11 results about "Thermal perception" patented technology

The invention relates to the technical field of hot melting salt storage, and discloses a photo-thermal system mirror surface micro-angle linkage adjustment method based on thermal perception, which breaks through the limitation of traditional single data monitoring and experience judgment through multi-source data fusion and triple physical constraint decoupling, and remarkably improves the fault diagnosis accuracy. And secondly, the problem of poor model timeliness caused by traditional static partitioning and long-period updating is solved through dynamic partitioning deployment and real-time contribution identification, so that the contribution degree matrix can be updated in real time along with the mirror surface state. Accurate tracking of parameters of different time scales and safe rehearsal of a control strategy are achieved through parameter hierarchical updating and digital twinborn verification, and parameter aliasing and optimization failure caused by a traditional single updating strategy are avoided.

The invention discloses a three-dimensional chip thermal perception optimization method based on tensor data flow, and relates to the field of integrated circuits, and the method comprises the steps: firstly, starting from the angle of tensor data flow, providing a collaborative thermal optimization process combining a network, an architecture, physics and packaging; then, based on a quantitative modeling tensor data flow energy optimization method, automatically identifying the most energy-saving tensor data flow under given three-dimensional chip constraints by using a heuristic optimization method; and finally, providing a thermal sensing three-dimensional architecture spatial layout method, simplifying the spatial layout problem of the SRAM macro block into an affine mapping optimization problem, and optimizing an affine mapping relationship through a heuristic optimization method to obtain the spatial layout with optimal thermal performance. According to the method, the tensor calculation characteristics are optimized in a targeted manner, so that the three-dimensional stacked SRAM chip meets the thermal reliability requirement in a high-density integration scene.

Embodiments of the present application provide an air conditioner control method, a vehicle and electronic equipment, which comprises the following steps: obtaining an initial physiological characteristic sequence and an initial environment temperature characteristic sequence according to a preset time cycle; constructing an individual dynamic thermal comfort evaluation model based on the initial physiological characteristic sequence and the initial environment temperature characteristic sequence which are synchronized by a time stamp; calculating a real-time thermal comfort index of a target user by the individual dynamic thermal comfort evaluation model; determining a target control parameter for the air conditioner based on the real-time thermal comfort index, an instant physiological characteristic sequence and an instant environment temperature characteristic sequence which are synchronized by the time stamp; generating a control command for the air conditioner by using the target control parameter, and controlling the air conditioner based on the control command. Thus, by time sequence synchronization modeling of physiological and environmental characteristics, the hysteresis effect of human physiological feedback can be effectively eliminated, and precise dynamic matching of the air conditioner control parameter and the passenger transient thermal perception demand can be realized.

The invention relates to a user-side load-oriented optimal scheduling method, which belongs to the technical field of energy scheduling and intelligent power grids, and comprises the following steps of: constructing a household micro-grid model containing electric equipment and HVAC, establishing an HVAC thermodynamic model based on building heat balance, describing room temperature change in combination with meteorological and building parameters, and establishing an HVAC thermodynamic model based on the HVAC thermodynamic model. Classification modeling of the rigid load and the flexible load is completed; a multi-objective optimization framework with thermal comfort, power consumption behavior deviation and economy as objectives is constructed, human body thermal perception is quantified by adopting a PMV index based on a Fanger model, and dynamic coupling of scheduling and comfort is realized; hVAC scheduling is modeled as MDP, and double TD3 is introduced to realize continuous power and mode control; an integer programming model is constructed for a flexible uninterrupted load, and a starting time slot is optimized through a rolling time domain strategy, so that overall load coordinated scheduling is realized. According to the invention, the resource utilization efficiency and the load scheduling flexibility of the household micro-grid are obviously improved.

The application discloses an indoor body temperature sensing detection device and method, and relates to the technical field of indoor environment monitoring. The device comprises a sensing and collecting module, a central processing module and a terminal module. The sensing and collecting module comprises an air dry bulb temperature sensor, a relative humidity sensor, a breeze speed sensor and a black ball radiation thermometer. The parallel sensing array composed of the air dry bulb temperature sensor, the relative humidity sensor, the breeze speed sensor and the black ball radiation thermometer is arranged to synchronously acquire multi-dimensional basic physical parameters, thereby overcoming the defect that the traditional device only relies on single temperature and humidity, which causes the detection value to be disengaged from the real thermal perception of the human body. The physical isolation structure corresponding to the independent signal transmission cable of each sensor is welded, thereby avoiding channel crosstalk during the concurrent transmission of multiple signals and ensuring the signal integrity of the underlying original physical data.

The invention relates to the technical field of automobile intelligent control, and discloses an in-vehicle temperature adjusting method, device and equipment and a readable storage medium, and the method comprises the steps that the actual air outlet temperature and the actual air outlet flow of each air outlet in a vehicle are calculated according to current environment temperature control parameters of the vehicle; on the basis of the thermal perception prediction model and the thermal perception correction model, according to the actual air outlet temperature, the actual air outlet flow and the environment temperature control parameters, local thermal perception values of the multiple body areas in the vehicle are determined, and according to the multiple local thermal perception values, the overall thermal perception value in the vehicle is determined; and the target air outlet temperature is calculated according to the overall heat sensing value, the working state of an air conditioner actuator is controlled according to the target air outlet temperature, and the working state of a local actuator is controlled according to the local heat sensing value. The thermal comfort state of the cabin environment can be accurately evaluated by taking the real feeling of passengers as a reference, so that the in-vehicle temperature actuator is dynamically adjusted to adapt to the complex and changeable in-vehicle and out-vehicle environments, and the overall experience of the passengers is improved.

