A set of electric hanging basket data acquisition system is developed to upgrade and reform the traditional hanging basket, the system has the function of networking and uploading data, and a 4G module is adopted for networking.
The internet of things equipment often uses an embedded system as a terminal, the embedded terminal uses an embedded microprocessor as a main control chip, acquires data of the equipment by accessing different sensors, and accesses a network module for networking. Some embedded microprocessors are also provided with a special image processing chip, and can compress and encode videos and audios collected by a camera and a microphone. And finally, the data are sent to the cloud end through a network interface of the embedded operating system. Compared with the traditional acquisition system, the embedded data acquisition system has the advantages of low cost, small volume, low power consumption, convenient use and the like.
Safety information such as a hanging basket main rope, a safety rope, an inclination angle and the like is acquired in real time through a series of sensors arranged on the hanging basket. The information is processed in a centralized way by an embedded system, closed-loop adjustment can be carried out, and an alarm can be sent out locally and rapidly to prevent accidents. Meanwhile, the safety information and the additional information such as videos and sounds can be uploaded to the cloud, so that managers can conveniently and comprehensively know the operation condition of the hanging basket, and the high-level business functions such as hanging basket leasing can be expanded. Therefore, the operation state of the hanging basket can be controlled in real time, potential safety hazards can be found in time, evidences can be reserved for responsibility confirmation, and upper-layer services can be enriched.
Disclosure of Invention
The invention provides an electric hanging basket data acquisition control method aiming at the problems in the prior art, and the technical scheme provides an electric hanging basket data acquisition control system which completes the safety measures of attendance check-in management, epidemic situation prevention and control management, improvement of safety belts, wearing of safety helmets, hanging basket inclination angle monitoring, hanging rope tension testing and the like for workers through an embedded Linux system, wherein the safety measures comprise serial port data, network data receiving and sending processing, a timed task service program, a video automatic lead-out program, an audio and video acquisition coding program, a streaming media service program operation environment configuration script, a GPIO (general purpose input/output) and a network monitoring program.
In order to achieve the purpose, the technical scheme of the invention is that the method for collecting and controlling the data of the electric hanging basket comprises the following steps:
step 1: video acquisition and picture analysis and processing are specifically as follows:
step 1.1: video acquisition and storage and data preprocessing;
step 1.2: face recognition analysis processing;
step 1.3: predicting the weight of the employee by the neural network;
step 1.4: self-adaptive repairing of network packet loss;
step 2: the real-time data analysis and processing of the parameters of the hanging basket are as follows:
step 2.1: estimating the total load of the motor;
step 2.2: self-adaptive starting strategy of the motor;
step 2.3: transmitting and processing basket parameters;
the video acquisition, storage and data preprocessing in the step 1.1 are specifically as follows:
step 1.1.1: setting the attribute of Internet protocol version 4(TCP/IPv4), matching the ip address with the ip address of an external camera, applying rtsp plug flow protocol to access a CMOS camera, coding a video signal into an analog signal, transmitting the analog signal into an analog video coding chip of a circuit board through a coaxial line and then accessing the analog video coding chip into a camera interface of an MCU (microprogrammed control Unit), and if strong electric interference is more in an industrial field, shielding an electromagnetic induction transmission signal by a twisted pair; meanwhile, the video is coded in an H.264 mode with low code rate, high quality, strong fault tolerance and strong network adaptability;
step 1.1.2: for the coded video and audio streams, the RTMP protocol which has better real-time performance in the streaming media protocol and supports the video streams and the audio streams of various coding modes is adopted; the network is connected to an NB-IoT module, and the video stream is stored in an SD card or uploaded to a background real-time monitoring interface;
step 1.1.3: performing cross compilation at a PC (personal computer) end, downloading the cross compilation into an embedded system, transmitting the cross compilation to a cloud end through a network serial port of an embedded operating system, and acquiring face photos, body shape photos and weight data of workers as training data of the step 2.3.2; converting the face photos into a pgm format, unifying the sizes of the photos, respectively placing 10 photos in a folder with a specified sequence number to generate a csv label file, marking the figure photos by an OpenCV function getOrientation and a PCA method, outputting the length and width of the figure outline, and generating the label file for storage;
wherein, the step 1.2 of face recognition analysis processing specifically comprises the following steps:
the method comprises the steps of judging names of workers through face recognition, checking in and calculating the total number of the workers in a hanging basket, training a model by using a Facerecognizer type Eigenfaces function in OpenCV, extracting a face csv label file processed in the step 1.1.3, training to generate a face xml model file, firstly opening a camera based on an MFC design monitoring platform, loading the face model generated above, detecting faces, checking in names of the workers, and recognizing a total number of people in the hanging basket.
Step 1.3, predicting the weight of the employee by using a neural network, wherein the method specifically comprises the following steps:
step 1.3.1: a worker stands at a position appointed by a camera, a system automatically shoots a figure photo, the figure photo needs to be marked out by an OpenCV function getOrientation and a PCA method, the length and the width of the figure outline are output, and for the mobility of a project, an LSTM neural network is trained to conveniently and directly estimate the weight according to the length and the width of the figure of a human body;
step 1.3.2: inputting a body shape training data set generated in the step 1.1.3 of the LSTM network, wherein the body shape training data set contains label files with length and width of the body shape outline, corresponding weight data is used as supervision data, and a neural network is trained to obtain a result. The weight of the person can be directly predicted from the length and the width of the figure picture;
step 1.3.3: inputting the length and width of the body shape collected in real time into the well-trained LSTM neural network, predicting the approximate weight range of the person by the network, recognizing the total number of people in the hanging basket through the step 1.2, shooting for multiple times by the system, and accumulating the weight of the workers predicted by the network.
