A kind of intelligent control system for monitoring breeding equipment

Abstract

The present invention relates to a breeding equipment monitoring intelligent control system, and relates to the field of digital control technology. The system monitors and performs blockage analysis on the feeding port of automatic feed equipment through a distributed digital monitoring module, obtains blockage morphological parameters, performs blockage feeding simulation analysis, obtains feed waste analysis results, monitors and performs bonding analysis on the feed at the feeding port of the distributed automatic feed equipment, obtains bonding ratios, corrects the feed waste analysis results, obtains corrected feed waste analysis results, monitors feeding control parameters on blocked feeding channels, obtains feeding control parameters, sets control steps according to the corrected feed waste analysis results, optimizes feeding control parameters, obtains optimized feeding control parameters, and controls blocked feeding channels through a distributed digital control module. The present invention achieves the technical effects of reducing the probability of feeding blockage of breeding automated feed equipment, reducing feed waste, and ensuring breeding benefits.

Description

Technical Field

[0001] The present invention relates to the field of digital control, and in particular to an intelligent control system for monitoring breeding equipment. Background Art

[0002] In modern agricultural breeding, automatic feed feeding equipment has been widely used to improve breeding efficiency and reduce labor costs. However, the existing automatic feed feeding equipment often faces some technical difficulties in practical application, especially the blockage problem. The blockage of the feed feeding port not only affects the smooth delivery of feed, but also may cause equipment operation failure and feed waste, thereby reducing the production efficiency of the farm.

[0003] The current automatic feed feeding equipment only involves digital control of the feeding amount, etc., and cannot perform intelligent digital control according to the blockage situation when blocked. Therefore, there is a technical problem in the prior art that the automatic feed feeding equipment cannot perform intelligent digital control when blocked, resulting in feed waste and reduced breeding efficiency. Summary of the invention

[0004] The present invention aims to solve the technical problem that the automatic control of automatic feed feeding equipment in the prior art only involves digital control of the feeding amount, etc., and cannot perform intelligent digital control when blocked, resulting in feed waste and reduced breeding efficiency.

[0005] In a first aspect, the present invention provides an intelligent control system for monitoring breeding equipment, the system includes a distributed digital monitoring module and a distributed digital control module, and the system also includes: a blockage identification unit, which is used to monitor multiple feeding ports of distributed automatic feed equipment through the distributed digital monitoring module, obtain a feeding port image array, and perform blockage analysis. After any feeding port is blocked, the blockage form of at least one blocked feeding port is identified to obtain at least one blockage form parameter;

[0006] A feed waste analysis unit, configured to perform a blockage feeding simulation analysis according to the at least one blockage morphology parameter, obtain at least one feeding area distribution, and obtain at least one feed waste analysis result according to at least one preset feeding area;

[0007] A feed waste correction unit, configured to monitor and analyze the adhesion of feed at the feed inlet of the distributed automatic feed equipment through the distributed digital monitoring module, obtain the adhesion ratio, and perform correction analysis on the at least one feed waste analysis result to obtain at least one corrected feed waste analysis result;

[0008] A feeding parameter acquisition unit, used to monitor the feeding control parameters of at least one blocked feeding channel of the distributed automatic feed equipment through the distributed digital monitoring module to obtain at least one feeding control parameter;

[0009] A feeding control unit is used to set at least one control step according to the at least one corrected feed waste analysis result, optimize the at least one feeding control parameter, obtain at least one optimized feeding control parameter, and control the at least one blocked feeding channel through the distributed digital control module.

[0010] In a second aspect, the present invention provides an electronic device, comprising: a memory for storing a computer software program; and a processor for reading and executing the computer software program, thereby realizing the intelligent control system for monitoring breeding equipment of the first aspect.

[0011] In a third aspect, the present invention provides a computer-readable storage medium, in which a computer software program is stored. When the computer software program is executed by a processor, the intelligent control system for monitoring breeding equipment of the first aspect is implemented.

[0012] The present invention performs blockage analysis through a distributed digital monitoring module and a feed port image array of multiple feed ports during the operation of the distributed automatic feed equipment. After the feed port is blocked, blockage morphology recognition is performed to obtain blockage morphology parameters, and the blockage situation of the feed port can be identified. Then, according to the blockage morphology parameters, a simulation analysis of the blocked feeding area is performed to obtain the distribution of the feeding area, and the feed waste analysis result is obtained by combining the preset feeding area; then, the feed at the feed inlet of the distributed automatic feed equipment is monitored and bonded through the distributed digital monitoring module to obtain the bonding ratio, and the feed waste analysis result is corrected to obtain the corrected feed waste analysis result, which can correct the feed waste analysis result in the case of feed bonding, and improve the accuracy of subsequent optimization control; further, the feed control parameter of the blocked feeding channel of the current distributed automatic feed equipment is monitored to obtain the feeding control parameter; according to the corrected feed waste analysis result, the control step length is set to improve the optimization accuracy, the feeding control parameter is optimized to obtain the optimized feeding control parameter, and the blocked feeding channel is controlled through the distributed digital control module to ensure the feeding efficiency under the premise of reducing the blockage probability. The present invention uses a distributed digital monitoring module to monitor the feeding port of the automatic feed equipment in real time, identify the blockage form and analyze the feeding area distribution and feed waste, make corrections based on the adhesion analysis results, and perform feeding control based on the optimized feeding control parameters. This solution effectively reduces feed waste, reduces the probability of blockage, optimizes feeding efficiency, and improves the stability of the operation of the breeding feed equipment and the level of intelligent control. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 A schematic diagram of the structure of an intelligent control system for monitoring breeding equipment provided by an embodiment of the present invention;

[0014] Figure 2 A functional block diagram of an electronic device provided by an embodiment of the present invention;

[0015] Figure 3 A principle block diagram of the computer-readable storage medium provided by the present invention.

