A non-invasive method and system for energy-saving operation of indoor electrical equipment without human presence

Abstract

The present invention provides a non-invasive method and system for energy-saving operation of indoor electrical equipment without human intervention, including establishing a rule library for the operating states of electrical equipment; and proposing a decomposition method based on a subsequence shapelet feature encoder to complete the decomposition of the total energy consumption data into the energy consumption data of the target equipment, and extracting the operating characteristics of the electrical equipment according to the decomposed data; then matching the operating state of the equipment in the state feature rule library according to the operating characteristics of the electrical equipment; finally, judging whether energy saving is required according to the existing operating state of the equipment, so as to adjust the operating state of the equipment. The present invention uses a non-invasive method to replace the traditional invasive method, can realize the monitoring, analysis and control of electrical equipment in the area, and automatically monitor and control the energy saving of electrical equipment in the area when there is no one, effectively reducing energy waste and realizing automatic and intelligent energy saving.

Description

Technical Field

[0001] The present invention belongs to the technical field of building energy conservation, and particularly relates to a non-invasive method and system for the energy-saving operation of indoor electrical equipment without human presence. Background Art

[0002] Energy is one of the material conditions for human survival and an essential material basis for social and economic development. Among them, building energy consumption, transportation energy consumption, and industrial energy consumption have become the three major "energy-consuming giants" side by side. Among them, building energy consumption accounts for about 30% of the total social energy consumption and shows an increasing trend year by year.

[0003] As the main building energy consumption, electricity conservation remains the main theme of energy conservation. Indoor occupants are the behavioral subjects of building energy consumption, and the change of indoor occupants is one of the key factors affecting building energy consumption. Due to certain reasons, the phenomenon of "electricity still on when people leave" often occurs, and the electrical equipment consumes electricity in vain when there is no one, which easily causes waste of electricity.

[0004] The existing energy-saving operation systems for electrical equipment without human presence have the following problems: (1) Monitoring each electrical equipment in an invasive manner not only increases the cost but also is not convenient for later maintenance. (2) The activation of the energy-saving system requires manual control, and the degree of automation is insufficient. (3) The existing energy-saving systems cannot classify and identify the necessity of equipment, and adopt a "one-size-fits-all" method to cut off power for energy conservation, and necessary equipment cannot maintain normal operation, lacking practicality and intelligence. Summary of the Invention

[0005] In order to solve the problems existing in the above-mentioned prior art, a non-invasive method and system for the energy-saving operation of indoor electrical equipment without human presence are provided.

[0006] The technical solution adopted by the present invention to solve the problems of the prior art is:

[0007] A non-invasive method for the energy-saving operation of indoor electrical equipment without human presence is proposed, including:

[0008] S1: Extract the operation characteristics of the equipment according to the electrical characteristics and working rules of the electrical equipment, and construct an operation state rule library of the electrical equipment according to the operation characteristics of the equipment;

[0009] S2: Construct a decomposition method based on the subsequence shapelet feature encoder to complete the decomposition of the total energy consumption data into the energy consumption data of the target equipment, and extract the operation characteristics of the electrical equipment according to the decomposed data;

[0010] S3: Match the operation state of the electrical equipment in the operation state rule library of the electrical equipment according to the operation characteristics of the electrical equipment;

[0011] S4: Determine whether energy conservation is required based on the operating status of existing electrical equipment, and then adjust the operating status of the electrical equipment to achieve energy-saving operation.

[0012] Preferably, in step S1, the method for constructing the operating status rule library of electrical equipment includes the following steps:

[0013] S11: Select several kinds of equipment electrical parameters to construct an equipment operating status feature vector. The status feature vector L(i) of electrical equipment i is expressed as:

[0014] L(i) = [U(i), I(i), P(i), Q(i), S(i), C(i), M(i), R(i)] (1)

[0015] S12: According to the different status feature vectors of electrical equipment i, construct the status rule library Z of electrical equipment i i , in the status rule library Z i , different status vectors L correspond to different working states of the equipment , and the rule Z i in the status rule library Z of equipment i i (m) is expressed as:

[0016] Z i (m): If the equipment status feature is L(m), then the equipment status is

[0017] where, L(m) is the status feature of equipment i, represents the equipment operating status, and the complete rule library is expressed as:

[0018] Z i = {Z i (1), Z i (2), …, Z i (m), …, Z i (n)} (3)

[0019] where, n represents the total number of rules, corresponding to the types of working states of the equipment.

