42results about How to "Realize dynamic prediction" patented technology

The invention relates to a method for predicting mine inflow and belongs to the field of mine mining, in particular to a method for predicting mine inflow based on a dynamic change of a permeability coefficient along with layered mining. The space of the mine excavation engineering is organically combined with the time distribution characteristics according to layered mining conditions, the influence of overburden failures on the permeability coefficient of an overlying aquifer caused by the mining is taken into consideration, the dynamic adjustment of the permeability coefficient matching with actual excavation activities in the research area is realized; a relationship between the permeability coefficient and the groundwater numerical model is built, the adjusted parameters can be imported into the simulation model in real time, the dynamic prediction of the water inflow along with coal mining is realized, and the prediction precision of the water inflow is improved.

The invention discloses a COVID-19 real-time risk prediction method, which comprises the following steps of: acquiring epidemic risk evaluation data and social vulnerability evaluation data; establishing an infection number attenuation equation and an epidemic situation propagation dynamics model; predicting epidemic risk evaluation data according to an infection number attenuation equation and anepidemic propagation dynamics model; establishing an entropy-analytic hierarchy process combined model; carrying out epidemic risk prediction according to the entropy-analytical hierarchy combined model, the epidemic risk evaluation data and the social vulnerability evaluation data; and drawing an epidemic risk prediction grade map. The method provides support for later epidemic prevention and control decisions, and can be used for emergency prediction of risks in new epidemic situation occurrence periods or other sudden public safety events in the future.

The invention discloses an intelligent decision-making method for controlling the pollutant discharge total amount of a coal-fired power plant. An initial distribution target of the discharge total amount is decomposed, the actual discharge amount is dynamically tracked, the capacity of the pollutant discharge total amount of a coal-fired unit is analyzed, a multi-target linear programming theory is applied, power generation enterprises and relative government sectors are guided to timely conduct and execute dynamic total amount control strategies such as internal regulation and regional balancing of the discharge total amount, and the target maximization of the power generation enterprises, regional environmental benefits and economic benefits is achieved.

Embodiments of the invention disclose an engineering project cost prediction method, apparatus and device, and a readable storage medium. The method comprises the steps of obtaining a BIM 5D management platform for calculating an engineering amount of an engineering project; inputting unit prices of labors, materials and machineries of a current market to a pre-built cost prediction model, therebyobtaining unit prices of the labors, the materials and the machineries when the engineering project is in different implementation stages, wherein the cost prediction model is obtained by learning and training sample data of multiple historical engineering projects by utilizing a rote learning algorithm; and according to the engineering amount and the unit prices of the labors, the materials andthe machineries in the corresponding implementation stages, calculating the cost of the engineering project, thereby realizing dynamic prediction of the cost of the engineering project. According to the technical scheme provided by the method, when the cost of the engineering project is calculated, the changes of the unit prices of the labors, the materials and the machineries along with the timeare taken into account; the deficiency of existing manufacturing cost prediction is made up for; the cost prediction accuracy and reliability of the engineering project are improved; and accurate costcontrol of the whole engineering project can be realized.

Aiming at the oil spillage accidents which possibly occur to ships conveying oil substances in inland river water bodies such as rivers and long and narrow reservoirs and the like, the invention provides a technology for predicting the oil film diffusion process after oil spillage occurs by utilizing combination of a dynamic model and an oil spilling diffusion model on the basis of a geographic information system (GIS). The technology has the characteristics that the response speed is fast and the cost is low, and has very important realistic significance for shipping management and environmental protection of the inland river water bodies.

The invention discloses a hydro-fracture propping agent parameter optimization method. The method comprises the following steps of: S1, establishing a physical model for reducing the real size of a propping agent; S2, applying closing pressure on the surfaces of an upper rock stratum and a lower rock stratum of the model, wherein the difference between average heights of fracture surface particles of the upper rock stratum and the lower rock stratum is a fracture closing width w; S3, carrying out flow field grid discretization on a filling layer to ensure that the propping agent is wrapped by a flow field, and setting a viscosity and a density of a fluid and pressure of the fluid at two ends of the flow field; S4, calculating a total flow q of the flow field; S5, calculating a penetration rate and a flow conductivity; S6, changing parameters of the propping agent, and calculating and comparing the flow conductivities under different parameter combinations; and S7, carrying out optimizing by combining laboratory flow conductivity effect evaluations and on-site model selection experiments to obtain a final parameter combination. The method disclosed by the invention has the beneficial effects of being low in laboratory cost, being capable of testing flow conductivities of different types of propping agents, really reflecting the flow conductivities of formation fluids, and correctly and efficiently predicting the flow conductivities of propping agents under different geological conditions.

