CN118112937A - Intelligent control system and method for micro-positive pressure of farm based on multisource fusion perception - Google Patents
Intelligent control system and method for micro-positive pressure of farm based on multisource fusion perception Download PDFInfo
- Publication number
- CN118112937A CN118112937A CN202410524516.XA CN202410524516A CN118112937A CN 118112937 A CN118112937 A CN 118112937A CN 202410524516 A CN202410524516 A CN 202410524516A CN 118112937 A CN118112937 A CN 118112937A
- Authority
- CN
- China
- Prior art keywords
- data
- sensor
- equipment
- farm
- value
- 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
- 238000000034 method Methods 0.000 title claims abstract description 30
- 230000004927 fusion Effects 0.000 title claims abstract description 21
- 230000008447 perception Effects 0.000 title claims abstract description 21
- 238000012549 training Methods 0.000 claims abstract description 45
- 238000009423 ventilation Methods 0.000 claims abstract description 45
- 238000011217 control strategy Methods 0.000 claims abstract description 24
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 24
- 230000035945 sensitivity Effects 0.000 claims description 34
- 238000001514 detection method Methods 0.000 claims description 26
- 238000004364 calculation method Methods 0.000 claims description 18
- 230000015654 memory Effects 0.000 claims description 17
- 238000012790 confirmation Methods 0.000 claims description 15
- 238000003860 storage Methods 0.000 claims description 15
- 230000008859 change Effects 0.000 claims description 14
- 230000002159 abnormal effect Effects 0.000 claims description 13
- 238000007405 data analysis Methods 0.000 claims description 13
- 238000007781 pre-processing Methods 0.000 claims description 13
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 claims description 12
- 230000007613 environmental effect Effects 0.000 claims description 12
- 238000011176 pooling Methods 0.000 claims description 11
- 238000012544 monitoring process Methods 0.000 claims description 8
- 230000004913 activation Effects 0.000 claims description 7
- 238000013500 data storage Methods 0.000 claims description 7
- 230000008034 disappearance Effects 0.000 claims description 7
- 238000004880 explosion Methods 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 7
- 238000001914 filtration Methods 0.000 claims description 7
- 230000009466 transformation Effects 0.000 claims description 7
- 229910002092 carbon dioxide Inorganic materials 0.000 claims description 6
- 239000001569 carbon dioxide Substances 0.000 claims description 6
- 238000004140 cleaning Methods 0.000 claims description 6
- 238000013515 script Methods 0.000 claims description 5
- 230000001174 ascending effect Effects 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 239000013618 particulate matter Substances 0.000 claims description 3
- 238000005096 rolling process Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000004088 simulation Methods 0.000 claims description 3
- 238000013499 data model Methods 0.000 abstract description 4
- 230000008569 process Effects 0.000 description 9
- 230000009286 beneficial effect Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 230000001276 controlling effect Effects 0.000 description 4
- 238000004904 shortening Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000009467 reduction Effects 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000003213 activating effect Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Landscapes
- Air Conditioning Control Device (AREA)
Abstract
The invention discloses a multisource fusion perception-based intelligent control system and method for micro positive pressure of a farm, relates to the technical field of micro positive pressure control of the farm, and aims to solve the problem that ventilation in the farm cannot be controlled. According to the training time-series data model, short-time prediction is carried out on a future control strategy, the control strategy adjustment delay time caused by instantaneous basic parameter changes such as weather dip is greatly shortened, the control parameters are learned and trained through the CNN convolutional neural network, the optimal equipment operation parameters are automatically obtained, the debugging efficiency is greatly improved, the problem that experience data cannot be copied due to inconsistency of different units is solved, meanwhile, the manual debugging time is shortened, the control strategy deviation degree caused by poor consistency of unit parameters and equipment performance parameters is reduced, and the maximum approaching target differential pressure environment in a house is maintained.
Description
Technical Field
The invention relates to the technical field of farm micro-positive pressure control, in particular to a farm micro-positive pressure intelligent control system and method based on multisource fusion perception.
Background
The micro-positive pressure control of the farm is a control method which enables the air pressure inside the farm to be slightly higher than the external environment pressure through a certain technical means and equipment.
The Chinese patent with publication number CN114428529A discloses a novel temperature detection and ventilation control system in a farm, wherein monitoring data in the farm is mastered in real time mainly by monitoring various environmental information in the farm, and the environmental data can be regulated automatically according to the change of the monitoring data; the system is also provided with an individual environment control module which can respectively control the environments of individuals of different age groups and pregnant individuals and control each terminal to adjust the environments of the individuals, and the problems of environmental control in a farm are solved, but the following problems exist in actual operation:
1. Before data acquisition is carried out on control equipment in a farm, the equipment is not further detected, so that errors of acquired data caused by abnormal equipment cannot be timely detected.
2. After the sensor data in the farm are acquired, the data are not effectively processed and stored, so that the safety of the data is reduced.
3. The plant parameters in the farm are not effectively trained, so that the optimal plant operation parameter data of the farm cannot be obtained.
4. The time series model training is not carried out on the equipment operation parameters in the farm, so that timely decision-making on the control parameters of the farm equipment cannot be realized according to future weather.
Disclosure of Invention
The invention aims to provide a multisource fusion perception-based intelligent control system and a multisource fusion perception-based intelligent control method for micro positive pressure of a farm, which are used for training a time-series data model, carrying out short-time prediction on future control strategies, greatly shortening control strategy adjustment delay time caused by instantaneous basic parameter changes such as weather dip and the like, learning and training control parameters through a CNN convolutional neural network, automatically acquiring optimal equipment operation parameters, greatly improving debugging efficiency, solving the problem that empirical data cannot be copied due to inconsistency of different units, simultaneously shortening manual debugging time, reducing control strategy deviation degree caused by poor consistency of unit parameters and equipment performance parameters, maintaining maximum approach to a target differential pressure environment in a house, and solving the problems in the prior art.
