A building data governance system and method based on multi-modal data fusion

Abstract

The application belongs to the technical field of building data governance, and relates to a building data governance system and method based on multi-modal data fusion, which acquires real-time operation data of a multi-source perception network of a target building, performs cross-domain data correlation analysis on the real-time operation data to generate an energy consumption distribution state diagram, retrieves a synergistic abnormal area of light intensity and medium through spatial coordinate mapping, generates an energy-saving early warning signal based on the topological position of the synergistic abnormal area, dynamically adjusts energy regulation strategy parameters distributed to different functional areas based on the energy consumption distribution state diagram and the energy-saving early warning signal, collects comparison feedback data before and after energy regulation operation, analyzes energy-saving effect performance corresponding to energy regulation, corrects energy regulation strategy parameters accordingly and triggers a new round of cross-domain data correlation analysis until a preset energy-saving target is reached, thereby significantly improving building data governance level and energy-saving and consumption-reducing capability.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of building data governance, and in particular, relates to a building data governance system and method based on multi-modal data fusion. BACKGROUND

[0002] Building energy management is the key to achieving energy saving and sustainable development. In large public buildings and commercial complexes, the optimization and fine management of energy consumption are particularly important. The core is to collect, analyze and control various energy data in buildings to improve utilization efficiency, reduce energy consumption and achieve energy saving and emission reduction goals. However, building energy systems involve complex interactions of multiple subsystems such as electricity, heating, ventilation, air conditioning, lighting, water supply and drainage, and the data types of each subsystem are diverse and scattered, which brings significant challenges to data integration and governance.

[0003] The prior art for building energy management often takes single data source monitoring and analysis as the core mode: through electric meters, water meters, gas meters and other devices, various energy use is independently monitored, trend prediction and energy saving optimization are carried out relying on historical data, and in the data processing process, discrete algorithm mode is often used. After separate analysis of different data sources, the data is processed.

[0004] However, the above technical means has certain limitations in dealing with complex building energy management requirements, which is specifically manifested as follows:

[0005] Building energy management involves a variety of data types and dynamic changes, and single data source monitoring and analysis methods are difficult to fully reflect the overall operation status of the building. The existing data processing method lacks adaptability when facing multi-modal data, and is prone to data island phenomenon, which cannot effectively mine the potential association between data.

[0006] The existing building energy saving optimization strategy is mostly based on static rules or experience models, which lacks the ability to respond quickly to real-time dynamic changes, which may lead to inaccurate energy regulation and even unnecessary energy waste, resulting in the overall effect of building energy management being restricted, and the potential of multi-modal data in energy saving optimization being difficult to fully play. SUMMARY

[0007] In view of this, in order to solve the problems raised in the background art, a building data governance system and method based on multi-modal data fusion are proposed.

[0008] The technical solution adopted by the application to solve its technical problems is as follows: In the first aspect, the application provides a building data governance system based on multi-modal data fusion, which comprises a data acquisition module, a data analysis module, an energy regulation module and an energy saving optimization module.

[0009] The data acquisition module is connected with a data analysis module, the data analysis module is connected with an energy regulation module, the energy regulation module is connected with an energy saving optimization module, and the energy saving optimization module is connected with the data analysis module.

[0010] The data acquisition module acquires real-time operation data of a multi-source perception network of the target building.

[0011] The data analysis module performs cross-domain data correlation analysis on the real-time operation data to generate an energy consumption distribution state diagram, and retrieves a synergistic abnormal area of light intensity and medium through spatial coordinate mapping, and generates an energy saving early warning signal based on the topological position of the synergistic abnormal area.

[0012] The energy regulation module dynamically adjusts energy regulation strategy parameters distributed to different functional areas by comprehensively considering the energy consumption distribution state diagram and the energy saving early warning signal.

[0013] The energy saving optimization module collects comparison feedback data before and after the energy regulation operation, analyzes the energy saving effect corresponding to the energy regulation, corrects the energy regulation strategy parameters accordingly, and triggers a new round of cross-domain data correlation analysis until the preset energy saving target is reached.

[0014] In a second aspect, the present application provides a building data governance method based on multi-modal data fusion, comprising: acquiring real-time operation data of a multi-source perception network of a target building.

[0015] Performing cross-domain data correlation analysis on the real-time operation data to generate an energy consumption distribution state diagram, and retrieving a synergistic abnormal area of light intensity and medium through spatial coordinate mapping, and generating an energy saving early warning signal based on the topological position of the synergistic abnormal area.

