Method and equipment for automatically extracting incidence relation between building energy consumption loop and equipment

Abstract

The invention provides a method and equipment for automatically extracting the incidence relation between a building energy consumption loop and equipment, and the method comprises the steps: collecting the power data of an electric meter and the operation state data of the equipment in the energy consumption loop, and carrying out the time alignment; calculating an association score based on a comparison result of the electricity meter power data and the equipment operation state data after time alignment; modeling of noise influence is carried out based on the correlation score; for the equipment and the electricity meters with significant association of the tasks, associating the equipment and the electricity meters to the corresponding same electricity meter loop, and constructing a loop path according to the descending order of association scores; combining the loop paths of the plurality of devices to construct a complete electricity meter tree structure; based on an ammeter tree structure, real-time alarm and fault location are carried out, and an equipment start-stop strategy is automatically optimized, so that the association relationship between the building energy consumption loop and the equipment can be quickly and accurately established, the energy management efficiency is improved, and the energy waste is reduced.

Description

Technical Field

[0001] The invention relates to a method and a device for automatically extracting the association relationship between a building energy circuit and equipment. Background Art

[0002] In large buildings, energy management is a key link. However, due to the complex design of energy loops, the wide variety of equipment, and the fact that design, construction, and equipment installation are completed by different teams, it is difficult to accurately establish the relationship between each device and its energy loop. This situation will lead to the following problems:

[0003] Difficulty in locating faults: When abnormal power consumption is detected in a certain energy-consuming circuit, it is impossible to quickly locate the specific device, delaying fault detection and processing.

[0004] Energy waste: Failure to promptly rectify abnormal energy usage results in energy waste and increased operating costs.

[0005] Inefficient operation and maintenance: The traditional method of sorting out the relationship between energy circuits and equipment requires manual shutdown of meters or circuit breakers to observe which equipment has stopped operating. This method is inefficient and affects the normal operation of the building. Summary of the invention

[0006] The object of the present invention is to provide a method and a device for automatically extracting the relationship between a building energy circuit and equipment.

[0007] In order to solve the above problems, the present invention provides a method for automatically extracting the relationship between building energy circuits and equipment, comprising:

[0008] Collect meter power data and equipment operating status data in the energy consumption circuit and perform time alignment;

[0009] Based on the comparison of the time-aligned meter power data and the equipment operation status data, the correlation score A is calculated. ij Calculation of

[0010] Based on the association score A ij , modeling of noise impact;

[0011] Based on the devices and meters that are significantly associated, a complete meter tree structure is constructed;

[0012] Based on the tree structure of the electricity meter, real-time alarm and fault location are carried out, and the start and stop strategies of the equipment are automatically optimized.

[0013] Furthermore, in the above method, based on determining the devices and meters that are significantly associated, a complete tree structure of the meters is constructed, including: for the devices e that are significantly associated with the tasks i With electric meter j, associated to the corresponding same meter circuit, from the association score A ij Sort from high to low to build a loop path; connect multiple devices i The loop paths are merged to build a complete meter tree structure.

[0014] Furthermore, in the above method, the power data of the electric meter and the equipment operation status data in the energy consumption circuit are collected and time-aligned, including:

[0015] Real-time monitoring of each electricity meter in the building j The power reading P j (t), where j represents the serial number of the meter, j = 1, 2, ..., k; each meter m j Installed in different energy consumption circuits, t represents the time point;

[0016] Real-time monitoring equipment i The operating status S i (t), where i represents the serial number of the device, i = 1, 2, ..., m, and the on state value is 1 or the off state value is 0;

[0017] Time-align the meter power data and equipment operating status data.

[0018] Furthermore, in the above method, the correlation score A is calculated based on the comparison result of the time-aligned meter power data and the equipment operation status data. ij Calculations include:

[0019] When the device i The operating status S i (t) When a state change occurs, record the time point t and the state change ΔS i ; Change formula: ΔS i (t) = S i (t)-S i (t-δt), where δt is the data sampling interval;

[0020] Within the preset time window ΔT after the state changes, calculate the j Power change ΔP j = P j (t+ΔT)- P j (t);

[0021] Update device i With electric meter j The corresponding correlation score A ij .

[0022] Furthermore, in the above method, the device e is updated i With electric meterj The corresponding correlation score A ij ,include:

[0023] If ΔS i ΔP j <0, then A ij =A ij +1;

[0024] If ΔS i ΔP j >=0, then A ij =A ij -1;

[0025] If ΔS i =0 or ΔP_j=0, the score of A_ij remains unchanged.

