A fuel intelligent sensing monitoring system based on human-computer interaction
The intelligent fuel sensing and monitoring system with human-computer interaction solves the problem of inaccurate fuel prediction, realizes accurate prediction and visualization of fuel status, and enhances the interactive experience.
Patent Information
- Application Number
- CN202411044897.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-01
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-08-01
AI Technical Summary
Existing fuel usage forecasting technologies are not accurate enough and cannot visually display forecast results, affecting fuel procurement timing and inventory levels.
The fuel intelligent sensing and monitoring system based on human-computer interaction uses a data acquisition module to acquire real-time fuel data, a data processing module to clean and fuse the data, a model building module to train a neural network model, and a model optimization module to make predictions and provide visualizations, thereby achieving accurate prediction and feedback of fuel status.
It improves the accuracy of fuel prediction results, enhances the human-computer interaction experience, and ensures the visualization and real-time monitoring of fuel usage status.
Smart Images

Figure CN118569449B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of human-computer interaction, and in particular to a fuel intelligent sensing monitoring system based on human-computer interaction. Background Art
[0002] The usage status of fuel will affect the fuel procurement time and fuel inventory, but most of the current fuel usage prediction technologies are not accurate enough and cannot intuitively display the prediction results.
[0003] Therefore, the present invention provides a fuel intelligent sensing monitoring system based on human-computer interaction. Summary of the invention
[0004] The present invention provides a fuel intelligent sensing monitoring system based on human-computer interaction, which is used to solve the defects of the fuel intelligent sensing monitoring system based on human-computer interaction.
[0005] In one aspect, the present invention provides a fuel intelligent sensing monitoring system based on human-computer interaction, comprising:
[0006] Data acquisition module: used to establish a fuel sensing monitoring system, obtain log data of the fuel intelligent sensing monitoring system, and extract data from the log data to obtain real-time data of the fuel to be monitored;
[0007] Data processing module: used to process the real-time data of the monitored fuel to obtain valid data;
[0008] Model building module: used to obtain fuel property data of corresponding fuel types in the fuel depot, determine fuel characteristic data based on the fuel property data, and input the data into the neural network model for data training to obtain a fuel prediction model;
[0009] Model optimization module: used to input valid data into the fuel prediction model to predict the fuel status, build a visual monitoring interface to display the prediction results, obtain feedback information on the display results, and update the visual monitoring interface based on the feedback information.
[0010] According to a fuel intelligent sensing monitoring system based on human-computer interaction provided by the present invention, the data acquisition module includes:
[0011] a sensor determination unit, configured to determine a fuel type of the fuel to be monitored, and determine sensor models of different types of sensors based on the fuel type;
[0012] Determine the fuel operation system of the fuel to be monitored, obtain different operation modules of the fuel operation system, and determine the type and quantity of sensors required for each operation module;
[0013] The system establishment unit is used to install corresponding types of sensors according to the number of sensors to the corresponding operation modules of the fuel operation system to obtain the fuel intelligent sensing system.
[0014] According to a fuel intelligent sensing monitoring system based on human-computer interaction provided by the present invention, the data acquisition module includes:
[0015] A data parsing unit, used to obtain log data of the fuel intelligent sensing monitoring system, and parse the log data based on the corresponding communication protocol to obtain corresponding information of each log data;
[0016] The data extraction unit is used to extract information from the corresponding information of all log data to obtain information related to the fuel to be monitored, and use the corresponding data of the related information as the real-time data of the fuel to be monitored.
[0017] According to a fuel intelligent sensing monitoring system based on human-computer interaction provided by the present invention, the data processing module includes:
[0018] A data cleaning unit, used to determine the sensor type of the sensor corresponding to each piece of real-time data;
[0019] Acquire data cleaning methods corresponding to different types of sensors, and perform data cleaning on corresponding data based on the data cleaning methods;
[0020] The data fusion unit classifies the cleaned real-time data according to the operation module, and performs data fusion according to the classification results to obtain separate valid data for each operation module;
[0021] Each individual valid data is fused twice to obtain the valid data of the fuel intelligent sensing monitoring system.
