33results about How to "Achieving Adaptive Optimization" patented technology

The invention provides an urban rail transit station simulation system and method based on dynamic driving. The system comprises a data access and verification module, a basic model and knowledge base module, a behavior modeling and adaptive optimization module, a dynamic driving simulation control module and a dynamic simulation decision optimization module. The data access and verification module is used for acquiring passenger flow arrival and departure, get-on and get-off data and station service personnel on-duty and distribution data required by modeling and simulation online and offline, and transmitting the data to the corresponding modules as required after cleaning and verification; the basic model and knowledge base module stores related historical data and empirical models according to business themes, and constructs corresponding resource knowledge bases and empirical data model bases; the behavior modeling and adaptive optimization module is used for establishing a behavior control model; the dynamic driving simulation control module is used for injecting data of the data access and verification module and adaptively correcting a simulation data model; the dynamic simulation decision optimization module is used for simulation deduction, analysis, prediction and evaluation, and provides quantitative index support for station operation management.

The invention discloses a control method of an air-conditioner expansion valve, a control device and an air-conditioner. The method comprises the steps of conducting operation of air conditioning refrigeration, and obtaining a current indoor environment temperature, a current return air deviation of degree of superheat, and a current deviation variation amount of the degree of superheat; executing the following valve adjusting procedures if the current return air deviation of degree of superheat and the current deviation variation amount of the degree of superheat satisfy the conditions of degree of superheat and satisfy the condition of the indoor environment temperature at the same time, wherein the procedures specifically comprise the step of magnifying the current return air deviation of degree of superheat to be larger than zero and using the current return air deviation of degree of superheat as an actual return air deviation of degree of superheat, obtaining an actual deviation variation amount of the degree of superheat according to the actual return air deviation of degree of superheat, and then adjusting the opening degree of the expansion valve according to the actual return air deviation of degree of superheat and the actual return air deviation variation amount of the degree of superheat. According to the control method of the air-conditioner expansion valve, the control device and the air-conditioner, the self-adaptability of the adjustment of the expansion valve along with the changing of the temperature is achieved, and the operating performance of the system when the valve is adjusted in a degree of superheat and the comfort of the temperature adjustment can be improved.

The invention is a switched reluctance motor intelligent control system based on WIFI, which includes a server, a WIFI communication module, a mobile terminal, multiple bottom intelligent switched reluctance motor controllers and multiple switched reluctance motors. Each bottom intelligent switched reluctance motor controller is connected with one switched reluctance motor on the production line,and includes a monitoring module, a drive module, a control module and a WIFI module. The mobile terminal accesses and modifies various functions in the server, and can receive fault warning information. The switched reluctance motors feed back the operation parameters of the switched reluctance motors in real time during normal operation and send the data to the server periodically. The server optimizes and records the operation parameters of the switched reluctance motors according to a lot of received data and a reinforcement learning algorithm, and optimizes and updates the control parameters of the control modules when the collected parameters of the switched reluctance motors deviate from a normal value and reach a safety threshold, so as to better control the operation of the switched reluctance motors.

The invention discloses a remote-sensing image feature selection method based on an early land thematic map; the method comprises the steps of selecting a remote-sensing image partitioning algorithm and an image feature set of partitioning objects, and setting corresponding partitioning parameters and a feature extraction algorithm; segmenting the early land thematic map into a corresponding amount of layers according to plot classes, and extracting an influence strength map and a feature weight matrix to generate a feature weight distribution map; partitioning a remote-sensing image to be analyzed, and extracting image features of the partitioning objects; selecting an image classifier, modifying weight settings of classification features of the image classifier, and constructing a feature weighted classifier; setting feature weights for the objects in the partitioned maps, inputting image features and weights of the objects into the feature weighted classifier, and calculating land classes of the partitioning objects to generate a land thematic map. The method of the invention has the advantages that feature weights of local areas of a remote-sensing image can be adaptively optimized and adjusted, and remote-sensing image analysis precision can be improved.

The invention discloses an image denoising method and device and terminal equipment. The method comprises the steps that size information of a to-be-processed image is acquired; transforming referencefiltering parameters of a pre-obtained reference image based on a relative relation between the size information of the to-be-processed image and the size information of the pre-obtained reference image to obtain filtering parameters of the to-be-processed image; and denoising the to-be-processed image based on the filtering parameters of the to-be-processed image. In the embodiment of the specification, the invention also discloses a method for implementing the same, based on a relative relationship between the to-be-processed image and size information of a reference image obtained in advance, the filtering parameter of the reference image is transformed, and the transformed filtering parameter is used for denoising, so that adaptive optimization of the filtering parameter according tothe size information of the to-be-processed image is realized, the denoising effect of the image is ensured, and the operation is simplified.