Step 1.4 is self-adaptive repairing of network packet loss, and the specific method is as follows:
step 1.4.1: the bandwidth of the blocked link is properly increased in order to reduce the network packet loss caused by the network link blockage;
step 1.4.2: real-time applications are prioritized using Qos (traffic priority and resource reservation control mechanism) and although this approach does not alleviate network link congestion, it can prioritize voice and video to reduce the likelihood of disconnection.
Wherein, step 2.1 estimates the total load of the motor, and the specific method is as follows:
and (3) accumulating the weight of the worker delivered by the predicted device in the step (1.2) by the total weight of the hanging basket device, wherein the total weight of the hanging basket device comprises the self weight of the hanging basket, the total weight of a rope connected with a motor, a speed sensor, a frequency converter, an inclinometer, an accelerometer, a hanging rope force sensor and other devices, the wind speed sensor measures the wind speed, the force exerted on the hanging basket by the external environment at the moment is obtained through buoyancy and pressure formula calculation, and therefore the total load on the motor is obtained through calculation.
Step 2.2, a self-adaptive starting strategy of the motor is specifically as follows:
step 2.2.1: a closed-loop control algorithm is adopted to quickly and stably start the motor. And when the total load weight is more than 200kg, adopting a PI control algorithm. When the load is increased, that is, the armature circuit series resistance R1 is increased, the armature voltage Ua and the magnetic flux phi are kept unchanged, and the load torque T is kept unchanged1And when the speed is not changed, the electromagnetic torque T and the armature current Ia are also not changed, and the mechanical characteristic equation after speed regulation is as follows:
where n is the speed, Ua is the voltage of the armature, R1 is the series resistance of the armature circuit, phi is the magnetic flux, T1T is a load torque, Ia is an armature current, Ce is an electromotive constant, and CTIs the torque constant, Ra is the armature resistance;
it can be seen that when R1 increases, the motor speed n decreases, and when Ua and Ia are not changed, the total power P1 input from the power supply is not changed to UaIa. With the increase of the voltage dependent resistor R1, the copper loss Pcu ═ Ia of the winding2(Ra + R1) is followed byIncreasing the electromagnetic power P thus converted into mechanical powerMWill decrease when P1-Pcu is equal toMWhen the load torque is unchanged, the rotating speed of the motor is reduced along with the increase of the armature loop series resistor R1, after the rotating speed of the motor is reduced, the PI parameter is adjusted, the rotating speed can be increased by increasing the armature power supply voltage stage or reducing the magnetic flux phi, and the motor is adjusted to operate at different powers;
step 2.2.2: the total load weight is less than 200kg, the system is easy to be interfered, a fuzzy PI control algorithm is adopted, the fuzzy PI control algorithm combines the advantages of a fuzzy controller and a traditional PI controller, the system has good dynamic and steady-state performance and anti-interference capability, the fuzzy controller takes errors and error changes as input, PI parameters kp and ki are modified in real time, and the requirement of self-tuning the PI parameters at different moments can be met;
step 2.2.3: the fuzzy PI algorithm of the system takes a feedback speed omega and a speed set value omega ref of a motor encoder as input quantities, outputs output quantities to a current loop through input quantity fuzzification, a data expert base, fuzzy reasoning and defuzzification, and enables a three-way resistance method to sample three-phase current to obtain A-phase current ia and B-phase current ib in a control system, and time-varying current values i alpha and i beta which are orthogonal to each other are obtained through Clarke transformation according to the three-phase current characteristics of ia + ib + ic which is 0; obtaining orthogonal current constants id and iq under a rotating coordinate system by performing Park transformation on the ia and the i beta; and comparing id and iq with a speed variation i x d and a target reference value i x q to be used as error input quantity of a current loop PI algorithm, converting the error input quantity into a static reference coordinate system after inverse Park conversion to obtain v alpha and v beta, obtaining va, vb and vc after the v alpha and v beta are subjected to inverse Clarke conversion, calculating a new PWM duty ratio value according to the three-phase voltage value, and generating an expected voltage vector, thereby completing the control of the speed and current double closed loop by the system.
And 2.3, transmitting and processing basket parameters specifically as follows:
step 2.3.1: file transmission is carried out by adopting an FTP protocol, a set of FTP server is built, and a file uploading function can be conveniently realized by adopting a corresponding FTP tool at a device end;
step 2.3.2: considering that the FTP protocol is only responsible for the transmission of files and the server cannot know whether newly uploaded files exist, a reporting process is also required to be designed for the data needing to be replied in the voice talkback mode, and when the files are successfully uploaded, the newly uploaded file names are reported to the background in a common data mode and an instruction is sent to the background;
step 2.3.3: the background receives the instructions and responds.