[0016] In the accompanying drawings, the components represented by the reference numerals are listed as follows:

[0017] Blockage identification unit 10, feed waste analysis unit 20, feed waste correction unit 30, feeding parameter collection unit 40, feeding control unit 50, electronic device 100, memory 110, processor 120, computer program 111, computer readable storage medium 200, computer program 211. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0019] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, "plurality" means two or more, unless otherwise clearly and specifically defined.

[0020] In the description of the present application, the term "for example" is used to mean "used as an example, illustration or description". Any embodiment described as "for example" in the present application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid unnecessary details to obscure the description of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present application.

[0021] Example 1

[0022] like Figure 1 As shown, an embodiment of the present invention provides an intelligent control system for monitoring breeding equipment, including a distributed digital monitoring module and a distributed digital control module, and also includes a blockage identification unit 10, a feed waste analysis unit 20, a feed waste correction unit 30, a feeding parameter collection unit 40 and a feeding control unit 50.

[0023] The blockage identification unit 10 is used to monitor multiple feeding ports of distributed automatic feed equipment through the distributed digital monitoring module, monitor and obtain a feeding port image array, and perform blockage analysis. After any feeding port is blocked, the blockage morphology of at least one blocked feeding port is identified to obtain at least one blockage morphology parameter.

[0024] In an embodiment of the present application, the breeding equipment monitored and controlled by the breeding equipment monitoring intelligent control system is a distributed automatic feeding equipment, which includes a plurality of feeding ports and a plurality of feeding channels arranged in a distributed manner. The feeding channels convey the feed to the feeding ports through conveying mechanisms such as conveyor belts. The speed and amount of the conveyed feed can be set for automatic feeding. The plurality of feeding ports are respectively arranged at a plurality of locations where feed needs to be delivered, such as a plurality of livestock areas.

[0025] The distributed digital monitoring module includes multiple monitors, which are arranged at multiple feeding ports and the feeding ports of distributed automatic feed equipment, such as monitoring cameras, for monitoring and collecting images of the feeding ports and the feeding ports to analyze whether feed blockage occurs and analyze feed quality, etc. The distributed digital control module is respectively connected to multiple feeding channels and feeding ports, and the feeding speed or feeding amount of the feeding channels is set and controlled according to the instructions of the breeding equipment monitoring intelligent control system. It can be set based on PLC control in the prior art, for example, by controlling the transmission speed of the conveyor belt, or the feeding amount of the feeding port.

[0026] The blockage identification unit monitors multiple feeding ports of distributed automatic feed equipment in real time through a distributed digital monitoring module. The distributed digital module is equipped with multiple cameras or image sensors to capture the working status of each feeding port, obtain images of multiple feeding ports, and form a feeding port image array. Through these image arrays, the system can monitor the feeding situation of each feeding port in real time and perform automated blockage analysis. Blockage analysis uses image processing and machine learning algorithms to identify and classify key features in the image. When any feeding port is blocked, the system can detect it in time and trigger blockage identification.

[0027] After detecting that any feeding port is blocked, the blockage identification unit further identifies the blockage form of the feeding port. The blockage form refers to the specific parameters of the feed accumulation method or blockage position when the feeding port is blocked, which may include the distribution form of the feed, the size and position of the blockage, etc. This is achieved by extracting different morphological features from the feeding image, such as the shape, area, and position distribution of the blockage, to obtain at least one blockage morphology parameter.

[0028] At least one blockage morphology parameter provides technical support for subsequent analysis of feed waste caused by blockage.

[0029] In the embodiment of the present application, the steps performed by the blocking identification unit 10 include:

[0030] Through the distributed digital monitoring module, multiple feeding ports of distributed automatic feed equipment are monitored to obtain a feeding port image array;

[0031] Collect a set of sample feeding port images, mark each sample feeding port image based on whether it is blocked, and obtain a set of sample blockage classification results;

[0032] Using the sample feeding port image set and the sample blockage classification result set to train a blockage classifier;

[0033] The blockage classifier is used to perform blockage identification and classification on the feed port image array to obtain a plurality of blockage classification results. When any blockage classification result indicates that a blockage has occurred, at least one blocked feed port is obtained.

[0034] In the embodiment of the present application, a distributed digital monitoring module is used to perform real-time image monitoring of multiple feeding ports of distributed automatic feed equipment, and a feeding port image of each feeding port is obtained to obtain a feeding port image array.

[0035] Furthermore, a sample feed port image set is obtained from the feed port image data collected at different time points and in different environments. These sample images are annotated manually or automatically, and divided into two categories of "blockage" and "no blockage" according to whether there is blockage in each image, forming a sample blockage classification result set. This annotation process provides accurate label data for training the blockage classifier.

[0036] After obtaining the sample feed port image set and its corresponding blockage classification results, the sample feed port image set and the sample blockage classification result set are used to train the blockage classifier. The blockage classifier is a classification model built based on machine learning or deep learning models. It learns the features in a large number of annotated sample images to identify whether the feed port is blocked. The training process involves extracting features from the image, such as the feed port shape, feed distribution status, etc., and performing supervised learning based on the annotation results so that the classifier can automatically identify blockages in future feed port images.

[0037] For example, a convolutional neural network (CNN) is used to construct a congestion classifier. The labeled sample image set is input into the classifier through supervised learning. The congestion classifier gradually learns the relationship between the features in the image and the classification results during the training process. The error between the model prediction value and the actual label is calculated through a loss function (such as cross entropy loss). The weight parameters of the congestion classifier are adjusted according to the error through the back propagation algorithm to gradually reduce the classification error. Repeated training is performed until the performance of the congestion classifier meets the requirements, for example, the accuracy rate meets 92%.