The invention discloses an air-conditioner expansion valve control method and device based on the superheat degree. The method includes the steps that air-conditioner refrigeration running is carried out, and the current indoor environment temperature, the current compressor exhaust gas temperature, the current air return superheat degree deviation and the current air return superheat degree deviation variable quantity are obtained; if the current air return superheat degree deviation and the current air return superheat degree deviation variable quantity meet the superheat degree conditions and the temperature conditions are met at the same time, the following valve regulating processes are executed; and the current air return superheat degree deviation is increased to be larger than zero to serve as an actual air return superheat degree deviation, the current air return superheat degree deviation variable quantity is increased to be larger than zero to serve as an actual air return superheat degree deviation variable quantity, and the opening degree of an expansion valve is adjusted according to the actual air return superheat degree deviation and the actual air return superheat degree deviation variable quantity. By means of the air-conditioner expansion valve control method and device, the adaptivity of the expansion valve adjustment along with temperature changes is achieved, the running performance of the system and the comfort of temperature adjustment during superheat degree valve regulation are improved.

The invention discloses a remote sensing image feature selection method based on the previous land thematic map, comprising the following steps: selecting a remote sensing image segmentation algorithm and an image feature set of the segmentation object, and setting corresponding segmentation parameters and feature extraction algorithms; The plot category divides the previous land thematic map into a corresponding number of layers, and extracts the influence intensity map and feature weight matrix to generate a feature weight distribution map; segment the remote sensing image to be analyzed, and extract the image features of the segmented object; select a An image classifier, and improve the weight setting of its classification features, construct a feature weighted classifier; set feature weights for the objects in the segmentation map according to the feature weight distribution map, and input its image features and weights into the feature weighted classifier , calculate the land category of the segmented object, and generate the land thematic map. The invention can realize the self-adaptive optimization and adjustment of the feature weight of the local region of the remote sensing image, and can improve the analysis precision of the remote sensing image.

The invention relates to a rule-based hybrid energy storage system load adaptive real-time energy management system, including a hybrid energy storage system model module (1), a rule-based three-mode power distribution strategy module (2) and a load condition adaptive Parameter determination module (3). Among them, the hybrid energy storage system model module (1) includes the system topology, lithium battery and super capacitor model; the rule-based three-mode power distribution strategy module (2) includes three working modes and protection mechanisms; load condition self-adaptation The parameter determination module (3) constructs intermediate variables, then determines the relationship between parameters and intermediate variables and summarizes them into empirical formula expressions, and realizes high-efficiency and high-quality power distribution under unknown working conditions through iterative optimization in online use.

The invention provides a mesh distortion optimization algorithm based on discrete attribute data, comprising the steps of: 1, adopting a seed point method to perform region searching aiming at one discrete attribute value of a mesh to generate a region mesh set; 2, tracking the boundary of the region to generate a region mesh boundary segment set, analyzing the adjacency relations of the segments according to segment coordinates, sequentially connecting the segments end to end, tracking the region mesh boundary to form a polygonal boundary set, and removing the inner boundary by an area sorting method; 3, adopting a five point method to smoothen the curve to form a smooth region boundary under the premise of using the region area as the weight to restrain the smooth movement of the shared region boundary; and 4, adjusting mesh angular points to the corresponding smooth polygon to self-adaptively optimize the mesh angular points. The mesh distortion optimization algorithm based on discrete attribute data implements the self-adaptive optimization of the original mesh angular points, and can be used for solving the problem that the boundary of the model mesh region is saw-toothed and unsmooth.

The embodiment of the invention discloses an audio injection updating method, a control device and an audio injection extractor hood, the updating method is applied to audio injection updating of the extractor hood in the use stage, and the updating method comprises the following steps: acquiring working noise of the extractor hood under different gears when the audio injection is started; comparing the annoyance degree of the working noise with the annoyance degree of the last working noise, determining a final regulation and control sound according to a difference value between the annoyance degree of the working noise and the annoyance degree of the last working noise, and performing audio injection updating on the last regulation and control sound of the range hood under different gears by using the final regulation and control sound. The self-checking configuration updating of the audio injection function is completed, self-adaptive optimization of the regulation sound is realized, the audio injection function is played, the hearing feeling of a user is effectively improved, the service life of the audio injection function is prolonged, and the effect of the audio injection function is perfectly played.

The invention discloses a phased array antenna directional pattern performance parameter optimization method and an electronic device. The method includes the following steps that: the performance parameter of the directional pattern of a phased array antenna is determined; an objective function corresponding to the performance parameter of the directional pattern of the phased array antenna is determined; the attenuation value of the T / R component of the phased array antenna is obtained, and the change value of the objective function corresponding to the attenuation value is obtained; and an optimal amplitude weighted iteration direction is determined based on the change value of the objective function. With the method provided by the invention adopted, the technical problem that the directional pattern performance parameter of an active phased array antenna in the prior art does not satisfy actual demands can be solved.