[0026] Furthermore, in the above method, based on the correlation score A ij , modeling of noise impacts, including:

[0027] Introducing additional noise variance Correlation score A ij The total variance for:

[0028]

[0029] Based on the association score A ij and the total variance of the association score Conduct statistical tests of association.

[0030] Furthermore, in the above method, based on the correlation score A ij and the total variance of the association score Conduct statistical tests of association, including:

[0031] Calculate the standardized Z value Z ij :

[0032]

[0033] Set the significance level α, the corresponding critical value is:

[0034] Z α =Φ -1 (1-α)≈3.09

[0035] Among them, Φ -1 is the inverse function of the standard normal distribution; α is 0.001, and the confidence level is 99.9%;

[0036] If Z ij >Z α , then reject the null hypothesis and think that the correlation score Aij Corresponding device i With electric meter j There is a significant association;

[0037] If Z ij <=Z α , then accept the original hypothesis and think that the correlation score A ij Corresponding device i With electric meter j No significant association.

[0038] Furthermore, in the above method, according to the total variance of the association score (significance level and total variance) and critical value Z α , calculate the threshold θ of the association score ij :

[0039]

[0040] n represents the number of observations;

[0041] When the correlation score A ij ≥θ ij When the correlation score A ij Corresponding device i With electric meter j There is a significant correlation.

[0042] Furthermore, in the above method, multiple devices e i The loop paths are merged, including the following merging principles:

[0043] The nodes of devices with high monitoring power are located in the upper layer of the loop path, closer to the main electricity meter;

[0044] The loop paths with high association scores are located in the lower layers of the meter tree structure and are closer to the devices;

[0045] When the loop paths of different devices have common loop nodes, the loop paths are merged to reflect the sharing and branching relationship of the meters.

[0046] Furthermore, in the above method, based on the tree structure of the electric meter, real-time alarm and fault location and automatic optimization of equipment start-stop strategy are performed, including:

[0047] When an abnormal power is detected in a certain meter, the system quickly locates the relevant equipment based on the association relationship in the meter tree structure and prompts the operation and maintenance personnel to conduct an inspection;

[0048] Based on the usage patterns of the devices and the associations in the meter tree structure, the system can automatically optimize the device start and stop strategies.

[0049] According to another aspect of the present invention, there is further provided a computer-readable storage medium having computer-executable instructions stored thereon, wherein when the computer-executable instructions are executed by a processor, the processor is caused to: adopt any of the methods described above.

[0050] The purpose of the present invention is to provide an automatic association method based on meter power monitoring and equipment operation status monitoring, quickly and accurately establish the association relationship between the building energy circuit and the equipment, improve energy management efficiency, and reduce energy waste. The present invention successfully realizes the automatic association between the equipment and the meter circuit in the office building. Through real-time data analysis and noise processing, not only the energy management efficiency is improved, but also the operation and maintenance time and energy waste are significantly reduced.

[0051] The present invention has the following advantages: high degree of automation: automatically establishing the association between equipment and energy consumption circuits without manual intervention; efficient and accurate: utilizing real-time data and statistical methods to improve the accuracy of association judgment; no impact on normal operation: avoiding the need to shut down the meter or equipment in traditional methods, and not affecting the normal operation of the building; strong scalability: suitable for building energy management systems of various sizes, and can be integrated with existing building management systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 The present invention is a flowchart of a method and device for automatically extracting the relationship between building energy circuits and equipment according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0054] like Figure 1 As shown, the present invention provides a method for automatically extracting the relationship between building energy circuits and equipment, comprising:

[0055] Step S1: Data preparation: Collect meter power data and equipment operation status data in the energy consumption circuit, and perform time alignment;

[0056] Step S11: Electric meter power data collection: Real-time monitoring of each electric meter m in the building j The power reading P j (t), where j represents the serial number of the meter, j = 1, 2, ..., k; each meter m j Installed in different energy consumption circuits, t represents the time point;

[0057] Specifically, in an office building with 6 floors above ground and 2 floors underground, the method of the present invention is implemented to monitor the operating status of key equipment such as air-conditioning system, lighting system, elevator system, water pump system and fan system, and the power data of each circuit in the building is monitored through the installed smart meter.

