CN-122015177-A - Intelligent water supply temperature regulation and control method based on thermal characteristic analysis and migration correction

Abstract

The invention relates to an intelligent water supply temperature regulation and control method based on thermal characteristic analysis and migration correction, which comprises the steps of S1, thermal engineering principle feature screening, S2, data acquisition and processing, S3, data set segmentation, S4, basic model determination, S5, thermal characteristic identification, S6, model combination, and S7, model application. The invention takes the load predicted value and the hydraulic flow as core input, expresses the water supply temperature into a concise function form, and transfers the water supply temperature to a real-time running state for correction and optimization by excavating key thermal characteristics of pipe network characteristics in historical data. The invention takes the thermodynamic mechanism architecture as a leading part, characterizes uncertainty factors, introduces an intelligent algorithm and adopts a modularized design to form a lightweight modeling scheme, can keep steady under data fluctuation and working condition change, and has the capabilities of interpretation, self-learning, self-adaption and continuous optimization, thereby realizing dynamic prediction and quick response of water supply temperature.

Inventors

• Peng Mengbo
• YANG JUNHONG
• ZHU JUNDA
• LI YANSONG
• WANG TIANYU
• LIANG XINYUE
• Ben Chaoran
• SUN YAOGUO
• WANG JIAHUI

Assignees

• 天津大学

Dates

Publication Date

20260512

Application Date

20251208

Claims (2)

1. 1. The intelligent water supply temperature regulation and control method based on thermal characteristic analysis and migration correction is characterized in that the regulation and control method combines a thermal engineering basic principle with a thermal characteristic recognition mechanism, establishes a simplified and efficient water supply temperature prediction and control model, and realizes rapid modeling and regulation from two core quantities of load and flow; the control system comprises a peripheral module, a control module and a control object, wherein the peripheral module is used for providing historical/real-time data input, the control module is used for completing model construction and calibration, model application and outputting a regulating instruction, the control object is used for responding to the instruction to execute operation regulation and form operation demand feedback, and the data acquisition in the peripheral module relies on a flow sensor, a heat meter and a water supply temperature sensor which are arranged at a key node of a thermal gateway to acquire the operation data of the heating system in real time through the data acquisition module, wherein the data acquisition module comprises a water supply temperature t g , a thermal load Q and a flow G, and the acquired data are transmitted to an upper computer and stored in a database to serve as a basis for modeling and prediction; the regulation and control method comprises the following steps: S1, screening the characteristics of the thermal engineering principle, namely searching more general input characteristics from the basic heat balance principle of a heating system, and determining the input characteristics by taking parameter simplification as a target; S2, data acquisition and processing, namely, forming a parameter set with consistent time scale according to the actual heat supply network operation history data by reading a database in a peripheral module, wherein the parameter set comprises water supply temperature t g , flow G and heat load Q; s3, determining a basic model, namely taking parameter set data for basic model parameter calibration, and establishing a functional relation between water supply temperature and heat load and flow by combining a multiple regression or integrated learning method to determine a basic model f 0 : T_static=f 0 (Q,G); S4, thermal characteristic identification, namely inputting a parameter set into a basic model to obtain a simulated water temperature T_static, defining deviation from an actual water supply temperature T g as a residual term representing system thermal characteristic time-varying drift, and establishing a residual mapping model f 1 : ε=f 1 (Q,G); introducing a multi-model fusion learning framework to train and integrate f 1 , realizing self-adaptive characterization of uncertain disturbance and drift characteristics of working conditions, and constructing a thermal characteristic recognition module; S5, combining the basic model f 0 with the residual mapping model f 1 to obtain a water supply temperature prediction and control model: T_pred=T_static+ε=f 0 (Q,G)+f 1 (Q,G); s6, model application, namely dividing the method into two stages according to the actual conditions of a heating season: Stage ①, entering an initial operation stage of a new heating season, inputting a flow limit value based on planned heat load set in a peripheral module and hydraulic regulation of heating system operation, and outputting a water supply temperature set value T_set of a control unit after inputting a water supply temperature prediction and control model; Stage ② -entering a fresh heating season operation stage, namely collecting actual operation data of last days (more than 2 days) or weeks, checking and resetting a set value T_set based on actual load and flow, introducing actual water supply temperature, adopting a rolling window retraining and automatic parameter updating strategy to identify and absorb new heat characteristics, constructing a residual correction function e=f 2 (Q, G), and using the residual correction function e=f 2 (Q, G) to perform real-time correction optimization on the T_set to form T_set=f 0 (Q,G)+f 1 (Q,G)+f 2 (Q,G); Wherein f 2 represents an incremental correction term of the run-time relative to f 1 for compensating for additional drift and disturbances under new conditions.
2. 2. The intelligent water supply temperature regulation and control method based on thermal characteristic analysis and migration correction according to claim 1, wherein the S1 specifically comprises the following steps: Based on the basic heat balance principle of the heating system, in an ideal stable operation state (without considering heat loss in the heat supply network conveying process), the heat output by the heat source is completely matched with the actual heat load demand of the user side, and the heat conservation relationship is as follows: ; (1) ; (2) ; (3) ; (4) Wherein Q 1 、Q 2 and Q 3 respectively represent the actual heat load of a building, the heat dissipation capacity of heat dissipation equipment and the load of an energy station side, Q is the building heat load index, W/(m 3 . DEG C), V is the building exterior structure volume, m 3 ;t in and t out are the indoor and outdoor temperatures respectively, DEG C, K is the heat transfer coefficient of the heat dissipation equipment, W/(m 2 . DEG C), A is the heat dissipation area of the heat dissipation equipment, m 2 , C is the hot water specific heat capacity, J/(kg. DEG C), G is the energy station flow, kg/h, t g and t h are the water supply temperature and the return water temperature respectively, DEG C, t p is the average temperature of water supply and return water, and the calculation formula is t p =（t g + t h )/2; When a certain heating system is determined, the formulas (1) - (4) are simplified as follows: ; (5) in order to further explore the correlation degree of the water supply temperature and other parameters, a more general input characteristic is searched, the pearson correlation coefficient is adopted to analyze the correlation degree between the water supply temperature and the parameters in the formula (5), and the aim of simplification is to ensure that for the water return temperature, the correlation coefficient between the water supply and return temperature difference and the load reaches high correlation, so that the load can reflect the change of the water supply and return temperature difference, and the water supply temperature is described by the lightweight relational expression presented in the formula (6): ; (6)。

