CN-121578658-B - Cloud-coordinated power consumption dynamic sensing and energy-saving optimal control method and system

CN121578658BCN 121578658 BCN121578658 BCN 121578658BCN-121578658-B

Abstract

The invention provides a cloud cooperative power consumption dynamic sensing and energy-saving optimization control method and a cloud cooperative power consumption dynamic sensing and energy-saving optimization control system, which relate to the technical field of energy-saving control and comprise the steps of performing local deployment of a heterogeneous sensing array and constructing a plurality of edge intelligent sensing nodes; the method comprises the steps of receiving a plurality of multi-source space-time data streams, carrying out load stream coupling prediction to generate a plurality of initial regulation strategies, comparing a plurality of regulation feedback data with the plurality of initial regulation strategies to generate a regional energy consumption deviation matrix, correcting the regional energy consumption deviation matrix according to a plurality of thermodynamic coupling coefficients to obtain a plurality of neighborhood compensation strategies, carrying out space-time collaborative optimization to generate a plurality of dynamic optimization strategies, and carrying out regional autonomous closed-loop control. The invention solves the technical problems that the prior art mainly controls the regional power consumption based on a fixed mode or timing control, and the power consumption of partial regions is too high or too low to reduce the overall efficiency of the system because the power consumption of the regions cannot be dynamically adjusted according to the actual use conditions of the regions in the building.

Inventors

DUAN GANG
LI HUINAN
WANG TIANHENG
Lv Fuchen

Assignees

荣科科技股份有限公司
荣科科技股份有限公司郑州第二分公司

Dates

Publication Date: 20260505
Application Date: 20260127

Claims (7)