[0038] The trained blocking classifier is used to perform blocking identification and classification on the images of multiple feeding ports in the feeding port image array to obtain multiple blocking classification results. When any blocking classification result is blocking, at least one blocked feeding port is obtained.

[0039] Through the above steps, the blocked feeding port can be identified quickly and accurately during the automatic feeding process. The classification result of the feeding port image recognition is binary classification with high recognition efficiency, which significantly improves the efficiency of identifying the blocked feeding port of automatic feed equipment.

[0040] In the embodiment of the present application, the steps performed by the blocking identification unit 10 include:

[0041] Collect a set of sample blocking feed port images, annotate the blocking morphology in each sample blocking feed port image, and obtain a set of sample blocking morphology parameters;

[0042] Using the sample blocking feed port image set and the sample blocking morphology parameter set, training a blocking morphology identifier;

[0043] The blocking morphology identifier is used to perform blocking morphology identification on the feeding port image of the at least one blocked feeding port to obtain at least one blocking morphology parameter.

[0044] In the embodiment of the present application, after identifying at least one blocked feeding port, the feeding channel corresponding to the at least one blocked feeding port can be controlled to suspend feeding, and subsequent blockage morphology identification and control optimization can be performed to avoid continuous blockage.

[0045] In identifying the blockage form of at least one blocked feeding port, different blockage forms may cause the feed to be affected by the blockage and change the feeding position away from the feeding port, such as deviating from the livestock feed container area, resulting in feed waste. Therefore, it is necessary to identify the blockage form.

[0046] First, a rich blockage morphology dataset is established by collecting a collection of sample blocked feed port images. These sample images come from the blockage conditions under different environments and at different times collected during the previous operation of the distributed automatic feed equipment, covering various forms of feed port blockage. After the data collection is completed, each sample image is annotated in detail, with an emphasis on the blockage morphology features shown in the image. These blockage morphology features include but are not limited to the accumulation shape of the feed, the location of the blockage, the size of the blockage, the morphological boundaries, etc. This step is completed through image processing or manual annotation to ensure a high-precision set of sample blockage morphology parameters. Blockage morphology parameters are key data that describe the blockage state, such as the geometric characteristics of the morphology, the location of the blockage, the height and width of the blockage, etc.

[0047] Furthermore, based on the convolutional neural network, a blockage morphology identifier for identifying blockage morphology is constructed. The sample blocked feed port image set and the sample blockage morphology parameter set are used to train the blockage morphology identifier. The blockage morphology identifier is a blockage morphology recognition model based on the convolutional neural network in deep learning, which can automatically extract and analyze the blockage morphology features from the image. During the training process, the model establishes the association between the blockage morphology and the image features by learning the blockage morphology parameters of a large number of labeled samples. The training process includes feature extraction, model construction, error feedback, parameter optimization and other steps to ensure that the model can accurately identify different types of blockage morphology until its accuracy meets the use requirements. After the training is completed, the blockage morphology identifier can also be verified to verify whether its accuracy meets the requirements.

[0048] After the blockage morphology identifier is trained and verified, it can be used for actual blockage identification tasks. When a feed port blockage is detected, the blockage morphology identifier is used to further analyze the feed port blockage image of at least one blocked feed port. The morphological features of the blockage are automatically analyzed to obtain at least one blockage morphology parameter. These blockage morphology parameters provide accurate data support for subsequent feed port dredging strategies and equipment adjustments, such as determining the location of the blockage, the height of the blockage, the geometry of the blocked feed pile, etc. It can be used for subsequent analysis of the location distribution of feed delivery deviations caused by blockages.

[0049] The feed waste analysis unit 20 is used to perform a blockage feeding simulation analysis according to the at least one blockage morphology parameter, obtain at least one feeding area distribution, and obtain at least one feed waste analysis result according to at least one preset feeding area.

[0050] In the embodiment of the present application, after obtaining at least one blocking morphological parameter, a simulation analysis of feeding under blocking is performed. Under the influence of blocking, the feed may deviate from the preset feeding direction of the feeding port and deviate from the preset area where the feed is to be fed, resulting in feed waste.

[0051] The blockage feeding simulation analysis simulates the flow of feed during the feeding process based on the blockage morphology parameters, which can help accurately determine the impact of blockage on the feeding path and feed distribution. During the simulation analysis, the system can predict the flow trajectory and accumulation pattern of feed under different feeding conditions through computational fluid dynamics (CFD) or physics-based simulation models, thereby obtaining the feeding area distribution.

[0052] The feeding area distribution includes the area distribution of the feed falling from the feeding port under the influence of the feeding port blockage, which is specifically manifested as which areas the feed will fall into after being affected by the blockage and the feed proportion in these areas. Through simulation analysis, the area distribution of the feed after leaving the feeding port under the influence of the blockage can be identified. The area distribution may include the preset areas where feeding is required, such as the livestock feeding area, and may also include deviated areas, such as the aisle area of ​​the farm.

[0053] According to at least one preset feeding area for feeding with the feeding port blocked, that is, the area where the feed should fall after feeding (such as the feeding area of ​​livestock), processing is performed to obtain at least one feed waste analysis result. Specifically, the proportion of feed in the area deviating from at least one preset feeding area in the distribution of at least one feeding area is screened, and the proportion of feed that has not been fed into at least one preset feeding area is obtained as at least one feed waste analysis result.

