Smart AI energy-saving method based on comprehensive energy management

Abstract

The present invention relates to the technical field of integrated energy management, and discloses an intelligent AI energy-saving method based on integrated energy management, the method comprising: collecting signal strengths at different fixed positions to form fingerprint information and perform fingerprint expansion; collecting signal strength information of mobile devices in the internal area of ​​a building for fingerprint positioning; clustering personnel position information, constructing an air-conditioning adaptive control model, and adaptively adjusting air-conditioning parameters in combination with personnel position distribution cluster information and air-conditioning position information. The present invention combines the gray wolf optimization algorithm to implement iterative optimization of interpolation parameters, uses interpolation parameters to perform fingerprint information interpolation calculation on building positions that lack fingerprint information, implements expansion processing of fingerprint information in the building, uses fingerprint information to locate personnel positions in the building, and adjusts air-conditioning adjustment parameters of the air-conditioning in combination with personnel position distribution characteristics near different air-conditioning positions, thereby implementing air-conditioning energy management based on personnel position distribution.

Description

Technical Field

[0001] The present invention relates to the technical field of integrated energy management, and in particular to an intelligent AI energy-saving method based on integrated energy management. Background Art

[0002] With the rapid development of the global economy and the acceleration of urbanization, the demand for energy in the construction industry is also growing. Buildings generate a large amount of carbon dioxide emissions in the process of using energy, becoming one of the main sources of global greenhouse gas emissions. According to statistics, energy consumed in the construction field accounts for about 40% of the world's total energy consumption, and greenhouse gases emitted by the construction industry account for about 1 / 3 of the world's total emissions. Therefore, how to improve the energy efficiency of buildings and reduce energy consumption has become an important issue that needs to be solved urgently. In response to this problem, the present invention proposes an intelligent AI energy-saving method based on comprehensive energy management, which comprehensively optimizes the energy utilization efficiency and energy-saving effect of buildings through the coordinated control of the intelligent air-conditioning system. Summary of the invention

[0003] In view of this, the present invention provides an intelligent AI energy-saving method based on integrated energy management, the purpose of which is to: 1) use a signal collector to collect fingerprint information at a fixed position in a building, and combine the gray wolf optimization algorithm to implement iterative optimization of interpolation parameters, use the interpolation parameters to perform fingerprint information interpolation calculation on the building position where fingerprint information is missing, implement expansion processing of fingerprint information in the building, collect signal strength information of mobile devices in the internal area of ​​the building, combine the neighboring search algorithm to match and search the fingerprint information in the building fingerprint library, implement positioning processing of mobile devices in the current internal area of ​​the building, and obtain personnel location information inside the building; 2) use an adaptive density clustering method, The local distance density and gravity value of different personnel location information are calculated to realize adaptive screening and clustering processing of cluster centers, and an air-conditioning adaptive control model is constructed. The personnel location distribution cluster information, the number of people in the cluster, and the air-conditioning position are mapped to high-dimensional space respectively, and the personnel location distribution cluster vector and the air-conditioning position vector are fused to obtain the corresponding relationship between the air-conditioning position vector and the personnel location distribution cluster vector in space, that is, the personnel location distribution characteristics of different air-conditioning locations. The air-conditioning adjustment parameters are adjusted in combination with the personnel location distribution characteristics. The denser the personnel, the higher the energy consumption of the corresponding air-conditioning after adjustment, so as to realize air-conditioning energy-saving control based on personnel location distribution.

