CN-121995987-A - Heating system self-adaptive temperature control method based on data acquisition

CN121995987ACN 121995987 ACN121995987 ACN 121995987ACN-121995987-A

Abstract

The invention relates to the technical field of building thermal response characteristic analysis and self-adaptive temperature, and discloses a self-adaptive temperature control method of a heating system based on data acquisition, wherein the self-adaptive temperature control method is used for carrying out seasonal decomposition treatment on multi-period acquired data to separate long-term degradation trend by establishing a time sequence thermal response fingerprint vector database, adopting track fitting and time extrapolation to predict and control mismatch moment, establishing a parameter sensitivity mapping matrix to realize parameter self-adaptive adjustment, and solving the temperature control problem of the heating system under a scene of slow degradation of building thermal characteristics.

Inventors

XU FUSHENG
ZHANG DONGMEI

Assignees

济南浩星新能源有限公司

Dates

Publication Date: 20260508
Application Date: 20260206

Claims (10)

1. The self-adaptive temperature control method for the heating system based on data acquisition is characterized by comprising the following steps of: Collecting indoor temperature response data of a building under a standard thermal excitation test, performing feature extraction and dimension reduction on the temperature response data, and generating a time sequence thermal response fingerprint vector database; Performing time sequence decomposition processing on fingerprint vectors in the time sequence thermal response fingerprint vector database, separating periodic fluctuation components and trend components, and removing the periodic fluctuation components to generate a seasonal fingerprint trend sequence; performing track fitting on the seasonal fingerprint trend sequence, predicting an evolution function of each feature dimension along with time, performing time extrapolation according to a preset temperature control mismatch threshold, calculating the moment when the fingerprint feature is predicted to exceed the mismatch threshold, and generating a predicted mismatch time point and a future fingerprint predicted value; Calculating the difference value between the future fingerprint predicted value and the initial reference fingerprint vector, inputting the difference value into a parameter sensitivity mapping matrix to calculate a temperature control parameter correction increment, and generating and outputting an adaptive parameter adjustment scheme; the parameter sensitivity mapping matrix is used for representing the sensitivity coefficient of each temperature control parameter to each fingerprint feature dimension change.
2. The method for controlling the self-adaptive temperature of the heating system based on data acquisition according to claim 1, wherein the time sequence decomposition process adopts an STL decomposition algorithm, and each dimension component of the fingerprint vector is separated into a season term, a trend term and a remainder by iterative local weighted regression.
3. The method for adaptive temperature control of a heating system based on data acquisition according to claim 1, further comprising, prior to the performing of the trajectory fitting on the de-seasonal fingerprint trend sequence: carrying out sliding window statistical analysis on the seasonal fingerprint trend sequence, and calculating the mean value and variance of each characteristic dimension in each window; And counting the variation of the mean and the variance of each dimension between adjacent windows, calculating the slope of the variation trend, and generating a fingerprint degradation trend feature set.
4. The data acquisition-based heating system self-adaptive temperature control method according to claim 1, wherein the track fitting adopts a polynomial regression algorithm, and the time evolution of each characteristic dimension is fitted respectively to obtain a polynomial function of each dimension changing with time.
5. The method for controlling the self-adaptive temperature of the heating system based on data acquisition as set forth in claim 1, wherein the parameter sensitivity mapping matrix is obtained by performing a disturbance test on each temperature control parameter in an initial debugging stage of the heating system, recording corresponding thermal response fingerprint changes, and establishing a linear mapping relationship between the fingerprint changes and the parameter changes by using least square fitting.
6. The method for controlling the adaptive temperature of a heating system based on data collection according to claim 1, further comprising, before generating and outputting the adaptive parameter adjustment scheme: Calculating the interval between the current moment and the time of the expected mismatch time point; comparing the interval between the times with a preset self-adaptive adjustment advance; And if the interval between the time is smaller than or equal to the adaptive adjustment advance, triggering a temperature control parameter adaptive adjustment flow.
7. The method for controlling the adaptive temperature of a heating system based on data collection according to claim 1, wherein the output adaptive parameter adjustment scheme comprises: Calculating the deviation between the target parameter and the current temperature control operation parameter in the self-adaptive parameter adjustment scheme; linearly decomposing the deviation into a plurality of fine tuning increments in time; And sequentially outputting parameter fine-tuning instructions according to intervals between preset times, so that the current parameters are smoothly transited to the target parameters.
8. The data acquisition-based heating system adaptive temperature control method according to claim 1, further comprising: Continuously collecting indoor temperature response data and extracting a current thermal response fingerprint; Calculating the deviation between the current actual fingerprint and the predicted fingerprint of the fitting track at the corresponding moment; if the deviation exceeds the preset track deviation tolerance, updating a track fitting function by using current actual fingerprint data, recalculating a predicted mismatching time point and a future fingerprint predicted value, and updating a self-adaptive parameter adjustment scheme.
9. The method for controlling the self-adaptive temperature of a heating system based on data acquisition according to claim 1, wherein the dimension reduction process adopts a principal component analysis method, and principal components with accumulated contribution rate exceeding a preset threshold are reserved as dimension components of the fingerprint vector.
10. A data acquisition-based heating system adaptive temperature control system for performing the data acquisition-based heating system adaptive temperature control method according to any one of claims 1 to 9, comprising: The fingerprint generation module is used for collecting indoor temperature response data of the building under a standard thermal excitation test, performing feature extraction and dimension reduction on the temperature response data, and generating a time sequence thermal response fingerprint vector database; The seasonal decomposition module is used for carrying out time sequence decomposition processing on the fingerprint vectors in the time sequence thermal response fingerprint vector database, separating periodic fluctuation components and trend components and generating a seasonal fingerprint trend sequence; the track prediction module is used for carrying out track fitting on the seasonal fingerprint trend sequence, carrying out time extrapolation according to a temperature control mismatch threshold value and generating a predicted mismatch time point and a future fingerprint predicted value; And the parameter adjustment module is used for calculating the difference value between the future fingerprint predicted value and the initial reference fingerprint vector, inputting the difference value into the parameter sensitivity mapping matrix to calculate the temperature control parameter correction increment, and generating and outputting an adaptive parameter adjustment scheme.