[0016] The energy regulation module dynamically adjusts energy regulation strategy parameters distributed to different functional areas by comprehensively considering the energy consumption distribution state diagram and the energy saving early warning signal.

[0017] The energy saving optimization module collects comparison feedback data before and after the energy regulation operation, analyzes the energy saving effect corresponding to the energy regulation, corrects the energy regulation strategy parameters accordingly, and triggers a new round of cross-domain data correlation analysis until the preset energy saving target is reached.

[0018] Compared with the prior art, the embodiments of the present application have at least the following advantages or beneficial effects:

[0019] The present application divides the target area into each energy consumption intensity level area and locates the synergistic abnormal area triggering the energy saving early warning signal through multi-modal data fusion, which not only realizes rapid response and accurate tracing of energy abnormality and avoids data island phenomenon, but also provides accurate basis for differentiated energy saving regulation.

[0020] The application formulates a zoning energy regulation strategy, according to the energy consumption intensity level and energy saving early warning characteristics of different regions, accurately configures differentiated regulation instructions and regulation rates, not only fully releases the energy saving potential of different regions, but also realizes the dynamic balance of energy supply and demand, effectively improves the scientificity and effectiveness of the regulation measures, and promotes the building energy management from extensive to fine.

[0021] The application analyzes the energy regulation corresponding energy saving effect, corrects the energy regulation strategy parameters and triggers a new round of cross-domain data correlation analysis, so as to realize the closed-loop optimization mechanism to continuously promote the energy saving target, and significantly improve the building data governance level and energy saving and consumption reduction ability. BRIEF DESCRIPTION OF DRAWINGS

[0022] The application is further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the application. For ordinary skilled in the art, other drawings can be obtained without creative labor on the basis of the following drawings.

[0023] Figure 1 The system module structure block diagram provided for the first embodiment of the application.

[0024] Figure 2 The logic flow chart of the data analysis module in the first embodiment of the application.

[0025] Figure 3 The method step flow chart provided for the second embodiment of the application. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the application will be described clearly and completely with reference to the accompanying drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all. Based on the embodiments in the application, all other embodiments obtained by ordinary skilled in the art without creative labor are within the protection scope of the application.

[0027] Embodiment one

[0028] As shown in Figure 1 The first embodiment of the application provides a building data governance system based on multi-modal data fusion, which comprises a data acquisition module, a data analysis module, an energy regulation module and an energy saving optimization module.

[0029] The data acquisition module is connected with the data analysis module, the data analysis module is connected with the energy regulation module, the energy regulation module is connected with the energy saving optimization module, and the energy saving optimization module is connected with the data analysis module.

[0030] The data acquisition module acquires real-time operation data of the target building multi-source perception network.

[0031] In a preferred embodiment of the present application, the real-time operation data acquisition of the target building multi-source perception network includes: based on the current sensors, temperature and humidity sensors, light sensors and fluid flow sensors deployed in the target building, collecting monitoring data of various sensors in the detection range covered by their fixed layout positions.

[0032] Extract the minimum detection coverage range of each sensor, and discretize the target building space into several spatial coordinate units accordingly.

[0033] It should be noted that the detection coverage range parameters of the above-mentioned various sensors can refer to the standard detection space range defined in the technical specification provided by the sensor manufacturer. In order to ensure the standardization and uniformity of space division, the coverage range of the standard detection space range needs to be standardized in the form of regular cuboid grid, which provides accurate and consistent reference units for subsequent space discretization operations.

[0034] Collect each sensor monitoring data according to the spatial mapping relationship to the corresponding spatial coordinate unit, and output the current fluctuation data, environmental change data, light intensity distribution data and medium flow data of each spatial coordinate unit in real time.

[0035] It should be noted that the specific example of the above real-time operation data is explained as follows: the current fluctuation data is used to directly reflect the running load of the electrical equipment of the target building such as air conditioner, lighting, office equipment, etc., at least including real-time current effective value, instantaneous power peak value, power factor and current harmonic distortion rate, which is the core index for determining the electrical energy consumption.

[0036] The environmental change data is related to the operation intensity of the heating ventilation air conditioner of the target building, at least including environmental temperature and environmental humidity, when the environmental temperature and humidity deviate from the comfort interval, the air conditioner load will increase significantly, and it has a strong coupling relationship with the electrical energy consumption.