The invention discloses a machine learning-based elastic resource expansion method and system, and belongs to the technical field of cloud computing and the field of deep learning, and the method comprises the steps: calculating the minimum total resource amount required for completing a task by utilizing a regression model under the condition that the operation cut-off time limit td and the taskcalculation amount of the to-be-operated task are known; continuously collecting the current running state and the resource utilization rate of the task in the running process of the task, and inputting the minimum total resource amount, the current running state of the task, the resource utilization rate and the task calculation amount into a prediction model for prediction to obtain the task completion time Tc; if Tc is larger than td, calculating the minimum total resource amount enabling the final task completion time Tc 'to be smaller than td; and if the task is not completed, continuingcollection, and if the task is completed, stopping collection. According to the method, the minimum total resource amount is calculated through the regression model to ensure that the task can be completed on time, the completion time is predicted in the operation process, and when the completion time exceeds the operation cut-off time limit, elastic expansion of the calculation resources is automatically carried out.

The invention discloses a method for dynamic prediction of coal mine gas emission amount based on GIS. The method uses the GIS as a platform, and on the basis of coal mine gas geological dynamic analysis, combining with high negative pressure and low negative pressure gas extraction data of a working face, a different-source prediction method is used to dynamically predict coal-mass gas emission amount in front of a coal mine excavation working face. The method gives full consideration on influence of gas geologic dynamic variation and high and low negative pressure gas extraction on gas emission prediction, and the gas emission different-source prediction method and a gas geologic data model processed by rasterization by the GIS platform are organically combined, dynamic prediction of mine gas emission amount is realized, and prediction precision is improved, and the method is beneficial for prevention and cure of mine gas disasters.

The invention discloses a dynamic water environment management method. The method comprises the following steps: firstly, collecting and monitoring water environment related information and analyzing water quality and specific pollutants; calibrating / training a water body inlet boundary condition model; then calibrating or estimating the roughness of the hydrodynamic water quality model and the degradation coefficient of pollutants; forecasting the flow or water quality data on the inlet boundary of the research area by adopting the calibrated / trained boundary condition model in combination with other related information; and finally, utilizing the determined or estimated roughness of the hydrodynamic water quality model, the degradation coefficient of pollutants and a water body boundary condition forecast result, and combining with water system section data and latest monitored point source emission measured data to forecast the hydrodynamic water quality of the research area; carrying out water quality early warning and discharge capacity emission reduction feedback according to the forecast result. According to the method, a large amount of actually measured data and information are included, and dynamic monitoring and dynamic prediction of the water environment and more scientific dynamic water environment management are achieved.

The invention discloses a dynamic prediction method for a flow duration curve under a climatic change condition. The method comprises steps of analyzing the correlation degree of runoff and rainfall in good and bad years by adopting a linked list inspection method; extracting one-dimensional and two-dimensional principal components of the actually measured runoff volume by adopting an empirical orthogonal function, identifying the high correlation of rainfall and runoff in magnitude through fluctuation change characteristics of the principal components in runoff and rainfall good and bad years, and further establishing the correlation of rainfall-runoff good and bad changes; and predicting a flow comprehensive duration curve under a specified weather condition by taking a future runoff good and bad division ratio speculated based on a predicted future rainfall time sequence as an adjustment coefficient and based on flow duration curves of good-diameter and bad-diameter years. Accordingto the method, rainfall information of various climate conditions can be flexibly described by introducing different climate scenes, non-stationary change characteristics of flow duration curves of different drainage basins are identified, dynamic prediction of the flow duration curves is realized, and the method has very high practicability and wide applicability.