1. The cloud cooperative power consumption dynamic sensing and energy-saving optimal control method is characterized by comprising the following steps of: According to a partition topological structure of a central air conditioner, carrying out local deployment of a heterogeneous sensing array, and constructing a plurality of edge intelligent sensing nodes corresponding to a plurality of independent control areas, wherein the plurality of edge intelligent sensing nodes and an energy-saving optimized cloud are in bidirectional communication connection through an RS485 bus; The energy-saving optimization cloud receives a plurality of multi-source space-time data streams returned by the plurality of edge intelligent sensing nodes, carries out load people stream coupling prediction, and generates a plurality of initial regulation strategies; Generating a regional energy consumption deviation matrix by comparing the plurality of regulation feedback data with the plurality of initial regulation strategies; correcting the regional energy consumption deviation matrix according to a plurality of thermodynamic coupling coefficients of the plurality of independent control regions to obtain a plurality of neighborhood compensation strategies; Performing space-time collaborative optimization of the plurality of initial regulation strategies based on the plurality of neighborhood compensation strategies to generate a plurality of dynamic optimization strategies; The energy-saving optimization cloud end transmits the dynamic optimization strategies to the intelligent edge sensing nodes to conduct the regional autonomous closed-loop control of the independent control areas; Generating a regional energy consumption bias matrix by comparing the plurality of regulatory feedback data with the plurality of initial regulatory strategies, comprising: After topology mapping of the plurality of regulation feedback data and the plurality of initial regulation strategies is carried out according to the plurality of independent control areas, a plurality of energy consumption integral deviations are calculated; constructing a zero matrix based on the plurality of independent control regions; filling main diagonal elements of the zero matrix by adopting the energy consumption integration deviations; Carrying out neighborhood energy consumption interference intensity quantification of the plurality of independent control areas according to the partition topological structure, and then carrying out off-diagonal element filling of the zero matrix to complete construction of the area energy consumption deviation matrix; Correcting the regional energy consumption deviation matrix according to a plurality of thermodynamic coupling coefficients of the plurality of independent control regions to obtain a plurality of neighborhood compensation strategies, wherein the method comprises the following steps: Extracting a plurality of groups of adjacent control areas of the plurality of independent control areas based on the partition topology; extracting a plurality of groups of neighborhood interface heat conductivity coefficients and a plurality of groups of shared interface areas of the plurality of groups of adjacent control areas from a building BIM model; taking the plurality of groups of adjacent control areas as matrix element coordinates, and constructing a symmetrical thermal coupling matrix based on the plurality of groups of neighborhood interface heat conductivity coefficients and the plurality of groups of shared interface areas; carrying out Hadamard product correction on the area energy consumption deviation matrix by adopting the symmetrical thermal coupling matrix to generate a thermal coupling correction deviation matrix; Decomposing the thermal coupling correction deviation matrix to extract a plurality of row vector coupling strengths of the plurality of independent control areas; performing gain adjustment fitting on the coupling strengths of the plurality of row vectors, and outputting the plurality of neighborhood compensation strategies; performing gain adjustment fitting on the plurality of row vector coupling strengths, and outputting the plurality of neighborhood compensation strategies, including: Configuring a plurality of region gain coefficients according to the heat capacities of the plurality of independent control regions; performing overall gain scaling of the coupling strengths of the plurality of row vectors by adopting the plurality of regional gain coefficients, and outputting a plurality of gain adjustment row vectors; scalar aggregation is carried out on the gain adjustment row vectors to generate a plurality of compensation reference quantities; And carrying out double-thread decoupling conversion on the compensation reference quantities, outputting a plurality of temperature compensation strategies and a plurality of fan compensation strategies, and forming the neighborhood compensation strategies.
2. The cloud-coordinated power consumption dynamic sensing and energy-saving optimization control method of claim 1, wherein the energy-saving optimization cloud receives a plurality of multi-source space-time data streams returned by the plurality of edge intelligent sensing nodes, performs load people stream coupling prediction, and generates a plurality of initial regulation strategies, and comprises the following steps: performing space-time fusion modeling on a first multi-source space-time data stream by taking a first independent control area as a space boundary to generate a first environment state tensor; Performing space-time characteristic decoupling extraction on the first environmental state tensor to obtain a first region residence index, a first cross-region migration intensity, a first air conditioner power change gradient and a first thermal inertia coefficient; And taking the first regional residence index, the first cross-region migration intensity, the first air conditioner power change gradient and the first thermal inertia coefficient as physical information decision factors to perform load people stream coupling prediction, so as to generate a first initial regulation strategy.
3. The cloud-coordinated power consumption dynamic sensing and energy saving optimization control method of claim 2, wherein the load induced abortion coupling prediction is performed by taking the first regional residence index, the first cross-region migration intensity, the first air conditioner power variation gradient and the first thermal inertia coefficient as physical information decision factors, and generating a first initial regulation strategy comprises: a load stream of people coupling prediction model is pre-constructed, wherein the load stream of people coupling prediction model comprises a time convolution branch, a drawing meaning branch and an attention fusion layer, the time convolution branch and the drawing meaning branch are connected in parallel, and an output end is converged into the attention fusion layer; Inputting the first air conditioner power variation gradient and a first thermal inertia coefficient into the time convolution branch, and extracting a dynamic load response vector and a thermal inertia time code through an expansion causal convolution layer; inputting the first region resident index and the first cross-region migration intensity into the graph meaning force branch to perform space neighborhood aggregation, and outputting a space people stream coupling matrix and a region thermal potential field, wherein the time convolution branch and the graph meaning force branch execute feature extraction operation in parallel; The weighted energy fusion of the dynamic load response vector, the thermal inertia time code, the space stream coupling matrix and the regional thermal potential field is carried out on the attention fusion layer, and a predicted load time sequence curve is output; And performing rolling time domain optimization output of the first initial regulation strategy based on the predicted load time sequence curve.
4. The cloud-coordinated power consumption dynamic sensing and energy-saving optimization control method of claim 2, wherein performing space-time fusion modeling of a first multi-source space-time data stream with a first independent control region as a spatial boundary to generate a first environmental state tensor comprises: Extracting first flow thermodynamic diagram stream, first device power timing stream, first ambient temperature and humidity, and first CO 2 concentration data from a first multi-source spatiotemporal data stream; Performing space-time point cloud rigid registration of the first thermodynamic flow and the first equipment power time sequence flow by taking a first independent control area as a space boundary to generate a first fusion space-time grid; And on the first fusion space-time grid, fusing the first environment temperature and humidity and the first CO 2 concentration data through tensor channel stacking, and constructing the first environment state tensor.
5. The cloud-coordinated power consumption dynamic sensing and energy-saving optimization control method of claim 1, further comprising: Synchronously sensing real-time dynamic running states of the plurality of independent control areas in the process of executing the plurality of dynamic optimization strategies by the plurality of edge intelligent sensing nodes to obtain a plurality of multi-source incremental data streams; According to the abnormal fluctuation characteristics of the multiple multi-source incremental data streams, performing control deviation identification of the multiple independent control areas, and positioning P abnormal control areas; and executing fault positioning operation and maintenance of the power equipment on the P abnormal control areas.
6. The cloud-coordinated power consumption dynamic sensing and energy-saving optimization control method of claim 1, further comprising: performing crisis level quantification according to the composite degradation trend of the multiple multi-source space-time data streams, and outputting multiple regional degradation indexes; Traversing the region degradation indexes by adopting a preset dynamic degradation threshold value to screen K high-risk control regions; carrying out short-time domain trend prediction on K multi-source space-time data streams to obtain K groups of running state extremum; And matching K limit regulation strategies according to the K groups of running state extremum, and performing high-response priority strong intervention control on the K high-risk control areas.
7. The cloud-coordinated power consumption dynamic sensing and energy-saving optimization control system is characterized by being used for implementing the cloud-coordinated power consumption dynamic sensing and energy-saving optimization control method according to any one of claims 1-6, and comprises the following steps: the local deployment module is used for carrying out local deployment of the heterogeneous sensing array according to the partition topological structure of the central air conditioner, and constructing a plurality of edge intelligent sensing nodes corresponding to a plurality of independent control areas, wherein the plurality of edge intelligent sensing nodes and the energy-saving optimized cloud are in bidirectional communication connection through an RS485 bus; The coupling prediction module is used for receiving a plurality of multi-source space-time data streams returned by the plurality of edge intelligent sensing nodes by the energy-saving optimization cloud, carrying out load people stream coupling prediction and generating a plurality of initial regulation strategies; The deviation matrix generation module is used for generating a regional energy consumption deviation matrix by comparing the plurality of regulation feedback data with the plurality of initial regulation strategies; The deviation matrix correction module is used for correcting the regional energy consumption deviation matrix according to a plurality of thermodynamic coupling coefficients of the independent control regions to obtain a plurality of neighborhood compensation strategies; The space-time collaborative optimization module is used for carrying out space-time collaborative optimization of the plurality of initial regulation strategies based on the plurality of neighborhood compensation strategies to generate a plurality of dynamic optimization strategies; And the closed-loop control module is used for the energy-saving optimization cloud to issue the dynamic optimization strategies to the intelligent edge sensing nodes and conduct the regional autonomous closed-loop control of the independent control areas.