[0004] To achieve the above objectives, the present invention provides an intelligent AI energy-saving method based on comprehensive energy management, comprising the following steps:

[0005] S1: deploy a signal collector in the building, and use the signal collector to collect signal strength at different fixed positions to form fingerprint information, and perform fingerprint expansion on the collected fingerprint information to obtain a building fingerprint library, wherein the improved grey wolf algorithm is the main implementation method of the fingerprint expansion;

[0006] S2: Collect signal strength information of mobile devices in the internal area of ​​the building, perform fingerprint matching in combination with the building fingerprint database, obtain location information of mobile devices in the building, and use the location information of mobile devices as personnel location information;

[0007] S3: Clustering the personnel location information to obtain personnel location distribution clusters inside the building, extracting the cluster center location information and cluster radius of each personnel location distribution cluster, wherein adaptive density clustering is the main implementation method of the clustering;

[0008] S4: Construct an air conditioning adaptive control model, and adaptively adjust the air conditioning parameters in combination with the personnel location distribution cluster information and the air conditioning location information. The air conditioning adaptive control model takes the personnel location distribution cluster information, the number of people in the cluster and the air conditioning location as input, and takes the current air conditioning adjustment parameters as output, wherein the personnel location distribution cluster information includes the cluster center location information and radius of the personnel location distribution cluster.

[0009] As a further improvement method of the present invention:

[0010] Optionally, in step S1, a signal collector is deployed in the building, and the signal collector is used to collect signal strengths at different fixed positions to form fingerprint information, including:

[0011] M signal collectors are deployed in the building, and N fixed positions are set in the building. The signal collectors are used to collect the signal strength at different fixed positions to form fingerprint information. The signal collector can send out WI-FI signals. The fingerprint information collection process at different fixed positions is as follows:

[0012] Place the mobile device with WI-FI turned on in a fixed position to receive the WI-FI signal from the signal collector;

[0013] The mobile device records the wireless signal addresses and corresponding signal strengths of different WI-FI signals received, and sends the recorded content to the data processing center for aggregation;

[0014] The data center matches the wireless signal address of the WI-FI signal with the wireless signal address of the signal collector, and summarizes the signal strengths of the WI-FI signals emitted by the M signal collectors recorded by different mobile devices;

[0015] In the embodiment of the present invention, the signal collector is a WIFI AP device;

[0016] The signal strength of the Wi-Fi signals emitted by M signal collectors recorded by mobile devices at N fixed locations constitutes fingerprint information, where the fingerprint information is expressed in the form of:

[0017]

[0018] in:

[0019] L n represents the fingerprint information at the nth fixed position, Indicates the signal strength of the Wi-Fi signal emitted by M signal collectors at the nth fixed position, Indicates the signal strength of the Wi-Fi signal emitted by the mth signal collector at the nth fixed position;

[0020] The collected fingerprint information is expanded to obtain a building fingerprint library.

[0021] Optionally, the fingerprint expansion of the collected fingerprint information to obtain a building fingerprint library includes:

[0022] The fingerprint expansion process is as follows:

[0023] S11: Initialize and generate the expansion step parameter vector in is the expanded step size parameter vector θ 0 The three parameters in;

[0024] S12: Set the current iteration number of the extended step parameter vector to t, the initial value of t is 0, and the maximum value is Max. Then the tth iteration result of the extended step parameter vector is

[0025] S13: construct an iterative objective function of the expanded step parameter vector, wherein the iterative objective function takes the expanded step parameter vector and the fingerprint information as input, takes the objective function value corresponding to the expanded step parameter vector as output, and takes the expanded step parameter vector θ as output. t The iterative objective function with input is:

[0026]

[0027] in:

[0028] F(θ t ) represents the expanded step size parameter vector θ t is the input iterative objective function, that is, the expanded step parameter vector θ t The corresponding objective function value;

[0029] L n represents the fingerprint information at the nth fixed position, Represents the expanded step size parameter vector θ t Fingerprint information at the nth fixed position obtained by interpolation calculation;

[0030] ||·|| 2 represents the L2 norm;

[0031] L near (n) represents the fingerprint information at the fixed position closest to the nth fixed position;

[0032] loc n Indicates the position coordinates of the nth fixed position, loc n (near) represents the position coordinates of the fixed position closest to the nth fixed position among the N fixed positions, dis(loc n ,loc n (near)) represents the position coordinate loc n ,loc n (near) the distance between;

[0033] represents the iterative objective function F(θ t ) Intermediate parameter variables in the calculation process;