[0037] The light intensity distribution data directly corresponds to the lighting energy consumption of the target building, at least including illumination intensity, illumination uniformity and natural light penetration rate, artificial lighting load increases when the illumination is insufficient, and excessive illumination will cause energy waste, which is the key basis for determining the lighting energy consumption characteristics.

[0038] The medium flow data involves parameters such as cold and hot water flow, pipeline pressure and fluid temperature, which reflects the running state of water supply and drainage or air conditioning water circulation in the target building, and abnormal fluctuation of flow often means inefficient operation or leakage of equipment, which is a direct manifestation of water energy consumption.

[0039] For example, Figure 2The data analysis module performs cross-domain data correlation analysis on the real-time operation data to generate an energy consumption distribution state diagram, and retrieves a synergistic abnormal area of light intensity and medium through spatial coordinate mapping, and generates an energy saving warning signal based on the topological position of the synergistic abnormal area.

[0040] In a preferred embodiment of the present application, the cross-domain data correlation analysis on the real-time operation data to generate an energy consumption distribution state diagram comprises: performing spatio-temporal consistency verification on current fluctuation data and environmental change data, identifying abnormal data points and applying sliding window mean replacement.

[0041] It should be noted that the spatio-temporal consistency verification process is: aligning the current fluctuation data sequence and the environmental change data sequence of the same spatial coordinate unit according to the collection time stamp, quantifying the data mean and data standard deviation in the sliding time window, and if the absolute deviation of a data point value in the window relative to the mean exceeds a preset multiple of the standard deviation, then the data point is marked as an abnormal data point.

[0042] According to the mapping relationship between the verified current fluctuation data and environmental change data, energy consumption intensity level division is performed on each spatial coordinate unit of the target building, and connected domain search is performed on the same level units to form an initial energy consumption intensity level area.

[0043] It should be noted that the energy consumption intensity level division process of each spatial coordinate unit of the target building is: performing correlation analysis of each parameter in the current fluctuation data and the environmental change data through Pearson correlation coefficient calculation, constructing a Pearson correlation coefficient matrix containing all parameter combinations between the two data, based on the actual influence degree of different parameter combinations on energy consumption, constructing a weight coefficient matrix consistent with the dimension of the Pearson correlation coefficient matrix, the weight coefficient assignment process can be determined by analytic hierarchy process or historical data sensitivity analysis, multiplying the Pearson correlation coefficient matrix and the weight coefficient matrix to obtain a fusion matrix, and the cumulative value of the elements of the fusion matrix is used as the comprehensive correlation strength of the current fluctuation data and the environmental change data.

[0044] The comprehensive correlation strength of the spatial coordinate unit and the current fluctuation data are integrated by fuzzy logic method to map the energy consumption intensity level, the specific steps are as follows: for the comprehensive correlation strength reflecting the matching rationality of energy consumption and environment and the current fluctuation data reflecting the size of the power load, according to their typical performance characteristics in the actual scene, several fuzzy categories are divided, these fuzzy categories need to completely cover the common value range of the parameters, and a reasonable overlapping interval is reserved between adjacent categories to effectively handle the fuzziness when the parameters are in a critical state, the corresponding fuzzy category of the spatial coordinate unit is retrieved based on the actual parameter value of the spatial coordinate unit, and a fuzzy category set composed of the comprehensive correlation strength fuzzy category and the current fluctuation parameter fuzzy category is formed.

[0045] Based on the fuzzy category set obtained above, corresponding trigger rules are retrieved and matched in a preset fuzzy rule base, which is a condition-conclusion database constructed according to professional experience and actual scene characteristics of energy consumption analysis, and the core construction points include:

[0046] i. Rule antecedent composition: formed by the combination of fuzzy categories of comprehensive correlation strength and fuzzy categories of current fluctuation parameters, which needs to completely cover all possible performance combinations of the two types of parameters. ii. Rule consequent determination: the consequent is the corresponding energy consumption intensity level, and the level division needs to reflect both energy consumption scale and matching rationality. For example, when the comprehensive correlation strength is in a higher category and the current fluctuation parameter is in a lower category, a lower energy consumption intensity level is corresponded. When the comprehensive correlation strength is in a lower category and the current fluctuation parameter is in a higher category, a higher energy consumption intensity level is corresponded.

[0047] iii. Rule set property requirement: all parameter combinations have corresponding rules and no mutually contradictory rule conclusions.