[0020] Preferably, the equipment electrical parameters include the selected effective voltage value, effective current value, active power, reactive power, current shape factor, current crest factor, current pulse factor, and starting peak current.

[0021] Preferably, in step S2, the decomposition method includes the following steps:

[0022] (11) Cut the input data P Σ (X) into multiple subsequences:

[0023] The data after cutting is expressed as:

[0024] P Σ (X) = P Σ (x 1 , x 2 , …, x v ) (4)

[0025] Among them, v represents the number of subsequences;

[0026] (12) Extract the subsequence P Σ (X) = P Σ (x 1 , x 2 , …, x v )'s similarity features:

[0027] The similarity features of the subsequence include the Euclidean feature vector E o , the Mahalanobis feature vector E m and the cosine feature vector E t ,

[0028]

[0029] Among them, x a and x b represent subsequences, σ is a non-linear function, G is the covariance matrix of the distribution to which the subsequence belongs, and the subsequence similarity feature T is expressed as:

[0030] T = [E o , E m , E t (6)

[0031] (13) Construct an encoder for the subsequence shapelets features:

[0032] The encoder takes the subsequence data and its similarity feature T as inputs and outputs a coding vector J. This process is expressed as:

[0033]

[0034] Among them, W = [w q , w k , w v is the attention parameter, f att is the initialization function, σ is a non-linear function, D is the feature extracted from the subsequence, f e represents the coding function, and J is the coding vector;

[0035] (14) Construct the decoding function F:

[0036] The function F calculates and outputs the energy consumption data y i of the target device i based on the known coding vector J. This process is expressed as:

[0037] y i = F(J) (8).

[0038] Preferably, in the step S3, the operation status matching method includes: According to the decomposed data y i Calculate the current state feature vector L'(i) of the device, and then match it with the operation status rule library of the electrical equipment. If the current state feature vector L'(i) meets the rule trigger condition L(i), the device status is output according to the rule corresponding to L(i) The trigger condition is:

[0039]

[0040] where L'(i) is the current device operation status vector, L(i) is the rule trigger status vector, and δ is the rule trigger threshold.

[0041] Preferably, in the step S4, the energy-saving operation method includes:

[0042] (21) Classify the electrical equipment in advance:

[0043] Necessary electrical equipment:

[0044] Non-necessary electrical equipment:

[0045] (22) Divide the system operation status into two modes:

[0046] Unmanned mode: When the last mobile device in the area disconnects from the WiFi in the area for more than a predetermined delay time, the system determines that there are no people in the area and turns on the unmanned mode. The smart meter sends the current operation status of the electrical equipment to the application layer through WiFi. According to the pre-classification of the necessity of the electrical equipment, the application layer interacts with the smart relay through WiFi to adjust the operation status of the non-necessary electrical equipment from the current status to disconnected, and the necessary electrical equipment remains unchanged;

[0047] Occupied mode: When a mobile device accesses the WiFi in the area, the system determines that there are people in the area, and the system maintains the occupied mode operation. The operation status of all electrical equipment in the area remains unchanged.

[0048] A non-intrusive indoor electrical equipment unmanned energy-saving operation method and system, including:

[0049] Operation status rule library of electrical equipment: Realize the mutual correspondence between the operation characteristics of electrical equipment and the working status of electrical equipment;

[0050] Non-invasive load decomposition module: Construct a decomposition mapping function F to achieve the mapping from the total energy consumption data on the household side to the energy consumption data of the target device;

[0051] Load status recognition module: After the non-invasive load decomposition module outputs the energy consumption data of the target device, the load status recognition module calculates the current status feature vector L'(i) of the device through the decomposed data y i and matches it with the rules in the operation status rule library of the electrical equipment;

[0052] Energy-saving control module: After the load status recognition module outputs the current operation status of the electrical equipment, it judges whether there are people in the room by the connection situation between the mobile devices in the area and WiFi, and turns on the corresponding mode, and realizes the energy-saving control of the electrical equipment through the smart meter and the smart relay.

