Research On Load Balance Control Strategy Of Electric Heating Group

With the increase of heating demand,residents pay more and more attention to the improvement of heating mode quality.Radiant floor heating is one of the main heating methods for residents,including electric floor heating and water floor heating.Among them,electric floor heating is more and more widely used because of its space saving and convenient operation.With the wide application of electric floor heating,many negative effects and technical problems are formed.For example,the unbalance of three phases has many adverse effects on the operation of the system.Firstly,the voltage deviation at the client end causes voltage instability,and the non-uniform circulation of transformer triangular winding increases the transformer loss.Second,unbalanced three-phase voltage output causes voltage offset at the neutral point,and if the voltage offset exceeds the allowable range,the user's equipment will be damaged.Load balance control can effectively improve the three phase unbalance phenomenon,which plays an important role in the normal operation of the whole electric heating system.This paper analyzes the corresponding structure and basic principle of load balance control system,and proposes a load balance control system based on group intelligent optimization algorithm.To realize load balance control,this article is based on electric heating load of residential buildings in a district in Changchun to carry out the following load balance control strategy research: First of all,based on the basic idea of multivariate adaptive regression splines algorithm,electric heating value of load modeling is built.Through the corresponding simulation experiment,the corresponding model fitting and prediction accuracy is effectively proved.Secondly,based on genetic algorithm to optimize the BP neural network and leapfrog algorithm to optimize the BP neural network,the electric heating value of load forecasting model is established.Through the corresponding simulation experiment,the corresponding algorithm convergence rate and prediction precision is effectively proved.Then,based on the idea of genetic algorithm and leapfrog algorithm,a three-phase unbalanced phase conversion algorithm was established.Through corresponding simulation experiments,the unbalanced degree of three-phase current was effectively reduced.Finally,the corresponding software design and hardware design of the system is carried out,based on the research results of this paper and the actual engineering requirements,and through VC++ 6.0 MFC programming for the corresponding softwaredevelopment is carried out.BP neural network was optimized based on genetic algorithm and leapfrog algorithm,and the load value prediction model of electric heating group was established to effectively predict the system load value,so as to overcome the hysteresis of system load conversion.Based on the idea of improved genetic algorithm and improved leapfrog algorithm,the three-phase unbalanced phase commutation algorithm is established,and the system load imbalance is effectively improved.Therefore,the above research has a strong engineering background,and the designed electric heating group load balance control system has certain practicability and popularization value.