2results about How to "Improve the mismatch" patented technology

The utility model belongs to the technical field of thermal power equipment, and relates to a high-efficiency combined preheater system of a thermal power plant boiler, which comprises a tubular air preheater arranged at the tail part of the boiler and used for heating primary air and a rotary air preheater used for heating secondary air, flue gas inlets of the tubular air preheater and the rotary air preheater are connected with a flue outlet of the denitration system, flue gas outlets of the tubular air preheater and the rotary air preheater are connected with the dust remover, and a water supply heat exchanger is further arranged between the flue gas inlet of the tubular air preheater and the flue outlet of the denitration system. The rotary preheater can further improve the operation efficiency of a coal power plant, solves the problems of high air leakage, loss and catalyst blockage of a traditional rotary preheater, has the advantages of energy conservation, economy and reliability, and meets the technical requirements of efficient and clean operation of a thermal power plant.

The invention discloses an intelligent wheelchair multi-mode control method and system, and particularly relates to the technical field of wheelchair control. The method comprises the following steps: acquiring voice data, head posture data and residual limb movement data of a user in real time, and establishing an initial association relationship between a user control intention and each modal feature; continuously tracking a sound production mode, head posture control stability and time evolution characteristics of residual limb actions, generating functional state evolution trend data and posture control drift characteristic data, and executing dynamic weighting compensation on multi-modal fusion parameters to obtain a self-adaptively updated multi-modal fusion model; according to the method, intention recognition and wheelchair motion control are carried out on the current real-time multi-modal features by using the self-adaptively updated multi-modal fusion model, so that the consistency and reliability of control intention recognition results are kept under the condition that the function state of a user is slowly changed, and the risk that the same motion intention is misjudged as different control instructions in different periods is reduced.