[0053] Compared with the prior art, the present invention has the following advantages:

[0054] 1. Through the load decomposition, device operation status recognition and personnel monitoring in the area, the present invention realizes the automatic monitoring of electrical equipment in the area and automatic energy-saving control when there is no one, effectively reducing energy waste.

[0055] 2. The present invention adopts the non-invasive load decomposition and recognition technology, greatly reducing the monitoring cost and maintenance cost. In the non-invasive load decomposition module, a new decomposition method based on the subsequence shapelet feature encoder is used, and the energy consumption data of the target device is output by calculating the subsequence similarity to achieve the purpose of energy consumption decomposition.

[0056] 3. The present invention also constructs an operation status rule library for electrical equipment to realize the detection of the operation status of electrical equipment, judges whether there are people in the area by the WiFi access of personal devices, and cuts off the energy-saving operation of unnecessary devices according to the current operation status of electrical equipment and the personnel situation to realize the energy-saving control of electrical equipment when there is no one. Brief Description of the Drawings

[0057] The above and / or additional aspects and advantages of the present invention will become obvious and easy to understand from the description of the embodiments in conjunction with the following drawings, where:

[0058] Figure 1 is the overall flowchart of the present invention. Detailed Embodiments

[0059] Embodiments of the present invention will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and are only used to explain the present invention and should not be construed as a limitation of the present invention.

[0060] As Figure 1 shown, this embodiment proposes a non-invasive method for energy-saving operation of indoor electrical equipment without human presence, including:

[0061] S1: Extract the operation characteristics of the equipment according to the electrical characteristics and working rules of the equipment, and construct an operation status rule library for the electrical equipment based on the operation characteristics of the equipment to achieve a one-to-one correspondence between the operation characteristics of the electrical equipment and the working status of the electrical equipment;

[0062] S2: Construct a decomposition method based on the subsequence shapelet feature encoder to complete the decomposition of the total energy consumption data into the energy consumption data of the target equipment, and extract the operation characteristics of the electrical equipment according to the decomposed data;

[0063] S3: Match the operation status of the electrical equipment in the operation status rule library of the electrical equipment according to the operation characteristics of the electrical equipment;

[0064] S4: Judge whether energy saving is required according to the existing operation status of the electrical equipment, so as to adjust the operation status of the electrical equipment and achieve energy-saving operation.

[0065] The operation status rule library of the electrical equipment aims to extract a set of status features through the electrical characteristics and operation modes of the equipment, and construct a mapping relationship between the set of equipment operation status features and the actual status of the equipment.

[0066] In step S1, the method for constructing the operation status rule library of the electrical equipment includes the following steps:

[0067] S11: Select several kinds of equipment electrical parameters to construct an operation status feature vector of the equipment. The status feature vector L(i) of the electrical equipment i is expressed as:

[0068] L(i) = [U(i), I(i), P(i), Q(i), S(i), C(i), M(i), R(i)] (1)

[0069] S12: Construct a status rule library Z i of the electrical equipment i according to different status feature vectors of the electrical equipment i. In the status rule library Z i , different status vectors L correspond to different working statuses of the equipment. The rule Z in the status rule library Z i of the equipment i is expressed as: i (m) is expressed as:

[0070] Z i (m): If the device status feature is L(m), then the device status is

[0071] where L(m) is the status feature of device i, indicating the device operation status. The complete rule base is expressed as:

[0072] Z i ={Z i (1), Z i (2), …, Z i (m), …, Z i (n)} (3)

[0073] where n represents the total number of rules, corresponding to the types of device working status.

[0074] Among them, the electrical parameters of the device include the selected effective voltage (U), effective current (I), active power (P), reactive power (Q), current shape factor (S), current crest factor (C), current pulse factor (M), and starting peak current (R).