[0048] Based on the membership degree calculation result of the trigger rules, the energy consumption intensity levels corresponding to each rule and their membership degrees are integrated, the energy consumption intensity quantitative index is calculated by the gravity method, and the calculated energy consumption intensity quantitative index is corresponded to a specific energy consumption intensity level by combining a preset level mapping standard. The mapping standard clearly defines the quantitative index interval corresponding to each level.

[0049] The spatial gradient feature of the superimposed light intensity distribution data performs sub-grid level correction on the boundary of the initial energy consumption intensity level region.

[0050] It should be noted that the boundary correction process of the initial energy consumption intensity level region is as follows:

[0051] The light intensity distribution data is preprocessed, the spatial illuminance data collected by the light intensity sensor is interpolated and mapped to the spatial coordinate grid system of the target building, a light intensity distribution matrix aligned with the initial energy consumption intensity level region grid is generated, and on this basis, sub-grid level light intensity data is generated through spatial interpolation technology, realizing high-precision representation of light intensity details inside the grid cell.

[0052] Based on the sub-grid level light intensity data, the light intensity gradient amplitude of each sub-grid cell is calculated using the Sobel operator or gradient operator to quantify the spatial gradient feature of the light intensity distribution.

[0053] Locating the to-be-corrected region by correlation analysis of the light intensity gradient feature and the initial energy consumption boundary: extracting the boundary line of the initial energy consumption level region, synchronously identifying the sub-grid cells on both sides of the boundary line, calculating the light intensity gradient amplitude of the sub-grid cells on both sides of the boundary line, if the gradient amplitude exceeds the light intensity gradient threshold value set based on the building lighting design standard, and there is a difference in the initial energy consumption level on both sides, then mark the boundary segment as the to-be-corrected region.

[0054] Performing fine adjustment on the marked to-be-corrected boundary segment: tracking the continuous distribution trajectory of the high-gradient sub-grid cells along the light intensity gradient direction, determining the actual physical boundary of the light intensity mutation, and taking the actual physical boundary as the reference, if there is a deviation between the physical boundary pointed by the light intensity gradient and the initial energy consumption level boundary, correcting the boundary line to the light intensity mutation position, if the energy consumption levels on both sides need to be further subdivided due to the light intensity difference, then adding a subdivision boundary in the sub-grid cell, and realizing the accurate division of the energy consumption level region.

[0055] After the correction, the rationality of the results is verified in two aspects: one is to check whether the corrected boundary is consistent with the physical characteristics of the light intensity distribution, and the other is to compare the consistency of the energy consumption level regions before and after the correction with the actual electricity data, if there is a deviation, then dynamically adjusting the light intensity gradient threshold value or the correction amplitude, and finally outputting the energy consumption intensity level region corrected at the sub-grid level.

[0056] Outputting a three-dimensional energy consumption distribution state diagram marking the topological boundaries of the high-energy consumption area, the medium-energy consumption area and the low-energy consumption area.

[0057] In a preferred embodiment of the present application, the cooperative abnormal region retrieval process comprises: performing fluctuation feature extraction and event detection on the time sequence composed of the light intensity distribution data and the medium flow data of each space coordinate unit respectively.

[0058] Establishing an event correlation rule, and the rule content is: when the light intensity distribution data detects an illumination sudden change event and the medium flow data synchronously detects a flow rate mutation event, the time difference between the two is within a preset time window and occurs in the same space coordinate unit, the space coordinate unit is marked as a cooperative abnormal unit.

[0059] It should be noted that the specific detection process of the illumination sudden change event is: performing first-order derivative operation on the light intensity distribution data to obtain the light intensity change rate, if there is a space coordinate point with an absolute value of light intensity change rate greater than a preset light intensity change rate threshold value, then it indicates that there is an illumination sudden change event.

[0060] The specific detection process of the flow rate mutation event is that: the time series second derivative operation is performed on the medium flow data to obtain the flow change acceleration, and if the flow change acceleration of a space coordinate point exceeds the preset amplitude of the baseline value, it indicates that there is a medium mutation event, wherein the baseline value refers to the running mean value of the medium flow change acceleration calculated from the historical normal operation data, and the preset amplitude can be exemplarily a double value of the standard deviation.

[0061] The connected domain search is performed on the cooperative abnormal units, the cooperative abnormal units that are spatially continuous are merged to form a cooperative abnormal region, and the minimum circumscribed rectangle boundary coordinate set thereof is output.