Compared with the prior art, the invention has the advantages that,
1) the invention mainly adopts the LSTM visual neural network to estimate the weight according to the length and the width of the body contour of the person, thereby realizing the self-adaptive estimation of the complex weight from the length and the width of the person in the picture; the effect of load evaluation is achieved, the motor starting strategy is adjusted in a self-adaptive manner, the use of an expensive gravity sensor is avoided, and the cost in industrial production is reduced;
2) the invention adopts the Fuzzy PID algorithm to improve the response speed and the control precision of the control algorithm, is convenient for mass production, prepares for the design of a high-precision servo control system, and in a linear control theory, the integral control action can eliminate steady-state errors, but the dynamic response is slow, the proportional control action has fast dynamic response, and the proportional integral control action can obtain higher steady-state precision and has faster dynamic response, so that a P control strategy is introduced into a Fuzzy controller to form Fuzzy-PI composite control, thereby improving the steady-state performance of the Fuzzy controller;
3) the invention adopts a video monitoring processing scheme of 'video acquisition module-image compression module + DE storage module + network module + embedded processor', mainly realizes an embedded network video terminal with video acquisition function and MPEG-4 hardware coding, is accessed into the cloud for management, can utilize the NPU module of the Haisi processor to perform mode identification, and completes high-level functions such as video analysis and the like;
4) the invention adopts a self-adaptive motor exception handling strategy, when the hanging basket is detected to be overweight, the inclination angle is overlarge and the acceleration is overlarge, a circuit is disconnected, the motor is stopped to rotate, a cable is tensioned and locked, and multiple detection and confirmation can be carried out at the moment; if the circuit faults such as overlarge inclination angle, overlarge acceleration and the like are confirmed, restarting the main circuit and reporting an abnormal fault to a background; if the hanging basket is determined to be overweight, a hanging basket loudspeaker prompts that the personnel are overweight, reports an abnormal fault to a background, and restarts the main circuit when the tension of the main lifting rope is restored to a normal value;
5) since the device requires adaptive motor control, the present invention provides for script auto-configuration runtime environments, such as: dynamic link library, connection network, synchronization time and system upgrade; because the system keeps stable operation for a long time, a series of protection programs such as GPIO related programs, network monitoring programs and other system environment monitoring repair programs are set to automatically detect and repair system faults or indicate the working state of the system so as to restore the normal working state as soon as possible. Such as: a network monitoring program is needed to monitor the on-off of the network and restart the network connection module when the network is disconnected; when the motor is restarted, the environment detection program automatically configures the operation environment, and the motor is ensured to be started in a more stable state.
Drawings
FIG. 1 is a flow chart of the working principle of the present invention;
fig. 2 is a schematic diagram of a process of an adaptive starting strategy of the motor.
The specific implementation mode is as follows:
for the purpose of enhancing an understanding of the present invention, the present embodiment will be described in detail below with reference to the accompanying drawings.
Example 1: referring to fig. 1 and 2, a method for controlling data acquisition of an electric basket comprises the following steps:
step 1: video acquisition and picture analysis and processing are specifically as follows:
step 1.1: video acquisition and storage and data preprocessing;
step 1.2: face recognition analysis processing;
step 1.3: predicting the weight of the employee by the neural network;
step 1.4: self-adaptive repairing of network packet loss;
step 2: the real-time data analysis and processing of the parameters of the hanging basket are as follows:
step 2.1: estimating the total load of the motor;
step 2.2: self-adaptive starting strategy of the motor;
step 2.3: transmitting and processing basket parameters;
the video acquisition, storage and data preprocessing in the step 1.1 are specifically as follows:
step 1.1.1: setting the attribute of Internet protocol version 4(TCP/IPv4), matching the ip address with the ip address of an external camera, applying rtsp plug flow protocol to access a CMOS camera, coding a video signal into an analog signal, transmitting the analog signal into an analog video coding chip of a circuit board through a coaxial line and then accessing the analog video coding chip into a camera interface of an MCU (microprogrammed control Unit), and if strong electric interference is more in an industrial field, shielding an electromagnetic induction transmission signal by a twisted pair; meanwhile, the video is coded in an H.264 mode with low code rate, high quality, strong fault tolerance and strong network adaptability;
step 1.1.2: for the coded video and audio streams, the RTMP protocol which has better real-time performance in the streaming media protocol and supports the video streams and the audio streams of various coding modes is adopted; the network is connected to an NB-IoT module, and the video stream is stored in an SD card or uploaded to a background real-time monitoring interface;
step 1.1.3: performing cross compilation at a PC (personal computer) end, downloading the cross compilation into an embedded system, transmitting the cross compilation to a cloud end through a network serial port of an embedded operating system, and acquiring face photos, body shape photos and weight data of workers as training data of the step 2.3.2; converting the face photos into a pgm format, unifying the size of the photos, respectively placing 10 photos in a folder with a specified sequence number to generate a csv label file, marking the figure photos by an OpenCV function getOrientation and a PCA method, outputting the length and width of the figure outline, and generating the label file for storage.
Wherein, the step 1.2 of face recognition analysis processing specifically comprises the following steps:
the method comprises the steps of judging names of workers through face recognition, checking in and calculating the total number of the workers in a hanging basket, training a model by using a Facerecognizer type Eigenfaces function in OpenCV, extracting a face csv label file processed in the step 1.1.3, training to generate a face xml model file, firstly opening a camera based on an MFC design monitoring platform, loading the face model generated above, detecting faces, checking in names of the workers, and recognizing a total number of people in the hanging basket.
Step 1.3, predicting the weight of the employee by using a neural network, wherein the method specifically comprises the following steps:
step 1.3.1: a worker stands at a position appointed by a camera, a system automatically shoots a figure photo, the figure photo needs to be marked out by an OpenCV function getOrientation and a PCA method, the length and the width of the figure outline are output, and for the mobility of a project, an LSTM neural network is trained to conveniently and directly estimate the weight according to the length and the width of the figure of a human body;
step 1.3.2: inputting a body shape training data set generated in the step 1.1.3 of the LSTM network, wherein the body shape training data set contains label files with length and width of the body shape outline, corresponding weight data is used as supervision data, and a neural network is trained to obtain a result. The weight of the person can be directly predicted from the length and the width of the figure picture;
step 1.3.3: inputting the length and width of the body shape collected in real time into the well-trained LSTM neural network, predicting the approximate weight range of the person by the network, recognizing the total number of people in the hanging basket through the step 1.2, shooting for multiple times by the system, and accumulating the weight of the workers predicted by the network.