[0058] For example, 50 devices such as air conditioners, lighting, and elevators are connected to the building automation system (BAS). The switch status of the device (1 for on, 0 for off) is collected once a minute and uploaded to the energy management system. Data format: device ID, timestamp, status;

[0059] Step S12: Equipment operation status data collection: real-time monitoring of equipment e i The operating status S i (t), where i represents the serial number of the device, i = 1, 2, ..., m, and the on state value is 1 or the off state value is 0;

[0060] For example, 30 meters are installed in different energy circuits, including main circuits, floor circuits and equipment partition circuits. The meter records power data every 10 seconds and uploads the data to the energy management system. Data format: meter ID, timestamp, power;

[0061] Step S13: Time-aligning the meter power data and the equipment operation status data;

[0062] Here, since the meter data collection frequency is higher than the device status data, the two need to be time-aligned. Aggregate the meter data by time window (such as 10 seconds), take the maximum, minimum, and average values ​​within the window, and ensure that the meter and device data are synchronized on the same time axis.

[0063] Step S2: Based on the comparison result of the time-aligned meter power data and the equipment operation status data, the correlation score A is calculated. ij Calculation of

[0064] Step S21: When the device e i The operating status S i (t) When a state change occurs, record the time point t and the state change ΔS i ; Change formula: ΔS i (t) = S i (t)-S i (t-δt), where δt is the data sampling interval;

[0065] For example, when the device e i At a certain time point t, a state change occurs (such as the air conditioner is turned on or off), and the state change is recorded. Change formula: ΔS i (t) = S i (t)-Si (t-δt), where δt is the data sampling interval.

[0066] Step S22: Calculate the value of each meter m within the preset time window ΔT after the state changes. j Power change ΔP j =P j (t+ΔT)-P j (t);

[0067] For example, the power change calculation of each meter is as follows: within the 10-second time window after the device status changes, the power change of each meter m is calculated. j Power change: ΔP j (t) = P j (t+ΔT)-P j (t) where ΔT is the time window size.

[0068] Step S23: Update device e i With electric meter j The corresponding correlation score A ij :

[0069] Preferably, step S231: if ΔS i ΔP j <0, then A ij =A ij +1;

[0070] If ΔS i ΔP j >=0, then A ij =A ij -1;

[0071] If ΔS i =0 or ΔP_j=0, the score of A_ij remains unchanged.

[0072] Step S3: Based on the association score A ij , modeling of noise impact;

[0073] For example, calculation and update of association score: Update the association score A between the device and the meter according to the following rules: ij :

[0074] A ij =A ij +f(ΔS i ΔP j ); if ΔS i ΔP j <0, then A ij =A ij +1 (such as power reduction when the air conditioner is turned off); if ΔS i ΔPj >0, then A ij =A ij -1; if ΔS i =0 or ΔP j =0, then A ij The score remains unchanged.

[0075] Step S3.1: Consider the impact of noise on the power measurement of the meter and introduce additional noise variance Correlation score A ij The total variance for:

[0076]

[0077] Among them, the variance Var[A ij ] indicates A ij The degree of dispersion or uncertainty of this indicator is given in the formula Var[A ij ] = 0.5n, n represents the number of observations;

[0078] Here, noise level estimation: Calculate the noise level in the meter data and get the noise variance

[0079] Step S3.2: Based on the association score A ij and the total variance of the association score Conduct statistical tests of associations;

[0080] Preferably, step S3.2 comprises:

[0081] To test whether the association score is significantly higher than the expected value in the absence of association, the standardized Z value Z is calculated. ij :

[0082]

[0083] Set the significance level α, the corresponding critical value is:

[0084] Z α =Φ -1 (1-α)≈3.09

[0085] Among them, Φ -1 is the inverse function of the standard normal distribution; α can be, for example, 0.001, with a confidence level of 99.9%;

[0086] If Z ij >Z α , then reject the null hypothesis and think that the correlation score A ij Corresponding device i With electric meter j There is a significant association;

[0087] If Z ij <=Z α , then accept the original hypothesis and think that the correlation score A ij Corresponding device i With electric meter j No significant association;

[0088] For example, the significance test of the association score: Calculate the standardized Z value of the association score:

[0089]

[0090] Where n is the number of observations. ij >3.09 (99.9% confidence level), the correlation score is considered to be A ij Corresponding device i With electric meter j There is a significant correlation.

[0091] Preferably, step S3.2 comprises:

[0092] According to the total variance of the association score (significance level and total variance) and critical value Z α , calculate the threshold θ of the association score ij :

[0093]

[0094] n represents the number of observations;

[0095] When the correlation score A ij ≥θ ij When the correlation score A ij Corresponding device i With electric meter j There is a significant correlation.

