CN-122015261-A - Central air conditioner operation control method based on load prediction reliability judgment

CN122015261ACN 122015261 ACN122015261 ACN 122015261ACN-122015261-A

Abstract

The invention provides a central air conditioner operation control method based on load prediction reliability judgment, which comprises data acquisition, multi-time scale load prediction, prediction reliability evaluation, load fluctuation characteristic analysis, prediction reliability gating, operation mode arbitration decision, hierarchical control execution and safety verification, and improves energy efficiency and reduces equipment loss on the premise of ensuring safety and comfort through reliability prediction driving control.

Inventors

WANG ZHAO
LI YONG
LI JIANFU
Zheng Chaoqin
Hu Zhouhao

Assignees

广州市铭汉科技股份有限公司

Dates

Publication Date: 20260512
Application Date: 20260414

Claims (8)

1. A central air conditioner operation control method based on load prediction reliability judgment is characterized by comprising the following steps: S1, data acquisition, namely acquiring operation and environment data of the central air conditioner; S2, multi-time scale load prediction is carried out, a multi-time scale load prediction model is constructed, load changes of different time scales in the future are predicted based on operation and environment data, and load prediction intervals of different time scales are generated; S3, evaluating prediction reliability, namely calculating the comprehensive prediction reliability of a load prediction interval; s4, load fluctuation characteristic analysis, namely calculating the current fluctuation intensity of the load based on the operation and environment data, and extracting future fluctuation trend information based on a load prediction interval; S5, prediction reliability gating, including safety gating and information gating, safety gating, judging whether comprehensive prediction reliability is lower than a preset safety threshold, if so, using a preset safety operation mode to drive the central air conditioner to operate, if not, performing information gating, judging whether future fluctuation trend information is allowed to participate in control decision, and then performing S6; S6, performing operation mode arbitration decision, namely introducing fluctuation information, trend information and margin information as arbitration input, selecting degradation arbitration or look-ahead arbitration according to an information gating result, dividing different working conditions and matching corresponding operation modes, and outputting an engineering action set comprising a target set value, action constraint and host start-stop pre-judgment; S7, performing hierarchical control and safety verification, establishing equivalent time delay parameters of system cold energy transportation, performing hierarchical execution according to a time-scale engineering action set, setting the constraint of starting and stopping of a host, performing safety verification on a control result, correcting instructions in case of violation, and returning to a safe operation mode.
2. The method for controlling operation of a central air conditioner based on reliability determination of load prediction according to claim 1, wherein the operation and environment data includes load data, environment data and time data, and the load prediction intervals of different time scales include a short-term load prediction interval and a medium-term load prediction interval.
3. The central air conditioner operation control method based on the load prediction reliability judgment according to claim 2, wherein S3 specifically comprises the following steps: s3.1, calculating the credibility of each load prediction interval; S3.2, judging the consistency of different load prediction intervals; And S3.3, fusing the reliability of each load prediction interval and the consistency results of different load prediction intervals to obtain comprehensive prediction reliability.
4. The central air conditioner operation control method based on the load prediction reliability judgment according to claim 3, wherein S3.1 comprises the steps of evaluating whether each load prediction interval has coverage on a historical real load value, evaluating whether each load prediction interval has effective information amount, evaluating whether each load prediction interval has stability with time change, and evaluating whether the load prediction interval has abnormal jump.
5. The method for controlling operation of a central air conditioner based on load prediction reliability determination as set forth in claim 3, wherein S3.2 comprises determining whether the short-term load prediction interval falls within the intermediate load prediction interval, and determining a degree of coincidence of the short-term load prediction interval and the intermediate load prediction interval in a risk of load change.
6. The central air conditioner operation control method based on the load prediction reliability judgment according to claim 2, wherein the fluctuation information comprises current fluctuation intensity and future fluctuation trend, the trend information comprises short-term trend and medium-term trend, and the margin information is capability margin M.
7. The central air conditioner operation control method based on load prediction reliability judgment according to claim 1, wherein the degradation arbitration is a decision mode based on current real-time information only, three working conditions of low fluctuation steady state L1, medium fluctuation transition state L2 and high fluctuation information shortage L3 are divided, only fine adjustment of the frequency of a freezing pump and the temperature of water outlet of the freezing water is allowed, and starting and stopping of a host are forbidden; the look-ahead arbitration is to fuse the current fluctuation intensity and the future fluctuation trend information of the current load, and divide five working conditions of a steady-state predictable F1, a storm pre-warning F2, a continuous high fluctuation F3, a short-time disturbance impact F4 and an uncertain transition state F5.
8. The central air conditioner operation control method based on load prediction reliability judgment according to claim 1, wherein the host start-stop strong constraint comprises a minimum continuous operation time constraint, a minimum downtime constraint, a multi-time scale prediction consistency constraint and an execution limit check constraint, the host start-stop judgment is allowed to be triggered only when all the constraints are met, the content of the safety check comprises whether a control instruction violates an equipment operation boundary, a change rate constraint and a start-stop time constraint, and if the control instruction violates the equipment operation boundary, the change rate constraint and the start-stop time constraint, the execution of the corresponding instruction is automatically corrected or refused, and the safety operation control mode is returned to a preset safety operation control mode.

