5results about How to "Avoid Conversion Errors" patented technology

The invention belongs to the technical field of electronic circuits, and provides a temperature detection circuit and a temperature compensation system. The temperature detection circuit comprises a voltage generation module, a temperature acquisition module, a selection module, a conversion module and a control module, the selection module is respectively connected with the voltage generation module, the temperature acquisition module, the conversion module and the control module, the control module is connected with the conversion module, and the control module and the conversion module are both connected with the power supply module. The voltage generation module generates a reference voltage; the temperature acquisition module acquires environment temperature; the control module outputs a first control signal to the selection module and outputs a first trigger signal to the conversion module; the selection module transmits the reference voltage to the conversion module according to the first control signal; the conversion module starts to work according to the first trigger signal and converts the reference voltage to obtain a first signal; the control module also determines a target error according to the first signal, and determines whether to control the selection module to transmit the environment temperature to the conversion module according to the target error.

This invention discloses a paragraph-level generated text detection method based on deep maximum mean difference (MDD) metric. The method includes: modeling generated text detection as a nonparametric two-sample test problem; designing a deep composite kernel to fuse shallow syntactic features and deep semantic features; training a deep kernel network with the goal of maximizing unbiased estimation of test power; in the testing phase, introducing a Wild Bootstrap mechanism to generate random weight sequences with autoregressive structures; performing a weighted quadratic transformation on the MMD kernel difference matrix; preserving paragraph sentence sequence correlation; accurately estimating the null distribution of the test statistic; and completing the detection by comparing the observed statistic with the test threshold. This method improves sensitivity to high-dimensional semantic differences, alleviates the problem of decreased test power caused by sequence correlation, and can efficiently distinguish between human and AI-generated text. It is suitable for generated text detection in scenarios such as news and academic writing.

The invention provides a data type conversion control method and device, equipment and a computer readable medium. Obtaining a plurality of data types corresponding to the modeling simulation tool and a numerical range of each data type, determining a numerical range sequence of the plurality of data types according to the numerical range of each data type, constructing a plurality of nodes of a type lattice according to the plurality of data types, constructing a connection line sequence of the type lattice according to the numerical range sequence, according to the method, a plurality of data types and numerical range sorting are utilized to construct a type lattice, the type lattice can reflect whether data type conversion can be carried out among the plurality of data types or not and the sequence of data type conversion, all modules can be traversed, and the data type corresponding to each module can be controlled according to the type lattice. Therefore, multiple data types can be smoothly subjected to data type conversion according to the type lattice, data type conversion errors are avoided, and data distortion or precision reduction is avoided.

The embodiment of the invention discloses a building model data conversion method, and relates to the technical field of data processing, and the building model data conversion method comprises the steps: obtaining a building data format identifier of each piece of to-be-converted building data in to-be-converted building model data; the to-be-converted building data comprises structured building data and unstructured building data. According to the building data format identifier, a target conversion function of each piece of building data to be converted is determined, and the target conversion function comprises a structured data conversion function and an unstructured data conversion function. And converting the structured building data through the structured data conversion function to obtain structured conversion data corresponding to the structured building data. And converting the unstructured building data through the unstructured data conversion function to obtain unstructured conversion data corresponding to the unstructured building data. According to the embodiment of the invention, the conversion precision and accuracy of data conversion can be improved.

The utility model discloses a fixed pollution source waste gas sampling port sealing device, including waste gas sampling gun, the fixed connection of the outer wall of the sampling pipe department of waste gas sampling gun has the threaded sleeve pole, the outer wall of threaded sleeve pole has seted up the thread, the outer wall threaded connection of threaded sleeve pole has the threaded sleeve piece, and the one side of inflatable ring is attached with the annular protruding department of threaded sleeve piece, the other side of inflatable ring is provided with the trachea, and the inner bag of trachea and inflatable ring are linked together. The device is applied to the sampling link of waste gas sampling gun, when the barrel department of waste gas sampling gun is inserted into the fixed pollution source pipeline, through the trachea of inflatable equipment connection inflatable ring one end, the input air of it makes it expand, seals the pipeline mouth, avoids the outside air and will flow into the pipeline through the gap reversely, improves the sealing effect is strong, avoids and will cause the discharge concentration calculation because of oxygen content conversion error, makes the actual emission of pollutant be underestimated, ensures the real and reliable of waste gas sampling data.