[0022] According to a fuel intelligent sensing monitoring system based on human-computer interaction provided by the present invention, the model building module also includes:
[0023] a fuel characteristic determination unit, configured to obtain fuel characteristic data corresponding to the fuel type based on a fuel reservoir, and determine the fuel characteristic based on the fuel characteristic data;
[0024] Among them, the fuel characteristics include: different combustion stages of the fuel, the fuel state at different combustion stages, the combustion time of a unit weight of fuel at different combustion stages, the combustion temperature and the heat generated.
[0025] According to a fuel intelligent perception monitoring system based on human-computer interaction provided by the present invention, the model building module includes:
[0026] A first training unit, used for inputting different combustion stages of the fuel into the neural network model to perform a first main training;
[0027] At the same time, the fuel states at different combustion stages are input into the neural network model for first auxiliary training;
[0028] A second training unit is used to input the heat generated by the unit weight of the fuel at different combustion stages into the neural network model to perform a second main training;
[0029] At the same time, the combustion time and combustion temperature of the unit weight of fuel at different combustion stages are input into the neural network model for second auxiliary training;
[0030] The model building unit is used to obtain the fuel prediction model after all the training of the neural network model is completed.
[0031] According to a fuel intelligent perception monitoring system based on human-computer interaction provided by the present invention, the model optimization module includes:
[0032] A state prediction unit, used for inputting corresponding valid data of each operation module into the fuel prediction model to perform a first state prediction, and obtain a first operation result of each operation module;
[0033] Inputting valid data of the fuel intelligent sensing monitoring system into the fuel prediction model to perform a second state prediction, and obtaining a second operation result of the fuel intelligent sensing monitoring system;
[0034] a similarity analysis unit, configured to obtain a second sub-operation result of each operation module in the second operation result, and perform a similarity analysis between each second sub-operation result and the corresponding first operation result;
[0035] ;in, is the similarity between the first operation result of the kth operation module and the corresponding second sub-operation result, n is the total number of fuel features in each operation module, is the i-th fuel feature in the k-th operating module, is the characteristic weight of the i-th fuel characteristic, is the number of data collections per unit time in the kth running module, is the first running result of the i-th fuel feature of the k-th running module collected at the jth time, The second operation sub-result is obtained by collecting the i-th fuel characteristic of the k-th operation module at the j-th time;
[0036] A state determination unit is used to determine the operation state of the corresponding operation module based on the similarity analysis result. If the similarity exceeds a specified threshold, the operation state of the corresponding operation module is deemed to be normal operation, and a similarity level based on the similarity is obtained to determine an optimization plan;
[0037] A result optimization unit, configured to perform a first optimization on the second operation result based on the first operation result according to the optimization scheme to obtain a first optimization result;
[0038] Otherwise, it is determined that the corresponding operation module has an operation fault, the corresponding operation module is cleared of the fault, and data collection and data processing are performed again to obtain new independent valid data;
[0039] Inputting the new independent valid data into the fuel prediction model to perform a third state prediction, and obtaining a third operation result of the fault operation module;
[0040] A similarity analysis is performed on each third operation result and the corresponding second operation sub-result, and a second optimization is performed on the first optimization result to obtain a prediction result.
[0041] According to a fuel intelligent perception monitoring system based on human-computer interaction provided by the present invention, the model optimization module includes:
[0042] A prediction display unit, used to construct a visual monitoring interface to display the first operation result and prediction result of each operation module;
[0043] A feedback collection unit, used to determine a collection period for collection result feedback information, and to regularly collect collection result feedback information according to the collection period;
[0044] A feedback sorting unit is used to classify the result feedback information according to different feedback directions, sort the feedback information of the same feedback direction, and obtain an update list;
[0045] The periodic updating unit is used to update the visual monitoring interface according to the collection cycle of the collection result feedback information based on the update list.