The application discloses a kind of based on three-dimensional form constraint's thermal exposure field identification method, device and medium, the method includes the pre-processing of target city's multi-source basic data, according to the grid unit corresponding to the target city of processed multi-source basic data is constructed;According to the digital surface model of fusion topography, building and vegetation, the sky view factor of pedestrian height and three-dimensional shelter angle are calculated per pixel, to construct three-dimensional radiation geometric relationship;According to grid unit and three-dimensional radiation geometric relationship generates three-dimensional city form constraint field;According to three-dimensional city form constraint field and the solar position parameter of current time generates pedestrian layer thermal exposure field.Based on grid unit, rely on the form constraint field generated by the digital surface model of fusion topography, building and vegetation, combine solar position parameter to reconstruct pedestrian layer thermal exposure field, consider the radiation interaction of building, vegetation, topography, can solve the defect that two-dimensional remote sensing evaluation microcosmic thermal heterogeneity is lost, and with human thermal perception deviation is big.

The invention relates to a steel rail crack three-dimensional reconstruction and parameter quantification method based on TA-LNO. The method comprises the following steps: acquiring a steel rail surface temperature sequence through a multi-frequency eddy current pulse thermal imaging technology; inverting three-dimensional heat source distribution in the steel rail from the surface temperature sequence by using a thermal perception latent neural operator model; processing the three-dimensional heat source distribution to generate a three-dimensional defect point cloud; and carrying out feature extraction and regression calculation on the three-dimensional defect point cloud by adopting a point cloud feature Transform network to obtain geometric parameters of the steel rail crack, and the geometric parameters comprise a crack angle, a crack length and a crack depth. According to the method, multi-frequency information is fused to overcome the skin effect, thermal image inverse problem morbidity is reduced through TA-LNO, submillimeter-level parameter quantification is achieved through point cloud Transform, the problems that in the prior art, only two-dimensional thermal images are relied on, precision is low, and generalization is poor are solved, and reliable data support is provided for steel rail safety assessment.

The application discloses a cabinet thermal environment hierarchical regulation method and system based on channel multi-point thermal perception, and belongs to the technical field of data center thermal environment perception and operation regulation, and comprises the following contents: a plurality of temperature measuring points are arranged in cold and hot channels, multi-point temperature data of the cold and hot channels and time series thereof are collected and pretreated; a cabinet thermal environment equivalent state vector is constructed; short-term disturbance features, medium-term cumulative features and long-term offset features are extracted from the cabinet thermal environment equivalent state vector in a short time scale window, a medium time scale window and a long time scale window; a hierarchical mutual exclusion judgment criterion is used to uniquely judge the current thermal environment state as an offset type, a cumulative type or a disturbance type; end cooling device regulation instructions are generated for different thermal environment state levels; and the channel temperature response is continuously evaluated after regulation execution. The application adopts the above method to realize in-situ real-time multi-time scale thermal environment state perception and hierarchical fine regulation.

The application discloses an infrared thermal imaging diabetic foot early risk assessment system based on biological heat conduction constraints and belongs to the technical field of artificial intelligence auxiliary diagnosis, and comprises the following modules: a data acquisition module used for acquiring a foot bottom pseudo-color infrared image and an original temperature matrix; a thermal perception region segmentation module used for outputting a pseudo-color infrared image containing only a foot region and a foot bottom temperature matrix; a feature extraction module used for extracting temperature features, physical parameter features and high-dimensional thermal texture features; a double-channel adaptive feature fusion module used for adaptively fusing the temperature features, the physical parameter features and the thermal texture features; a risk assessment module used for performing classification reasoning based on the fused feature vectors and outputting a diabetic foot risk assessment grade; and a man-machine interaction module used for showing a risk assessment reasoning process and results to a user. According to the application, accurate foot segmentation can be realized without pre-processing of an original image, and the reliability and the automation degree of subsequent analysis are improved.

The invention discloses a thermal perception influence factor identification method for heterogeneous multi-modal data. The method comprises the following steps: S1, synchronously acquiring multi-dimensional heterogeneous data through a static experiment and a dynamic experiment; s2, preprocessing the collected multi-dimensional heterogeneous data; s3, constructing a thermal perception data set containing environment-physiology-behavior multi-dimensional features; s4, adopting an XGBoost model to perform classification prediction on thermal sensation, dividing the constructed multi-modal data set into a training set and a test set according to 8: 2, inputting model XGBoost training for a resting state and an active state, and verifying the effectiveness of the model in combination with a multi-dimensional index of model performance evaluation; and S5, introducing an SHAP method to analyze the contribution degree of each multi-dimensional feature, and respectively obtaining the influence degree of each multi-dimensional feature in a resting state and an active state on the thermal perception prediction result. According to the method, the problems that factor identification is not comprehensive enough and the contribution degree of each feature to a prediction result cannot be quantified due to single-dimension data acquisition of an existing environment data sensing model are solved.