Description

Cloud-coordinated power consumption dynamic sensing and energy-saving optimal control method and system Technical Field The invention relates to the technical field of energy-saving control, in particular to a cloud-coordinated power consumption dynamic sensing and energy-saving optimal control method and system. Background Energy conservation and emission reduction have become key problems to be solved in various industries, especially in the building industry, and the energy consumption of equipment such as an air conditioning system, a lighting system, a ventilation system and the like of a building occupies a large part of the total energy consumption of the building. The traditional method is mainly based on a fixed mode or timing control to control regional power consumption, generally does not have real-time monitoring and feedback functions, and the control mode lacking real-time feedback cannot be dynamically adjusted according to the actual use condition of each region in a building, so that the energy consumption of partial regions is excessively high or excessively low, unbalanced distribution of energy efficiency is finally caused, the energy consumption of partial regions is excessively high, energy waste is caused, the power consumption is increased, and the overall efficiency of the system is reduced. Disclosure of Invention The application provides a cloud cooperative power energy consumption dynamic sensing and energy saving optimization control method and system, and aims to solve the technical problems that in the prior art, regional power energy consumption control is mainly performed based on a fixed mode or timing control, dynamic adjustment cannot be performed according to actual use conditions of all regions in a building, so that energy consumption of partial regions is too high or too low, and overall efficiency of the system is reduced. The first aspect of the application discloses a cloud coordinated power consumption dynamic sensing and energy-saving optimization control method, which comprises the steps of carrying out local deployment of a heterogeneous sensing array according to a partition topological structure of a central air conditioner, constructing a plurality of edge intelligent sensing nodes corresponding to a plurality of independent control areas, wherein the plurality of edge intelligent sensing nodes and the energy-saving optimization cloud are in bidirectional communication connection through an RS485 bus, the energy-saving optimization cloud receives a plurality of multi-source space-time data streams returned by the plurality of edge intelligent sensing nodes, carries out load people stream coupling prediction, generates a plurality of initial regulation strategies, corrects the area energy consumption deviation matrix according to a plurality of thermal coupling coefficients of the plurality of independent control areas, generates a plurality of dynamic optimization strategies based on space-time cooperative optimization of the plurality of initial regulation strategies, and sends the plurality of dynamic optimization strategies to the plurality of edge intelligent sensing nodes to carry out closed-loop neighborhood compensation of the plurality of independent control areas. The application discloses a second aspect of the cloud collaborative power energy consumption dynamic sensing and energy saving optimization control system, which is used for carrying out load human flow coupling prediction to generate a plurality of initial regulation strategies, a bias matrix generation module used for generating an area energy consumption bias matrix by comparing a plurality of regulation feedback data with the plurality of initial regulation strategies, a bias matrix correction module used for correcting the area energy consumption bias matrix according to a plurality of thermal coupling coefficients of a plurality of independent control areas to obtain a plurality of neighborhood compensation strategies, a coupling prediction module used for receiving a plurality of multi-source space-time data flows returned by the plurality of edge intelligent sensing nodes by the energy saving optimization cloud, and a bias matrix generation module used for carrying out closed-loop dynamic optimization on the plurality of independent control strategies under the control of the cloud. The one or more technical schemes provided by the application have at least the following beneficial effects: By arranging a plurality of edge intelligent sensing nodes in each independent control area of the central air conditioner, the running state and the energy consumption condition of each area can be sensed in real time, the sensing nodes are in bidirectional communication with the energy-saving optimized cloud through an RS485 bus, real-time transmission and feedback of data are ensured, and highly accurate area sensing and dynamic control are realized; the energy-saving optimization cloud end