[0034] S14: Based on the improved grey wolf algorithm, the step size parameter vector θ is expanded t Iterate and get θ t+1 , where the iterative process is:

[0035] S141: Input the t expanded step length parameter vectors obtained in the current iteration into the iterative objective function, and use the expanded step length parameter vector with the smallest objective function value as the prey coordinate vector of this iteration

[0036] S142: Based on prey coordinate vector The expansion step parameter vector is iterated, where the iteration formula is:

[0037]

[0038] in:

[0039] rand(0,1) represents a random number between 0 and 1;

[0040] a t Represents the proportional coefficient of the objective function value used for iteration, a max Represents the preset maximum objective function value proportional coefficient, a min Indicates the preset minimum objective function value proportional coefficient;

[0041] ε represents the convergence factor;

[0042] b t represents the dynamic weight factor used for iteration, b max Indicates the preset maximum weight factor, b min Indicates the preset minimum weight factor;

[0043] S15: Let t = t + 1, return to step S14, until the maximum number of iterations is reached, and get θ Max ;

[0044] S16: Select the location in the building where no fingerprint information is generated, combine the fingerprint information of all currently known locations, and expand the step length parameter vector θ Max And the calculation formula in step S13 is used for interpolation calculation to obtain the fingerprint information of the selected position;

[0045] S17: Repeat step S16 to calculate the fingerprint information of H positions in the building to form a building fingerprint library, where the H positions include N fixed positions and HN positions selected in step S16. The representation form of the building fingerprint library is:

[0046] {L h =(r 1 (h),r 2 (h),...,r m (h),...,r M (h))|h∈[1,H]}

[0047] in:

[0048] L h represents the fingerprint information of the hth position in the building, (r 1 (h),r 2 (h),...,r m (h),...,r M (h)) represents the signal strength of the WI-FI signal emitted by M signal collectors at the h-th position in the building, where the position coordinate of the h-th position in the building is loc(h).

[0049] Optionally, collecting signal strength information of mobile devices in the internal area of ​​the building in step S2 includes:

[0050] Collect the signal strength information of mobile devices in the internal area of ​​the building. The collection process of the mobile device signal strength information is as follows:

[0051] The mobile device receives the WI-FI signals from the M signal collectors and records the signal strength of each WI-FI signal to form the signal strength information of the mobile device;

[0052] The mobile device sends the signal strength information to the data processing center in real time;

[0053] The data processing center forms a collection of signal strength information of all mobile devices currently in the building:

[0054] {L(g)|g∈[1,G]}

[0055] in:

[0056] L(g) represents the signal strength information of the gth mobile device in the internal area of ​​the building, and G represents the total number of all mobile devices currently in the building;

[0057] Combined with the building fingerprint database, the signal strength information of the mobile device is fingerprint matched to obtain the location information of the mobile device inside the building, and the location information of the mobile device is used as the personnel location information. The location information of the mobile device obtained by matching the signal strength information L(g) is loc g ,loc g is the position coordinate of the g-th mobile device in the internal area of ​​the building corresponding to the signal strength information L(g).

[0058] Optionally, the fingerprint matching of the signal strength information of the mobile device in combination with the building fingerprint library to obtain the location information of the mobile device inside the building includes:

[0059] Combined with the building fingerprint library, the signal strength information of the mobile device is fingerprint matched to obtain the location information of the mobile device inside the building. The fingerprint matching process of the signal strength information L(g) is as follows:

[0060] Calculate the similarity between the signal strength signal L(g) and the fingerprint information corresponding to H positions in the building fingerprint database, where the signal strength signal L(g) and the fingerprint information L h The similarity calculation formula is:

[0061]

[0062] in:

[0063] Sim(L(g),L h ) represents the signal strength signal L(g) and the fingerprint information L h Similarity:

[0064] ||·|| 2 represents the L2 norm;

[0065] Select the K position coordinates corresponding to the K fingerprint information with the highest similarity as the neighboring position coordinates of the signal strength signal L(g):