[0096] Specific application examples are as follows:

[0097] Assumptions: Number of observations: n = 100; Noise variance: Significance level: α =

[0098] 0.001, Z α ≈3.09. Calculate the total variance

[0099]

[0100] Calculate the relevance score threshold:

[0101]

[0102] If the observed correlation score A ij =50, then calculate the Z value:

[0103]

[0104] Because Z ij =5.0>Z α , Z α ≈3.09, so with a confidence level of 99.9%, it is believed that device e i With electric meter j There is a significant correlation.

[0105] Step 4: Establishing association relationships, based on determining the devices and meters that have significant associations, building a complete meter tree structure;

[0106] Preferably, there is a significant correlation between the task and the device e i With electric meter j , associated to the corresponding same meter circuit, from the association score A ij Sort from high to low to build a loop path; connect multiple devices i The loop paths are merged to build a complete meter tree structure;

[0107] Preferably, the merging principle is:

[0108] Power size: The device nodes that monitor high power are located in the upper layer of the loop path, closer to the main meter;

[0109] Hierarchical relationship: The loop paths with high association scores are located in the lower layers of the meter tree structure and are closer to the devices;

[0110] Path merging: When the loop paths of different devices have common loop nodes, the paths are merged to reflect the sharing and branching relationship of the meters.

[0111] Preferably, the steps include: building an associated path and merging the meter tree:

[0112] Step S4.1: Construct the association path: ij The devices and meters are sorted by power to form a loop path; the nodes with high device power are located at the upper layer of the loop path, closer to the main meter;

[0113] Step S4.2: Merge the meter tree structure: If the associated paths of multiple devices have overlapping nodes, merge the paths to construct a complete meter tree structure.

[0114] Merging principle: high-power devices are preferentially located at the upper layer of the loop path; paths of shared nodes are sorted by correlation scores to ensure path rationality.

[0115] During the merging process, a certain loop path may be judged to belong to a different level, or be associated with multiple unrelated devices. In this case, it is necessary to introduce code logic to check the integrity and accuracy of the data. When necessary, manual judgment of the contradictions should be carried out based on the actual situation and professional knowledge.

[0116] Step S5: Result application and optimization: Based on the tree structure of the electric meter, real-time alarm and fault location are performed and the start and stop strategies of the equipment are automatically optimized.

[0117] Step S5.1 Real-time alarm and fault location: When an abnormal power is detected in a certain meter, the system quickly locates the relevant equipment according to the association relationship in the tree structure of the meter, and prompts the operation and maintenance personnel to check.

[0118] Step S5.2 Energy-saving strategy optimization: Based on the usage pattern of the equipment and the association relationship in the meter tree structure, the system can automatically optimize the equipment start and stop strategy to reduce unnecessary energy consumption.

[0119] In summary, the purpose of the present invention is to provide an automatic association method based on meter power monitoring and equipment operation status monitoring, quickly and accurately establish the association relationship between the building energy circuit and the equipment, improve energy management efficiency, and reduce energy waste. The present invention successfully realizes the automatic association between the equipment and the meter circuit in the office building. Through real-time data analysis and noise processing, not only the energy management efficiency is improved, but also the operation and maintenance time and energy waste are significantly reduced.

[0120] The present invention has the following advantages: high degree of automation: no human intervention is required, and the association relationship between equipment and energy consumption circuits is automatically established; efficient and accurate: real-time data and statistical methods are used to improve the accuracy of association judgment; no impact on normal operation: the need to shut down the meter or equipment in traditional methods is avoided, and the normal operation of the building is not affected; strong scalability: it is suitable for building energy management systems of various sizes, and can be integrated with existing building management systems.

[0121] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0122] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0123] Obviously, those skilled in the art can make various changes and modifications to the invention without departing from the spirit and scope of the invention. Thus, if these modifications and variations of the invention fall within the scope of the claims of the invention and their equivalents, the invention is also intended to include these modifications and variations.

Claims

1. A method for automatically extracting the relationship between building energy circuits and equipment, characterized in that: include: Collect meter power data and equipment operating status data in the energy consumption circuit and perform time alignment; Based on the comparison of the time-aligned meter power data and the equipment operation status data, the correlation score A is calculated. ij Calculation of Based on the association score A ij , modeling of noise impacts to identify equipment and meters with significant correlations; Based on the devices and meters that are significantly associated, a complete meter tree structure is constructed; Based on the tree structure of the electricity meter, real-time alarm and fault location are carried out, and the start and stop strategies of the equipment are automatically optimized.

