CN210226693U

CN210226693U - Intelligent bird repelling device

Info

Publication number: CN210226693U
Application number: CN201920697297.XU
Authority: CN
Inventors: Chun Shi; 时春; Haima Liu; 刘海马; Jiangwei Chen; 陈江伟; Jun Lin; 林俊; Fenglei Wang; 王凤雷; Yixiao Wu; 吴义孝
Original assignee: Cixi Langhua Machinery Technology Co ltd
Current assignee: Cixi Langhua Machinery Technology Co ltd
Priority date: 2019-05-15
Filing date: 2019-05-15
Publication date: 2020-04-03
Anticipated expiration: 2029-05-15

Abstract

The utility model discloses an intelligence bird repellent device belongs to birds and drives technical field, including terminal, sound collection system, driver and play device, sound collection system and driver respectively with the terminal connection, the driver still is connected with play device, the terminal includes neural network module, bird sound database and instruction module. The utility model discloses a neural network algorithm difference birds are called for the sound, and the accuracy is to the driving of realization to photovoltaic power station birds to different birds through playing the mode of like tragic sound, natural enemy are called for the sound, ultrasonic wave etc.. The utility model discloses accurate high efficiency, energy-concerving and environment-protective, with low costs, the practicality is strong, easily promotes.

Description

Intelligent bird repelling device

Technical Field

The utility model relates to a birds drives technical field, especially relates to an intelligence bird repellent device.

Background

The photovoltaic power generation is an important component of the power generation industry in China, the generated energy is 1755 hundred million kilowatt hours in 2018, the equivalent growth is about 50%, and the development potential is huge; dust is easily accumulated on the surface of the photovoltaic module, and the cleanliness of the photovoltaic module greatly influences the power generation efficiency. Photovoltaic power generation station area is great, and birds prefer here to stop more, leads to birds excrement and urine more to, for the dust, birds excrement and urine influences the cleanliness on photovoltaic module surface more seriously, so must wash the generating efficiency who guarantees photovoltaic module to birds excrement and urine on the photovoltaic module.

Photovoltaic board generally adopts high-pressure squirt to clear up, and the range is far, and it is effectual to wash the dust, and personnel factor of safety is high. However, the bird feces are difficult to clean by the method, manual cleaning is needed, time and labor are consumed for manual washing, the method belongs to high-altitude operation, potential safety hazards exist, and the photovoltaic panel can be damaged by trampling. The most effective method is to treat from the source and reduce the stay of birds in the photovoltaic power station, thereby reducing the excrement of the birds.

The existing bird repelling technology is mainly divided into the following categories:

(1) some conventional bird repellers can be classified as action type, for example, common farmers hang on a bamboo pole by using a plastic bag and blow the plastic bag by using wind, and use a windmill bird repeller to drive a windmill to rotate by using wind, so that birds are frightened to repel birds. The two modes are greatly influenced by weather, and cannot play a role in windless or windless weather.

(2) The bird is driven by a smell method, and the bird driving agent is matched with a windmill to emit uncomfortable smell of the birds so as to drive the birds. However, the bird repellent is not environment-friendly, and brings discomfort to birds and human beings.

(3) Utilize the camera to shoot and carry out image processing after judging birds kind, recycle the ultrasonic wave and drive the bird, this kind of method is influenced seriously by the illumination, can't use evening to single camera can't satisfy the scope and cover, needs a plurality of cameras, and the cost is higher.

(4) The occupation method drives birds, utilizes the habit that birds like standing on the edge, sets up the silk thread that birds can't grasp on the subassembly to the purpose that the bird is driven is realized. However, the birds can still approach to each other, and the aim of thoroughly repelling the birds cannot be fulfilled.

(5) The sound bird repelling method is used for repelling birds through continuous high-noise playing or natural enemy sounds of the birds, but in a long-time high-noise and ultrasonic environment, a human body can be injured, then single sound is played for a long time, the bird repelling effect can not be achieved, the birds can only be fatigued, and the birds cannot be repelled.

SUMMERY OF THE UTILITY MODEL

Problem to prior art exists, the utility model provides a can be accurate to the expulsion of different birds, and the intelligent bird repellent device of high-efficient environmental protection. The utility model comprises the following technical scheme:

the utility model provides an intelligence bird repellent device, its characterized in that includes terminal, sound collection system, driver and play device, sound collection system and driver are connected with the terminal respectively, the driver still is connected with play device, the terminal includes neural network module, bird sound database and instruction module.

The working method of the utility model is as follows:

the method comprises the following steps that a neural network module of a terminal collects different types of sounds in advance and stores the sounds into a bird sound database;

after the sound collection device collects the sound of birds, the sound collection device transmits the sound to the terminal, the neural network module of the terminal extracts characteristic parameters of the sound, bird repelling sound stored in a bird sound database is found, and the terminal sends an instruction to control a driver to drive a sound playing device to play the bird repelling sound.