[0054] In the embodiment of the present application, the steps performed by the feed waste analysis unit 20 include:

[0055] Collect a set of sample blocking morphological parameters, and mark and obtain a set of sample feeding area distribution according to a set of sample areas for feeding under each blocking morphological parameter and a feeding ratio in each sample area;

[0056] Using the sample blocking morphology parameter set and the sample feeding area distribution set, training a feeding area distribution generator;

[0057] Using the feeding area distribution generator, performing a blocking feeding simulation analysis on the at least one blocking morphological parameter to obtain at least one feeding area distribution;

[0058] Acquire at least one preset feeding area of ​​the at least one blocked feeding port;

[0059] The feeding areas that do not overlap with the at least one preset feeding area are screened within the at least one feeding area distribution to obtain at least one feed waste area distribution, and at least one feed waste ratio is calculated as at least one feed waste analysis result.

[0060] In the embodiment of the present application, a machine learning method is used according to the blockage morphology parameters to predict the distribution of feed feeding areas caused by the influence of the blockage morphology parameters.

[0061] Among them, according to the recorded data when the same feeding port was previously blocked, such as the recorded blocking morphological parameters and the feeding area distribution of the feed after feeding, a sample blocking morphological parameter set is collected. The blocking morphological parameters are key data describing the blocking characteristics of the feeding port, such as the blocking position, size, shape, etc. Then, a sample feeding area set is formed according to the multiple feeding areas where the feed leaves the feeding port under each blocking morphological parameter, and according to the feeding ratio in each sample area, that is, the ratio of the feed falling in each sample area, it is marked, and finally a sample feeding area distribution set is obtained. These sample feeding area distribution sets can reflect the feeding distribution state of the feed under different blocking conditions. Exemplarily, under the influence of a blocking morphological parameter, the flow direction of the feed is affected, deviates from the preset feeding area of ​​the feeding port, and falls into two areas of the same area on the left and right sides of the feeding port, and the feed feeding ratios in the two areas are 40 and 60%, forming a sample feeding area distribution.

[0062] Furthermore, the sample blockage morphology parameter set and the sample feeding area distribution set are used as training data to train the feeding area distribution generator. For example, a BP neural network in machine learning is used to construct it, and the sample blockage morphology parameter set and the sample feeding area distribution set are used as supervised training data to perform iterative supervised training. According to the deviation between the data output by the feeding area distribution generator and the sample feeding area distribution, the parameters of the feeding area distribution generator are adjusted until the deviation is reduced to meet the requirements, for example, the deviation is less than 10%. After the training of all the training data is completed, a trained feeding area distribution generator is obtained.

[0063] After training, the feeding area distribution generator is used to perform a blocking feeding simulation analysis on at least one newly detected blocking morphological parameter to generate at least one feeding area distribution. The feeding area distribution shows the area where the feed falls after the feeding port is blocked, as well as the proportion of feed falling into different areas, reflecting the deviation distribution of the feed feeding direction caused by the blocking of the feeding port.

[0064] Furthermore, at least one preset feeding area of ​​at least one blocked feeding port is obtained. The preset feeding area is pre-set according to a normal feeding state, that is, a feeding area where an ideal feeding path of feed falls when the feed is not blocked, such as the area directly in front of the feeding port, such as a feeding area for livestock.

[0065] The distribution of the feeding area of ​​at least one blocked feeding port is compared with the preset feeding area, and the feeding area that does not overlap with the preset feeding area is screened out and determined as the feed waste area distribution, that is, the feed in this part of the area does not fall into the preset feeding area, and the livestock cannot eat the feed that falls into this part of the area, resulting in feed waste. These waste areas are usually caused by deviations caused by abnormal feed flow due to blockage.

[0066] Further, the feed waste ratio is calculated by adding up the feed feeding ratio in the waste area within the at least one feeding area distribution. The feed waste ratio refers to the proportion of feed in the waste area relative to the total feeding amount, which can quantify the impact of blockage on the feeding feed waste. In this way, the feed waste ratio of the feed feeding deviation caused by the blockage of at least one blocked feeding port is obtained as at least one feed waste analysis result.

[0067] The embodiment of the present application identifies the blockage form of the feed port, further analyzes the distribution of the feed area formed by the deviation of the feed feeding caused by different blockage forms, and combines the preset feed area to screen and calculate the proportion of feed waste caused by the deviation of the feed due to the blockage, thereby completing the analysis of the impact of the feed port blockage on the feed waste, which serves as the basis for subsequent automatic feed equipment control, and can improve the intelligence and accuracy of equipment control.

[0068] The feed waste correction unit 30 is used to monitor and analyze the feed at the feed inlet of the distributed automatic feed equipment through the distributed digital monitoring module to obtain the adhesion ratio, and to perform correction analysis on the at least one feed waste analysis result to obtain at least one corrected feed waste analysis result;

[0069] In the embodiment of the present application, the feed inlet of the distributed automatic feed equipment is monitored in real time through a distributed digital monitoring module, and feed adhesion analysis is performed. Feed adhesion may be caused by moisture or deterioration. This type of feed has a low nutritional value during feeding, and it can be considered that no feed waste occurs when it is not put into the preset feeding area. Therefore, the adhesion of the feed at the feed inlet is monitored to correct the feed waste analysis results obtained by the above analysis.

[0070] The feed image at the feed inlet is monitored by sensors or cameras in the distributed digital monitoring module to identify feed adhesion in real time. Specifically, the adhesion ratio of the feed in the feed is identified. Based on the adhesion ratio, at least one feed waste analysis result is corrected and analyzed to obtain at least one corrected feed waste analysis result. By introducing feed adhesion analysis to correct the feed waste analysis result, the accuracy and rationality of the feed waste analysis can be improved, thereby improving the precision and accuracy of the subsequent control and adjustment of the automatic feed equipment.