[0046] The present invention provides a fuel intelligent sensing monitoring system based on human-computer interaction. The system establishes a fuel sensing monitoring system and obtains log data for data extraction to obtain real-time data, and then obtains valid data through data processing. The fuel characteristic data is determined through fuel characteristic data and data training is performed to obtain a fuel prediction model. The valid data is input for prediction and displayed through a visual monitoring interface. At the same time, display feedback information is obtained to update the monitoring interface, which can ensure the accuracy of the fuel prediction results and improve the interactive experience of human-computer interaction. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0048] Figure 1 It is a structural diagram of a fuel intelligent sensing monitoring system based on human-computer interaction provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0049] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0050] like Figure 1 As shown, an embodiment of the present invention provides a fuel intelligent sensing monitoring system based on human-computer interaction, including:
[0051] Data acquisition module: used to establish a fuel sensing monitoring system, obtain log data of the fuel intelligent sensing monitoring system, and extract data from the log data to obtain real-time data of the fuel to be monitored;
[0052] Data processing module: used to process the real-time data of the monitored fuel to obtain valid data;
[0053] Model building module: used to obtain fuel property data of corresponding fuel types in the fuel depot, determine fuel characteristic data based on the fuel property data, and input the data into the neural network model for data training to obtain a fuel prediction model;
[0054] Model optimization module: used to input valid data into the fuel prediction model to predict the fuel status, build a visual monitoring interface to display the prediction results, obtain feedback information on the display results, and update the visual monitoring interface based on the feedback information.
[0055] In this embodiment, the fuel sensing monitoring system monitors the temperature, pressure and liquid level of the fuel through different types of sensors.
[0056] In this embodiment, the log data of the fuel monitoring system includes data collected by each sensor, and the log data can be parsed through the communication protocol to obtain specific fuel information corresponding to each piece of data.
[0057] In this embodiment, the real-time data of the fuel refers to the latest data related to the fuel information within a unit time obtained from the log data.
[0058] In this embodiment, valid data refers to data obtained after data processing is performed on real-time data, and the data processing includes data cleaning and data fusion.
[0059] In this embodiment, the fuel library contains fuel types of all fuels and corresponding fuel characteristic data. The fuel type refers to the type of fuel, such as coal, coke, natural gas, biogas, etc. The fuel characteristic data is data that can reflect the characteristics of the fuel. For example, the fuel characteristic data of fuel A is 030217, where 03 represents coal, 02 represents the solid of the fuel, and 17 represents the heat generated by the combustion of the fuel. Fuel A is solid coal fuel, and the combustion of unit weight of fuel A can fully generate 17 calories.
[0060] In this embodiment, the fuel characteristic data is data that can reflect the characteristics of the fuel, including characteristic data that can reflect the different combustion stages of the fuel, the fuel state at different combustion stages, the combustion time of unit weight of fuel at different combustion stages, the combustion temperature and the heat generated.
[0061] In this embodiment, the fuel prediction model can predict the fuel state through valid data, and the fuel state includes the combustion time of the fuel, the heat that can be generated by the remaining fuel, and the combustion temperature of the fuel.
[0062] In this embodiment, the visual monitoring interface can intuitively reflect the fuel status, and different fuel statuses can be viewed by selection. The feedback information is provided by the user, and feedback opinions can be obtained by collating the feedback information.
[0063] The beneficial effects of the above technical solution are: by establishing a fuel sensing monitoring system and obtaining log data for data extraction to obtain real-time data, then obtaining valid data through data processing, determining fuel characteristic data through fuel characteristic data and performing data training to obtain a fuel prediction model, inputting valid data for prediction and displaying it through a visual monitoring interface, and obtaining display feedback information to update the monitoring interface, the accuracy of the fuel prediction results can be ensured, and the interactive experience of human-computer interaction is improved.
[0064] The embodiment of the present invention provides a fuel intelligent sensing monitoring system based on human-computer interaction, and a data acquisition module, including:
[0065] a sensor determination unit, configured to determine a fuel type of the fuel to be monitored, and determine sensor models of different types of sensors based on the fuel type;
[0066] Determine the fuel operation system of the fuel to be monitored, obtain different operation modules of the fuel operation system, and determine the type and quantity of sensors required for each operation module;
[0067] The system establishment unit is used to install corresponding types of sensors according to the number of sensors to the corresponding operation modules of the fuel operation system to obtain the fuel intelligent sensing system.
[0068] In this embodiment, the fuel type refers to the fuel type of the fuel. Different fuel types require different sensor models. For example, liquid fuel requires a contact level meter to count the remaining fuel amount.
[0069] In this embodiment, the fuel operation system refers to a complete operation system for fuel use, including different operation modules, such as: a conveying module, an abrasive module, a combustion module, a smoke treatment module, and an ash destruction module.
[0070] The beneficial effects of the above technical solution are: determining the sensor models of different types of sensors by the fuel type of the fuel to be monitored, obtaining the different operating modules of the fuel operation system and determining the required sensor types and number of sensors for installation, and obtaining a fuel intelligent sensing system, thereby ensuring that the fuel intelligent sensing system can comprehensively collect information.