2. The method for automatically extracting the relationship between building energy circuits and equipment according to claim 1, characterized in that: Collect meter power data and equipment operating status data in the energy consumption circuit and perform time alignment, including: Real-time monitoring of each electricity meter in the building j The power reading P j (t), where j represents the serial number of the meter, j = 1, 2, ..., k; each meter m j Installed in different energy consumption circuits, t represents the time point; Real-time monitoring equipment i The operating status S i (t), where i represents the serial number of the device, i = 1, 2, ..., m, and the on state value is 1 or the off state value is 0; Time-align the meter power data and equipment operating status data.

3. The method for automatically extracting the relationship between building energy circuits and equipment according to claim 1, characterized in that: Based on the comparison of the time-aligned meter power data and the equipment operation status data, the correlation score A is calculated. ij Calculations include: When the device i The operating status S i (t) When a state change occurs, record the time point t and the state change ΔS i ; Change formula: ΔS i (t) = S i (t)-S i (t-δt), where δt is the data sampling interval; Within the preset time window ΔT after the state changes, calculate the j Power change ΔP j = P j (t+ΔT)-P j (t); Update device i With electric meter j The corresponding correlation score A ij .

4. The method for automatically extracting the relationship between building energy circuits and equipment according to claim 3, characterized in that: Update device i With electric meter j The corresponding correlation score A ij ,include: If ΔS i ΔP j <0, then A ij =A ij +1; If ΔS i ΔP j >=0, then A ij =A ij -1; If ΔS i =0 or ΔP_j=0, the score of A_ij remains unchanged.

5. The method for automatically extracting the relationship between building energy circuits and equipment according to claim 1, characterized in that: Based on the association score A ij , modeling of noise impacts, including: Introducing additional noise variance Correlation score A ij The total variance for: Based on the association score A ij and the total variance of the association score Conduct statistical tests of association.

6. The method for automatically extracting the relationship between building energy circuits and equipment according to claim 5, characterized in that: Based on the association score A ij and the total variance of the association score Conduct statistical tests of association, including: Calculate the standardized Z value Z ij : Set the significance level α, the corresponding critical value is: Z α =Φ -1 (1-a)≈3.09 Among them, Φ -1 is the inverse function of the standard normal distribution; α is 0.001, and the confidence level is 99.9%; If Z ij >Z α , then reject the null hypothesis and think that the correlation score A ij Corresponding device i With electric meter j There is a significant association; If Z ij <=Z α , then accept the original hypothesis and think that the correlation score A ij Corresponding device i With electric meter j No significant association.

7. The method for automatically extracting the relationship between building energy circuits and equipment according to claim 5, characterized in that: According to the total variance of the association score and critical value Z α , calculate the threshold θ of the association score ij : n represents the number of observations; When the correlation score A ij ≥θ ij When the correlation score A ij Corresponding device i With electric meter j There is a significant correlation.

8. The method for automatically extracting the relationship between building energy circuits and equipment according to claim 1, characterized in that: Based on the identification of the devices and meters with significant association, a complete meter tree structure is constructed, including: For devices and meters with significant task association, they are associated with the same corresponding meter circuit, and then the associated score A is used to calculate the number of devices and meters. ij Sort from high to low to build a loop path; connect multiple devices i The loop paths are merged to build a complete meter tree structure; Among them, multiple devices e i The loop paths are merged, including the following merging principles: The nodes of devices with high monitoring power are located in the upper layer of the loop path, closer to the main electricity meter; The loop paths with high association scores are located in the lower layers of the meter tree structure and are closer to the devices; When the loop paths of different devices have common loop nodes, the loop paths are merged to reflect the sharing and branching relationship of the meters.

9. The method for automatically extracting the relationship between building energy circuits and equipment according to claim 1, characterized in that: Based on the tree structure of the electric meter, real-time alarm and fault location are carried out, and the start and stop strategies of the equipment are automatically optimized, including: When an abnormal power is detected in a certain meter, the system quickly locates the relevant equipment based on the association relationship in the meter tree structure and prompts the operation and maintenance personnel to conduct an inspection; Based on the usage patterns of the devices and the associations in the meter tree structure, the system can automatically optimize the device start and stop strategies.

10. A computer-readable storage medium having computer-executable instructions stored thereon, wherein: When the computer executable instructions are executed by a processor, the processor is caused to: adopt the method according to any one of claims 1 to 9.

Images