Step 1.4 is self-adaptive repairing of network packet loss, and the specific method is as follows:
step 1.4.1: the bandwidth of the blocked link is properly increased in order to reduce the network packet loss caused by the network link blockage;
step 1.4.2: real-time applications are prioritized using Qos (traffic priority and resource reservation control mechanism) and although this approach does not alleviate network link congestion, it can prioritize voice and video to reduce the likelihood of disconnection.
Wherein, step 2.1 estimates the total load of the motor, and the specific method is as follows:
and (3) accumulating the weight of the worker delivered by the predicted device in the step (1.2) by the total weight of the hanging basket device, wherein the total weight of the hanging basket device comprises the self weight of the hanging basket, the total weight of a rope connected with a motor, a speed sensor, a frequency converter, an inclinometer, an accelerometer, a hanging rope force sensor and other devices, the wind speed sensor measures the wind speed, the force exerted on the hanging basket by the external environment at the moment is obtained through buoyancy and pressure formula calculation, and therefore the total load on the motor is obtained through calculation.
Step 2.2, a self-adaptive starting strategy of the motor is specifically as follows:
step 2.2.1: a closed-loop control algorithm is adopted to quickly and stably start the motor. And when the total load weight is more than 200kg, adopting a PI control algorithm. When the load is increased, that is, the armature circuit series resistance R1 is increased, the armature voltage Ua and the magnetic flux phi are kept unchanged, and the load torque T is kept unchanged1And when the speed is not changed, the electromagnetic torque T and the armature current Ia are also not changed, and the mechanical characteristic equation after speed regulation is as follows:
where n is the speed, Ua is the voltage of the armature, R1 is the series resistance of the armature circuit, phi is the magnetic flux, T1T is a load torque, Ia is an armature current, Ce is an electromotive constant, and CTIs the torque constant, Ra is the armature resistance;
it can be seen that when R1 increases, the motor speed n decreases, and when Ua and Ia are not changed, the total power P1 input from the power supply is not changed to UaIa. With the increase of the voltage dependent resistor R1, the copper loss Pcu ═ Ia of the winding2(Ra + R1) is also accompanied by an increase in the electromagnetic power P thus converted into mechanical powerMWill decrease when P1-Pcu is equal toMWhen the load torque is unchanged, the rotating speed of the motor is reduced along with the increase of the armature loop series resistor R1, after the rotating speed of the motor is reduced, the PI parameter is adjusted, the rotating speed can be increased by increasing the armature power supply voltage stage or reducing the magnetic flux phi, and the motor is adjusted to operate at different powers;
step 2.2.2: the total load weight is less than 200kg, the system is easy to be interfered, a fuzzy PI control algorithm is adopted, the fuzzy PI control algorithm combines the advantages of a fuzzy controller and a traditional PI controller, the system has good dynamic and steady-state performance and anti-interference capability, the fuzzy controller takes errors and error changes as input, PI parameters kp and ki are modified in real time, and the requirement of self-tuning the PI parameters at different moments can be met;
step 2.2.3: the fuzzy PI algorithm of the system takes a feedback speed omega and a speed set value omega ref of a motor encoder as input quantities, outputs output quantities to a current loop through input quantity fuzzification, a data expert base, fuzzy reasoning and defuzzification, and enables a three-way resistance method to sample three-phase current to obtain A-phase current ia and B-phase current ib in a control system, and time-varying current values i alpha and i beta which are orthogonal to each other are obtained through Clarke transformation according to the three-phase current characteristics of ia + ib + ic which is 0; obtaining orthogonal current constants id and iq under a rotating coordinate system by performing Park transformation on the ia and the i beta; and comparing id and iq with a speed variation i x d and a target reference value i x q to be used as error input quantity of a current loop PI algorithm, converting the error input quantity into a static reference coordinate system after inverse Park conversion to obtain v alpha and v beta, obtaining va, vb and vc after the v alpha and v beta are subjected to inverse Clarke conversion, calculating a new PWM duty ratio value according to the three-phase voltage value, and generating an expected voltage vector, thereby completing the control of the speed and current double closed loop by the system.
And 2.3, transmitting and processing basket parameters specifically as follows:
step 2.3.1: file transmission is carried out by adopting an FTP protocol, a set of FTP server is built, and a file uploading function can be conveniently realized by adopting a corresponding FTP tool at a device end;
step 2.3.2: considering that the FTP protocol is only responsible for the transmission of files and the server cannot know whether newly uploaded files exist, a reporting process is also required to be designed for the data needing to be replied in the voice talkback mode, and when the files are successfully uploaded, the newly uploaded file names are reported to the background in a common data mode and an instruction is sent to the background;
step 2.3.3: the background receives the instructions and responds.
The specific embodiment is as follows: in the example, the Haisi processor Hi3516 is used as the main control chip of the embedded system, and the version of the Linux kernel is 2.6.35. A data acquisition control method for an electric hanging basket comprises the following steps:
step 1: video acquisition and picture analysis and processing are specifically as follows:
step 1.1: video acquisition and storage and data preprocessing;
step 1.1.1: setting the attribute of Internet protocol version 4(TCP/IPv4), matching the ip address with the ip address of an external camera, applying rtsp plug flow protocol to access a CMOS camera, coding a video signal into an analog signal, transmitting the analog signal into an analog video coding chip of a circuit board through a coaxial line and then accessing the analog video coding chip into a camera interface of an MCU (microprogrammed control Unit), and if strong electric interference is more in an industrial field, shielding an electromagnetic induction transmission signal by a twisted pair; meanwhile, the video is coded in an H.264 mode with low code rate, high quality, strong fault tolerance and strong network adaptability;
step 1.1.2: for the coded video and audio streams, the RTMP protocol which has better real-time performance in the streaming media protocol and supports the video streams and the audio streams of various coding modes is adopted; and accessing to NB-IoT module networking, and storing the video stream to the SD card or uploading to a background real-time monitoring interface.