In order to achieve the above purpose, the present invention provides the following technical solutions:
Intelligent control system of plant micro-positive pressure based on multisource fusion perception includes:
A sensor confirmation unit for:
Firstly, confirming a sensor in a farm and equipment corresponding to the sensor, and detecting the use state of each equipment when the equipment is started;
the sensor data acquisition unit is used for:
Receiving data acquired by a sensor corresponding to the equipment which is qualified in detection, respectively storing the data according to the type of the acquired data, and simultaneously, when the data is stored, correspondingly storing a memory according to the storage of the data and marking the acquired data in the memory as standard control data;
The acquired data analysis unit is used for:
calculating data operation parameters of standard control data through a CNN convolutional neural network, and judging the optimal equipment operation parameter value of each sensor according to a calculation result;
a sensor parameter control unit for:
and carrying out time sequence model training on the optimal equipment operation parameter value of each sensor, and carrying out strategy prediction on the farm according to the model training result.
Preferably, the sensor confirmation unit includes:
a sensor and device confirmation module for:
the intelligent control equipment in the farm and the corresponding sensor of the equipment are confirmed;
The sensor comprises an inclination angle sensor, a distance sensor, a current detection sensor, a meteorological sensor, an air pressure sensor and an air speed sensor;
the equipment corresponding to the sensor comprises an operation fan set, a small window, a curtain, a weather station, a heater and a pressure-building collector;
wherein,
The small window is provided with an inclination sensor for acquiring the opening and closing angles of the small window;
A distance sensor is arranged at the top of the curtain and used for collecting the actual ventilation area in the farm;
The current detection sensor is used for collecting actual operation parameters of the fan, and the wind speed sensor is arranged at an air inlet and an air outlet of the operation fan and used for detecting actual ventilation quantity of the fan;
A weather sensor is arranged in the weather station and comprises a temperature sensor, a humidity sensor, a carbon dioxide concentration sensor, a rainfall sensor, an air pressure sensor, a wind speed and direction sensor and a particulate matter sensor;
The pressure sensor is arranged on the pressure collector and comprises a pressure sensor in the house and a pressure sensor outside the house and is used for collecting pressure difference values inside the house and outside the house.
Preferably, the sensor confirmation unit further includes:
The equipment detection module is used for:
performing state evaluation on equipment of a farm;
Firstly monitoring and recording the using time of the equipment, then acquiring a sensitivity change curve of the equipment, and confirming the current sensitivity of each equipment according to the using time on the sensitivity change curve;
confirming a historical fault record of each device, wherein the historical fault record is retrieved from a database, and confirming fault sensitivity in the historical fault record;
comparing the current sensitivity with the fault sensitivity by a threshold value;
If the current sensitivity threshold is not in the fault sensitivity threshold range, the corresponding equipment is in a normal working state;
if the current sensitivity threshold is within the fault sensitivity threshold range, the corresponding equipment is in an abnormal working state;
and reporting the equipment in the abnormal working state, and maintaining the equipment by staff.
Preferably, the sensor confirmation unit further includes:
The device parameter adjustment module is used for:
parameter adjustment is carried out on parameters of the equipment, wherein the adjusted parameters are debugged according to training scripts in a database;
and obtaining sensor data under different equipment parameters after the equipment parameters are adjusted, and taking the sensor data as environmental data of the farm.
Preferably, the sensor data acquisition unit includes:
The sensor data classification module is used for:
receiving environment data, and preprocessing the data after the data is received;
the data preprocessing comprises data cleaning and data standardization;
After the data preprocessing, classifying the data types according to the sensors corresponding to the environmental data;
the data types are classified to obtain inclination angle sensor data, distance sensor data, current detection sensor data, weather sensor data, air pressure sensor data and wind speed sensor data;
and labeling the data with the classified data types as target acquisition data.
Preferably, the sensor data acquisition unit further includes:
A sensor data storage module for:
Confirming the data stock of each target acquisition data;
The data stock is obtained by dividing each target acquisition data into data segments, wherein the divided data segments have the same length, and the stock data of each target acquisition data is confirmed according to the number of the data segments;
each target acquisition data corresponds to an independent memory, and each independent memory is provided with a plurality of sub-memories;
when the target acquisition data is stored, the corresponding independent memory is firstly confirmed, and then the residual capacity of the sub-memory is confirmed;
And confirming the stored sub-memory according to the stock of the target acquisition data, wherein the stock of the target acquisition data is smaller than the residual capacity of the sub-memory.
Preferably, the collected data analysis unit is further configured to:
When the standard control data carries out data operation parameter calculation of the CNN convolutional neural network, firstly, confirming equipment parameters in the standard control data, wherein the data parameters are a temperature value, a humidity value, a differential pressure value, an operation fan parameter, a curtain opening value, a small window opening value, a heater operation parameter, a carbon dioxide value and a wind speed value;
Noise filtering is carried out on the equipment parameters according to the rolling layer and the pooling layer, and local characteristic data of the equipment parameters are obtained after noise filtering;
carrying out gradient explosion or gradient disappearance on the local characteristic data according to the residual error structure;
Carrying out depth feature extraction on local feature data of gradient explosion or gradient disappearance, and confirming feature values of the data after the depth feature extraction through Inception;
And (3) carrying out dimension ascending on the confirmed characteristic value through convolution of 1x1, and obtaining the optimal equipment operation parameter value of the target pressure difference under different ventilation levels through softmax.
Preferably, the collected data analysis unit is further configured to:
when the ventilation quantity of the running fan set is mapped to the fan rotating speed of the running fan set, the running fan set is subjected to analog output signal control and running control;
The analog output signal is controlled to set the control analog quantity of the frequency converter of the running fan set to be a 0-10V level interval value;
Collecting the wind speeds of an air inlet and an air outlet of the running fan set according to a wind speed sensor, and calculating real-time ventilation quantity according to the current curtain opening and the small window opening;
Collecting the numerical value of a current detection sensor in the running fan set in real time, and reversely calculating the running state deviation of the running fan set according to the current parameters;
the operation control is to control and execute equipment according to the deviation value, wherein the equipment for controlling and executing comprises an operation fan set, a small window and a curtain;
The deviation value calculation formula is as follows:
E=SV-PV
e is expressed as a deviation value, and the deviation value comprises a historical deviation, a current deviation and a latest deviation; SV is expressed as a device parameter target set value; PV represents the actual value of the current running wind turbine unit acquired data through the current detection sensor;
calculating a difference value between the target ventilation quantity and the actual ventilation quantity of the farm, wherein the difference value = the target ventilation quantity-the actual ventilation quantity, and the target ventilation quantity is retrieved from a database;
and obtaining the optimal analog quantity parameter of the farm after the difference value is calculated.