[0066] (loc g (1),locg (2),...,loc g (k),...,loc g (K))

[0067] in:

[0068] loc g (k) represents the kth neighbor position coordinate of the signal strength signal L(g), k∈[1,K];

[0069] Perform weighted calculation on the K nearest neighbor position coordinates to obtain the mobile device location information loc corresponding to the signal strength information L(g) g :

[0070]

[0071] in:

[0072] w F Represents the weighted weight of the kth neighbor position coordinate;

[0073] Represents the kth neighbor position coordinates to the mobile device location information loc g The intensity weight of

[0074] Represents the kth neighbor position coordinates to the mobile device location information loc g The distance weight of

[0075] e represents a natural constant;

[0076] d 0 Indicates the reference distance of signal propagation, d 1 Represents the attenuation coefficient of signal propagation;

[0077] L g (k) represents the kth neighbor position coordinate loc g (k) Fingerprint information in the building fingerprint database.

[0078] Optionally, clustering the personnel location information in step S3 to obtain personnel location distribution clusters inside the building includes:

[0079] The personnel location information is clustered to obtain the personnel location distribution clusters inside the building. The personnel location information clustering process is as follows:

[0080] S31: Extract the current location information of G persons in the building:

[0081] {loc g |g∈[1,G]}

[0082] in:

[0083] loc g Indicates the location information of the g-th person;

[0084] S32: Calculate the Euclidean distance between any two different personnel location information;

[0085] S33: Calculate the local distance density of each person's location information, where the person's location information loc g The local distance density is:

[0086]

[0087] in:

[0088] ρ g Indicates the location information of the person loc g The local distance density of

[0089] dist represents the preset cutoff distance;

[0090] dis(loc g ,loc j ) indicates the personnel location information loc g ,loc j The Euclidean distance between

[0091] S34: Calculate the gravity value of each person's location information, where the person's location information loc g The gravitational value of is:

[0092]

[0093] in:

[0094] f g Indicates the location information of the person loc g The gravitational value of

[0095] S35: Calculate the cluster center weight of each person's location information, where the person's location information loc g The cluster center weight is:

[0096]

[0097] in:

[0098] weight g Indicates the location information of the person loc g The cluster center weights of

[0099] S36: Select the personnel location information with the highest weights among U cluster centers as the cluster centers of the personnel location distribution cluster, where U represents the number of personnel location distribution clusters, merge the personnel location information of non-cluster centers into the personnel location distribution cluster where the cluster center with the closest Euclidean distance is located, and extract the cluster center location information and cluster radius of each personnel location distribution cluster, where the extracted cluster center location information of the u-th personnel location distribution cluster is δ(u), and the cluster radius is R(u), the cluster radius is the maximum Euclidean distance from the non-cluster center to the cluster center in the personnel location distribution cluster, and the cluster center location information is the personnel location information as the cluster center in the personnel location distribution cluster.

[0100] Optionally, the step S4 of constructing an air conditioning adaptive control model includes:

[0101] Construct an air conditioning adaptive control model, and use the air conditioning adaptive control model to perform adaptive air conditioning parameter adjustment in combination with personnel position distribution cluster information and air conditioning position information. The air conditioning adaptive control model takes personnel position distribution cluster information, the number of personnel in the cluster, and the air conditioning position as input, and takes the current air conditioning adjustment parameters as output. The personnel position distribution cluster information includes the cluster center position information and radius of the personnel position distribution cluster, wherein the air conditioning adaptive control model includes an input layer, a feature mapping layer, a feature information fusion layer, and an output layer;

[0102] The input layer is used to receive the personnel position distribution cluster information, the number of people in the cluster and the air-conditioning position. The feature mapping layer is used to map the personnel position distribution cluster information, the number of people in the cluster and the air-conditioning position to the high-dimensional space respectively to obtain the personnel position distribution cluster vector and the air-conditioning position vector. The feature information fusion layer is used to fuse the personnel position distribution cluster vector and the air-conditioning position vector to obtain the corresponding relationship between the air-conditioning position vector and the personnel position distribution cluster vector in space, that is, the personnel position distribution characteristics of different air-conditioning positions. The output layer calculates the air-conditioning adjustment parameters based on the personnel position distribution characteristics of different air-conditioning positions to obtain the current air-conditioning adjustment parameters of different air-conditioning positions.