[0075] By constructing the operation status rule base of the electrical equipment, the detection of the operation status of the electrical equipment is realized. By judging the presence or absence of personal device WiFi access, the presence or absence of personnel in the area is determined, and according to the current operation status of the electrical equipment and the personnel situation, the energy-saving operation of non-essential equipment is cut off, realizing the energy-saving control of electrical equipment when there is no one.

[0076] It also includes a non-intrusive load decomposition module. The non-intrusive load decomposition module aims to construct a decomposition mapping function F to realize the mapping from the total energy consumption data P Σ at the household entrance side to the energy consumption data P i of the target device. This process can be expressed as: F: P Σ →P i .

[0077] In this application, the non-intrusive load decomposition module adopts a decomposition method based on subsequence shapelet feature encoder. This method first extracts the similar features of the subsequence, then constructs a subsequence shapelet feature encoding function to realize the encoding of the original subsequence and the similar features. Finally, a decoding function is used for decoding to output the decomposition data of the target device.

[0078] In step S2, the decomposition method includes the following steps:

[0079] (11) The input data P Σ (X) is sliced into multiple subsequences:

[0080] The sliced data is expressed as:

[0081] P Σ (X) = P Σ (x 1 , x 2 , …, x v ) (4)

[0082] Among them, v represents the number of subsequences;

[0083] (12) Extract subsequence P Σ (X) = P Σ (x 1 , x 2 , …, x v )'s similar features:

[0084] Among them, the similar features of the subsequence include three different feature vectors, namely the Euclidean feature vector E o , the Mahalanobis feature vector E m and the cosine feature vector E t , and the three vectors can be respectively expressed as:

[0085]

[0086] Among them, x a and x b represent subsequences, σ is a non-linear function, G is the covariance matrix of the distribution to which the subsequence belongs, and the subsequence similar feature T is expressed as:

[0087] T = [E o , E m , E t (6)

[0088] (13) Construct an encoder for subsequence shapelets features:

[0089] The encoder takes the subsequence data and its similar feature T as inputs, realizes feature fusion and re-extraction based on the attention mechanism, and outputs the encoded vector J. This process is expressed as:

[0090]

[0091] Among them, W = [w q , w k , w v is the attention parameter, f att is the initialization function, σ is a non-linear function, D is the feature extracted from the subsequence, f e represents the encoding function, and J is the encoded vector;

[0092] (14) Construct the decoding function F:

[0093] The function F calculates and outputs the energy consumption data y of the target device i based on the known coding vector J i , and this process is expressed as:

[0094] y i = F(J) (8).

[0095] In this application, by adopting the non-intrusive load decomposition and identification technology, the monitoring cost and maintenance cost are greatly reduced. In the non-intrusive load decomposition module, a new decomposition method based on the subsequence shapelet feature encoder is used. By calculating the subsequence similarity, the energy consumption data of the target device is output to achieve the purpose of energy consumption decomposition.

[0096] It also includes a load status identification module.

[0097] In step S3, the operating state matching method includes: after the non-intrusive load decomposition module outputs the energy consumption data y of the target device i , the load status identification module aims to calculate the current state feature vector L'(i) of the device according to the decomposition data y i , and then match it with the operating state rule library of the electrical equipment. If the current state feature vector L'(i) meets the rule trigger condition L(i), the device state is output according to the rule corresponding to L(i) The trigger condition is:

[0098]

[0099] where L'(i) is the current device operating state vector, L(i) is the rule trigger state vector, and δ is the rule trigger threshold.

[0100] It also includes an energy-saving control module. The overall process of the energy-saving control module is: after the electrical equipment state identification module outputs the current operating state of the electrical equipment, it judges whether there are people in the room by the connection situation between the personal mobile devices in the area and WiFi, turns on the occupied or unoccupied mode, and realizes the energy-saving control of the electrical equipment through the smart meter (embedded with a non-intrusive load monitoring module) and the smart relay.