Preferably, the sensor parameter control unit is further configured to:
Performing time sequence model training on the optimal equipment operation parameter value;
Firstly, confirming n pieces of time slice data in the optimal equipment operation parameter values, wherein when the n pieces of time slice data are recorded at the sampling frequency of 0.2Hz for the equipment operation parameter, each time interval comprises data of 0.2 x n seconds, and in each time interval, an n x 3 matrix is obtained according to the real-time temperature and humidity of a weather station, the network prediction temperature and humidity and the equipment operation parameter;
carrying out feature training on n pieces of time slice data through time series prediction;
The feature training is carried out on two network layers, a feature detector in a first network layer detects a single feature, the single feature is input into a second network layer, and the second network layer trains the single feature;
performing nonlinear transformation and pooling operation on the trained single characteristic;
And after nonlinear transformation and pooling operation, performing connection layer activation by using Softmax, and obtaining a control strategy decision within one minute of the future of the farm after the connection layer activation.
The invention provides another technical scheme, namely a farm micro-positive pressure intelligent control method based on multisource fusion perception, which comprises the following steps of:
The first step: the method comprises the steps of confirming a sensor and equipment in a farm according to a sensor confirming unit, and detecting the use state of each equipment;
And a second step of: the sensor data acquisition unit is used for carrying out data preprocessing and data storage on the data acquired in real time and the data subjected to equipment parameter adjustment;
And a third step of: performing data operation parameter calculation on the stored data according to the acquired data analysis unit, and obtaining an optimal operation parameter value of equipment in the farm after the data operation parameter calculation;
fourth step: and finally, carrying out time sequence model training on the optimal operation parameter value of the equipment in the farm by using a sensor parameter control unit, and obtaining an optimal control strategy within one minute of the farm in the future after the time sequence simulation training.
Compared with the prior art, the invention has the following beneficial effects:
1. Training a time-series data model, carrying out short-time prediction on a future control strategy, greatly shortening the control strategy adjustment delay time caused by instantaneous basic parameter changes such as weather dip, learning and training control parameters through a CNN convolutional neural network, automatically acquiring optimal equipment operation parameters, greatly improving the debugging efficiency, solving the problem that empirical data cannot be copied due to inconsistency of different units, simultaneously shortening the manual debugging time, and reducing the control strategy deviation caused by poor consistency of unit parameters and equipment performance parameters, thereby further improving the control efficiency of a target differential pressure environment in a farm.
2. The data cleaning process can remove noise, abnormal values or repeated data in the original data, ensure the accuracy and reliability of subsequent analysis, clearly distinguish different types of sensor data through data type classification, enable the data to be easier to understand and apply, dynamically allocate storage space according to actual needs, meet scenes of different data amounts and storage requirements, confirm corresponding independent memories, and check the residual capacity of the sub-memories. This ensures that data is stored in a sub-memory of sufficient capacity to avoid loss or corruption of data due to insufficient storage space.
3. The using time length and sensitivity change of the equipment are monitored regularly, the equipment can be found and processed in time before the equipment fails, the influence on the production of the farm caused by sudden shutdown or performance reduction of the equipment is avoided, the equipment parameters are adjusted according to the training script, sensor data under different equipment parameters can be obtained, and the data can reflect the states of the farm under different environmental conditions. By analyzing and comparing these data, an optimal combination of device parameters can be found, creating an environment most conducive to animal growth.
Drawings
FIG. 1 is a schematic diagram of an intelligent control module according to the present invention;
Fig. 2 is a schematic diagram of a CNN convolutional neural network control parameter training process according to the present invention;
FIG. 3 is a schematic diagram of a control flow of the operational fan set of the present invention;
FIG. 4 is a schematic diagram of a training process of a time series model according to the present invention;
FIG. 5 is a schematic diagram of the training process of the optimal device operation parameter value according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In order to solve the problem that in the prior art, before data acquisition is performed on control equipment in a farm, the equipment is not further detected, so that data acquisition errors caused by abnormal equipment cannot be timely checked, please refer to fig. 1-5, the embodiment provides the following technical scheme:
Intelligent control system of plant micro-positive pressure based on multisource fusion perception includes:
A sensor confirmation unit for:
Firstly, confirming a sensor in a farm and equipment corresponding to the sensor, and detecting the use state of each equipment when the equipment is started;
the sensor data acquisition unit is used for:
Receiving data acquired by a sensor corresponding to the equipment which is qualified in detection, respectively storing the data according to the type of the acquired data, and simultaneously, when the data is stored, correspondingly storing a memory according to the storage of the data and marking the acquired data in the memory as standard control data;
The acquired data analysis unit is used for:
calculating data operation parameters of standard control data through a CNN convolutional neural network, and judging the optimal equipment operation parameter value of each sensor according to a calculation result;
a sensor parameter control unit for:
and carrying out time sequence model training on the optimal equipment operation parameter value of each sensor, and carrying out strategy prediction on the farm according to the model training result.
Specifically, the sensor data acquisition unit is used for carrying out data segment division on each target acquisition data and confirming the stock of each data segment, so that the storage space can be managed more accurately, the optimal equipment operation parameters are automatically acquired through the acquisition data analysis unit, the debugging efficiency is greatly improved, the problem that experience data cannot be copied due to inconsistency of different units is solved, short-time prediction is carried out on future control strategies through the sensor parameter control unit, and the control strategy adjustment hysteresis time caused by instantaneous basic parameter changes such as weather dip is greatly shortened.