Step 1.1.3: and (4) performing cross compiling at the PC terminal, downloading the compiled data into an embedded system, transmitting the compiled data to the cloud terminal through a network serial port of the embedded operating system, and collecting the face photos, the body shape photos and the weight data of the workers as training data in the step 4. The face photos are converted into a pgm format, the photos are unified in size, and 10 photos are respectively placed in a folder with a specified serial number. And generating a csv label file. The figure photo needs to be marked out by an OpenCV function getOrientation and a PCA method, and the length and the width of the figure outline are output to generate a label file for storage.
Step 1.2: face recognition analysis processing;
and judging the name of the worker by face recognition to sign in and calculating the total number of the workers in the hanging basket. And (3) training a model by using a facerecognizers Eigenfaces function in OpenCV, extracting the face csv label file processed in the step 1.1.3, and training to generate a face xml model file. Based on an MFC design monitoring platform, a camera is opened firstly, a face model generated on the camera is loaded, face detection is carried out, the name of a worker is checked in, and a plurality of persons are identified in a hanging basket.
Step 1.3: predicting the weight of the employee by the neural network;
step 1.3.1: when a worker stands at a position appointed by a camera, the system automatically shoots a figure photo, the figure photo needs to be marked out by an OpenCV function getOrientation and a PCA method, and the length and the width of the figure outline are output. Training the LSTM neural network facilitates direct weight estimation based on the length and width of the human body for item mobility.
Step 1.3.2: inputting a body shape training data set generated in the step 1.1.3 of the LSTM network, wherein the body shape training data set contains label files with length and width of the body shape outline, corresponding weight data is used as supervision data, and a neural network is trained to obtain a result. The weight of the person can be directly predicted from the length and the width of the figure picture.
Step 1.3.3: the length and width of the body shape collected in real time are input into a trained LSTM neural network, and the network predicts the approximate weight range of the person. And (4) shooting for multiple times by the system according to the total number of people in the hanging basket identified in the step 1.3, and accumulating the weight of the workers predicted by the network.
Step 1.4: self-adaptive repairing of network packet loss;
step 1.4.1: the bandwidth of the blocked link should be increased appropriately to reduce the network packet loss caused by the network link blocking.
Step 1.4.2: real-time applications are prioritized using Qos (traffic priority and resource reservation control mechanism) and although this approach does not alleviate network link congestion, it can prioritize voice and video to reduce the likelihood of disconnection.
Step 2: the real-time data analysis and processing of the parameters of the hanging basket are as follows:
step 2.1: estimating the total load of the motor;
step 2.1.1: and (3) accumulating the weight of the worker delivered by the predicted worker in the step (1.2) by the total weight of the hanging basket equipment, wherein the total weight of a rope connected with a motor, a speed sensor, a frequency converter, an inclinometer, an accelerometer, a lifting rope force sensor and the like are included. The wind speed sensor measures the wind speed, and the force exerted on the hanging basket by the external environment at the moment is obtained through buoyancy and pressure formula calculation. From this, the total load on the motor is calculated.
Step 2.2: self-adaptive starting strategy of the motor;
step 2.2.1: a closed-loop control algorithm is adopted to quickly and stably start the motor. And when the total load weight is more than 200kg, adopting a PI control algorithm. When the load is increased, that is, the armature circuit series resistance R1 is increased, the armature voltage Ua and the magnetic flux phi are kept unchanged, and the load torque T is kept unchanged1And when the speed is not changed, the electromagnetic torque T and the armature current Ia are also not changed, and the mechanical characteristic equation after speed regulation is as follows:
where n is the speed, Ua is the voltage of the armature, R1 is the series resistance of the armature circuit, phi is the magnetic flux, T1T is a load torque, Ia is an armature current, Ce is an electromotive constant, and CTIs the torque constant, Ra is the armature resistance;
it can be seen that when R1 increases, the motor speed n decreases, and when Ua and Ia are not changed, the total power P1 input from the power supply is not changed to UaIa. With the increase of the voltage dependent resistor R1, the copper loss Pcu ═ Ia of the winding2(Ra + R1) is also accompanied by an increase in the electromagnetic power P thus converted into mechanical powerMWill decrease when P1-Pcu is equal toMWhen the load torque is unchanged, the rotating speed of the motor is reduced along with the increase of the armature loop series resistor R1, after the rotating speed of the motor is reduced, the PI parameter is adjusted, the rotating speed can be increased by increasing the armature power supply voltage stage or reducing the magnetic flux phi, and the motor is adjusted to operate at different powers;
step 2.2.2: the total load weight is less than 200kg, the system is easy to be interfered, and a fuzzy PI control algorithm is adopted. The fuzzy PI control algorithm combines the advantages of a fuzzy controller and a traditional PI controller, and the system has good dynamic and steady-state performance and anti-interference capability. The fuzzy controller takes the error and the error change as input, modifies the PI parameters kp and ki in real time, and can meet the requirement of self-tuning the PI parameters at different moments.
Step 2.2.3: the fuzzy PI algorithm of the system takes the feedback speed omega and the set speed value omega ref of a motor encoder as input quantities, and outputs the output quantities to a current loop through input quantity fuzzification, a data expert base, fuzzy reasoning and defuzzification. In a control system, a three-way resistance method is used for sampling three-phase current to obtain ia and ib, and according to the characteristics of the three-phase current of which the ia + ib + ic is 0, time-varying current values i alpha and i beta which are orthogonal to each other are obtained through Clarke transformation; obtaining orthogonal current constants id and iq under a rotating coordinate system by performing Park transformation on the ia and the i beta; and comparing id and iq with a speed variation i x d and a target reference value i x q to be used as error input quantity of a current loop PI algorithm, converting the error input quantity into a static reference coordinate system after inverse Park conversion to obtain v alpha and v beta, obtaining va, vb and vc after the v alpha and v beta are subjected to inverse Clarke conversion, calculating a new PWM duty ratio value according to the three-phase voltage value, and generating an expected voltage vector, thereby completing the control of the speed and current double closed loop by the system.