Description

Intelligent water supply temperature regulation and control method based on thermal characteristic analysis and migration correction Technical Field The invention belongs to the technical field of central heating, and particularly relates to an intelligent water supply temperature regulation and control method based on thermal characteristic analysis and migration correction. Background The central heating system is widely applied to building heating and industrial heating in northern areas of China. The 'heat supply on demand' is a basic path for guaranteeing the heat demand of users and realizing remarkable energy conservation and emission reduction. Because of the large thermal inertia of the system, the thermal inertia of the building and the time-varying property of the heat mode used by users, the transient adjustment response capability of the system is limited, and the actual heating process has supply and demand mismatch and energy waste with different degrees. Along with the continuous expansion of the scale of the central heating system in northern areas of China and the continuous improvement of the comfort requirements of users, the intelligent and energy-saving regulation and control of the operation of the heating system become an important direction for transformation of urban energy systems. Heat load prediction is generally regarded as a key element in accurate heating management and plays an important role in evaluating the most efficient energy saving strategy [1]. Recent advances in internet of things technology and automation control systems have increased the level of intellectualization of heating systems, thereby enabling more precise regulation to meet demand for on-demand heating [2]. Numerous studies have improved the accuracy of thermal load prediction models [3-4]. Even with accurate load predictions, many current heating systems still face challenges in selecting an appropriate regulation strategy that can match fluctuating load demands. This often leads to energy inefficiency caused by excessive reliance on manual control [5]. Accurate load prediction is critical to guiding the operation of the heating system, and therefore, in order to achieve intelligent heating regulation, it is necessary to decompose the load prediction result into appropriate control parameters. The main purpose of central heating operation regulation is to enable a heating system to avoid energy waste caused by excessive heating on the premise of meeting the heat demands of users. Many scholars have made corresponding research work on the operation regulation of central heating systems. The specific operation regulation modes can be basically divided into three types, namely, the quantity regulation of only changing the system flow, the quality regulation of only changing the water supply temperature and the quality regulation based on the staged flow regulation. For example, the load on the heat source side of the energy station is relatively large, the number of heating users is large, the load fluctuation is large during the whole heating period, and the operation adjustment is generally performed by mass adjustment based on staged flow adjustment. In the process, the water supply temperature is used as a core parameter for regulating and controlling the heat transfer efficiency and is also used as the most basic record data of daily operation and maintenance, and the heat comfort of the tail end of a user, the heat transfer efficiency of a pipe network and the energy consumption of heat source operation are directly determined. With regard to the operation regulation of the central heating system, many researches show that the operation regulation strategy of the central heating system has no direct relation with the design heat load index of the building, and great difference exists between the actual parameters of central heating and the design parameters. Therefore, the actual operation adjustment scheme of the central heating system is different from the theoretical operation adjustment scheme, and the operation adjustment scheme of the central heating system needs to be formulated according to the actual parameters. In the actual operation of the central heating system, the traditional actual operation is mainly determined according to the history operation experience and the change of the future outdoor air temperature, for example, an empirical climate compensation curve is adopted to set the water supply temperature according to the outdoor temperature, and the method has simple structure and convenient implementation, but does not consider dynamic factors such as building thermal inertia, meteorological disturbance coupling, actual thermal load change, user side temperature feedback and the like, so that the heating system is easy to overheat or underheat, the energy waste is serious, and the comfort level is difficult to guarantee. Chen Tingting et al [6] provides a prediction of the heat