A sensor confirmation unit comprising:
a sensor and device confirmation module for:
the intelligent control equipment in the farm and the corresponding sensor of the equipment are confirmed;
The sensor comprises an inclination angle sensor, a distance sensor, a current detection sensor, a meteorological sensor, an air pressure sensor and an air speed sensor;
the equipment corresponding to the sensor comprises an operation fan set, a small window, a curtain, a weather station, a heater and a pressure-building collector;
wherein,
The small window is provided with an inclination sensor for acquiring the opening and closing angles of the small window;
A distance sensor is arranged at the top of the curtain and used for collecting the actual ventilation area in the farm;
The current detection sensor is used for collecting actual operation parameters of the fan, and the wind speed sensor is arranged at an air inlet and an air outlet of the operation fan and used for detecting actual ventilation quantity of the fan;
A weather sensor is arranged in the weather station and comprises a temperature sensor, a humidity sensor, a carbon dioxide concentration sensor, a rainfall sensor, an air pressure sensor, a wind speed and direction sensor and a particulate matter sensor;
The pressure sensor is arranged on the pressure collector and comprises a pressure sensor in the house and a pressure sensor outside the house and is used for collecting pressure difference values inside the house and outside the house.
The equipment detection module is used for:
performing state evaluation on equipment of a farm;
Firstly monitoring and recording the using time of the equipment, then acquiring a sensitivity change curve of the equipment, and confirming the current sensitivity of each equipment according to the using time on the sensitivity change curve;
confirming a historical fault record of each device, wherein the historical fault record is retrieved from a database, and confirming fault sensitivity in the historical fault record;
comparing the current sensitivity with the fault sensitivity by a threshold value;
If the current sensitivity threshold is not in the fault sensitivity threshold range, the corresponding equipment is in a normal working state;
if the current sensitivity threshold is within the fault sensitivity threshold range, the corresponding equipment is in an abnormal working state;
and reporting the equipment in the abnormal working state, and maintaining the equipment by staff.
The device parameter adjustment module is used for:
parameter adjustment is carried out on parameters of the equipment, wherein the adjusted parameters are debugged according to training scripts in a database;
and obtaining sensor data under different equipment parameters after the equipment parameters are adjusted, and taking the sensor data as environmental data of the farm.
Specifically, all control devices in the farm and sensors corresponding to the devices are confirmed through a sensor confirmation unit, various sensors can accurately acquire required data in real time, the environment in the farm can be optimized through real-time monitoring and regulation, the use state of the control devices in the farm is detected through a device detection module, the use time and sensitivity change of the devices can be timely found and processed before the devices are in failure, the influence on the production of the farm caused by sudden shutdown or performance reduction of the devices is avoided, the sensitivity of the current devices can be compared with a historical failure record of the devices, so that whether the devices are in an abnormal working state is judged, potential problems of the devices can be accurately identified, misjudgment or misjudgment is avoided, a device parameter adjustment module is provided for adjusting the device parameters in the farm, sensor data under different device parameters can be obtained according to training scripts, and the data can reflect the states of the farms under different environment conditions. The optimal equipment parameter combination can be found by analyzing and comparing the data, thereby creating the environment which is most beneficial to animal growth, the equipment parameters can be flexibly adjusted to adapt to the changes, the best environmental condition of the farm is ensured to be kept all the time, meanwhile, the air pressure sensor adopts a separated high-precision mes air pressure sensor, one air pressure sensor can be arranged inside and outside a house, the model of the air pressure sensor can be BMP280, the small window and the curtain are respectively added with feedback signals through the sensors, the small window and the curtain are used as feedback through the inclination sensor and the distance sensor, the accurate control of the small window and the curtain is realized, the fan operation current detection sensor is used for controlling the fan operation parameters to realize the target ventilation quantity, and the large deviation between the actual ventilation quantity and the target ventilation quantity is avoided, wherein the data of the transmitter is subjected to data acquisition through an RS485 bus, the inclination sensor of the small window and the curtain end distance detection sensor are fed back to the control system through an independent RS485 bus at a certain frequency, and the target control parameters of the small window and the curtain are compensated according to a PID algorithm.
In order to solve the problem that in the prior art, after sensor data in a farm are acquired, the data are not effectively processed and stored, so that the security of the data is reduced, referring to fig. 1-5, the embodiment provides the following technical scheme:
A sensor data acquisition unit comprising:
The sensor data classification module is used for:
receiving environment data, and preprocessing the data after the data is received;
the data preprocessing comprises data cleaning and data standardization;
After the data preprocessing, classifying the data types according to the sensors corresponding to the environmental data;
the data types are classified to obtain inclination angle sensor data, distance sensor data, current detection sensor data, weather sensor data, air pressure sensor data and wind speed sensor data;
and labeling the data with the classified data types as target acquisition data.
A sensor data storage module for:
Confirming the data stock of each target acquisition data;
The data stock is obtained by dividing each target acquisition data into data segments, wherein the divided data segments have the same length, and the stock data of each target acquisition data is confirmed according to the number of the data segments;
each target acquisition data corresponds to an independent memory, and each independent memory is provided with a plurality of sub-memories;
when the target acquisition data is stored, the corresponding independent memory is firstly confirmed, and then the residual capacity of the sub-memory is confirmed;
And confirming the stored sub-memory according to the stock of the target acquisition data, wherein the stock of the target acquisition data is smaller than the residual capacity of the sub-memory.
Specifically, the sensor data acquisition unit is used for preprocessing the data fed back by the sensor, noise, abnormal values or repeated data in the original data can be removed in the data cleaning process, and the accuracy and reliability of subsequent analysis are ensured. The data format is further standardized, so that data of different sources and scales can be compared and analyzed, different types of sensor data are clearly distinguished through data type classification, the data are easier to understand and apply, the data fed back by each sensor are classified and stored through the sensor data storage module, the storage space can be managed more accurately through dividing data segments of each target acquisition data and confirming the stock of each data segment, each target acquisition data corresponds to an independent storage, and each independent storage comprises a plurality of sub-storages. The structure enables the storage to be more flexible, the storage space can be dynamically allocated according to actual needs, scenes of different data amounts and storage requirements are met, corresponding independent memories are confirmed, and the residual capacity of the sub-memories is checked. Thus, the data can be ensured to be stored in the sub-memories with sufficient capacity, the data loss or damage caused by insufficient storage space is avoided, and the backup, recovery and maintenance of the data can be more easily carried out.