[0071] The embodiment of the present invention provides a fuel intelligent sensing monitoring system based on human-computer interaction, and a data acquisition module, including:
[0072] A data parsing unit, used to obtain log data of the fuel intelligent sensing monitoring system, and parse the log data based on the corresponding communication protocol to obtain corresponding information of each log data;
[0073] The data extraction unit is used to extract information from the corresponding information of all log data to obtain information related to the fuel to be monitored, and use the corresponding data of the related information as the real-time data of the fuel to be monitored.
[0074] In this embodiment, information extraction refers to the process of screening out fuel-related information from all information. For example, if information A includes machine operation time and fuel combustion time, the fuel combustion time is extracted from information A.
[0075] The beneficial effects of the above technical solution are: by acquiring the log data of the fuel intelligent sensing monitoring system and parsing the log data through the corresponding communication protocol to obtain the corresponding information, by extracting the information to obtain the information related to the fuel to be monitored and using the corresponding data as real-time data, it is ensured that the real-time data are all fuel-related data, thereby improving the efficiency of data processing.
[0076] The embodiment of the present invention provides a fuel intelligent sensing monitoring system based on human-computer interaction, and a data processing module, including:
[0077] A data cleaning unit, used to determine the sensor type of the sensor corresponding to each piece of real-time data;
[0078] Acquire data cleaning methods corresponding to different types of sensors, and perform data cleaning on corresponding data based on the data cleaning methods;
[0079] The data fusion unit classifies the cleaned real-time data according to the operation module, and performs data fusion according to the classification results to obtain separate valid data for each operation module;
[0080] Each individual valid data is fused twice to obtain the valid data of the fuel intelligent sensing monitoring system.
[0081] In this embodiment, different types of sensors have corresponding different collection noises, and different data cleaning methods need to be selected to clean the data collected by different types of sensors. Data cleaning is actually the process of denoising the data.
[0082] In this embodiment, data fusion refers to the process of fusing similar real-time data together and removing duplicate values and invalid values to obtain the operating status of the corresponding operating module. The real-time data is fused according to the different operating modules to which it belongs to obtain separate valid data.
[0083] In this embodiment, secondary fusion refers to the process of fusing all the individual valid data together and deleting duplicate data to obtain data that can reflect the operating status of the fuel intelligent sensing monitoring system. The individual valid data is obtained after secondary fusion to obtain valid data.
[0084] The beneficial effects of the above technical solution are: by confirming the type of sensor and determining the data cleaning method to clean the real-time data, the accuracy of the real-time data is guaranteed; by fusing the cleaned data to obtain separate valid data, and then fusing the separate valid data for the second time to obtain valid data, the accuracy of the valid data is guaranteed.
[0085] The embodiment of the present invention provides a fuel intelligent sensing monitoring system based on human-computer interaction, and the model building module also includes:
[0086] a fuel characteristic determination unit, configured to obtain fuel characteristic data corresponding to the fuel type based on a fuel reservoir, and determine the fuel characteristic based on the fuel characteristic data;
[0087] Among them, the fuel characteristics include: different combustion stages of the fuel, the fuel state at different combustion stages, the combustion time of a unit weight of fuel at different combustion stages, the combustion temperature and the heat generated.
[0088] The beneficial effect of the above technical solution is: the fuel characteristic data of the corresponding fuel type is obtained through the fuel library, and the fuel characteristics are determined through the fuel characteristic data, which ensures the accuracy of the fuel characteristics and facilitates the subsequent acquisition of the fuel prediction model.
[0089] The embodiment of the present invention provides a fuel intelligent sensing monitoring system based on human-computer interaction, and a model building module, including:
[0090] A first training unit, used for inputting different combustion stages of the fuel into the neural network model to perform a first main training;
[0091] At the same time, the fuel states at different combustion stages are input into the neural network model for first auxiliary training;
[0092] A second training unit is used to input the heat generated by the unit weight of the fuel at different combustion stages into the neural network model to perform a second main training;
[0093] At the same time, the combustion time and combustion temperature of the unit weight of fuel at different combustion stages are input into the neural network model for second auxiliary training;
[0094] The model building unit is used to obtain the fuel prediction model after all the training of the neural network model is completed.
