An Industrial Internet of Things system for abnormal analysis, the Industrial Internet of Things system including a management platform, a user platform, a service platform, a sensor network platform, and one or more object platforms, wherein the user platform, the service platform, the management platform, the sensor network platform, and the one or more object platforms are interacted sequentially from top to bottom, wherein
- the management platform is configured to:
- obtain an ordered set of instructions for a preset time period, wherein the ordered set of instructions includes a plurality of instructions, and instruction content of each instruction of the plurality of instructions includes at least one of instruction type, instruction parameter, and instruction execution time;
- determine whether there is an abnormal instruction based on the ordered set of instructions;
- and
- analyze and modify the abnormal instruction in response to a determination that there is the abnormal instruction;
- the service platform and the sensor network platform are both arranged in an independent layout, and the management platform is arranged in a rear sub platform layout;
- the independent layout means that the service platform and the sensor network platform are provided with a plurality of independent sub platforms, and the plurality of independent sub platforms respectively perform data storage, data processing and/or data transmission for different data;
- the rear sub platform layout means that the management platform is provided with a general platform of the management platform and a plurality of sub platforms of the management platform, the plurality of sub platforms of the management platform respectively store and process data of different types or different receiving objects sent by the service platform, and the general platform of the management platform stores and processes data of the plurality of sub platforms of the management platform after summarizing, and transmits the data of the plurality of sub platforms of the management platform to the sensor network platform;
- and the one or more object platforms is configured as one or more intelligent manufacturing device;
- when the user platform sends a manufacturing task instruction, a corresponding sub platform of the service platform receives the manufacturing task instruction, converts the manufacturing task instruction into a configuration file recognized by the management platform, and sends the configuration file to a corresponding sub platform of the management platform, wherein the manufacturing task instruction includes at least one manufacturing task and an execution time corresponding to the manufacturing task;
- the corresponding sub platform of the management platform receives the configuration file, uses a cyclic neural network model to carry out a sequence modeling for data in the configuration file in time order to form sequence modeling data, and sends the sequence modeling data together with the configuration file to the general platform of the management platform;
- and
- after receiving and storing the corresponding sequence modeling data, the general platform of the management platform decomposes the sequence modeling data sent by all the sub platforms of the management platform according to execution times, and sorts all decomposed sequence modeling data according to a sequence logic order from beginning to end to form a series of event sequence data, wherein each sequence corresponds to a sequence point, the general platform of the management platform analyzes and predicts the event sequence data through a probability suffix tree model, and executes operations based on an analysis and prediction result, the operations including:
- when the analysis and prediction result is that the task is normal, sending received configuration files to a corresponding sub platform of the sensor network platform, the corresponding sub platform of the sensor network platform receiving the configuration files and converting the configuration files into instruction files recognizable by the one or more object platforms and sending the instruction files to the one or more object platforms, and the one or more object platforms receiving the instruction files and executing intelligent manufacturing based on the instruction files;
- or
- when the analysis and prediction result is that the task is abnormal, refusing to receive all the configuration files corresponding to the manufacturing task, deleting the corresponding sequence modeling data, obtaining associated manufacturing data corresponding to abnormal timing points, and sending the associated manufacturing data to a corresponding sub platform of the management platform, the corresponding sub platform of the management platform compiling the associated manufacturing data and sending the associated manufacturing data to a corresponding sub platform of the service platform, the corresponding sub platform of the service platform storing the associated manufacturing data and sending a task failure instruction to the user platform;
- wherein the associated manufacturing data includes associated sequence modeling data corresponding to the abnormal timing points, sub platform information of the management platform corresponding to the sequence modeling data, and sub platform information of the service platform.