[0071] In the embodiment of the present application, the steps performed by the feed waste correction unit 30 also include:

[0072] The feed at the feed inlet of the distributed automatic feed equipment is monitored and analyzed for adhesion by the distributed digital monitoring module to obtain the adhesion ratio;

[0073] According to the bonding ratio, a correction calculation and analysis is performed on the at least one feed waste analysis result to obtain at least one corrected feed waste analysis result, wherein each corrected feed waste analysis result includes a corrected feed waste ratio.

[0074] In the embodiment of the present application, the distributed digital monitoring module is used to monitor the feed inlet of the distributed automatic feed equipment, obtain the feed image, and identify the bonding of the feed image. Bonding analysis refers to identifying the proportion of feed blocks that are bonded due to moisture, deterioration, etc. in the feed particles by means of image processing and other means. Through analysis, the bonding ratio is obtained, that is, the proportion of feed particles that are bonded.

[0075] Exemplarily, based on the convolutional neural network in image processing and the previous feed adhesion recognition data, a set of sample feed images in the feed inlet can be collected, and then according to the feed adhesion ratio in each sample feed image (for example, the weight ratio of the feed that has been bonded), manual annotation is performed to obtain a set of sample bonding ratios, and the set of sample feed images is used as supervised training input data, and the set of sample bonding ratios is used as supervised training output data. A recognition model for the feed bonding ratio is constructed through a convolutional neural network, and supervised training is performed. Through gradient descent training, until the model meets the accuracy requirements, the trained model is used to identify the feed image at the feed inlet to obtain the bonding ratio. Feed images at multiple moments can also be identified, multiple bonding ratios are obtained, and the mean is calculated. Optionally, the proportion of bonded feed in the feed inlet can be identified based on other image processing methods in the prior art.

[0076] Further, according to the bonding ratio, at least one feed waste analysis result is corrected and calculated and analyzed. Exemplarily, the bonding ratio is used to multiply the feed waste ratio in the at least one feed waste analysis result to obtain at least one corrected feed waste ratio as at least one corrected feed waste analysis result, thereby completing the correction of the feed waste analysis result based on the bonding feed ratio.

[0077] The feeding parameter acquisition unit 40 is used to monitor the feeding control parameters of at least one blocked feeding channel of the distributed automatic feed equipment through the distributed digital monitoring module to obtain at least one feeding control parameter;

[0078] In an embodiment of the present application, a distributed digital monitoring module is used to monitor the feeding control parameters of at least one blocked feeding channel of a distributed automatic feeding equipment. At least one blocked feeding channel corresponds to at least one blocked feeding port, that is, at least one feeding channel connected to a blocked feeding port transmits the feed to the feeding port through a transmission mechanism for automatic feed delivery.

[0079] The distributed digital monitoring module can be connected to at least one controller of the blocked feeding channel, such as a PLC controller, to monitor and collect the operating parameters of the blocked feeding channel, such as feed flow rate, feeding amount, opening of the connecting feed inlet, etc. By monitoring these feeding control parameters, it serves as the basis for subsequent adjustment and control of the blocked feeding channel to avoid blockage. In this way, at least one feeding control parameter is obtained.

[0080] Optionally, an image monitoring method may be used to monitor the image in the feeding channel, identify the image to obtain the feed movement speed and the feed delivery amount as at least one feeding control parameter.

[0081] The feeding control unit 50 is used to set at least one control step according to the at least one corrected feed waste analysis result, optimize the at least one feeding control parameter, obtain at least one optimized feeding control parameter, and control the at least one blocked feeding channel through the distributed digital control module.

[0082] In an embodiment of the present application, at least one control step is set based on at least one corrected feed waste analysis result. The control step refers to the amplitude or gradient of parameter adjustment during the optimization process, which represents the precision of the adjustment of the feeding control parameters. The control step is set according to at least one corrected feed waste analysis result. When the corrected feed waste ratio is large, a larger step can be set to improve control efficiency and quickly reduce feed waste. When the corrected feed waste ratio is small, a smaller step can be set to improve control accuracy. The size of the control step will determine the amplitude of each feeding control parameter adjustment.

[0083] Further, according to at least one control step, at least one corresponding feeding control parameter is adjusted and optimized to obtain at least one optimized feeding control parameter. Among them, an optimization algorithm can be used for optimization to obtain the optimized feeding control parameter, so that the feed delivery gradually tends to the optimal state, reduces the accumulation of feed in the blocked channel, reduces waste, and at the same time ensures the feeding amount to avoid affecting the livestock breeding effect.

[0084] After the optimization is completed, at least one blocked feeding channel is controlled by using at least one optimized feeding control parameter through the distributed digital control module, and the specific control operation may include adjusting the opening and closing degree of the connection between the feeding channel and the feeding port, adjusting the feeding amount, adjusting the feed flow rate, etc. The distributed digital control module can be connected to the feeding channel through PLC control or other computer control methods for control.

[0085] By setting the control step size according to the corrected feed waste analysis results and optimizing at least one feeding control parameter, the optimization accuracy of the feeding control parameter can be improved, and feeding control can be performed to reduce feed waste caused by feed sticking and unreasonable feeding control parameters, and ensure the feeding amount to ensure the feeding effect.