In order to solve the problem that in the prior art, the device parameters in the farm are not effectively trained, so that the best device operation parameter data of the farm cannot be obtained, referring to fig. 1-5, the present embodiment provides the following technical scheme:
The acquired data analysis unit is further used for:
When the standard control data carries out data operation parameter calculation of the CNN convolutional neural network, firstly, confirming equipment parameters in the standard control data, wherein the data parameters are a temperature value, a humidity value, a differential pressure value, an operation fan parameter, a curtain opening value, a small window opening value, a heater operation parameter, a carbon dioxide value and a wind speed value;
Noise filtering is carried out on the equipment parameters according to the rolling layer and the pooling layer, and local characteristic data of the equipment parameters are obtained after noise filtering;
carrying out gradient explosion or gradient disappearance on the local characteristic data according to the residual error structure;
Carrying out depth feature extraction on local feature data of gradient explosion or gradient disappearance, and confirming feature values of the data after the depth feature extraction through Inception;
And (3) carrying out dimension ascending on the confirmed characteristic value through convolution of 1x1, and obtaining the optimal equipment operation parameter value of the target pressure difference under different ventilation levels through softmax.
The acquired data analysis unit is further used for:
when the ventilation quantity of the running fan set is mapped to the fan rotating speed of the running fan set, the running fan set is subjected to analog output signal control and running control;
The analog output signal is controlled to set the control analog quantity of the frequency converter of the running fan set to be a 0-10V level interval value;
Collecting the wind speeds of an air inlet and an air outlet of the running fan set according to a wind speed sensor, and calculating real-time ventilation quantity according to the current curtain opening and the small window opening;
Collecting the numerical value of a current detection sensor in the running fan set in real time, and reversely calculating the running state deviation of the running fan set according to the current parameters;
the operation control is to control and execute equipment according to the deviation value, wherein the equipment for controlling and executing comprises an operation fan set, a small window and a curtain;
The deviation value calculation formula is as follows:
E=SV-PV
e is expressed as a deviation value, and the deviation value comprises a historical deviation, a current deviation and a latest deviation; SV is expressed as a device parameter target set value; PV represents the actual value of the current running wind turbine unit acquired data through the current detection sensor;
calculating a difference value between the target ventilation quantity and the actual ventilation quantity of the farm, wherein the difference value = the target ventilation quantity-the actual ventilation quantity, and the target ventilation quantity is retrieved from a database;
and obtaining the optimal analog quantity parameter of the farm after the difference value is calculated.
Specifically, the CNN convolutional neural network is used for learning and training the control parameters, so that the optimal equipment operation parameters are automatically obtained, the debugging efficiency is greatly improved, the problem that experience data cannot be copied due to inconsistency of different units is solved, meanwhile, the manual debugging time can be shortened, the deviation degree of a control strategy caused by poor consistency of the unit parameters and the equipment performance parameters is reduced, the maximum approach target pressure difference environment in a house is maintained, the noise filtering is carried out on the equipment parameters through the convolution and pooling layers, irrelevant information and noise in the data can be effectively removed, and the local characteristic data of the equipment parameters are extracted. The method is beneficial to reducing the complexity of model learning, improving the generalization capability of the model, and effectively solving the problem of gradient explosion or gradient disappearance by adopting a residual structure, so that the model can learn deep features better in the training process. The training efficiency and accuracy of the model are improved, the deep relation among the parameters of the equipment can be further excavated through depth feature extraction and Inception structure feature value confirmation, so that more comprehensive and accurate feature representation is obtained, the Inception structure increases the width of the network and the adaptability of the network to the scale, the receptive fields of different branches have different multi-scale information inside, more feature values are obtained, the complexity of the model can be increased and the representation capability of the model is improved through 1x1 convolution. The method is characterized in that the method comprises the steps of combining a softmax classifier, accurately classifying the optimal equipment operation parameters of target differential pressure under different ventilation levels, simultaneously, performing three aspects of consideration on historical deviation, current deviation and latest deviation, namely, performing consideration on the current data of the deviation value through consideration on the historical data of the deviation value, obtaining future development trend of a control object through analysis on the latest data, comprehensively analyzing and predicting the possible future change of the control object through the three aspects of data, obtaining the optimal analog quantity parameters aiming at the possible future change, adjusting and outputting the optimal analog quantity parameters through a control system, mapping the ventilation quantity of an operation fan set into the fan rotating speed, combining the wind speed data of an air inlet and an air outlet acquired by a wind speed sensor, accurately controlling the ventilation quantity, and reversely calculating the operation state deviation of the operation fan set through collecting the numerical value of a current detection sensor in real time. The method is beneficial to timely finding out abnormal operation of a fan set, avoiding energy waste, improving energy utilization efficiency, carrying out difference calculation on the target ventilation quantity and the actual ventilation quantity of a farm, wherein the difference = the target ventilation quantity-the actual ventilation quantity, combining a historical deviation value, a current deviation value and a nearest deviation value, and realizing dynamic adjustment on the ventilation quantity, wherein a target ventilation quantity set value and the actual ventilation quantity difference value are calculated, three aspects of historical deviation, current deviation and nearest deviation are considered by PID control, namely, the current data of the deviation value are considered by taking the historical data of the deviation value into consideration, future development trend of a control object is obtained by analyzing the nearest data, future possible change of the control object is comprehensively analyzed and predicted, and optimal analog quantity parameters are obtained and adjusted and output for the future possible change of the control object.
In order to solve the problem that in the prior art, the time series model training is not performed on the equipment operation parameters in the farm, so that timely decision cannot be realized on the control parameters of the farm equipment according to future weather, please refer to fig. 1-5, the present embodiment provides the following technical scheme:
the sensor parameter control unit is further used for:
Performing time sequence model training on the optimal equipment operation parameter value;
Firstly, confirming n pieces of time slice data in the optimal equipment operation parameter values, wherein when the n pieces of time slice data are recorded at the sampling frequency of 0.2Hz for the equipment operation parameter, each time interval comprises data of 0.2 x n seconds, and in each time interval, an n x 3 matrix is obtained according to the real-time temperature and humidity of a weather station, the network prediction temperature and humidity and the equipment operation parameter;
carrying out feature training on n pieces of time slice data through time series prediction;
The feature training is carried out on two network layers, a feature detector in a first network layer detects a single feature, the single feature is input into a second network layer, and the second network layer trains the single feature;
performing nonlinear transformation and pooling operation on the trained single characteristic;
And after nonlinear transformation and pooling operation, performing connection layer activation by using Softmax, and obtaining a control strategy decision within one minute of the future of the farm after the connection layer activation.