Step 2.3: transmitting and processing basket parameters;
step 2.3.1: and file transmission is carried out by adopting an FTP protocol, a set of FTP server is built, and a file uploading function can be conveniently realized by adopting a corresponding FTP tool at a device end.
Step 2.3.2: considering that the FTP protocol is only responsible for file transmission, and the server cannot know whether a newly uploaded file exists, a reporting process needs to be designed for the data needing to be replied for voice talkback, and when the file is successfully uploaded, the newly uploaded file name is reported to the background in a common data form, and an instruction is sent to the background.
Step 2.3.3: the background receives the instructions and responds.
It should be noted that the above-mentioned embodiments are not intended to limit the scope of the present invention, and all equivalent modifications and substitutions based on the above-mentioned technical solutions are within the scope of the present invention as defined in the claims.
Claims (8)
1. A data acquisition control method for an electric hanging basket is characterized by comprising the following steps:
step 1: video acquisition and picture analysis and processing are specifically as follows:
step 1.1: video acquisition and storage and data preprocessing;
step 1.2: face recognition analysis processing;
step 1.3: predicting the weight of the employee by the neural network;
step 1.4: self-adaptive repairing of network packet loss;
step 2: the real-time data analysis and processing of the parameters of the hanging basket are as follows:
step 2.1: estimating the total load of the motor;
step 2.2: self-adaptive starting strategy of the motor;
step 2.3: and (5) transmitting and processing basket parameters.
2. The method for controlling data acquisition of the electric hanging basket according to the claim 1, wherein the step 1.1 comprises video acquisition, storage and data preprocessing, and specifically comprises the following steps:
step 1.1.1: setting the attribute of Internet protocol version 4, namely TCP/IPv4, matching the ip address with the ip address of an external camera, applying rtsp plug flow protocol to access a CMOS camera, coding a video signal into an analog signal, transmitting the analog signal into an analog video coding chip of a circuit board through a coaxial line and then accessing the analog video coding chip into a camera interface of an MCU (microprogrammed control Unit), and if strong electric interference is more in an industrial field, shielding an electromagnetic induction transmission signal by using a twisted pair; meanwhile, the video is coded in an H.264 mode with low code rate, high quality, strong fault tolerance and strong network adaptability;
step 1.1.2: for the coded video and audio streams, the RTMP protocol which has better real-time performance in the streaming media protocol and supports the video streams and the audio streams of various coding modes is adopted; the network is connected to an NB-IoT module, and the video stream is stored in an SD card or uploaded to a background real-time monitoring interface;
step 1.1.3: performing cross compilation at a PC (personal computer) end, downloading the cross compilation into an embedded system, transmitting the cross compilation to a cloud end through a network serial port of an embedded operating system, and acquiring face photos, body shape photos and weight data of workers as training data of the step 2.3.2; converting the face photos into a pgm format, unifying the size of the photos, respectively placing 10 photos in a folder with a specified sequence number to generate a csv label file, marking the figure photos by an OpenCV function getOrientation and a PCA method, outputting the length and width of the figure outline, and generating the label file for storage.
3. The method for controlling data acquisition of the electric hanging basket according to claim 2, wherein the step 1.2 of face recognition analysis processing comprises the following specific steps:
the method comprises the steps of judging names of workers through face recognition, checking in and calculating the total number of the workers in a hanging basket, training a model by using a Facerecognizer type Eigenfaces function in OpenCV, extracting a face csv label file processed in the step 1.1.3, training to generate a face xml model file, firstly opening a camera based on an MFC design monitoring platform, loading the face model generated above, detecting faces, checking in names of the workers, and recognizing a total number of people in the hanging basket.
4. The data acquisition control method for the electric hanging basket according to the claim 3, characterized in that the step 1.3 neural network predicts the weight of the staff as follows:
step 1.3.1: a worker stands at a position appointed by a camera, a system automatically shoots a figure photo, the figure photo needs to be marked out by an OpenCV function getOrientation and a PCA method, the length and the width of the figure outline are output, and an LSTM neural network is trained to conveniently and directly estimate the weight according to the length and the width of the figure of the human body;
step 1.3.2: inputting a figure training data set generated in the step 1.1.3 of the LSTM network, wherein the figure training data set comprises label files of the length and the width of the figure outline, corresponding weight data is used as supervision data, and a neural network is trained to obtain a result, so that the weight of the person can be directly predicted from the length and the width of the figure picture;
step 1.3.3: inputting the length and width of the body shape collected in real time into the well-trained LSTM neural network, predicting the approximate weight range of the person by the network, recognizing the total number of people in the hanging basket through the step 1.2, shooting for multiple times by the system, and accumulating the weight of the workers predicted by the network.
5. The method for data acquisition and control of the electric hanging basket according to claim 4, wherein step 1.4 is a step of adaptive repair of network packet loss, and the specific method is as follows:
step 1.4.1: increasing the bandwidth of the blocked link to reduce the network packet loss caused by the network link blocking;
step 1.4.2: real-time applications are prioritized using Qos, i.e. traffic priority and resource reservation control mechanism.