[0086] The steps performed by the feeding control unit 50 in the embodiment of the present application also include:

[0087] setting at least one control step size according to the at least one corrected feed waste analysis result;

[0088] Get the feeding control parameter space;

[0089] In the feeding control parameter space, the at least one control step length is used to adjust the at least one feeding control parameter to obtain at least one first feeding control parameter;

[0090] According to the at least one first feeding control parameter and in combination with the bonding ratio, a feeding port blockage prediction is performed to obtain at least one first blockage probability;

[0091] At least one feeding amount of the at least one first feeding control parameter is obtained, and combined with the at least one first blocking probability, a first control fitness is calculated, as shown in the following formula:

[0092] ;

[0093] Among them, DAF is the control fitness, and is the weight, M is the number of at least one blocked feeding port, is the first blocking probability of the i-th blocked feeding port, is the feeding amount of the i-th blocked feeding port, is the preset feeding amount of the i-th blocked feeding port;

[0094] Continue to adjust and optimize the at least one first feeding control parameter within the feeding control parameter space using the at least one control step length until convergence, and output at least one optimized feeding control parameter with the maximum control fitness.

[0095] In the embodiment of the present application, at least one control step is first set according to at least one corrected feed waste analysis result, wherein the size of the corrected feed waste analysis result is positively correlated with the size of the control step.

[0096] According to the at least one corrected feed waste analysis result, at least one control step is set, comprising:

[0097] Get the preset control adjustment step size;

[0098] Calculating an average corrected feed waste analysis result based on the at least one corrected feed waste analysis result;

[0099] According to each corrected feed waste analysis result and the average corrected feed waste analysis result, the preset control adjustment step is adjusted to obtain at least one control step.

[0100] In the embodiment of the present application, a preset control adjustment step is obtained. The preset control adjustment step is the default adjustment range set by the system in the initial stage to optimize the feeding control parameters, which is used to provide a basic adjustment strategy at the beginning of the optimization process. For example, the preset control adjustment step is a 5% adjustment range for adjusting the feeding speed and feeding amount of the feeding channel. Reasonable step size setting can help the system gradually approach the optimal feeding control state when performing parameter optimization.

[0101] Furthermore, based on at least one corrected feed waste analysis result, an average corrected feed waste analysis result is calculated, and specifically, an average of at least one corrected feed waste analysis result is calculated as the average corrected feed waste analysis result.

[0102] Further, based on the ratio of each corrected feed waste analysis result and the average corrected feed waste analysis result, the preset control adjustment step is adjusted and calculated. Specifically, the ratio is multiplied by the preset control adjustment step to complete the adjustment calculation and obtain at least one control step.

[0103] After obtaining at least one control step, a feeding control parameter space is obtained, which includes all control parameters within the control parameter range of the feeding channel, such as the parameter range of the feeding speed and the feeding amount, and optimization is performed within the feeding control parameter space as the optimized data space.

[0104] In the feeding control parameter space, at least one control step is used to adjust at least one feeding control parameter currently monitored and collected, and at least one adjusted first feeding control parameter is obtained. Wherein, at least one control step is used to adjust the feeding speed and feeding amount in at least one feeding control parameter, and the adjustment direction is random, such as increasing or decreasing, and at least one adjusted first feeding control parameter is obtained as the result of the first round of optimization.

[0105] Based on the first feeding control parameter obtained, combined with the bonding ratio obtained by the previous analysis, the feeding port blockage prediction is performed. The purpose of the blockage prediction is to predict the possibility of blockage by analyzing the bonding of the feed in the feeding port, the feeding speed and the feeding amount. At least one first blockage probability will be analyzed, which reflects the risk of blockage of the feeding port under the current feeding control parameter conditions. Among them, the greater the bonding ratio, the greater the feeding speed, and the greater the feeding amount, the greater the first blockage probability.

[0106] In the embodiment of the present application, according to the at least one first feeding control parameter and in combination with the bonding ratio, the feeding port blockage prediction is performed, including:

[0107] Obtain a set of sample feeding control parameters and a set of sample bonding ratios, perform feeding tests according to different sample feeding control parameters and sample bonding ratios, and obtain a set of sample blocking probabilities;

[0108] The sample feeding control parameters, the sample bonding ratio set and the sample blocking probability set are used to train a feeding blocking predictor;

[0109] The feed port blockage predictor is used to predict the blockage of the feed port by combining the at least one first feed control parameter with the bonding ratio to obtain at least one first blockage probability.

[0110] In the embodiment of the present application, a sample feeding control parameter set and a sample bonding ratio set are collected. The sample feeding control parameter set is a set of multiple feeding control parameters extracted from historical operation data, covering feeding control parameters of different feeding channels, such as feeding speed, feeding amount, feeding port opening, etc. The sample bonding ratio set refers to data obtained by monitoring the bonding state of feed (such as the proportion of feed that is bonded).

[0111] The feeding test is carried out according to different sample feeding control parameters and bonding ratio sets. The purpose of the test is to evaluate the blocking probability under these parameter combinations. During the test, the system actually operates each set of feeding control parameters and bonding ratio combinations, observes and records the blocking of the feeding port. If the feeding port is blocked, it is marked as blocked, otherwise it is marked as non-blocked. Through multiple tests, a sample blocking probability set is generated, which contains the corresponding blocking probability under each sample feeding control parameter and bonding ratio.

[0112] After completing the data collection, the sample data is used to train the feed blockage predictor. The feed blockage predictor is a prediction tool based on machine learning or deep learning models, which is trained by using a set of sample feed control parameters, a set of sample bonding ratios, and a set of corresponding sample blockage probabilities. The training process includes: providing the sample feed control parameters, bonding ratios, and blockage probabilities as input data to the feed blockage predictor, and the feed blockage predictor learns how different feed control parameters and bonding ratios affect the blockage probability by analyzing the relationship between these input data. Then, according to the deviation between the output and the sample blockage probability, the hyperparameters of the feed blockage predictor (such as learning rate, regularization parameters, etc.) are adjusted to improve the prediction accuracy of the model and ensure that the predictor can be generalized and applied to new data under different operating conditions.