Specifically, when the Softmax is used for activating the connecting layer, the obtained vector is reduced to a vector with the length of P, the size of P is generally calculated by combining the type of the culture unit where the vector is positioned through unit basic parameters such as the area, the quantity of cultured organisms and the like, real-time weather station data and network weather parameter data are imported into the model, a control strategy within 1 minute in the future is predicted, and the equipment operation state of the farm can be predicted more accurately through time sequence model training of the optimal equipment operation parameter value. The prediction is based on multi-dimensional data such as equipment operation parameters, real-time temperature and humidity of a weather station, network prediction temperature and humidity and the like, and various influencing factors are considered, so that the accuracy of prediction is improved, and a control strategy decision within one minute in the future can be obtained through time sequence model training. The fine control can timely adjust equipment operation parameters according to the real-time environment state and the equipment operation state of the farm so as to meet the actual requirements of the farm, the feature training adopts a two-layer network layer structure, the first layer detects single features, and the second layer carries out training. The structure can extract key features in the operation parameters of the equipment and train the key features, so that the adaptability and the robustness of the system are improved, the extracted features can be reduced and compressed through nonlinear transformation and pooling operation, the calculation complexity is reduced, and the calculation efficiency is improved. The method is beneficial to realizing real-time control and optimizing the cultivation environment, and realizes the automatic generation of control strategy decisions within one minute of the cultivation farm in the future through the activation of a connecting layer and a Softmax classifier. The method can lighten the burden of manual decision, improve the running efficiency and the automation level of a farm, train a time-series data model, predict the future control strategy in short time, and greatly shorten the control strategy adjustment lag time caused by instantaneous basic parameter changes such as weather dip and the like on the premise of 95% of prediction accuracy.
A farm micro-positive pressure intelligent control method based on multisource fusion perception comprises the following steps:
The first step: the method comprises the steps of confirming a sensor and equipment in a farm according to a sensor confirming unit, and detecting the use state of each equipment;
the method comprises the steps of monitoring the using time and sensitivity change of equipment periodically, finding and processing the equipment in time before the equipment fails, and avoiding the influence of sudden shutdown or performance reduction of the equipment on the production of farms;
And a second step of: the sensor data acquisition unit is used for carrying out data preprocessing and data storage on the data acquired in real time and the data subjected to equipment parameter adjustment;
The data cleaning process can remove noise, abnormal values or repeated data in the original data, so that the accuracy and reliability of subsequent analysis are ensured;
And a third step of: performing data operation parameter calculation on the stored data according to the acquired data analysis unit, and obtaining an optimal operation parameter value of equipment in the farm after the data operation parameter calculation;
the CNN convolutional neural network is used for learning and training the control parameters, so that the optimal equipment operation parameters are automatically acquired, the debugging efficiency is greatly improved, and the problem that experience data cannot be copied due to inconsistency of different units is solved;
Fourth step: finally, performing time sequence model training on the optimal operation parameter value of equipment in the farm through a sensor parameter control unit, and obtaining an optimal control strategy within one minute of the farm in the future after the time sequence simulation training;
wherein, through time series model training, can obtain the control strategy decision within one minute in the future. Such fine control may be based on the real-time environmental conditions of the farm and the equipment operating conditions.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (10)
1. Farm micro-positive pressure intelligent control system based on multisource fusion perception, which is characterized by comprising:
A sensor confirmation unit for:
Firstly, confirming a sensor in a farm and equipment corresponding to the sensor, and detecting the use state of each equipment when the equipment is started;
the sensor data acquisition unit is used for:
Receiving data acquired by a sensor corresponding to the equipment which is qualified in detection, respectively storing the data according to the type of the acquired data, and simultaneously, when the data is stored, correspondingly storing a memory according to the storage of the data and marking the acquired data in the memory as standard control data;
The acquired data analysis unit is used for:
calculating data operation parameters of standard control data through a CNN convolutional neural network, and judging the optimal equipment operation parameter value of each sensor according to a calculation result;
a sensor parameter control unit for:
and carrying out time sequence model training on the optimal equipment operation parameter value of each sensor, and carrying out strategy prediction on the farm according to the model training result.
2. The intelligent control system for micro-positive pressure of a farm based on multi-source fusion perception according to claim 1, wherein: the sensor confirmation unit includes:
a sensor and device confirmation module for:
the intelligent control equipment in the farm and the corresponding sensor of the equipment are confirmed;
The sensor comprises an inclination angle sensor, a distance sensor, a current detection sensor, a meteorological sensor, an air pressure sensor and an air speed sensor;
the equipment corresponding to the sensor comprises an operation fan set, a small window, a curtain, a weather station, a heater and a pressure-building collector;
wherein,
The small window is provided with an inclination sensor for acquiring the opening and closing angles of the small window;
A distance sensor is arranged at the top of the curtain and used for collecting the actual ventilation area in the farm;
The current detection sensor is used for collecting actual operation parameters of the fan, and the wind speed sensor is arranged at an air inlet and an air outlet of the operation fan and used for detecting actual ventilation quantity of the fan;
A weather sensor is arranged in the weather station and comprises a temperature sensor, a humidity sensor, a carbon dioxide concentration sensor, a rainfall sensor, an air pressure sensor, a wind speed and direction sensor and a particulate matter sensor;
The pressure sensor is arranged on the pressure collector and comprises a pressure sensor in the house and a pressure sensor outside the house and is used for collecting pressure difference values inside the house and outside the house.