The present invention discloses a double-factor adaptive judgment system of telecontrol uplink communication enabling time. The system includes a data object model defining module, a database real-time maintenance module and an uplink channel protocol module; the data object model defining module is used for adding a database construction preparation attribute in the initialization of a data object model; the database real-time maintenance module is responsible for maintaining the database construction preparation attribute of the data object; and the uplink channel protocol module is used for scanning the database construction preparation attributes of all related data objects at a current channel, performing overtime calculation according to the uplink communication enabling time threshold value of the channel, and deciding whether to enable uplink communication. With the system adopted, problems such as low efficiency, low adaptability and requirements for extra configuration work of a telecontrol uplink communication device can be solved. The invention also discloses a double-factor adaptive judgment method of telecontrol uplink communication enabling time.

The invention discloses a multi-dimensional dynamic interweaving device for multi-antenna scattering communication, which belongs to the technical field of wireless communication and scattering communication. The present invention performs adaptive coding modulation, interleaving, space mapping, frequency conversion, and amplification at the sending end to realize the conversion from baseband channel to analog signal; The proportional relationship and distribution relationship between the channel quality and SNR of each channel can be obtained; then the dynamic interleaving rule planner can be used to obtain the interleaving / deinterleaving matrix, and finally the optimization of the interleaving and deinterleaving rules can be realized, which can increase the bit error rate by 2 order of magnitude to achieve a good transmission effect. The invention can be used for high-speed information transmission under the fading channel beyond the line of sight, and is especially suitable for a multi-antenna scattering communication system with multiple transmission and multiple reception.

The invention provides a high-frequency CRM Boost PFC converter for adaptively optimizing input current THD, which belongs to the field of AC-DC power conversion. It reduces the input current THD based on the variable conduction time control method: the digital look-up table method is used to generate the modulated wave signal, and the variable conduction time control is realized through the preset conduction time data table to solve the problem of poor control accuracy of the analog control method and digital real-time calculation The cost of the method is high; other circuits except the modulation wave signal generation circuit are realized by analog to overcome the limitation of the digital controller’s response speed to the MHz high-frequency signal and reduce the performance requirements of the digital controller; through the data sheet The limited data preset in the system can be converted, and the on-time control can be changed without output power detection to follow the input voltage and output power for adaptive adjustment, so as to adaptively optimize the input current THD and significantly reduce the system cost.

The present invention is a WIFI-based switched reluctance motor intelligent control system. The system includes a server, a WIFI communication module, a mobile terminal, a plurality of underlying intelligent switched reluctance motor controllers, and a plurality of switched reluctance motors. Each underlying intelligent The switched reluctance motor controller is connected to a switched reluctance motor on the production line, including a monitoring module, a driving module, a control module, and a WIFI module; the mobile terminal can access and modify various functions in the server, and can receive fault warning information . Multiple switched reluctance motors feed back the operating parameters of switched reluctance motors in real time during normal operation, and periodically send the data to the server; the server optimizes the control parameters of switched reluctance motors using a large amount of received data combined with reinforcement learning algorithms. Record, and optimize and update the control parameters of the control module when the collected switched reluctance motor parameters deviate from the normal value and reach the safety threshold, so as to better control the operation of the switched reluctance motor.

The invention discloses a control method of an air-conditioner expansion valve, a control device and an air-conditioner. The method comprises the steps of conducting operation of air conditioning refrigeration, and obtaining a current indoor environment temperature, a current return air deviation of degree of superheat, and a current deviation variation amount of the degree of superheat; executing the following valve adjusting procedures if the current return air deviation of degree of superheat and the current deviation variation amount of the degree of superheat satisfy the conditions of degree of superheat and satisfy the condition of the indoor environment temperature at the same time, wherein the procedures specifically comprise the step of magnifying the current return air deviation of degree of superheat to be larger than zero and using the current return air deviation of degree of superheat as an actual return air deviation of degree of superheat, obtaining an actual deviation variation amount of the degree of superheat according to the actual return air deviation of degree of superheat, and then adjusting the opening degree of the expansion valve according to the actual return air deviation of degree of superheat and the actual return air deviation variation amount of the degree of superheat. According to the control method of the air-conditioner expansion valve, the control device and the air-conditioner, the self-adaptability of the adjustment of the expansion valve along with the changing of the temperature is achieved, and the operating performance of the system when the valve is adjusted in a degree of superheat and the comfort of the temperature adjustment can be improved.

The invention discloses a structural damage identification method and system based on a deep extreme learning machine, and the method combines an advanced signal processing technology, the deep extreme learning machine and a particle swarm optimization algorithm, and builds a structural damage identification method which can achieve the self-adaptive hyper-parameter optimization, is high in anti-noise performance and is high in damage identification robustness. Advanced signal processing technology Hilbert-Huang transform is adopted to reduce the influence of measurement noise on vibration signals, and the noise immunity of a damage identification algorithm is improved. And a particle swarm optimization algorithm is adopted to optimize hyper-parameters based on a deep extreme learning machine, so that the accuracy of structural damage identification is improved. And finally, the performance of the damage identification algorithm is tested by using ten-fold cross experiment verification, so that the robustness of the damage identification algorithm is ensured. Meanwhile, the structure damage data monitored on line is continuously used for updating the structure damage identification data set, so that the damage identification precision is continuously improved.