[0095] In this embodiment, the first main training is the process of training the neural network model through data from different combustion stages of the fuel, and the first auxiliary training is the process of training through the fuel status at different combustion stages, so that the neural network model can determine the remaining combustion stage and combustion time of the fuel through the fuel data.
[0096] In this embodiment, the neural network model can obtain the heat that can be generated by fuel combustion based on the effective data of the fuel through the second main training, and the second auxiliary training refers to the process in which the neural network model is trained to obtain the combustion stage of the fuel through the current combustion state.
[0097] In this embodiment, the fuel prediction model can obtain the combustion state of the fuel through the effective data of the fuel, and determine the combustion stage of the fuel, and obtain the fuel state and combustion duration of each remaining combustion stage.
[0098] The beneficial effects of the above technical solution are: by inputting the different combustion stages of the fuel into the neural network model for the first main training, the fuel status at different combustion stages is input into the neural network model for the first auxiliary training, so as to realize the prediction of the combustion status of the fuel at different combustion stages, and by performing the second main training and the second auxiliary training of the neural network model through the heat, combustion time and combustion temperature of the fuel, it is possible to determine the current combustion stage of the fuel through the combustion status of the fuel.
[0099] The embodiment of the present invention provides a fuel intelligent perception monitoring system based on human-computer interaction, a model optimization module, including:
[0100] A state prediction unit, used for inputting the corresponding separate valid data of each operation module into the fuel prediction model to perform a first state prediction, and obtain a first operation result of each operation module;
[0101] Inputting valid data of the fuel intelligent sensing monitoring system into the fuel prediction model to perform a second state prediction, and obtaining a second operation result of the fuel intelligent sensing monitoring system;
[0102] a similarity analysis unit, configured to obtain a second sub-operation result of each operation module in the second operation result, and perform a similarity analysis between each second sub-operation result and the corresponding first operation result;
[0103] ;in, is the similarity between the first operation result of the kth operation module and the corresponding second sub-operation result, n is the total number of fuel features in each operation module, is the i-th fuel feature in the k-th operating module, is the characteristic weight of the i-th fuel characteristic, is the number of data collections per unit time in the kth running module, is the first running result of the i-th fuel feature of the k-th running module collected at the jth time, The second operation sub-result is obtained by collecting the i-th fuel characteristic of the k-th operation module at the j-th time;
[0104] A state determination unit is used to determine the operation state of the corresponding operation module based on the similarity analysis result. If the similarity exceeds a specified threshold, the operation state of the corresponding operation module is deemed to be normal operation, and a similarity level based on the similarity is obtained to determine an optimization plan;
[0105] A result optimization unit, configured to perform a first optimization on the second operation result based on the first operation result according to the optimization scheme to obtain a first optimization result;
[0106] Otherwise, it is determined that the corresponding operation module has an operation fault, the corresponding operation module is cleared of the fault, and data collection and data processing are performed again to obtain new independent valid data;
[0107] Inputting the new independent valid data into the fuel prediction model to perform a third state prediction, and obtaining a third operation result of the fault operation module;
[0108] A similarity analysis is performed on each third operation result and the corresponding second operation sub-result, and a second optimization is performed on the first optimization result to obtain a prediction result.
[0109] In this embodiment, the first state prediction refers to the process of inputting separate valid data into the fuel prediction model to determine the operating state of each operating module, and the predicted operating state is the first operating result of each operating module.
[0110] In this embodiment, the second state prediction refers to the process of inputting valid data into the fuel prediction model to determine the operating state of the fuel intelligent sensing monitoring system, and the predicted operating state is the second operating result of the fuel intelligent sensing monitoring system.
[0111] In this embodiment, the second operation result is disassembled, and the operation status of each operation module corresponding to the obtained second operation result is the second sub-operation result.
[0112] In this embodiment, similarity analysis is a process of calculating the similarity between the second sub-operation result and the corresponding first operation result. The similarity level refers to the level to which the similarity belongs. For example, if the similarity is 95%, the similarity level is level one.
[0113] In this embodiment, the optimization schemes for different similarity levels are different. For example, when the similarity level is level one, the optimization scheme is to take the average of the value that can characterize the fuel combustion state in the first operation result and the corresponding value of the second operation result. When the similarity level is level two, the optimization scheme is to multiply the value that can characterize the fuel combustion state in the first operation result by 0.6, multiply the corresponding value of the second operation result by 0.4 and then add the two together.