The invention relates to a one-day operation accounting system and method for a coal-fired power generation enterprise. The system comprises a data acquisition module, a cost calculation module, an income calculation module, a one-day operation accounting module and a display output module. The data acquisition module is used for acquiring power generation enterprise operation system information according to the actual situation of the coal-fired power generation enterprise and transmitting various types of data to the cost calculation module or the income calculation module; the cost calculation module is used for calculating the day degree, the real-time change cost data and the fixed cost data of the coal-fired power generation enterprise; the income calculation module is used for calculating daily income data, monthly income data and annual income data of the coal-fired power generation enterprise; the one-day operation accounting module is used for carrying out statistical accounting on data such as enterprise daily operation cost and profits and providing plant-level decision support; and the display output module is used for displaying the data calculated by the one-day operation accounting module. According to the invention, the total cost of kilowatt-hour electricity can be calculated in real time, so the lean production management and quality and efficiency improvement are realized.

The invention discloses a modeling method for an in-furnace ash deposition model based on direct simulation of Monte Carlo, and belongs to the field of in-furnace ash deposition predication of thermal power plants. The method comprises the following steps that 1, a computational fluid dynamics method is used for calculating a flow field near an ash deposition sampling point; 2, flow field conditions obtained through calculation are imported into a particle motion equation; 3, a direct simulation Monte Carlo method is adopted for calculating collisions among flying ash particles; 4, wall heat flux is provided through a sampling gun, then a flying ash adherence rate equation is solved, and the deposition process of the flying ash particles on a sampling face is calculated. The calculation result indicates that the constructed in-furnace ash deposition model can accurately predict the deposition rate of the flying ash particles in a furnace, detailed ash deposition inner layer flying ash particle size distribution is given, and dynamic prediction of the in-furnace ash deposition process is successfully achieved.

The invention discloses a vehicle positioning method and system based on roadside unit machine learning. The method comprises the steps that a vehicle sends traffic information to RSUb; a machine learning system collects the traffic information which is received by the RSUb and sent by the vehicle; steering of the vehicle at the first junction after passing the RSUb, RSUc which is going to be passed and the travelling track between RSUa and the RSUc are predicted sequentially. The system comprises a sending module and the machine learning system. Therefore, according to the vehicle positioningmethod and system based on machine learning, by means of cooperation of a roadside unit and the vehicle, the real-time traffic information is collected and uploaded to a database, the received real-time data is analyzed and processed by means of the machine learning system on the basis of the historical traffic information, dynamic prediction of the traffic state of a target vehicle is achieved,and then dynamic positioning of the target vehicle located in a communication blind district is achieved; under the condition of different traffic flows, the data delivery rate can be obviously increased, timely message delivery is guaranteed, and control overhead is reduced.

The invention relates to a self-rotation stabilized satellite attitude prediction method based on equal-inclination-angle attitude control. According to the method, by taking an initial attitude, a target attitude and an attitude control amount of a self-rotation stabilized satellite in a reference coordinate system as input, a satellite attitude is predicted, an Mercato graph coordinate is calculated, a sun angle, the north infrared outer chord width and the south infrared outer chord width are predicted, according to a calculated attitude control parameter (the attitude control amount) and agiven control mode, the attitude changing trend of the satellite attitude caused by the control is calculated, and meanwhile, the changing trends of the infrared outer chord wide angel and the sun angle due to the attitude change are calculated. The method is suitable for attitude prediction of the attitude control of a self-rotation stabilized satellite in an equal-inclination-angle precession manner.

The invention relates to a modular cold and hot water air conditioning unit intelligent joint control system and a control method. According to the system and method, temperature and humidity parameters of indoor and outdoor areas of an office building are acquired, precise group control is performed on the operating power of an air conditioning unit according to the cold (hot) air enthalpy valuesrequired by the office building areas and the work efficiency of the air conditioning unit, the optimal energy consumption operating mode is established by means of the unique power-saving intelligent control procedure based on acquired data including the operating temperature, pressure, flow rate and the like of the air conditioning unit as well as a series of parameters such as the height, orientation, materials, the thermal load, indoor and outdoor temperature and humidity conditions of the building and the user using habit, building load variation is automatically tracked, dynamic predicting, advanced adjustment and synchronous optimization are achieved, finally, the central air conditioning parameters of temperature, pressure, flow rate and the like are adjusted, the power curve is made to approach the power curve required by an actual load as close as possible under the condition that the requirements of an end system for the temperature, pressure, flow rate and the like are met, the energy-saving effect of air conditioners is pushed to the limit, and the purpose of system energy-saving optimization is achieved.