Description

Central air conditioner operation control method based on load prediction reliability judgment Technical Field The invention relates to the technical field of central air conditioner operation control, in particular to a central air conditioner operation control method based on load prediction reliability judgment. Background The central air conditioner is core energy consumption equipment in scenes such as large commercial buildings, office buildings, industrial parks and the like, and the rationality of operation control of the central air conditioner directly influences the energy efficiency of the system, the service life of the equipment and the indoor comfort. The traditional central air conditioner operation control method is based on current load, takes real-time load feedback as the only adjustment basis, and triggers preset adjustment actions only by comparing load related parameters (such as temperature, pressure and flow) acquired in real time with a set threshold value, so that future loads cannot be predicted. There is a problem of wasting energy, and the response is delayed when the load suddenly changes, affecting comfort. The prior art introduces a load prediction technology to realize the prospective control of the central air conditioner, but the prior prediction control scheme has the obvious technical defects that firstly, single load values are output in a multi-point prediction mode, uncertainty of prediction results cannot be described, a judgment link of prediction quality is not arranged, the prediction results with errors are used blindly, system operation fluctuation is easily caused, even the situation of insufficient cooling or overcooling occurs, secondly, control decisions are only single reference load fluctuation information, key dimensions such as load trend, system capacity margin and the like are not fused, aggressive control or hysteresis control is easily occurring, energy consumption increase or frequent equipment actions are caused, thirdly, engineering practical constraints of a central air conditioner system, such as characteristics of cold energy transmission delay, system thermal inertia, high loss of starting and stopping of a host machine and the like are not fully considered, theoretical design and engineering fall to the ground are disjointed, control oscillation is easily occurring when a control instruction is executed, and the service life of equipment is greatly reduced. Disclosure of Invention In view of the above, the present invention aims to provide a central air conditioner operation control method based on load prediction reliability determination, which improves energy efficiency and reduces equipment loss on the premise of ensuring safety and comfort by means of reliable prediction driving control. In order to solve the technical problems, the invention adopts the following technical scheme: the invention relates to a central air conditioner operation control method based on load prediction reliability judgment, which comprises the following steps: s1, data acquisition, namely acquiring operation and environment data of the central air conditioner. And S2, predicting the load of multiple time scales, constructing a load prediction model of multiple time scales, predicting the load change of different time scales in the future based on the operation and environment data, and generating load prediction intervals of different time scales. And S3, evaluating prediction reliability, and calculating the comprehensive prediction reliability of the load prediction interval. And S4, load fluctuation characteristic analysis, namely calculating the current fluctuation intensity of the load based on the operation and environment data, and extracting future fluctuation trend information based on a load prediction interval. S5, prediction reliability gating, including safety gating and information gating, safety gating, judging whether comprehensive prediction reliability is lower than a preset safety threshold, if so, driving the central air conditioner to operate by using a preset safety operation mode, if not, performing information gating, judging whether future fluctuation trend information is allowed to participate in control decision, and then performing S6. S6, operation mode arbitration decision, introducing fluctuation information, trend information and margin information as arbitration input, selecting degradation arbitration or look-ahead arbitration according to an information gating result, dividing different working conditions and matching corresponding operation modes, and outputting an engineering action set comprising a target set value, action constraint and host start-stop pre-judgment. S7, performing hierarchical control and safety verification, establishing equivalent time delay parameters of system cold energy transportation, performing hierarchical execution according to a time-scale engineering action set, setting the constraint of star