Short-term photovoltaic power forecasting is of great significance for improving the operation systems and increasing penetration power. To improve accuracy short-term forecasting, an ensemble-model-based prediction method proposed. Firstly, quartile used to process raw data, Pearson coefficient utilized assess multiple features affecting output. Secondly, structure ensemble model constructed, a k-fold cross-validation train submodels. The results each submodel are merged. Finally, validity...

Combined cooling, heating, and power (CCHP) systems is a distributed energy system that uses the station or heat engine to generate electricity useful simultaneously. Due its wide range of advantages including efficiency, ecological, financial, CCHP will be main direction integrated system. The accurate prediction gas, electrical loads plays an essential role in management systems. This paper combined long short-term memory (LSTM) network convolutional neural (CNN) design novel hybrid for...

The components of the rotor are subjected to a tremendous centrifugal force at high speed velocity. It can be cause damage deformation or even failure structure. In this paper, 300kW,30000r/min interior permanent magnet synchronous motor(HS-IPMSM) is designed. rotational reached 196m/s.The mechanical strength analyzed by finite element method(FEM) and also over speeding testing result show that operate reliably 1.1 times rated operating content design requirements.

Photovoltaic (PV) power short-term forecasting is of great significance to system control and operations. To improve the accuracy, this paper proposes a machine-learning-based approach for PV power. The proposed built on multiple sub-models which contribute high prediction accuracy. Firstly, training set used build with better performance. Secondly, each sub-model fused by an equal weight fusion method or least square method, integrated model generates results. Finally, effectiveness...

Model reference adaptation system(MRAS) is a relatively good-performance speed identification method for permanent magnet synchronous motor, but the accuracy of model's parameters will directly affect results. To solve this problem, paper proposes new model method. Based on Popov's superstability theory, adaptive law designed to reduce influence identification. Finally, applied vector control system motors. The effectiveness improved MRAS algorithm verified through simulations and...

Abstract The rotor of a magnetic suspension centrifugal compressor is supported by bearing, and power failure will cause the high‐speed to fall directly onto protective bearings, affecting reliability compressor. To solve this problem, paper proposes an abnormal shutdown control method with semi‐switching modulation inverter. Firstly, overall scheme designed, its feasibility analysed in detail through mathematical modelling. Secondly, aiming at non‐linear, time‐varying gain uncertainty...

In engineering application, the hot press assembly technology is often used to improve stability of rotor structure, but conventional design methods cannot effectively evaluate influence this process on strength, which easily causes strength exceed its safety margin range, and seriously it will lead failure structure. This paper takes cylindrical magnet surface-mounted high-speed permanent synchronous motor as research object. Firstly, pressing force stresses interference analyzed; then,...

Conventional magnetic bearing axial suspension control generally uses a single probe displacement sensor to detect the in real time. In practical applications, installation of is inclined due some factors, and there an error between detected actual value. view this situation, paper simulates analyzes influence on system, proposes new detection scheme eliminate error. Finally, feasibility verified by experiments.

The following topics are dealt with: permanent magnet motors; finite element analysis; machine control; rotors; synchronous stators; electric current magnetic flux; machines; torque.