3. The intelligent control system for micro-positive pressure of a farm based on multi-source fusion perception according to claim 2, wherein: the sensor confirmation unit further includes:
The equipment detection module is used for:
performing state evaluation on equipment of a farm;
Firstly monitoring and recording the using time of the equipment, then acquiring a sensitivity change curve of the equipment, and confirming the current sensitivity of each equipment according to the using time on the sensitivity change curve;
confirming a historical fault record of each device, wherein the historical fault record is retrieved from a database, and confirming fault sensitivity in the historical fault record;
comparing the current sensitivity with the fault sensitivity by a threshold value;
If the current sensitivity threshold is not in the fault sensitivity threshold range, the corresponding equipment is in a normal working state;
if the current sensitivity threshold is within the fault sensitivity threshold range, the corresponding equipment is in an abnormal working state;
and reporting the equipment in the abnormal working state, and maintaining the equipment by staff.
4. The intelligent control system for micro-positive pressure of a farm based on multi-source fusion perception according to claim 3, wherein: the sensor confirmation unit further includes:
The device parameter adjustment module is used for:
parameter adjustment is carried out on parameters of the equipment, wherein the adjusted parameters are debugged according to training scripts in a database;
and obtaining sensor data under different equipment parameters after the equipment parameters are adjusted, and taking the sensor data as environmental data of the farm.
5. The intelligent control system for micro-positive pressure of a farm based on multi-source fusion perception according to claim 4, wherein: the sensor data acquisition unit includes:
The sensor data classification module is used for:
receiving environment data, and preprocessing the data after the data is received;
the data preprocessing comprises data cleaning and data standardization;
After the data preprocessing, classifying the data types according to the sensors corresponding to the environmental data;
the data types are classified to obtain inclination angle sensor data, distance sensor data, current detection sensor data, weather sensor data, air pressure sensor data and wind speed sensor data;
and labeling the data with the classified data types as target acquisition data.
6. The intelligent control system for micro-positive pressure of a farm based on multi-source fusion perception according to claim 5, wherein: the sensor data acquisition unit further comprises:
A sensor data storage module for:
Confirming the data stock of each target acquisition data;
The data stock is obtained by dividing each target acquisition data into data segments, wherein the divided data segments have the same length, and the stock data of each target acquisition data is confirmed according to the number of the data segments;
each target acquisition data corresponds to an independent memory, and each independent memory is provided with a plurality of sub-memories;
when the target acquisition data is stored, the corresponding independent memory is firstly confirmed, and then the residual capacity of the sub-memory is confirmed;
And confirming the stored sub-memory according to the stock of the target acquisition data, wherein the stock of the target acquisition data is smaller than the residual capacity of the sub-memory.
7. The intelligent control system for micro-positive pressure of a farm based on multi-source fusion perception according to claim 6, wherein: the acquired data analysis unit is further configured to:
When the standard control data carries out data operation parameter calculation of the CNN convolutional neural network, firstly, confirming equipment parameters in the standard control data, wherein the data parameters are a temperature value, a humidity value, a differential pressure value, an operation fan parameter, a curtain opening value, a small window opening value, a heater operation parameter, a carbon dioxide value and a wind speed value;
Noise filtering is carried out on the equipment parameters according to the rolling layer and the pooling layer, and local characteristic data of the equipment parameters are obtained after noise filtering;
carrying out gradient explosion or gradient disappearance on the local characteristic data according to the residual error structure;
Carrying out depth feature extraction on local feature data of gradient explosion or gradient disappearance, and confirming feature values of the data after the depth feature extraction through Inception;
And (3) carrying out dimension ascending on the confirmed characteristic value through convolution of 1x1, and obtaining the optimal equipment operation parameter value of the target pressure difference under different ventilation levels through softmax.
8. The intelligent control system for micro-positive pressure of a farm based on multi-source fusion perception according to claim 7, wherein: the acquired data analysis unit is further configured to:
when the ventilation quantity of the running fan set is mapped to the fan rotating speed of the running fan set, the running fan set is subjected to analog output signal control and running control;
The analog output signal is controlled to set the control analog quantity of the frequency converter of the running fan set to be a 0-10V level interval value;
Collecting the wind speeds of an air inlet and an air outlet of the running fan set according to a wind speed sensor, and calculating real-time ventilation quantity according to the current curtain opening and the small window opening;
Collecting the numerical value of a current detection sensor in the running fan set in real time, and reversely calculating the running state deviation of the running fan set according to the current parameters;
the operation control is to control and execute equipment according to the deviation value, wherein the equipment for controlling and executing comprises an operation fan set, a small window and a curtain;
The deviation value calculation formula is as follows:
E=SV-PV
e is expressed as a deviation value, and the deviation value comprises a historical deviation, a current deviation and a latest deviation; SV is expressed as a device parameter target set value; PV represents the actual value of the current running wind turbine unit acquired data through the current detection sensor;
calculating a difference value between the target ventilation quantity and the actual ventilation quantity of the farm, wherein the difference value = the target ventilation quantity-the actual ventilation quantity, and the target ventilation quantity is retrieved from a database;
and obtaining the optimal analog quantity parameter of the farm after the difference value is calculated.
9. The intelligent control system for micro-positive pressure of a farm based on multi-source fusion perception according to claim 8, wherein: the sensor parameter control unit is further used for:
Performing time sequence model training on the optimal equipment operation parameter value;
Firstly, confirming n pieces of time slice data in the optimal equipment operation parameter values, wherein when the n pieces of time slice data are recorded at the sampling frequency of 0.2Hz for the equipment operation parameter, each time interval comprises data of 0.2 x n seconds, and in each time interval, an n x 3 matrix is obtained according to the real-time temperature and humidity of a weather station, the network prediction temperature and humidity and the equipment operation parameter;
carrying out feature training on n pieces of time slice data through time series prediction;
The feature training is carried out on two network layers, a feature detector in a first network layer detects a single feature, the single feature is input into a second network layer, and the second network layer trains the single feature;
performing nonlinear transformation and pooling operation on the trained single characteristic;
And after nonlinear transformation and pooling operation, performing connection layer activation by using Softmax, and obtaining a control strategy decision within one minute of the future of the farm after the connection layer activation.