By adopting the technical scheme, the following beneficial effects are achieved:

(1) the utility model identifies the bird type through the bird sound by the neural network algorithm, is simpler and faster than the image identification, is not influenced by the environment and the climate, has lower cost, does not have shooting dead angle and has high bird-driving efficiency;

(2) the utility model can accurately drive birds by playing the cry, alarm, natural enemy and ultrasonic which is not suitable for the birds of the same kind, thereby reducing the adaptive situation of the birds;

(3) only when the sound collector collects the bird-scaring sound, the driver can drive the playing device to play the bird-scaring sound instead of continuously making the noise, so that the environmental noise is greatly reduced, and the device is energy-saving and environment-friendly;

(4) the utility model discloses the cost is lower, and the practicality is strong, easily promotes.

Drawings

Fig. 1 is a schematic structural view of the present invention;

FIG. 2 is a block diagram of the operation process of the neural network module for distinguishing bird species by using the BP neural network algorithm;

FIG. 3 is a schematic view of the speech feature parameter extraction process of the present invention;

FIG. 4 is a schematic diagram of neural network recognition according to the present invention;

in the figure: 1. frame, 2, adapter, 3, computer, 4, driver, 5, loudspeaker.

Detailed Description

As shown in fig. 1, an intelligent bird repelling device is characterized by comprising a terminal, a sound collecting device, a driver 4 and a playing device, wherein the sound collecting device and the driver 4 are respectively connected with the terminal, the driver 4 is further connected with the playing device, and the terminal comprises a neural network module, a bird sound database and an instruction module.

The utility model discloses a terminal still includes human-computer interaction interface.

The utility model discloses a computer 3 is adopted at the terminal.

The utility model discloses a sound collection system adopts adapter 2, and sound collection system's quantity is 1 at least. Specifically, 2 microphones were used.

The utility model discloses a play device adopts loudspeaker 5, and play device's quantity is 1 at least. Specifically, 3 horns were used.

The utility model discloses a sound collection system and play device all set up in frame 1.

The operation process of distinguishing the bird species by the neural network module by using the BP neural network algorithm is shown in figure 2, firstly, the sound of different types of birds is collected, extracted and subjected to feature processing, and the sound is trained by the BP neural network algorithm and stored in a bird sound database; and then, extracting characteristic parameters of the bird sounds to be classified collected by the pickup 2, inputting the parameters into a neural network, and comparing the newly input parameters with the parameters stored in the bird sound database to judge the bird species.

The feature extraction step is shown in fig. 3, and includes reading sound, preprocessing the sound, extracting FFT parameters, spectral line energy, MEL (MEL) filtering energy, logarithm Discrete Cosine Transform (DCT), and outputting sound feature parameters.

The BP neural network recognition principle is shown in FIG. 4 and mainly comprises three parts, namely neural network building, neural network training and neural network classification. The neural network building comprises: modeling a system and building a neural network structure; the neural network training includes: initializing a neural network and training the neural network; the neural network classification includes: and extracting characteristic parameters of the sound to be classified and comparing and classifying the sound to be classified with a neural network.

To sum up, the utility model discloses the working method of driving birds is as follows:

the neural network module collects different types of sounds in advance and stores the sounds in a bird sound database;

when the sound pickup 2 receives the bird sounds in the surrounding environment, the sound pickup 2 transmits the bird sounds to the computer 3, the computer 3 converts bird sound files into a WAV format, the neural network module extracts characteristic parameters of the bird sound files and compares the bird sound files with existing parameters in a bird sound database, then the computer 3 can display a pie chart to indicate the maximum possible birds, bird repelling sounds stored in the bird sound database are found, at the moment, the computer sends an instruction to the driver 4, and the driver 4 receives the instruction and drives the loudspeaker 5 to play corresponding bird repelling sounds.

The bird repelling sound is obtained by preferentially playing a tragic cry, an alarm, a natural enemy and 17KHZ-45KHZ ultrasonic waves of the same bird according to the identification result.

Claims

1. The utility model provides an intelligence bird repellent device, its characterized in that includes terminal, sound collection system, driver and play device, sound collection system and driver are connected with the terminal respectively, the driver still is connected with play device, the terminal includes neural network module, bird sound database and instruction module.

2. The intelligent bird repelling device of claim 1, wherein the terminal further comprises a human-computer interaction interface.

3. An intelligent bird repelling device according to claim 1, wherein the terminal employs a computer.

4. The intelligent bird repelling device of claim 1, wherein the sound collecting devices are pickups, and the number of the sound collecting devices is at least 1.

5. An intelligent bird repelling device according to claim 1, wherein the number of the playing devices is at least 1, and the playing devices are loudspeakers.

6. An intelligent bird repelling device according to claim 1, wherein the sound collection device and the playing device are both disposed on a rack.