Load predicating system based on PLC and principal component analysis-RBF neural network

Abstract

The invention relates to the field of air conditioning system end load predication, and particularly to a load predicating system based on a PLC and a principal component analysis-RBF neural network.The system comprises a PLC logical controller, a touch screen, an upper computer, an ODBC data module, a load predicating module, an SQL database module and a mathematical software module. The PLC logical controller is connected with the upper computer and the touch screen. The upper computer is connected with the touch screen and the ODBC data module. The ODBC data module and the touch screen areconnected with the load predicating module. The load predicating module is connected with the mathematical software module through the SQL database module. The mathematical software module is connected with a principal component analysis module and an RBF network construction module. The load predicating system has beneficial effects of reducing dimensions of original data, reducing the requirement time of an end load algorithm, realizing important meaning for real-time operation state adjustment of end equipment, and realizing simple structure, high safety and high reliability.

Description

technical field [0001] The invention relates to the field of air-conditioning system terminal load forecasting, in particular to a load forecasting system based on PLC and principal component analysis-RBF neural network. Background technique [0002] In the field of heating and ventilation, the accurate prediction of end-use cooling load is a very difficult problem, but it is also a very important problem that needs to be solved. At present, due to the advanced manufacturing process, the efficiency of the equipment can basically reach the maximum value of the design condition. host, etc.), so as to minimize the overall energy consumption. The cooling load forecast at the end can effectively solve the above problems. Through the relevant algorithm, the time-series cold storage capacity of the end is known in advance, and the operation status of the equipment (the number of pumps started, the number of running frequency, number of cold machines started, operating frequency, ...

AI Technical Summary

Problems solved by technology

At present, there are many factors related to the terminal load in the HVAC system (solar radiation intensity, solar radiation angle, outdoor temperature and humidity, weather conditions, enclosure structure, building orientation, supply and return water temperature, supply and return water pressure, supply and return water flow, Time period, holidays, etc.), so it is very difficult to directly use a large number of initial data to realize the forecast of the terminal load, and due to the large amount of calculation, it will also lead to a large time lag in the forecast results of the terminal load, which is not conducive to the real-time performance of equipment adjustment

Images