[0113] The feed port blockage predictor is used to predict the blockage of the feed port by combining at least one current first feed control parameter with the bonding ratio to obtain at least one first blockage probability.

[0114] After obtaining at least one first blocking probability, at least one feeding amount within at least one first feeding control parameter is obtained, for example, the feeding amount per unit time, and fitness calculation of at least one first feeding control parameter is performed. As shown in the following formula:

[0115] ;

[0116] Among them, DAF is the control fitness, and is a weight, which can be set based on those skilled in the art, reflecting the importance of reducing the probability of blockage and increasing the feeding amount in the optimization, for example, 0.65 and 0.35. M is the number of at least one blocked feeding port, is the first blocking probability of the i-th blocked feeding port, is the feeding amount of the i-th blocked feeding port in the optimized feeding control parameters, is the preset feeding amount of the i-th blocked feeding port, for example, the preset feeding amount is the standard feeding amount for feeding aquaculture feed. By calculating the fitness of the feeding control parameters by the above formula, the blocking probability of feeding control can be reduced, and the feeding amount can be guaranteed, thereby ensuring the feeding effect. It takes into account both the blocking risk of the feeding port and the feeding efficiency.

[0117] Continue to use at least one control step in the feeding control parameter space to adjust and optimize at least one first feeding control parameter, obtain at least one second feeding control parameter, and continue to iterate and optimize until convergence. For example, if the control fitness no longer improves in 50 optimizations, or the optimization rounds reach a preset number, the optimization is considered to have converged.

[0118] After convergence, at least one optimized feeding control parameter with the maximum control fitness is output as the optimization result.

[0119] Through the above steps, it is possible to effectively balance feeding efficiency and blockage risk and select the optimal combination of feeding control parameters. The intelligent feeding control optimization method combines multiple factors such as bonding ratio, blockage probability and feeding amount. Through iterative adjustment, the system can ensure feeding efficiency while minimizing feed waste and blockage in actual application.

[0120] The embodiments of the present application achieve the following technical effects through the above specific implementation methods:

[0121] The present invention performs blockage analysis through a distributed digital monitoring module and a feed port image array of multiple feed ports during the operation of the distributed automatic feed equipment. After the feed port is blocked, blockage morphology recognition is performed to obtain blockage morphology parameters, and the blockage situation of the feed port can be identified. Then, according to the blockage morphology parameters, a simulation analysis of the blocked feeding area is performed to obtain the distribution of the feeding area, and the feed waste analysis result is obtained by combining the preset feeding area; then, the feed at the feed inlet of the distributed automatic feed equipment is monitored and bonded through the distributed digital monitoring module to obtain the bonding ratio, and the feed waste analysis result is corrected to obtain the corrected feed waste analysis result, which can correct the feed waste analysis result in the case of feed bonding, and improve the accuracy of subsequent optimization control; further, the feed control parameter of the blocked feeding channel of the current distributed automatic feed equipment is monitored to obtain the feeding control parameter; according to the corrected feed waste analysis result, the control step length is set to improve the optimization accuracy, the feeding control parameter is optimized to obtain the optimized feeding control parameter, and the blocked feeding channel is controlled through the distributed digital control module to ensure the feeding efficiency under the premise of reducing the blockage probability. The present invention uses a distributed digital monitoring module to monitor the feeding port of the automatic feed equipment in real time, identify the blockage form and analyze the feeding area distribution and feed waste, make corrections based on the adhesion analysis results, and perform feeding control based on the optimized feeding control parameters. This solution effectively reduces feed waste, reduces the probability of blockage, optimizes feeding efficiency, and improves the stability of the operation of the breeding feed equipment and the level of intelligent control.

[0122] See also Figure 2 , Figure 2 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 2 As shown, an embodiment of the present invention provides an electronic device 100, including a memory 110, a processor 120, and a computer program 111 stored in the memory 110 and executable on the processor 120. When the processor 120 executes the computer program 111, the steps of the method executed by the unit within the system in Example 1 are implemented.

[0123] See also Figure 3 , Figure 3 A schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present invention. Figure 3 As shown, this embodiment provides a computer-readable storage medium 200 on which a computer program 211 is stored. When the computer program 211 is executed by a processor, the steps of the method executed by the unit in Embodiment 1 are implemented.

[0124] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0125] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0126] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0127] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0128] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0129] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0130] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An intelligent control system for monitoring aquaculture equipment, characterized in that: The system includes a distributed digital monitoring module and a distributed digital control module, and the system also includes: A blockage identification unit is used to monitor multiple feeding ports of distributed automatic feed equipment through the distributed digital monitoring module, monitor and obtain a feeding port image array, and perform blockage analysis. After any feeding port is blocked, the blockage form of at least one blocked feeding port is identified to obtain at least one blockage form parameter; A feed waste analysis unit, configured to perform a blockage feeding simulation analysis according to the at least one blockage morphology parameter, obtain at least one feeding area distribution, and obtain at least one feed waste analysis result according to at least one preset feeding area; A feed waste correction unit, configured to monitor and analyze the adhesion of feed at the feed inlet of the distributed automatic feed equipment through the distributed digital monitoring module, obtain the adhesion ratio, and perform correction analysis on the at least one feed waste analysis result to obtain at least one corrected feed waste analysis result; A feeding parameter acquisition unit, used to monitor the feeding control parameters of at least one blocked feeding channel of the distributed automatic feed equipment through the distributed digital monitoring module to obtain at least one feeding control parameter; A feeding control unit is used to set at least one control step according to the at least one corrected feed waste analysis result, optimize the at least one feeding control parameter, obtain at least one optimized feeding control parameter, and control the at least one blocked feeding channel through the distributed digital control module.