[0114] In this embodiment, the first optimization refers to a process of determining a corresponding optimization scheme and optimizing the second operation result according to the optimization scheme, and the result obtained by the first optimization is the first optimization result.
[0115] In this embodiment, fault clearing needs to be performed by manpower. When it is determined that the corresponding operation module has an operation fault, the staff is notified to perform inspection and maintenance.
[0116] In this embodiment, the new independent valid data is the data obtained after clearing the fault of the operating module having the fault, re-collecting the data and processing the collected data.
[0117] In this embodiment, the third state prediction refers to the process of inputting new independent valid data into the fuel prediction model to determine the operating state of each operating module, and the predicted operating state is the third operating result of each operating module.
[0118] In this embodiment, the second optimization refers to a process of determining a corresponding optimization scheme by combining the third operation result and the corresponding second sub-operation result, and optimizing the second operation result according to the optimization scheme, and the result obtained by the second optimization is a prediction result.
[0119] The beneficial effect of the above technical solution is: a separate prediction result is obtained by inputting separate valid data into the fuel prediction model for state prediction, and a second operation result is obtained by state prediction through valid data at the same time, and the second operation result is optimized through similarity analysis, and the fault condition is determined, the similarity is recalculated and optimized to obtain the final prediction result, thereby ensuring the accuracy of the prediction result.
[0120] The embodiment of the present invention provides a fuel intelligent perception monitoring system based on human-computer interaction, a model optimization module, and further includes:
[0121] A prediction display unit, used to construct a visual monitoring interface to display the first operation result and prediction result of each operation module;
[0122] A feedback collection unit, used to determine a collection period for collection result feedback information, and to regularly collect collection result feedback information according to the collection period;
[0123] A feedback sorting unit is used to classify the result feedback information according to different feedback directions, sort the feedback information of the same feedback direction, and obtain an update list;
[0124] The periodic updating unit is used to update the visual monitoring interface according to the collection cycle of the collection result feedback information based on the update list.
[0125] In this embodiment, the collection period can be determined according to the average daily number of visits to the visual monitoring interface.
[0126] In this embodiment, the feedback direction is the type targeted by the feedback information. For example, if the feedback information is that the interface response speed is slow and the machine is stuck, the feedback direction is the machine hardware direction.
[0127] In this embodiment, the update list is obtained by sorting the feedback information, and includes all the issues that need to be updated that are targeted by the feedback information.
[0128] The beneficial effects of the above technical solution are: by constructing a visual monitoring interface to display the first operating result and the prediction result of each operating module, and determining the collection cycle of feedback information to regularly collect the collection result feedback information, classify and organize the collected feedback information, obtain an update list and regularly update the visual monitoring interface, thereby improving the user's experience of human-computer interaction.
[0129] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A fuel intelligent sensing monitoring system based on human-computer interaction, characterized in that: include: Data acquisition module: used to establish a fuel sensing monitoring system, obtain log data of the fuel intelligent sensing monitoring system, and extract data from the log data to obtain real-time data of the fuel to be monitored; Data processing module: used to process the real-time data of the monitored fuel to obtain valid data; Model building module: used to obtain fuel property data of corresponding fuel types in the fuel depot, determine fuel characteristic data based on the fuel property data, and input the data into the neural network model for data training to obtain a fuel prediction model; Model optimization module: used to input valid data into the fuel prediction model to predict the fuel status, build a visual monitoring interface to display the prediction results, obtain feedback information on the display results, and update the visual monitoring interface based on the feedback information; The model building module also includes: A first training unit, used for inputting different combustion stages of the fuel into the neural network model to perform a first main training; At the same time, the fuel states at different combustion stages are input into the neural network model for first auxiliary training; A second training unit is used to input the heat generated by the unit weight of the fuel at different combustion stages into the neural network model to perform a second main training; At the same time, the combustion time and combustion temperature of the unit weight of fuel at different combustion stages are input into the neural network model for second auxiliary training; A model building unit, used for obtaining a fuel prediction model after all training of the neural network model is completed; Among them, the model optimization module includes: A state prediction unit, used for inputting corresponding valid data of each operation module into the fuel prediction model to perform a first state prediction, and obtain a first operation result