The invention discloses a method for dynamically predicting and calculating the productivity of a compact conglomerate reservoir. The method comprises the following steps: determining quantitative characterization parameters of the productivity of the compact conglomerate reservoir according to a well closing pressure test and production data; analyzing various parameters influencing the reservoirproductivity, and optimizing four main control key parameters; constructing a comprehensive quantitative index for indicating the reservoir productivity by utilizing the key parameters of the mastercontrol; and finally, considering the influence of reservoir measure transformation factors, establishing a liquid production index by utilizing the comprehensive quantitative index and the reservoirtransformation factors according to the reservoir types, and realizing dynamic prediction of the productivity. According to the invention, accurate and dynamic prediction of the tight conglomerate reservoir productivity is realized.

The invention discloses an intelligent coal mining working face gas prediction and equipment linkage safety guarantee system and method. The system comprises a working face key procedure equipment automatic regulation and control system and a working face multi-source gas emission dynamic prediction system. The working face key procedure equipment automatic regulation and control system is composed of a coal mining machine regulation and control unit, a coal caving support regulation and control unit, a conveyor regulation and control unit and a ventilation facility regulation and control unit. The working face multi-source gas emission dynamic prediction system is divided into a basic parameter input module, an equipment operation parameter input module, a gas emission prediction and parameter calculation module and a parameter calculation result output module. According to the method corresponding to the system, the working face gas emission amount can be predicted according to the equipment operation state parameters collected by the system and the return air flow gas emission actual measurement value, the equipment reasonable operation state parameters are calculated, and it isguaranteed that working face gas emission is in the expected stable state. The system and the method can be widely applied to intelligent coal mining working face gas emission control and gas prevention and control work.

The invention discloses a grapefruit disease and insect pest monitoring method based on the Internet of Things technology, and belongs to the field of grapefruit disease and insect pest monitoring. The grapefruit disease and insect pest monitoring method based on the Internet of Things technology predicts the grapefruit pest and disease damage occurrence trend through the steps of image preprocessing, denoising, feature extraction, SVM model construction and the like, grapefruit pest and disease damage is effectively monitored, and the method can be used through combination of image processingand SVM. Image processing technologies such as image filtering and feature extraction are used for carrying out label setting and scientific classification on features, the number of independent variables during pest and disease damage prediction is effectively reduced under the condition that the accuracy is not affected, and pest and disease damage dynamic prediction affecting grapefruit growthis achieved.

The invention discloses a flexible baseline dynamic prediction method based on a fiber grating sensor and a wing mode. According to the invention, the method comprises: employing a modal superpositionprinciple based on structural strain response to calculate structural displacement and a base line variable between child nodes, and then predicting a base line quantity change condition at a futuremoment by a base line dynamic model established based on flexible base line real-time data; employing the modal method to reduce the calculation amount and employing dynamic modeling to predict the flexible baseline in advance so as to solve the problem of real-time transfer alignment time delay caused by time consumption of resolving from the actual measurement dependent variable to the base linevariable, thereby facilitating the realization of high-precision pose real-time transfer alignment of child nodes. The method can be used for remarkably improving the real-time performance of transfer alignment and making up for the defects of previous baseline solution and estimation method research.

The invention discloses a conveyor belt longitudinal damage vibration sensing method based on infrared computer vision, and belongs to the technical field of computer vision. The technical problem to be solved is to provide the improvement of the conveyor belt longitudinal damage vibration sensing method based on infrared computer vision. The method for solving the technical problems comprises the following steps: arranging a high-speed camera above a mining conveying belt to collect tiny vibration images of the conveying belt in the normal state, the abrasion state, the scratching state and the tearing state; respectively training and testing vibration frequency signals of the conveying belt by adopting a convolutional neural network of a variable convolution kernel to obtain a preliminarily trained convolutional neural network; obtaining a further trained convolutional neural network model through transfer learning; and inputting the acquired image data into the convolutional neural network model, and outputting a conveyor belt damage diagnosis result according to the mining conveyor belt longitudinal damage information corresponding to the amplitudes of different wavebands. The method is applied to damage judgment of the conveying belt.