Description

Heating system self-adaptive temperature control method based on data acquisition Technical Field The invention relates to the technical field of building thermal response characteristic analysis and self-adaptive temperature, in particular to a heating system self-adaptive temperature control method based on data acquisition. Background In the long-term operation process of the heating system, the building envelope structure can be slowly degraded in thermal performance due to factors such as material aging, heat preservation layer wetting, door and window sealing performance reduction and the like, and the original temperature control parameters and the actual building thermal characteristics can be obviously mismatched after accumulating for a plurality of years. The traditional heating temperature control method adopts fixed parameter operation, and has the main technical problems that short-term fluctuation (comprising weather change, use habit difference and seasonal temperature period) of collected data can mask the long-term degradation trend of thermal response characteristics of a building, the traditional method is difficult to distinguish normal seasonal fluctuation and thermal response characteristic degradation signals from temperature data collected in multiple periods, the control parameters can only be passively adjusted after the temperature control effect is obviously deteriorated, and active self-adaptive temperature control based on data collection cannot be realized. Disclosure of Invention The invention provides a heating system self-adaptive temperature control method based on data acquisition, which solves the technical problem of temperature control parameter mismatch caused by slow degradation of building thermal characteristics in the related technology. The invention provides a heating system self-adaptive temperature control method based on data acquisition, which comprises the following steps: Collecting indoor temperature response data of a building under a standard thermal excitation test, performing feature extraction and dimension reduction on the temperature response data, and generating a time sequence thermal response fingerprint vector database; Performing time sequence decomposition processing on fingerprint vectors in the time sequence thermal response fingerprint vector database, separating periodic fluctuation components and trend components, and removing the periodic fluctuation components to generate a seasonal fingerprint trend sequence; performing track fitting on the seasonal fingerprint trend sequence, predicting an evolution function of each feature dimension along with time, performing time extrapolation according to a preset temperature control mismatch threshold, calculating the moment when the fingerprint feature is predicted to exceed the mismatch threshold, and generating a predicted mismatch time point and a future fingerprint predicted value; Calculating the difference value between the future fingerprint predicted value and the initial reference fingerprint vector, inputting the difference value into a parameter sensitivity mapping matrix to calculate a temperature control parameter correction increment, and generating and outputting an adaptive parameter adjustment scheme; the parameter sensitivity mapping matrix is used for representing the sensitivity coefficient of each temperature control parameter to each fingerprint feature dimension change. Further, the time sequence decomposition process adopts an STL decomposition algorithm, and each dimension component of the fingerprint vector is separated into a season term, a trend term and a remainder through iterative local weighted regression. Further, before the track fitting is performed on the seasonal fingerprint trend sequence, the method further includes: carrying out sliding window statistical analysis on the seasonal fingerprint trend sequence, and calculating the mean value and variance of each characteristic dimension in each window; And counting the variation of the mean and the variance of each dimension between adjacent windows, calculating the slope of the variation trend, and generating a fingerprint degradation trend feature set. Further, the track fitting adopts a polynomial regression algorithm to fit the time evolution of each characteristic dimension respectively, and a polynomial function of each dimension changing along with time is obtained. Further, the parameter sensitivity mapping matrix is obtained by performing disturbance test on each temperature control parameter in the initial debugging stage of the heating system, recording corresponding thermal response fingerprint changes, and establishing a linear mapping relation between the fingerprint changes and the parameter changes by using least square fitting. Further, before generating and outputting the adaptive parameter adjustment scheme, the method further comprises: Calculating the interval between the current moment