[0062] The embodiment of the application divides the target region into regions of different energy consumption intensity levels and locates the cooperative abnormal region triggering the energy saving warning signal through multi-modal data fusion, realizes rapid response and accurate tracing of energy anomalies, avoids the phenomenon of data island, and provides accurate basis for differentiated energy saving regulation.

[0063] The energy regulation module dynamically adjusts the energy regulation strategy parameters distributed to different functional regions in combination with the energy consumption distribution state diagram and the energy saving warning signal.

[0064] In a preferred embodiment of the application, the dynamic adjustment process of the energy regulation strategy parameters of the different functional regions includes: allocating a load reduction instruction to the high energy consumption area marked in the energy consumption distribution state diagram, and the instruction includes device load reduction priority sorting and load reduction amplitude threshold.

[0065] It should be noted that the device load reduction priority sorting refers to the hierarchical arrangement of the devices that need to perform the load reduction operation in the load reduction instruction of the high energy consumption area according to the load reduction priority order. The sorting logic has the dual goals of maximizing energy saving benefits and minimizing functional impact, that is, the devices with high energy consumption proportion and small impact on the core function of the building after load reduction are preferentially executed to ensure maximum energy saving effect under the same reduction amount, while avoiding affecting the key function. The device load reduction priority sorting can be pre-calibrated according to the energy consumption contribution degree, functional importance, load reduction sensitivity and operation stability of the historical operation of the devices in different functional regions before system development, so as to be called in the execution process.

[0066] The load reduction amplitude threshold refers to the maximum load reduction proportion or absolute amount limit of a single device when performing the load reduction operation on the devices in the high energy consumption area, which can refer to the minimum load proportion allowed by the technical specification of the device.

[0067] The dynamic balance regulation instruction is allocated to the medium energy consumption area, and an adaptive adjustment mode is matched according to the time-varying characteristics of the functional area to which it belongs.

[0068] It should be noted that the adaptive adjustment mode matching process is specifically based on the current use period of the functional area, the people flow density and the environmental demand, for example, the office area needs to maintain a high lighting brightness (300-500 lux) and air conditioning comfort (24-26 DEG C) during the working period (9:00-18:00), and automatically reduces the lighting brightness (less than or equal to 100 lux) and increases the air conditioning set temperature (26-28 DEG C) during the non-working period to reduce the energy consumption.

[0069] For the low energy consumption area, the load distribution maintenance instruction is distributed, the power supply of the core device is locked, and the unnecessary load is stopped.

[0070] In a preferred embodiment of the present application, the dynamic adjustment process of the energy regulation strategy parameters of the different functional areas further comprises: analyzing the coordinated abnormal area coordinates in the energy saving warning signal, and associating the energy consumption intensity level label of the corresponding area in the energy consumption distribution state diagram.

[0071] If the associated area is a high energy consumption area, the load shedding rate is increased to a reference preset multiple to trigger a rapid response load shedding, and a standby energy supply mode is switched to.

[0072] If it is a medium energy consumption area, an elastic rate correction is performed for the current adaptive adjustment mode.

[0073] If it is a low energy consumption area, a stepwise load recovery is started and the power supply of the core device is maintained.

[0074] The embodiment of the present application formulates a partitioned energy regulation strategy, accurately configures differentiated regulation instructions and regulation rates according to the energy consumption intensity level and energy saving warning characteristics of different areas, not only fully releases the energy saving potential of different areas, but also realizes the dynamic balance of energy supply and demand, effectively improves the scientificity and effectiveness of the regulation measures, and promotes the building energy management from extensive to fine.

[0075] The energy saving optimization module collects comparison feedback data before and after the energy regulation operation, analyzes the energy saving effect performance corresponding to the energy regulation, corrects the energy regulation strategy parameters and triggers a new round of cross-domain data association analysis until the preset energy saving target is reached.

[0076] In a preferred embodiment of the present application, the comparison feedback data collection process before and after the energy regulation operation comprises: synchronously acquiring the current parameter set and the environmental parameter set of each space coordinate unit in the pre-regulation steady state period and the post-regulation intermittent period, the current parameter set comprising current effective value, harmonic distortion rate and power factor, and the environmental parameter set comprising temperature, humidity and personnel distribution density.

[0077] The current original feature and the current reconstructed feature are obtained by performing normalization processing on the current parameter set.