2. The intelligent control system for monitoring aquaculture equipment according to claim 1, characterized in that: Through the distributed digital monitoring module, multiple feeding ports of distributed automatic feed equipment are monitored, feeding port image arrays are obtained, and blockage analysis is performed, including: Through the distributed digital monitoring module, multiple feeding ports of distributed automatic feed equipment are monitored to obtain a feeding port image array; Collect a set of sample feeding port images, mark each sample feeding port image based on whether it is blocked, and obtain a set of sample blockage classification results; Using the sample feeding port image set and the sample blockage classification result set to train a blockage classifier; The blockage classifier is used to perform blockage identification and classification on the feed port image array to obtain a plurality of blockage classification results. When any blockage classification result indicates that a blockage has occurred, at least one blocked feed port is obtained.

3. The intelligent control system for monitoring aquaculture equipment according to claim 2, characterized in that: Identifying the blockage form of at least one blocked feeding port and obtaining at least one blockage form parameter includes: Collect a set of sample blocking feed port images, annotate the blocking morphology in each sample blocking feed port image, and obtain a set of sample blocking morphology parameters; Using the sample blocking feed port image set and the sample blocking morphology parameter set, training a blocking morphology identifier; The blocking morphology identifier is used to perform blocking morphology identification on the feeding port image of the at least one blocked feeding port to obtain at least one blocking morphology parameter.

4. The intelligent control system for monitoring aquaculture equipment according to claim 1, characterized in that: According to the at least one blocking morphological parameter, a blocking feeding simulation analysis is performed to obtain at least one feeding area distribution, and according to at least one preset feeding area, at least one feed waste analysis result is obtained by processing, including: Collect a set of sample blocking morphological parameters, and mark and obtain a set of sample feeding area distribution according to a set of sample areas for feeding under each blocking morphological parameter and a feeding ratio in each sample area; Using the sample blocking morphology parameter set and the sample feeding area distribution set, training a feeding area distribution generator; Using the feeding area distribution generator, performing a blocking feeding simulation analysis on the at least one blocking morphological parameter to obtain at least one feeding area distribution; Acquire at least one preset feeding area of ​​the at least one blocked feeding port; The feeding areas that do not overlap with the at least one preset feeding area are screened within the at least one feeding area distribution to obtain at least one feed waste area distribution, and at least one feed waste ratio is calculated as at least one feed waste analysis result.

5. The intelligent control system for monitoring aquaculture equipment according to claim 1, characterized in that: The distributed digital monitoring module is used to monitor and analyze the feed at the feed inlet of the distributed automatic feed equipment to obtain the bonding ratio, and to perform correction analysis on the at least one feed waste analysis result, including: The feed at the feed inlet of the distributed automatic feed equipment is monitored and analyzed for adhesion by the distributed digital monitoring module to obtain the adhesion ratio; According to the bonding ratio, a correction calculation and analysis is performed on the at least one feed waste analysis result to obtain at least one corrected feed waste analysis result, wherein each corrected feed waste analysis result includes a corrected feed waste ratio.

6. The intelligent control system for monitoring aquaculture equipment according to claim 1, characterized in that: According to the at least one corrected feed waste analysis result, at least one control step is set, and the at least one feeding control parameter is optimized to obtain at least one optimized feeding control parameter, including: setting at least one control step size according to the at least one corrected feed waste analysis result; Get the feeding control parameter space; In the feeding control parameter space, the at least one control step length is used to adjust the at least one feeding control parameter to obtain at least one first feeding control parameter; According to the at least one first feeding control parameter and in combination with the bonding ratio, a feeding port blockage prediction is performed to obtain at least one first blockage probability; At least one feeding amount of the at least one first feeding control parameter is obtained, and combined with the at least one first blocking probability, a first control fitness is calculated, as shown in the following formula: ； Among them, DAF is the control fitness, and is the weight, M is the number of at least one blocked feeding port, is the first blocking probability of the i-th blocked feeding port, is the feeding amount of the i-th blocked feeding port, is the preset feeding amount of the i-th blocked feeding port; Continue to adjust and optimize the at least one first feeding control parameter within the feeding control parameter space using the at least one control step length until convergence, and output at least one optimized feeding control parameter with the maximum control fitness.

7. The intelligent control system for monitoring aquaculture equipment according to claim 6, characterized in that: According to the at least one corrected feed waste analysis result, at least one control step is set, comprising: Get the preset control adjustment step size; Calculating an average corrected feed waste analysis result based on the at least one corrected feed waste analysis result; According to each corrected feed waste analysis result and the average corrected feed waste analysis result, the preset control adjustment step is adjusted to obtain at least one control step.

8. The intelligent control system for monitoring aquaculture equipment according to claim 6, characterized in that: According to the at least one first feeding control parameter and in combination with the bonding ratio, a feeding port blockage prediction is performed, including: Obtain a set of sample feeding control parameters and a set of sample bonding ratios, perform feeding tests according to different sample feeding control parameters and sample bonding ratios, and obtain a set of sample blocking probabilities; The sample feeding control parameters, the sample bonding ratio set and the sample blocking probability set are used to train a feeding blocking predictor; The feed port blockage predictor is used to predict the blockage of the feed port by combining the at least one first feed control parameter with the bonding ratio to obtain at least one first blockage probability.

9. An electronic device, characterized in that: include: Memory for storing computer software programs; A processor is used to read and execute the computer software program, thereby realizing the aquaculture equipment monitoring intelligent control system according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer software program, and when the computer software program is executed by the processor, the intelligent control system for monitoring breeding equipment according to any one of claims 1 to 8 is implemented.

Images