The invention discloses a thermal barrier coating service life measuring and calculating method. The method comprises the steps of obtaining historical use data of a user, component structure information of a thermal barrier coating and planned use working conditions of components; obtaining a life prediction model of the thermal barrier coating according to the obtained user historical use data, the component structure information of the thermal barrier coating and the planned use condition of the component; measuring a microstructure damage factor of the thermal barrier coating; correcting the fitting coefficient of the life prediction model through the measured microstructure damage factor of the thermal barrier coating, obtaining the corrected life prediction model, and obtaining the life of the thermal barrier coating. A thermal barrier coating nondestructive testing device is adopted to measure crack distribution and TGO layer key parameters of a service part, the measured data can be used for verifying and correcting a thermal barrier coating service life prediction algorithm and CFD boundary conditions, the accuracy of thermal barrier coating service life prediction is improved, and dynamic prediction of the thermal barrier coating service life in the part service period is achieved.

The invention discloses a method for predicting the variable time-space conductivity of coalbed methane well fracturing fractures, comprising the following steps: S1, collecting relevant parameters of the reservoir rock of the target coalbed gas well; The spatial distribution of capacity; S3. Establish the mechanism expression of the influence of pulverized coal deposition, proppant embedding and time on the conductivity of proppant-filled fractures; S4. According to the established conductivity change mechanism model, the calibrated spatially discretize the global diversion capacity to obtain the diversion capacity value of each discrete unit at different times; S5. Taking the calculation result of step S4 as the starting point, repeat steps S3-S4 until the calculation of the global diversion capacity is completed. The present invention overcomes the shortcomings of the existing laboratory experiments in predicting the conductivity, and after the fracturing of coalbed methane wells, the damage to the conductivity caused by coal powder deposition and proppant embedding is considered, and the time-varying space-time conductivity that can effectively predict the global proppant-filled fractures is achieved. flow capability.

The invention discloses a flexible baseline dynamic prediction method based on optical fiber grating sensors and airfoil modes. The baseline dynamic model established by real-time data predicts the change of the baseline quantity in the future; the modal method is used to reduce the calculation amount and dynamic modeling is used to predict the flexible baseline ahead of time to solve the real-time transfer error caused by the time-consuming calculation of the measured strain variable to the baseline quantity. The problem of quasi-time delay helps to achieve real-time transfer and alignment of sub-nodes with high-precision poses. The invention can be used to significantly improve the real-time performance of the transfer alignment, and make up for the deficiency of the previous research on the baseline solution and estimation method.

The invention discloses a method and system for expanding elastic resources based on machine learning, which belong to the field of cloud computing technology and the field of deep learning, including: d In the case of the calculation amount of the task, the regression model is used to calculate the minimum total amount of resources required to complete the task; the current running status and resource utilization of the task are continuously collected during the running of the task, and the minimum total amount of resources, the current running status of the task, and The resource utilization rate and task calculation amount are input into the prediction model for prediction, and the task completion time T is obtained c ; if T c >t d , then the calculation makes the completion time T of the final task c '<t d The minimum total amount of resources; if the task is not completed, the collection will continue, and if the task is completed, the collection will be stopped. The method of the invention calculates the minimum amount of resources through a regression model to ensure that tasks can be completed on time, predicts the completion time during operation, and automatically performs elastic expansion of computing resources when the completion time exceeds the operation deadline.

The invention discloses a forest fire prediction method and system, and the method and system collect the temperature data, humidity data, rainfall data, wind speed data and ground combustible moisture content data of a forest in real time, calculate a triangular fuzzy number of fire, judge the level of the fire, and achieves the dynamic prediction of forest fire. Meanwhile, the method and system also can input the triangular fuzzy numbers of season parameters, time parameters, population density parameters, road density parameters, population activity parameters, historical fire parameters and ground combustible type parameters, and bring environment factors and human factors into a fire prediction calculation model, and further improves the accuracy of fire prediction.