[0078] The environmental reconstruction features and the environmental original features are obtained by performing thermal comfort index calculation on the environmental parameter set.

[0079] It should be noted that the thermal comfort index is generally calculated by using the internationally recognized PMV-PPD model, and the temperature in the environmental parameter set is taken as the dry-bulb temperature, the humidity is converted into a correction coefficient for latent heat dissipation by calculating the difference between the saturation water vapor pressure and the actual water vapor pressure, and the personnel distribution density is input into the PMV-PPD model by correlating the human metabolic rate and the activity state to obtain the output thermal comfort index. The PMV-PPD model is a prior art and will not be described here.

[0080] Based on the comparison and analysis of the original features and the reconstruction features, a current optimization rate matrix and an environmental comfort change rate matrix are generated, which are indexed by space coordinates and time slices, respectively, and constitute the comparison and feedback data.

[0081] In a preferred embodiment of the present application, the energy-saving effect performance of the energy regulation is analyzed by assigning fusion weights to the elements of the current optimization rate matrix and the environmental comfort change rate matrix according to the element dispersion of the matrix, performing linear weighted fusion processing on the current optimization rate matrix and the environmental comfort change rate matrix, obtaining the energy-saving effect indicators of each space coordinate unit, extracting the energy-saving effect indicators of all space coordinate units in the same energy consumption intensity level region, and taking the minimum value as the energy-saving efficiency of the level region. The energy-saving effect performance of the energy regulation in the energy consumption distribution state diagram is analyzed.

[0082] In a preferred embodiment of the present application, the energy regulation strategy parameter correction process includes:

[0083] The energy-saving efficiency of the energy consumption region in the energy consumption distribution state diagram is compared with the baseline energy-saving efficiency threshold pre-marked for the corresponding energy consumption intensity level, and the energy-saving property of each energy consumption intensity level region is determined according to the energy-saving efficiency standard amplitude. The energy-saving property includes one of over-saturation, moderate and low efficiency.

[0084] It should be noted that the determination of the energy-saving property of each energy consumption intensity level region according to the energy-saving efficiency standard amplitude specifically refers to: when the energy-saving efficiency of the energy consumption region is less than or equal to the first preset percentage of the baseline energy-saving efficiency threshold pre-marked for the corresponding energy consumption intensity level, the energy-saving property is determined as the low efficiency property.

[0085] When the energy-saving efficiency of the energy consumption region is greater than or equal to the second preset percentage of the baseline energy-saving efficiency threshold pre-marked for the corresponding energy consumption intensity level, the energy-saving property is determined as the over-saturation property.

[0086] The rest is considered as moderate property, wherein the first preset percentage <1< the second preset percentage.

[0087] Differentiation adjustment is made to the key parameters in the energy regulation strategy according to different energy-saving properties and energy consumption area types, and the load regulation accuracy is cooperatively optimized based on the energy-saving property maintenance period.

[0088] It should be noted that the above differentiation adjustment specifically includes:

[0089] (A) High energy consumption area correction: When the energy-saving efficiency is lower than the threshold value: improve the device load reduction priority, increase the load reduction amplitude upper limit value, and shorten the standby energy switching time delay.

[0090] When the energy-saving efficiency exceeds the threshold value: reduce the load reduction priority, reduce the load reduction amplitude upper limit value, and switch back to the main energy supply.

[0091] (B) Medium energy consumption area correction: When the energy-saving efficiency is lower than the threshold value: reduce the deviation tolerance of dynamic balance regulation, and increase the adjustment frequency.

[0092] When the energy-saving efficiency exceeds the threshold value: expand the deviation tolerance of dynamic balance regulation, and reduce the adjustment frequency.

[0093] (C) Low energy consumption area correction: When the energy-saving efficiency is lower than the threshold value: improve the core device power lock precision, and expand the unnecessary load shutdown range.

[0094] When the energy-saving efficiency exceeds the threshold value: accelerate the stepwise load recovery rate, and dynamically relax the power lock threshold value.

[0095] (D) Global coordination correction: For areas that are still low efficiency for two consecutive regulation periods: improve the load regulation instruction execution accuracy level.

[0096] For areas that are oversaturated for three consecutive regulation periods: extend the regulation time interval.

[0097] The embodiment of the present application analyzes the energy regulation corresponding energy-saving effect, corrects the energy regulation strategy parameters and triggers a new round of cross-domain data correlation analysis, so as to realize the closed-loop optimization mechanism to continuously promote the energy-saving goal, and significantly improve the building data governance level and energy-saving and consumption-reducing ability.

