[0025] Next, the technical solutions in the embodiments of the present invention will be apparent from the embodiment of the present invention, and it is clearly described, and it is understood that the described embodiments are merely embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, there are all other embodiments obtained without making creative labor without making creative labor premises.
[0026] In order to make the above objects, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0027] like figure 1 , The present invention provides a mine artificial intelligence-based monitoring system environment, including the environment information acquisition module which are sequentially connected, the cloud server, smart detection module, a voice alarm module;
[0028] Environment information collecting means for collecting environmental data within the monitoring area, and environmental data to the cloud server;
[0029] Environmental data for the cloud server environment information acquisition module for storing transmitted; intelligent means for detecting environmental data retrieved from the cloud server and the intelligent data analysis and detection, and outputs the detection result of collation stored;
[0030] Environment information acquisition module includes a communication station and a plurality of sensors located in the monitored area, wherein the environmental data collected by sensors within the monitored area, and the collected data to the communication base station.
[0031] Sensors within the monitored region using soil monitoring sensors, and each sensor has a fixed soil monitoring the ID number, so that when the monitored data exceeding the environmental data which can be positioned to excessive sensors and from which the monitored region;
[0032] Each sensor within the monitored area automatic detection time interval of time, 24 hours a day and monitored, wherein the time interval can be set according to the detection of human needs, environment detection, testing conditions, etc., the present example of the setting time was 2 hours interval, i.e. sensor every 2 hours of environmental data in the monitored area for an automatic detection and data transfer; communication base station is responsible for receiving the sensor sampled ambient data, then the received environment data stored communication network transmitted via to the cloud server, so that realize the remote transmission of field data, without the need to monitor on-site personnel reside.
[0033] Artificial intelligence module includes a calculation processor detection, artificial intelligent processor to analyze the environmental data detected by the smart detection model; artificial intelligence computing systems and processors are deployed Ubuntu Tensorflow depth learning framework, wherein the framework is to ensure deep learning Tensorflow RNN model can work and run the necessary environmental conditions, Ubuntu Linux system belongs to a system that can help RNN model runs at a faster speed, depth learning model software to meet normal operating environment; artificial intelligence computing processor must meet 64G memory, 4T disk space, in order to ensure sufficient AI calculation processor having computing power and data storage space, while artificial intelligent processor is preferably provided with a plurality of the GPU, which can meet the demand for parallel calculation processing.
[0034] Artificial Intelligence (ArtificialIntelligence, AI) is involved in the research, design and application of intelligent machines machine learning technology, which can simulate the human brain work, belonging to a sub-field of machine learning. AI is performed by the artificial neural network modeling complex relationships between the data, which is formed by combining the low-level feature to a more abstract level features, thereby extracting feature data, with more modeling and reasoning. Unlike the conventional method, the mathematical equations AI without prior mapping relationship between input and output, only by training learning neural network itself can autonomously learned from the data useful features, so that at a given input value to get closest to the desired output. Artificial neural networks are interconnected by a large number of processing units constituting the neural network, which has a strong self-learning capability to automatically summarize the data obtained from the existing data law characteristic. Recurrent Neural Network RNN (recurrent neural network) is an artificial neural network which is trained using the back propagation algorithm, a neural network having a feedback structure. The preceding information memory and the neural network will be stored and used to calculate the current output, thereby improving the reliability and accuracy of the output of the network.
[0035] Smart Detection Model The system model is used in detecting the data classification based autonomous learning Recurrent Neural Network (RNN) obtained, and its basic structure includes an input layer, a hidden layer and output layer, the input layer is responsible for data input, the hidden layer is responsible for data calculation processing, outputs the calculation result output layer; intelligent detection model environmental data retrieved from the cloud server intelligent data analysis of the detection, if the environmental data detected non-compliance is determined, then the detection result to "current time-specific", "excessive type "," excessive value "," excessive level "and" occurrence of excessive sensor ID number "and stored in the form of a corresponding output, and trigger the start of a voice alarm module, wherein the" level exceeded "is divided into generally, serious and very serious third gear , and the threshold between adjacent two steps may be manually set in accordance with the environmental conditions and demand; environmental data is detected if it is determined that standard, the result will be detected as "current time-specific" and "detection monitoring all normal" in the form of the corresponding output and saved, but will not trigger the start of a voice alarm module.
[0036] Voice alarm module deployed in artificial intelligence computing processor, judged by intelligent detection model, in a case where to start a voice alarm trigger conditions, i.e. when initiating a voice alarm processor uses artificial intelligence to detect non-compliance model checking data.
[0037] Detection process of the present invention are:
[0038]The plurality of sensors set the setting time of the detection interval are deployed in the specified monitoring area, which is set to 2 hours, so that the sensor automatically detects the environmental data in the monitoring area every 2 hours, and then detects environmental data Transfer to the communication base station, the communication base station will then transmit the received environment data to the cloud server via the communication network, and the cloud server is saved for the received environmental data. The artificial intelligence calculation processor uses the intelligent detection model to transfer the saved environment data from the cloud server, while intelligent data analysis of environmental data. If the environmental data detection is not up to the standard, the test results are "current specific time" , "Exceeding the standard type", "exceeding the standard value", "exceeding the standard level" and "exceeding the sensor ID number" in the form of output and saving, and trigger the startup voice alarm module, where the "over-standard level" is general, very serious, very Serious three gears, and the thresholds between the adjacent two files can be set according to the environment, conditions and needs; if the environmental data detection is determined by the standard, the test results will be detected in "current specific time" and "all monitor points. Normal "The form corresponds to the output and saved, but the voice alarm module will not be triggered.
[0039] In the description of the present invention, it is to be understood that the term "longitudinal", "horizontal", "upper", "lower", "front", "post", "left", "right", "vertical", The orientation relationship between "horizontal", "top", "bottom", "inside", "outside", etc. is based on the orientation or positional relationship shown in the drawings, but is intended to facilitate the description of the present invention, not an indication or I suggest that the device or component must have a specific orientation, constructed and operated in a particular direction, and thus is not to be construed as limiting the invention.
[0040] The embodiments described above are merely described in the preferred embodiments of the present invention, and are not limited to the scope of the invention, and various articles of the present invention will have various techniques for the present invention without departing from the spirit of the invention. Deformation and improvement, should fall within the scope of protection determined by the claims of the present invention.