The invention discloses a hydro-fracture propping agent parameter optimization method. The method comprises the following steps of: S1, establishing a physical model for reducing the real size of a propping agent; S2, applying closing pressure on the surfaces of an upper rock stratum and a lower rock stratum of the model, wherein the difference between average heights of fracture surface particles of the upper rock stratum and the lower rock stratum is a fracture closing width w; S3, carrying out flow field grid discretization on a filling layer to ensure that the propping agent is wrapped by a flow field, and setting a viscosity and a density of a fluid and pressure of the fluid at two ends of the flow field; S4, calculating a total flow q of the flow field; S5, calculating a penetration rate and a flow conductivity; S6, changing parameters of the propping agent, and calculating and comparing the flow conductivities under different parameter combinations; and S7, carrying out optimizing by combining laboratory flow conductivity effect evaluations and on-site model selection experiments to obtain a final parameter combination. The method disclosed by the invention has the beneficial effects of being low in laboratory cost, being capable of testing flow conductivities of different types of propping agents, really reflecting the flow conductivities of formation fluids, and correctly and efficiently predicting the flow conductivities of propping agents under different geological conditions.

The invention discloses an intelligent coal mining working face gas prediction and equipment linkage safety guarantee system and method. The system comprises a working face key procedure equipment automatic regulation and control system and a working face multi-source gas emission dynamic prediction system. The working face key procedure equipment automatic regulation and control system is composed of a coal mining machine regulation and control unit, a coal caving support regulation and control unit, a conveyor regulation and control unit and a ventilation facility regulation and control unit. The working face multi-source gas emission dynamic prediction system is divided into a basic parameter input module, an equipment operation parameter input module, a gas emission prediction and parameter calculation module and a parameter calculation result output module. According to the method corresponding to the system, the working face gas emission amount can be predicted according to the equipment operation state parameters collected by the system and the return air flow gas emission actual measurement value, the equipment reasonable operation state parameters are calculated, and it isguaranteed that working face gas emission is in the expected stable state. The system and the method can be widely applied to intelligent coal mining working face gas emission control and gas prevention and control work.

The invention provides a dynamic prediction method for large raw material storage yard foundation deformation under graded preloading. The method includes the following steps of firstly, collecting and processing monitoring data of foundation deformation under preloading; secondly, establishing a finite element calculating model according to areal geology conditions and foundation processing requirements; thirdly, compiling a soil mass parameter inversion program through the combination with the Nelder-Mead algorithm of an exact penalty function; fourthly, conducting inversion analysis on sedimentation data through the inversion program to obtain post-surcharge-load soil mass parameters; fifthly, predicting the surcharge load foundation deformation amount of the next grade through the soil mass parameters obtained through inversion, and conducting surcharge load design of a raw material yard of the next grade on this basis. By means of the method, soil mass design parameters are unceasingly corrected through the real-time monitoring data in the surcharge load process, dynamic prediction of raw material storage yard foundation deformation under all grades of preloading is achieved, foundation is provided for foundation processing design and construction, the project investment cost is greatly saved, and high application and popularization value is achieved.

The invention relates to a mine water gushing amount prediction method, belongs to the field of mine roof water disaster prevention and control and water resource protection, and particularly relatesto a coal mine working face water gushing amount dynamic prediction method based on a multi-order power system model. The method mainly comprises the steps of constructing a mathematical model of theactual drainage condition water gushing process of the mining area mined working face roof water head, solving the mathematical model of the drainage-free condition water gushing process of the miningarea mined working face roof water head, analyzing main control factors of the working face push-mining roof water gushing process, and researching and developing a multi-order power system model forpredicting the water gushing amount in the working face push-mining process. Based on working face water gushing amount prediction software development and prediction of a multi-order power system model, accurate and dynamic prediction of the coal mine working face water gushing amount is achieved, and the maximum water gushing amount, the normal water gushing amount and the occurrence position most concerned in the field of coal mine water disaster prevention and control are obtained.