6. The method for data collection and control of the electric cradle according to claim 5, wherein the step 2.1 estimates the total load of the motor by the following specific method:
and (3) accumulating the weight of the worker delivered to the hanging basket predicted in the step (1.2) by the total weight of the hanging basket equipment, wherein the total weight of a rope connected with a motor, a speed sensor, a frequency converter, an inclinometer, an accelerometer and a hanging rope force sensor equipment is used, the wind speed is measured by a wind speed sensor, the force exerted on the hanging basket by the external environment at the moment is obtained through buoyancy and pressure formula calculation, and therefore the total load on the motor is obtained through calculation.
7. The method for collecting and controlling the data of the electric basket according to claim 6, wherein the step 2.2 is that the self-adaptive starting strategy of the motor is as follows:
step 2.2.1: adopting a closed-loop control algorithm, and when the total weight of the load is more than 200kg, adopting a PI control algorithm, and when the load is increased, namely the armature loop series resistance R1 is increased, the armature voltage Ua and the magnetic flux phi are kept unchanged, and the load torque T1 is also unchanged, the electromagnetic torque T and the armature current Ia are also unchanged, and the mechanical characteristic equation after speed regulation is as follows:
where n is the rotation speed, Ua is the armature voltage, R1 is the armature loop series resistance, phi is the magnetic flux, T1 is the load torque, T is the electromagnetic torque, Ia is the armature loop series resistanceArmature current, Ce is the electromotive constant, CTIs the torque constant, Ra is the armature resistance;
it can be known that, when R1 is increased, the motor speed n will decrease, and when Ua and Ia are unchanged, the total power P1 input from the power supply is unchanged as UaIa, and as the voltage dependent resistor R1 increases, the winding copper loss Pcu is changed as Ia2(Ra + R1) is also accompanied by an increase in the electromagnetic power P thus converted into mechanical powerMWill decrease when P1-Pcu is equal toMWhen the load torque is unchanged, the rotating speed of the motor is reduced along with the increase of the armature loop series resistance R1, after the rotating speed of the motor is reduced, the PI parameter is adjusted, the rotating speed is increased by increasing the armature power supply voltage table to reduce the magnetic flux phi, and the motor is further adjusted to operate at different power;
step 2.2.2: the total load weight is less than 200kg, the system is easy to be interfered, a fuzzy PI control algorithm is adopted, the fuzzy PI control algorithm combines the advantages of a fuzzy controller and a traditional PI controller, the system has good dynamic and steady-state performance and anti-interference capability, the fuzzy controller takes errors and error changes as input, PI parameters kp and ki are modified in real time, and the requirement of self-tuning of the PI parameters at different moments is met;
step 2.2.3: the fuzzy PI algorithm of the system takes a feedback speed omega and a speed set value omega ref of a motor encoder as input quantities, outputs output quantities to a current loop through input quantity fuzzification, a data expert base, fuzzy reasoning and defuzzification, and enables a three-way resistance method to sample three-phase current to obtain A-phase current ia and B-phase current ib in a control system, and time-varying current values i alpha and i beta which are orthogonal to each other are obtained through Clarke transformation according to the three-phase current characteristics of ia + ib + ic which is 0; obtaining orthogonal current constants id and iq under a rotating coordinate system by performing Park transformation on the ia and the i beta; and comparing id and iq with a speed variation i x d and a target reference value i x q to be used as error input quantity of a current loop PI algorithm, converting the error input quantity into a static reference coordinate system after inverse Park conversion to obtain v alpha and v beta, obtaining va, vb and vc after the v alpha and v beta are subjected to inverse Clarke conversion, calculating a new PWM duty ratio value according to the three-phase voltage value, and generating an expected voltage vector, thereby completing the control of the speed and current double closed loop by the system.
8. The method for data acquisition and control of an electric basket according to claim 7, wherein step 2.3 is transmission and processing of basket parameters, specifically as follows:
step 2.3.1: file transmission is carried out by adopting an FTP protocol, a set of FTP server is built, and a file uploading function is conveniently realized by adopting a corresponding FTP tool at a device end;
step 2.3.2: considering that the FTP protocol is only responsible for the transmission of files and the server cannot know whether newly uploaded files exist, a reporting process is also required to be designed for the data needing to be replied in the voice talkback mode, and when the files are successfully uploaded, the newly uploaded file names are reported to the background in a common data mode and an instruction is sent to the background;
step 2.3.3: the background receives the instructions and responds.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110362590.2A CN113277388B (en) | 2021-04-02 | 2021-04-02 | Data acquisition control method for electric hanging basket |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110362590.2A CN113277388B (en) | 2021-04-02 | 2021-04-02 | Data acquisition control method for electric hanging basket |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113277388A true CN113277388A (en) | 2021-08-20 |
| CN113277388B CN113277388B (en) | 2023-05-02 |
Family
ID=77276228
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110362590.2A Active CN113277388B (en) | 2021-04-02 | 2021-04-02 | Data acquisition control method for electric hanging basket |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113277388B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114590668A (en) * | 2022-05-10 | 2022-06-07 | 大汉科技股份有限公司 | Operation management system for unmanned elevator based on big data |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103701396A (en) * | 2013-12-13 | 2014-04-02 | 天津大学 | Motor rotating-speed tracking control method based on self-adaptive fuzzy neural network |