[0101] In step S4, the energy-saving operation method includes:

[0102] (21) Classify the electrical equipment in advance:

[0103] First, classify the electrical equipment in advance according to factors such as the usage habits of the electrical equipment in the area by the indoor personnel and the operating mode of the electrical equipment

[0104] Necessary electrical equipment:

[0105] Non-necessary electrical equipment:

[0106] The necessity of the electrical equipment can be adjusted according to the usage requirements.

[0107] (22) According to the connection status of personal electrical equipment and WiFi in the area, the system operation status is divided into two modes:

[0108] Unmanned mode: When the last personal mobile device in the area disconnects from the WiFi in the area for more than a predetermined delay time, which can be 15 minutes, the system determines that there are no people in the area and turns on the unmanned mode. The smart meter sends the current operation status of the electrical equipment to the application layer through WiFi. According to the pre-classification of the necessity of the electrical equipment, the application layer interacts with the smart relay through WiFi and adjusts the operation status of the non-essential electrical equipment from the current status to disconnected, and the operation status of the essential electrical equipment remains unchanged;

[0109] Occupied mode: When a personal mobile device accesses the WiFi in the area, the system determines that there are people in the area, and the system maintains the operation in the occupied mode. The operation status of all electrical equipment in the area remains the current status unchanged.

[0110] This application realizes the automatic monitoring of electrical equipment in the area and automatic energy-saving control when there is no one by decomposing the load in the area, identifying the operation status of the equipment, and monitoring the personnel, effectively reducing resource waste.

[0111] A non-intrusive method and system for unmanned energy-saving operation of indoor electrical equipment, including:

[0112] Operation status rule library of electrical equipment: realizing the mutual correspondence between the operation characteristics of electrical equipment and the working status of electrical equipment;

[0113] Non-intrusive load decomposition module: constructing a decomposition mapping function F to realize the mapping from the total energy consumption data on the household side to the energy consumption data of the target equipment;

[0114] Load status identification module: After the non-intrusive load decomposition module outputs the energy consumption data of the target equipment, the load status identification module calculates the current status feature vector L'(i) of the equipment through the decomposed data y i and matches it with the rules in the operation status rule library of electrical equipment;

[0115] Energy-saving control module: After the load status identification module outputs the current operation status of the electrical equipment, it judges whether there are people in the room according to the connection status of the mobile device and WiFi in the area, and turns on the corresponding mode, and realizes the energy-saving control of the electrical equipment through the smart meter and the smart relay.

[0116] Although embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the claims and their equivalents.

Claims

1. A non-invasive method for energy-saving operation of indoor electrical equipment without human intervention, characterized in that, it includes: S1: Extract the equipment operation characteristics according to the electrical characteristics and working rules of the electrical equipment, and construct an operation status rule library of the electrical equipment according to the equipment operation characteristics; S2: Construct a decomposition method based on the subsequence shapelet feature encoder to complete the decomposition of the total energy consumption data into the energy consumption data of the target equipment, and extract the equipment operation characteristics according to the decomposed data; S3: Match the operation status of the electrical equipment in the operation status rule library of the electrical equipment according to the equipment operation characteristics; S4: Judge whether energy saving is required according to the existing operation status of the electrical equipment, so as to adjust the operation status of the electrical equipment and achieve energy-saving operation; In the step S2, the decomposition method includes the following steps: (11) Split the input data P Σ (X) into multiple subsequences: The data after segmentation is expressed as: P Σ P(X) = Σ (x 1 , x 2 , …, x v ) (4) where v represents the number of subsequences; (12) Extract subsequence P Σ (X) = P Σ (x 1 , x 2 , …, x v )'s similar features: The similar features of the subsequence include the Euclidean feature vector E o , the Mahalanobis feature vector E m and the cosine feature vector E t , where x a and x b represent subsequences, σ is a non-linear function, G is the covariance matrix of the distribution to which the subsequence belongs, and the subsequence similarity feature T is expressed as: T = [E o , E m , E t (6) (13) Construct an encoder for subsequence shapelets features: The encoder takes the subsequence data and its similar feature T as inputs and outputs an encoded vector J, and this process is expressed as: Among them, W = [w q , w k , w v is the attention parameter, f att is the initialization function, σ is the non-linear function, D is the subsequence extraction feature, f e represents the encoding function, and J is the encoding vector; (14) Construct a decoding function F: The function F calculates and outputs the energy consumption data y of the target device i according to the known coding vector J i , and this process is expressed as: y i = F(J) (8).