10. The intelligent control method for the micro-positive pressure of the farm based on the multi-source fusion perception is realized based on the intelligent control system for the micro-positive pressure of the farm based on the multi-source fusion perception as claimed in claim 9, and is characterized by comprising the following steps:
the method comprises the steps of confirming a sensor and equipment in a farm according to a sensor confirming unit, and detecting the use state of each equipment;
The sensor data acquisition unit is used for carrying out data preprocessing and data storage on the data acquired in real time and the data subjected to equipment parameter adjustment;
Performing data operation parameter calculation on the stored data according to the acquired data analysis unit, and obtaining an optimal operation parameter value of equipment in the farm after the data operation parameter calculation;
And (3) performing time sequence model training on the optimal operation parameter value of the equipment in the farm by using the sensor parameter control unit, and obtaining an optimal control strategy within one minute of the farm in the future after the time sequence simulation training.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410524516.XA CN118112937B (en) | 2024-04-29 | 2024-04-29 | Intelligent control system and method for micro-positive pressure of farm based on multisource fusion perception |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410524516.XA CN118112937B (en) | 2024-04-29 | 2024-04-29 | Intelligent control system and method for micro-positive pressure of farm based on multisource fusion perception |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118112937A true CN118112937A (en) | 2024-05-31 |
| CN118112937B CN118112937B (en) | 2024-07-02 |
Family
ID=91219560
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410524516.XA Active CN118112937B (en) | 2024-04-29 | 2024-04-29 | Intelligent control system and method for micro-positive pressure of farm based on multisource fusion perception |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118112937B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110083090A (en) * | 2019-04-09 | 2019-08-02 | 华南农业大学 | A kind of livestock and poultry cultivation environmental parameter multipoint wireless intelligent monitor system and its method |
| CN210038479U (en) * | 2019-04-09 | 2020-02-07 | 华南农业大学 | Multi-point wireless intelligent monitoring system for livestock and poultry breeding environment parameters |
| CN112817354A (en) * | 2021-02-08 | 2021-05-18 | 中国农业大学 | Livestock and poultry house culture environment temperature prediction control system and regulation and control method thereof |
| CN114847168A (en) * | 2022-05-17 | 2022-08-05 | 四川华能宝兴河水电有限责任公司 | Intelligent breeding system for animal husbandry |
| CN115167585A (en) * | 2022-08-01 | 2022-10-11 | 王秀 | Livestock farm house environment control system |
| US11758887B1 (en) * | 2022-08-10 | 2023-09-19 | China Agricultural University | Method, system and apparatus for intelligently monitoring aquafarm with multi-dimensional panoramic perception |
-
2024
- 2024-04-29 CN CN202410524516.XA patent/CN118112937B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110083090A (en) * | 2019-04-09 | 2019-08-02 | 华南农业大学 | A kind of livestock and poultry cultivation environmental parameter multipoint wireless intelligent monitor system and its method |
| CN210038479U (en) * | 2019-04-09 | 2020-02-07 | 华南农业大学 | Multi-point wireless intelligent monitoring system for livestock and poultry breeding environment parameters |
| CN112817354A (en) * | 2021-02-08 | 2021-05-18 | 中国农业大学 | Livestock and poultry house culture environment temperature prediction control system and regulation and control method thereof |
| CN114847168A (en) * | 2022-05-17 | 2022-08-05 | 四川华能宝兴河水电有限责任公司 | Intelligent breeding system for animal husbandry |
| CN115167585A (en) * | 2022-08-01 | 2022-10-11 | 王秀 | Livestock farm house environment control system |
| US11758887B1 (en) * | 2022-08-10 | 2023-09-19 | China Agricultural University | Method, system and apparatus for intelligently monitoring aquafarm with multi-dimensional panoramic perception |
Non-Patent Citations (1)
| Title |
|---|
| 王艳;陈惠英;乔记平;杨丽华;王智鹏;刘儒平;: "基于LabVIEW的生猪养殖环境监控系统设计", 黑龙江畜牧兽医, no. 09, 10 May 2020 (2020-05-10) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118112937B (en) | 2024-07-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111539553B (en) | Wind turbine generator fault early warning method based on SVR algorithm and off-peak degree | |
| de la Hermosa González | Wind farm monitoring using Mahalanobis distance and fuzzy clustering | |
| CN106796423A (en) | Intelligent programmable logic controller | |
| CN115329812A (en) | Road infrastructure abnormity monitoring method based on artificial intelligence | |
| CN114399081A (en) | Photovoltaic power generation power prediction method based on weather classification | |
| CN113780375A (en) | Virtual-real interaction wind power plant wind power monitoring system based on digital twins and application | |
| CN115689524A (en) | Predictive maintenance system and method for electrical equipment of data center machine room | |
| CN117762913A (en) | Automatic identification method and system for abnormal power data of offshore wind power | |
| CN117032165A (en) | Industrial equipment fault diagnosis method | |
| CN117151684A (en) | Wind power fan data analysis early warning method, system, device and readable storage medium | |
| CN118112937B (en) | Intelligent control system and method for micro-positive pressure of farm based on multisource fusion perception | |
| CN117933531A (en) | Distributed photovoltaic power generation power prediction system and method | |
| US11687772B2 (en) | Optimization of cyber-physical systems | |
| KR20220089853A (en) | Method for Failure prediction and prognostics and health management of renewable energy generation facilities using machine learning technology | |
| CN116720073A (en) | Abnormality detection extraction method and system based on classifier | |
| CN116756505A (en) | Photovoltaic equipment intelligent management system and method based on big data | |
| CN113723670B (en) | Photovoltaic power generation power short-term prediction method with variable time window | |
| CN115356990A (en) | Material yard equipment fault prediction method and system based on deep learning and storage medium | |
| KR20220085213A (en) | Detection/diagnosis/prediction system of environment sensor failure for indoor smart cultivation system based on deep llearning | |
| CN117688464B (en) | Hidden danger analysis method and system based on multi-source sensor data | |
| CN114151291B (en) | Early fault monitoring method for wind turbine generator | |
| CN117933974B (en) | Remote maintenance method and system for water quality data acquisition monitoring equipment | |
| CN117951633B (en) | Photovoltaic power generation equipment fault diagnosis method and system | |
| CN117332360B (en) | Greenhouse equipment fault monitoring method and system based on 5G technology | |
| Eredics et al. | Data Cleaning and Anomaly Detection for an Intelligent Greenhouse |
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 |