Prediction method and system for regional cooling, heating, cooling and heating loads

Abstract

The invention provides a prediction method and system for regional cooling, heating, cooling and heating loads. The method comprises the steps that regional environment temperature and humidity data and heat meter data are collected; collecting historical data, preprocessing the acquired historical data, and generating a data training set and a test set; adopting an LSTM-based Seq2Seq recurrent neural network model, and utilizing the data training set and the test set to perform model training and testing; importing the trained model file by adopting a TensorFlow framework, and deploying and running the trained model file; and outputting a future cold and heat load prediction value by taking historical data before the current time period as input of model prediction. The LSTM-based self-learning multi-dimensional time sequence multi-step prediction method for the Seq2Seq recurrent neural network model provided by the invention has the advantages of better fitting and expansibility andhigher algorithm accuracy. A load prediction system is combined with an existing automatic control system, and data sharing is achieved. Prediction results are applied to operation, energy can be saved, consumption can be reduced, and operation cost is reduced.

Description

technical field [0001] The present invention relates to the technical field of heating, ventilating and air-conditioning, and in particular to a method and system for predicting cooling and heat loads of district cooling and heating (Disrtict Heating and Cooling, DHC for short). Background technique [0002] Due to the large number of district cooling and heating system equipment, the large number of equipment, and the wide distribution range, the system is complex and huge; the lag caused by the water storage and heat storage of the pipe network; the operator predicts the load based on the temperature difference, pressure difference and other data of the system, relying on personal experience, It is difficult to guarantee the accuracy; the delay caused by control adjustment; the actual use factors of building air conditioners, etc., make it difficult to match the cooling and heat load supply with the actual demand. [0003] Therefore, in the district cooling and heating sys...

AI Technical Summary

Problems solved by technology

[0002] Due to the large number of district cooling and heating system equipment, the large number of equipment, and the wide distribution range, the system is complex and huge; the lag caused by the water storage and heat storage of the pipe network; the operator predicts the load based on the tempera

Images