2. A non-invasive method for energy-saving operation of indoor electrical equipment without human intervention according to claim 1, characterized in that, in the step S1, the method for constructing the operation status rule library of the electrical equipment includes the following steps: S11: Select several kinds of equipment electrical parameters to construct an equipment operation status feature vector, and the status feature vector L(i) of the electrical equipment i is expressed as: L(i) = [U(i), I(i), P(i), Q(i), S(i), C(i), M(i), R(i)] (1) S12: Construct a state rule base Z for the electrical device i according to the different state feature vectors of the electrical device i i , in the state rule base Z i , different state vectors L correspond to different working states of the device , and the rule Z i in the state rule base Z of the device i i (m) is expressed as: Z i (m): If the device status feature is L(m), then the device status is where L(m) is the status feature of device i, indicating the device operating status. The complete rule base is expressed as: Z i = {Z i (1), Z i (2), …, Z i (m), …, Z i (n)} (3) where n represents the total number of rules, corresponding to the types of equipment working states.

3. A non-invasive method for energy-saving operation of indoor electrical equipment without human intervention according to claim 2, characterized in that, the equipment electrical parameters include the selected effective voltage value, effective current value, active power, reactive power, current shape factor, current crest factor, current pulse factor and starting peak current.

4. A non-invasive method for energy-saving operation of indoor electrical equipment without human intervention according to claim 1, characterized in that, In the step S3, the running state matching method includes: according to the decomposed data y i calculating the current state feature vector L'(i) of the device, and then matching it with the operation state rule base of the electrical device. If the current state feature vector L'(i) meets the rule triggering condition L(i), the device state is output according to the rule corresponding to L(i) The triggering condition is: where L'(i) is the current equipment operation status vector, L(i) is the rule trigger status vector, and δ is the rule trigger threshold.

5. A non-invasive method for energy-saving operation of indoor electrical equipment without human intervention according to claim 4, characterized in that, in the step S4, the energy-saving operation method includes: (21) Classify the electrical equipment in advance: Necessary electrical equipment: Non-essential electrical equipment: (22) Divide the system operation status into two modes: Unmanned mode: When the last mobile device in the area is disconnected from the WiFi in the area for more than the preset delay time, the system determines that there is no one in the area and turns on the unmanned mode. The smart meter sends the current operating status of the power-consuming equipment to the application layer via WiFi. According to the pre-classification of the necessity of the power-consuming equipment, the application layer interacts with the smart relay via WiFi to disconnect the non-essential power-consuming equipment. The operating status of the equipment is adjusted from the current status to disconnected, and necessary electrical equipment The status remains unchanged; Occupied mode: When a mobile device connects to the in-area WiFi, the system determines that there are people in the area. The system remains in the occupied mode, and the operating states of all electrical devices in the area remain unchanged.

6. A non-invasive system for energy-saving operation of indoor electrical equipment without human intervention, characterized in that, used to execute a non-invasive method for energy-saving operation of indoor electrical equipment without human intervention according to any one of claims 1-5, including: An operation status rule library of the electrical equipment: realizing the mutual correspondence between the equipment operation characteristics and the working status of the electrical equipment; A non-invasive load decomposition module: constructing a decomposition mapping function F to realize the mapping from the total energy consumption data on the household side to the energy consumption data of the target equipment; Load status recognition module: After the non-intrusive load decomposition module outputs the energy consumption data of the target device, the load status recognition module decomposes the data y i to calculate the current state feature vector L'(i) of the device and match it with the rules in the operating state rule library of the electrical equipment; Energy-saving control module: After the load status recognition module outputs the current operating status of the electrical equipment, it determines whether there are people in the room by judging the connection status between the mobile devices in the area and WiFi, and activates the corresponding mode, and realizes the energy-saving control of the electrical equipment through the smart meter and the smart relay.

Images