[0098] Embodiment two

[0099] As shown in Figure 3 The second embodiment of the present application provides a building data governance method based on multi-modal data fusion, which comprises: acquiring real-time running data of a target building multi-source perception network.

[0100] Performing cross-domain data correlation analysis on the real-time running data to generate an energy consumption distribution state diagram, and retrieving the synergistic abnormal area of light intensity and medium through spatial coordinate mapping, generating an energy-saving warning signal based on the topological position of the synergistic abnormal area.

[0101] The energy consumption distribution state diagram and the energy saving early warning signal are used to dynamically adjust the energy regulation strategy parameters distributed to different functional areas.

[0102] The comparison feedback data before and after the energy regulation operation is collected, the energy saving effect corresponding to the energy regulation is analyzed, the energy regulation strategy parameters are corrected, a new round of cross-domain data correlation analysis is triggered, and the preset energy saving target is reached.

[0103] The implementation principle and the technical effects of the building data governance method based on multi-modal data fusion provided by the embodiment of the application are the same as those of the foregoing system embodiment, and for brevity of description, the part not mentioned in the method embodiment can refer to the corresponding content in the foregoing system embodiment.

[0104] The foregoing is merely an example and description of the structure of the application, and those skilled in the art of the technical field to which the application belongs can make various modifications or supplements to the described specific embodiments or replace them with similar ways, as long as the modifications or supplements do not deviate from the structure of the application or exceed the scope defined by the application, and all of them shall belong to the protection scope of the application.

Claims

1. A building data governance system based on multi-modal data fusion, characterized in that, The method comprises the following steps: a data acquisition module acquires real-time operation data of a target building multi-source perception network; a data analysis module performs cross-domain data correlation analysis on the real-time operation data to generate an energy consumption distribution state diagram, and retrieves a synergistic abnormal area of light intensity and medium through spatial coordinate mapping, and generates an energy saving warning signal based on the topological position of the synergistic abnormal area; an energy regulation module dynamically adjusts energy regulation strategy parameters distributed to different functional areas based on the energy consumption distribution state diagram and the energy saving warning signal; an energy saving optimization module collects comparison feedback data before and after energy regulation operation, analyzes the energy saving effect of energy regulation, and accordingly corrects the energy regulation strategy parameters and triggers a new round of cross-domain data correlation analysis until the preset energy saving target is reached; The real-time operation data of the target building multi-source perception network is acquired by: based on the current sensor, temperature and humidity sensor, light sensor and fluid flow sensor deployed in the target building, collecting the monitoring data of each type of sensor in the detection range covered by its fixed layout position; the minimum detection coverage range in each type of sensor is extracted, and accordingly the target building space is discretized into a plurality of spatial coordinate units; the monitoring data of each sensor is collected to the corresponding spatial coordinate unit according to the spatial mapping relationship, and the current fluctuation data, environmental change data, light intensity distribution data and medium flow data of each spatial coordinate unit are output in real time; The synergistic abnormal area retrieval process includes: extracting fluctuation characteristics and event detection from the time series composed of the light intensity distribution data and the medium flow data of each spatial coordinate unit; establish event correlation rules, the rule content is: when the light intensity distribution data detects a sudden change event and the medium flow data detects a sudden change event, the time difference is within the preset time window and occurs in the same spatial coordinate unit, mark the spatial coordinate unit as a synergistic abnormal unit; perform connected component search on the synergistic abnormal unit, merge the spatially continuous synergistic abnormal units to form a synergistic abnormal area, and output its minimum bounding rectangle boundary coordinate set.

2. The building data governance system based on multi-modal data fusion according to claim 1, characterized in that, The cross-domain data correlation analysis on the real-time operation data to generate an energy consumption distribution state diagram includes: Performing spatio-temporal consistency check on current fluctuation data and environmental change data, identifying abnormal data points and applying sliding window mean replacement; According to the mapping relationship between the current fluctuation data and the environmental change data after verification, perform energy consumption intensity level division on each spatial coordinate unit of the target building, and perform connected component search on the same level unit to form an initial energy consumption intensity level area; Superimpose the spatial gradient characteristics of the light intensity distribution data on the boundary of the initial energy consumption intensity level area to perform sub-grid level correction; Output the three-dimensional energy consumption distribution state diagram labeled with the topological boundaries of high energy consumption area, medium energy consumption area and low energy consumption area.

