9results about How to "Lower quality requirements" patented technology

An intelligent compensation method for coupling errors of a dynamic gravimeter comprises the following steps: selecting typical environment interference influencing the dynamic gravimeter in a complex measurement environment, and analyzing an influence mechanism of each typical environment interference on the measurement accuracy of the dynamic gravimeter; according to the influence mechanism under each typical environment interference, analyzing a coupling error influence mechanism between the typical environment interferences; based on the coupling error influence mechanism, constructing a coupling error comprehensive compensation model; adding the coupling effect loss into a data-driven loss function, and defining a model coefficient constrained by physical information; and performing model training by using the training data, and calculating the total loss of the verification set on the obtained model to obtain a final coupling error comprehensive compensation model. According to the invention, the coupling error of the dynamic gravimeter in a complex environment is compensated based on coupling error mechanism analysis and a physical constraint long-short-term memory model, and the measurement precision and environmental adaptability of the dynamic gravimeter under complex multi-factor interference conditions are improved.

The application discloses a kind of weakly supervised image semantic understanding methods based on multi-task learning, comprising the following steps: obtaining task missing image, constructing multi-level task sharing encoder, extracting high-level semantic information layer by layer, input corresponding decoder branch;Construct public space-task space feature mapping module, through the unaligned task fusion module and task interaction mapping module, update each subtask feature by mapping;Task adaptive feature update module is constructed, and multi-level iterative update unaligned task feature;Task adaptive weakly supervised image semantic understanding framework is constructed, model loss function is established, image data with task missing is input into model, and obtains multi-task prediction result such as semantic segmentation, depth estimation, surface normal estimation.The application is according to the data information of task label unaligned, through the mapping interaction of public space and task space, fully fuses unaligned task feature, iteratively generates high-quality multi-task prediction result, can effectively handle weakly supervised problem with task missing, and simultaneously improves each task prediction accuracy.

The utility model provides a pile film bag earth and rock cofferdam suitable for a high-permeable foundation. The pile film bag earth and rock cofferdam comprises a high-strength woven bag filled earth and rock cofferdam body, a high-strength composite geomembrane bagged earth and rock slope protection and anti-seepage blanket, a geomembrane anti-seepage inclined wall, an earth and rock filling cofferdam body, a high-strength composite geomembrane bagged earth and rock slope protection, a cofferdam crest mud stone pavement, a permeable layer and the like. The upstream upstream face adopts a double-insurance combination form of a high-strength composite geomembrane bagged soil and stone anti-seepage blanket, a high-strength composite geomembrane bagged soil and stone slope protection and a geomembrane anti-seepage inclined wall, and the good water retaining, anti-seepage, anti-impact and anti-frost heaving effects are achieved. And meanwhile, the foundation adopts the steel sheet pile anti-seepage core wall, so that the foundation seepage stability problem of the cofferdam constructed on the high-permeability foundation can be effectively solved. The anti-seepage structure can be applied to various wading engineering construction scenes and has the advantages of guaranteeing the anti-seepage effect, being safe in structure, saving investment, shortening the construction period, being environment-friendly and the like.

This invention relates to a method for non-explosive pyrolysis to enhance coal permeability and implement forced roof caving, belonging to the field of coal mine safety technology. The method involves injecting a pyrolysis agent into a borehole, which reacts, solidifies, and expands, generating expansion pressure that causes the coal or rock mass to crack, forming numerous secondary fractures. This improves the permeability of the coal or reduces the strength of the roof strata. Application scenarios include permeability enhancement of inclined coal seams, and permeability enhancement and forced roof caving of near-horizontal or horizontal coal seams. This invention utilizes the property of the pyrolysis agent itself—reaction, solidification, and expansion—to create secondary fractures in the coal (rock) mass, significantly improving coal permeability or reducing rock strength. It involves minimal electromechanical equipment and offers significant technical and economic benefits. For coal permeability enhancement operations in longwall mining faces, it strictly distinguishes between inclined, gently inclined, or near-horizontal coal seams, and differentiates the arrangement of process boreholes and gas drainage boreholes. The engineering workload is low, eliminating the need for blasting operations, and safety is extremely high.

The invention relates to the technical field of slag treatment, in particular to a method for preparing high-purity silicon dioxide by using silicon slag, which comprises the following steps: S1, dissolving slag; s2, removing impurities; s3, separating out a product; the method has the beneficial effects of no need of a large amount of acid and alkali and environmental friendliness; quality requirements of raw materials are reduced, the raw materials are cheap and easily available, and production cost is low; the impurity removal step is designed, so that the product quality is improved; the method is simple in steps, adopts a two-step ammoniation step, and avoids a large number of filtering, washing and drying steps.

The invention relates to a process and a system for converting municipal solid wastes into sustainable aviation kerosene components. The system comprises a cracking device, a condensing device, a solid waste treatment device, a gas purification device, an extraction device, a hydrogenation reaction device and a rectification device. According to the system and the process provided by the invention, the municipal solid wastes are converted into high-added-value products, the environmental benefits and the economic benefits are improved, and meanwhile, the conversion efficiency of the municipal solid wastes into sustainable aviation kerosene components is also improved.

This invention discloses a method for producing feed-grade granular high water-soluble phosphorus MDCP, comprising the following steps: 1) mixing phosphoric acid with a portion of calcium carbonate emulsion; 2) the premixed slurry entering a stirred tank reactor for deep reaction neutralization; 3) the reaction slurry being fed into a second tank reactor for secondary neutralization reaction; 4) the secondary neutralization reaction being stirred and granulated to generate granular semi-finished product; 5) drying to obtain the material; 6) the dried material being screened. This invention premixes purified phosphoric acid with a portion of calcium carbonate emulsion and then fully stirs and reacts it in a reaction tank to generate a large amount of high water-soluble calcium dihydrogen phosphate (MCP) slurry. The slurry forms a mixture with the remaining unreacted portion of phosphoric acid, which then undergoes a secondary reaction with the remaining calcium carbonate emulsion to generate dicalcium phosphate (MDCP) slurry. After granulation, drying, screening, and cooling, a qualified granular MDCP product with high water-soluble phosphorus content is obtained.

The invention relates to the technical field of reservoir transformation and yield increase, in particular to an oil well yield unsupervised prediction method based on an S2S-F-A model, which comprises the following steps: inputting an oil well yield time sequence in a certain time period into an LSTM network for coding, and taking the coded oil well yield time sequence as an output vector of the LSTM network; compressing an output vector of the LSTM network into a weight vector by using an attention mechanism module; reconstructing the weight vector to obtain a reconstructed yield sequence; and inputting the reconstructed yield sequence into the S2S-F-A yield model, and outputting yield prediction data of the next time period. According to the method, the information of the unlabeled yield data set can be effectively utilized under the condition that the labeled data set is scarce, and the yield after the oil well fracturing measure is predicted. According to the method, the yield of the oil well after refracturing can still be accurately predicted under the condition that the data missing rate exceeds 20%, the workload of oil reservoir engineers is effectively reduced, and an important basis is provided for accurate selection of refracturing potential wells and economic evaluation.