| CN105577058A (en) * | 2015-12-28 | 2016-05-11 | 江苏大学 | Novel fuzzy active disturbance rejection controller based five-phase fault-tolerant permanent magnet motor speed control method |
| WO2018020275A1 (en) * | 2016-07-29 | 2018-02-01 | Unifai Holdings Limited | Computer vision systems |
| CN107678340A (en) * | 2017-10-10 | 2018-02-09 | 中建八局第二建设有限公司 | A kind of intelligent hanging basket monitoring system, comprehensive monitoring system and monitoring method |
| CN109507869A (en) * | 2018-11-15 | 2019-03-22 | 南京越博电驱动系统有限公司 | A kind of optimization method of the motor control PI parameter suitable for permanent magnet synchronous motor |
| CN109761118A (en) * | 2019-01-15 | 2019-05-17 | 福建天眼视讯网络科技有限公司 | Wisdom ladder networking control method and system based on machine vision |
| CN109829390A (en) * | 2019-01-09 | 2019-05-31 | 浙江新再灵科技股份有限公司 | A kind of elevator intelligent scheduling system and method based on deep learning |
| CN110311807A (en) * | 2019-06-06 | 2019-10-08 | 东南大学 | A kind of driven hanging basket data collection system Network status adaptive process monitoring method |
| CN110807434A (en) * | 2019-11-06 | 2020-02-18 | 威海若维信息科技有限公司 | Pedestrian re-identification system and method based on combination of human body analysis and coarse and fine particle sizes |
| CN111476195A (en) * | 2020-04-20 | 2020-07-31 | 安徽中科首脑智能医疗研究院有限公司 | Face detection method, face detection device, robot and computer-readable storage medium |
| WO2020258508A1 (en) * | 2019-06-27 | 2020-12-30 | 平安科技（深圳）有限公司 | Model hyper-parameter adjustment and control method and apparatus, computer device, and storage medium |
-
2021
- 2021-04-02 CN CN202110362590.2A patent/CN113277388B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103701396A (en) * | 2013-12-13 | 2014-04-02 | 天津大学 | Motor rotating-speed tracking control method based on self-adaptive fuzzy neural network |
| CN105577058A (en) * | 2015-12-28 | 2016-05-11 | 江苏大学 | Novel fuzzy active disturbance rejection controller based five-phase fault-tolerant permanent magnet motor speed control method |
| WO2018020275A1 (en) * | 2016-07-29 | 2018-02-01 | Unifai Holdings Limited | Computer vision systems |
| CN107678340A (en) * | 2017-10-10 | 2018-02-09 | 中建八局第二建设有限公司 | A kind of intelligent hanging basket monitoring system, comprehensive monitoring system and monitoring method |
| CN109507869A (en) * | 2018-11-15 | 2019-03-22 | 南京越博电驱动系统有限公司 | A kind of optimization method of the motor control PI parameter suitable for permanent magnet synchronous motor |
| CN109829390A (en) * | 2019-01-09 | 2019-05-31 | 浙江新再灵科技股份有限公司 | A kind of elevator intelligent scheduling system and method based on deep learning |
| CN109761118A (en) * | 2019-01-15 | 2019-05-17 | 福建天眼视讯网络科技有限公司 | Wisdom ladder networking control method and system based on machine vision |
| CN110311807A (en) * | 2019-06-06 | 2019-10-08 | 东南大学 | A kind of driven hanging basket data collection system Network status adaptive process monitoring method |
| WO2020258508A1 (en) * | 2019-06-27 | 2020-12-30 | 平安科技（深圳）有限公司 | Model hyper-parameter adjustment and control method and apparatus, computer device, and storage medium |
| CN110807434A (en) * | 2019-11-06 | 2020-02-18 | 威海若维信息科技有限公司 | Pedestrian re-identification system and method based on combination of human body analysis and coarse and fine particle sizes |
| CN111476195A (en) * | 2020-04-20 | 2020-07-31 | 安徽中科首脑智能医疗研究院有限公司 | Face detection method, face detection device, robot and computer-readable storage medium |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114590668A (en) * | 2022-05-10 | 2022-06-07 | 大汉科技股份有限公司 | Operation management system for unmanned elevator based on big data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113277388B (en) | 2023-05-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7949483B2 (en) | Integration of intelligent motor with power management device | |
| CN107168294B (en) | Unmanned inspection monitoring method for thermal power water system equipment | |
| CN113277388A (en) | Data acquisition control method for electric hanging basket | |
| Zare et al. | Low-cost ESP32, Raspberry Pi, Node-Red, and MQTT protocol based SCADA system | |
| CN103338259B (en) | Conference video equipment management and control method based on cloud technology | |
| CN112286150B (en) | Intelligent household equipment management method, device and system and storage medium | |
| EP1938446A1 (en) | Ac motor controller | |
| CN116136613A (en) | Automatic inspection method, device, equipment and medium for data center | |
| CN108317696B (en) | Control method and device of air conditioner and computer readable storage medium | |
| US11734932B2 (en) | State and event monitoring | |
| CN109066984A (en) | Transformer substation assisted devices monitoring system and monitoring method | |
| CN114938707A (en) | Method and system for processing sensor data for transmission | |
| CN116744357A (en) | Base station fault prediction method, device, equipment and medium | |
| CN106338934A (en) | Servo driver remote control method and apparatus | |
| US11005407B2 (en) | Data obtaining method, inverter, and rotating electric machine | |
| Ashmitha et al. | Real time monitoring IoT based methodology for fault detection in induction motor | |
| US11380173B2 (en) | Eliminating mechanical chime hum | |
| CN113965698A (en) | Monitoring image calibration processing method, device and system for fire-fighting Internet of things | |
| Chen et al. | A Fault Diagnosis Platform of Actuators on Embedded IoT Microcontrollers | |
| CN114249244B (en) | Target-free sling positioning method for automatic traveling crane closed-loop control | |
| CN117250900B (en) | Artificial intelligent motor control system based on automatic identification operation mode | |
| CN117154955B (en) | Cloud control method and system for intelligent power distribution room | |
| Aashika et al. | Smart Factory Using IOT | |
| CN117011092B (en) | Intelligent building equipment management monitoring system and method | |
| CN117792190A (en) | High-precision motor driving method |
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 |