3. The building data governance system based on multi-modal data fusion according to claim 2, characterized in that, The dynamic adjustment process of the energy regulation strategy parameters of the different functional areas includes: Assigning load reduction instructions to the high energy consumption area marked in the energy consumption distribution state diagram, the instructions include device load reduction priority and load reduction amplitude threshold; Assign dynamic balance regulation instructions to the medium energy consumption area, and match adaptive adjustment mode according to the time-varying characteristics of its functional area. For low energy consumption area distribution load maintenance instruction, lock core device power supply power and shut down unnecessary load.

4. The building data governance system based on multi-modal data fusion of claim 3, wherein, The dynamic adjustment process of the energy regulation strategy parameters of the different functional areas further includes: Resolving the coordinates of the synergistic abnormal area in the energy-saving warning signal, and associating the energy consumption intensity level label of the corresponding area in the energy consumption distribution state diagram; If the associated area is a high energy consumption area, the load shedding rate will be increased to a preset multiple of the baseline to trigger a rapid response load shedding, and the standby energy supply mode will be switched to; If it is a medium energy consumption area, perform flexible rate correction for the current adaptive adjustment mode; If it is a low energy consumption area, start the step-by-step load recovery and maintain the core device power supply power.

5. The building data governance system based on multi-modal data fusion of claim 2, wherein, The comparison feedback data collection process before and after the energy regulation operation includes: Synchronously acquiring the current parameter set and the environment parameter set of each spatial coordinate unit in the steady state period before regulation and the intermittent period after regulation, the current parameter set including current effective value, harmonic distortion rate and power factor, and the environment parameter set including temperature, humidity and personnel distribution density; By performing normalization processing on the current parameter set, the current original feature and the current reconstructed feature are obtained; By performing thermal comfort index calculation on the environment parameter set, the environment reconstructed feature and the environment original feature are obtained; Based on the comparison analysis of the original feature and the reconstructed feature, the current optimization rate matrix and the environment comfort change rate matrix are generated with spatial coordinate units as row indexes and time slices as column indexes, which together constitute the comparison feedback data.

6. The building data governance system based on multi-modal data fusion of claim 5, wherein, The analysis of the energy-saving effect performance corresponding to the energy regulation includes: According to the element dispersion of the current optimization rate matrix and the environment comfort change rate matrix, the fusion weight is assigned to each element of the matrix, and linear weighted fusion processing is performed on the current optimization rate matrix and the environment comfort change rate matrix to obtain the energy-saving effect index of each spatial coordinate unit. The energy-saving effect index of all spatial coordinate units in the same energy consumption intensity level area is extracted, and the minimum value is taken as the energy-saving efficiency of the level area. In this way, the energy-saving effect performance of different energy consumption areas in the energy consumption distribution state diagram after energy regulation is analyzed.

7. The building data governance system based on multi-modal data fusion of claim 6, wherein, The energy regulation strategy parameter correction process includes: Compare the energy-saving efficiency of different energy consumption areas in the energy consumption distribution state diagram with the baseline energy-saving efficiency threshold pre-marked for their corresponding energy consumption intensity level, and determine the energy-saving property of each energy consumption intensity level area according to the energy-saving efficiency amplitude value. The energy-saving property includes one of over-saturation, moderate and low efficiency; For different energy-saving properties and energy consumption area types, the key parameters in the energy regulation strategy are adjusted differently, and the energy-saving property is maintained for a period of time to optimize the load regulation accuracy.

8. A building data governance method based on multi-modal data fusion, the following steps are executed by the building data governance system based on multi-modal data fusion according to any one of claims 1-7, characterized in that, It includes: Obtain the real-time running data of the target building multi-source perception network; Perform cross-domain data association analysis on the real-time running data to generate an energy consumption distribution state diagram, and retrieve the synergistic abnormal area of light intensity and medium based on the topological position of the synergistic abnormal area to generate an energy-saving warning signal; Comprehensively adjust and distribute the energy regulation strategy parameters to different functional areas based on the energy consumption distribution state diagram and the energy-saving warning signal; The comparison feedback data before and after the energy regulation operation is collected, the energy saving effect performance corresponding to the energy regulation is analyzed, the energy regulation strategy parameters are corrected according to the energy saving effect performance, and a new round of cross-domain data correlation analysis is triggered until the preset energy saving target is reached.

Images