of each operation module; Inputting valid data of the fuel intelligent sensing monitoring system into the fuel prediction model to perform a second state prediction, and obtaining a second operation result of the fuel intelligent sensing monitoring system; a similarity analysis unit, configured to obtain a second sub-operation result of each operation module in the second operation result, and perform a similarity analysis between each second sub-operation result and the corresponding first operation result; ;in, is the similarity between the first operation result of the kth operation module and the corresponding second sub-operation result, n is the total number of fuel features in each operation module, is the i-th fuel feature in the k-th operating module, is the characteristic weight of the i-th fuel characteristic, is the number of data collections per unit time in the kth running module, is the first running result of the i-th fuel feature of the k-th running module collected at the jth time, The second operation sub-result is obtained by collecting the i-th fuel characteristic of the k-th operation module at the j-th time; A state determination unit is used to determine the operation state of the corresponding operation module based on the similarity analysis result. If the similarity exceeds a specified threshold, the operation state of the corresponding operation module is deemed to be normal operation, and a similarity level based on the similarity is obtained to determine an optimization plan; A result optimization unit, configured to perform a first optimization on the second operation result based on the first operation result according to the optimization scheme to obtain a first optimization result; Otherwise, it is determined that the corresponding operation module has an operation fault, the corresponding operation module is cleared of the fault, and data collection and data processing are performed again to obtain new independent valid data; Inputting the new independent valid data into the fuel prediction model to perform a third state prediction, and obtaining a third operation result of the fault operation module; A similarity analysis is performed on each third operation result and the corresponding second operation sub-result, and a second optimization is performed on the first optimization result to obtain a prediction result.
2. The fuel intelligent sensing monitoring system based on human-computer interaction according to claim 1 is characterized in that: Data acquisition module, including: a sensor determination unit, configured to determine a fuel type of the fuel to be monitored, and determine sensor models of different types of sensors based on the fuel type; Determine the fuel operation system of the fuel to be monitored, obtain different operation modules of the fuel operation system, and determine the type and quantity of sensors required for each operation module; The system establishment unit is used to install corresponding types of sensors according to the number of sensors to the corresponding operation modules of the fuel operation system to obtain the fuel intelligent sensing system.
3. The fuel intelligent sensing monitoring system based on human-computer interaction according to claim 1 is characterized in that: The data acquisition module also includes: A data parsing unit, used to obtain log data of the fuel intelligent sensing monitoring system, and parse the log data based on the corresponding communication protocol to obtain corresponding information of each log data; The data extraction unit is used to extract information from the corresponding information of all log data to obtain information related to the fuel to be monitored, and use the corresponding data of the related information as the real-time data of the fuel to be monitored.
4. The fuel intelligent sensing monitoring system based on human-computer interaction according to claim 3 is characterized in that: Data processing module, including: A data cleaning unit, used to determine the sensor type of the sensor corresponding to each piece of real-time data; Acquire data cleaning methods corresponding to different types of sensors, and perform data cleaning on corresponding data based on the data cleaning methods; The data fusion unit classifies the cleaned real-time data according to the operation module, and performs data fusion according to the classification results to obtain separate valid data for each operation module; Each individual valid data is fused twice to obtain the valid data of the fuel intelligent sensing monitoring system.
5. The fuel intelligent sensing monitoring system based on human-computer interaction according to claim 1 is characterized in that: Model building modules, including: a fuel characteristic determination unit, configured to obtain fuel characteristic data corresponding to the fuel type based on a fuel reservoir, and determine the fuel characteristic based on the fuel characteristic data; Among them, the fuel characteristics include: different combustion stages of the fuel, the fuel state at different combustion stages, the combustion time of a unit weight of fuel at different combustion stages, the combustion temperature and the heat generated.
6. The fuel intelligent sensing monitoring system based on human-computer interaction according to claim 1 is characterized in that: The model optimization module also includes: A prediction display unit, used to construct a visual monitoring interface to display the first operation result and prediction result of each operation module; A feedback collection unit, used to determine a collection period for collection result feedback information, and to regularly collect collection result feedback information according to the collection period; A feedback sorting unit is used to classify the result feedback information according to different feedback directions, sort the feedback information of the same feedback direction, and obtain an update list; The periodic updating unit is used to update the visual monitoring interface according to the collection cycle of the collection result feedback information based on the update list.
