Strategy Energy-saving Control Based On The Central Heating System Load Forecasting

Statistics show that in the three northern regions the building heating energyconsumption accounts for a quarter of total energy consumption, therefore the studyof building heating systems energy-saving control has a very important significanceto reduce building heating energy consumption. In this paper, energy controlstrategy based on the central heating system load forecasting, mainly from thefollowing four aspects:(1) Determine optimal scheduling model on the economyand energy consumption of heating systems with multiple heat sources to (2) Carryout the heating system forecasting based on the load of the heating needs of theheating system users.(3) Scheduling of distributed heat source to optimize theheating load distribution.(4) Monitor and control the heating parameters and signalsto update the heating system and to fulfill the goals of energy-saving respectively.This article uses the model of the minimax probability machine to establish theindoor heating load curve of18℃and20℃. The output of the heating loadrestrictions can prevent excessive thermal heating load error to the user discomfort.And then analyze the heating system optimization scheduling model, the minimumoverall energy consumption and operating costs a minimum of two indicators, theuse of entropy coefficient to evaluate the costs and energy costs to run thescheduling method for the optimization objectives and determine the weights ofexperts in charge of the scheduling and evaluation methods are combined, to obtainpractical combination of both subjective and objective weights.Load forecasting is a prerequisite for the heating system to optimize scheduling,while the heating system is the basis of actual operating data load forecasting, thisarticle has discussed the control during the operation of the heating system and alarge segment of individual complement missing data, and then use the informationgranulation methods for data processing and normalization. SVM has more superiorthan the good characteristics of artificial neural network is more suitable forhandling non-linear characteristics of the heating system, so using multiple supportvector machine regression of multiple-input single-output heating system to predictthe final simulation analysis.Multi-heat heating system is the key to the optimal energy saving scheduling,this article will first follow the condition of the national standards and heating loadforecasting heating combined heat given heat load to ensure the user’s heatingdemand, and then further design for the optimization of heating load. According tothe heating system optimization scheduling model, in considering thecomprehensive energy consumption of the economy and the best case, the use of particle swarm optimization calculations, avoiding the other initial calculationmethod selected this defect, the last two examples computational analysis, to theheating system to save energy.In order to offer better load forecast for heating system control, to achieveenergy efficiency goals, this article gives heat source heating system design formulti-monitoring device, including energy-saving heating control device includesand software design and part of hardware configuration for PLC control, and alsoincluding GPRS hardware and software configurations for the communication part,hardware configuration and software design for HCI touch-screen, and finallyoffering saving GPRS network integration control device system for heating.