[0103] Optionally, the adaptive air conditioning parameter adjustment using the air conditioning adaptive control model in combination with the personnel location distribution cluster information and the air conditioning location information includes:

[0104] The air conditioning adaptive control model is used to combine the personnel location distribution cluster information and the air conditioning location information to perform adaptive air conditioning parameter adjustment. The adaptive air conditioning parameter adjustment process is as follows:

[0105] S41: The input layer receives the personnel position distribution cluster information, the number of personnel in the cluster and the air conditioner position, where the number of personnel in the u-th personnel position distribution cluster is num(u), and the air conditioner position is:

[0106] {AC q|q∈[1,Q]}

[0107] in:

[0108] AC q represents the position coordinates of the qth air conditioner, and Q represents the total number of air conditioners;

[0109] S42: The feature mapping layer maps the personnel position distribution cluster information, the number of people in the cluster, and the air conditioner position to the high-dimensional space to obtain the personnel position distribution cluster vector Y 1 The air conditioning position vector Y 2 :

[0110] Y 1 =[(num(1),δ(1),R(1)),...,(num(u),δ(u),R(u)),...,(num(U),δ(U),R(U))]

[0111] Y 2 =(AC 1 ,...,AC q ,...,AC Q )

[0112] in:

[0113] δ(u) is the cluster center location information of the u-th personnel location distribution cluster, and R(u) is the cluster radius;

[0114] S43: The feature information fusion layer fuses the personnel position distribution cluster vector and the air conditioner position vector to obtain the spatial correspondence between the air conditioner position vector and the personnel position distribution cluster vector, that is, the personnel position distribution characteristics of different air conditioner positions, where the qth air conditioner position AC q The personnel location distribution characteristics are:

[0115] X q =[X q (1),X q (2),...,X q (u),...,X q (U)]

[0116]

[0117] in:

[0118] X q Indicates the qth air conditioner position AC q The personnel location distribution features include the mapping fusion information of U personnel location distribution clusters, X q (u) represents the mapping fusion information of the u-th personnel location branch cluster;

[0119] dis(δ(u),AC q ) represents δ(u),AC q The Euclidean distance between them, eup(·) represents the exponential function with the natural constant as the base;

[0120] S44: The output layer calculates the adjustment parameters of the air conditioner based on the personnel position distribution characteristics of different air conditioner positions, and obtains the current air conditioner adjustment parameters of different air conditioner positions, where the air conditioner position AC q The current air conditioning adjustment parameters of the air conditioner are:

[0121]

[0122] in:

[0123] Indicates the air conditioner position AC q The current air conditioning adjustment parameters of the air conditioner, W represents the adjustment parameter mapping matrix, and T represents transpose.

[0124] In order to solve the above problem, the present invention provides an electronic device, the electronic device comprising:

[0125] A memory storing at least one instruction;

[0126] Communication interface, enabling electronic equipment to communicate; and

[0127] A processor executes instructions stored in the memory to implement the above-mentioned intelligent AI energy-saving method based on integrated energy management.

[0128] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored, and the at least one instruction is executed by a processor in an electronic device to implement the above-mentioned intelligent AI energy-saving method based on integrated energy management.

[0129] Compared with the prior art, the present invention proposes an intelligent AI energy-saving method based on comprehensive energy management, which has the following advantages:

[0130] First, this scheme proposes a fingerprint information expansion method, which expands the collected fingerprint information to obtain a building fingerprint library, and initializes the generation of the expansion step parameter vector in is the expanded step size parameter vector θ 0 The three parameters in ; Set the current iteration number of the extended step parameter vector to t, the initial value of t is 0, and the maximum value is Max, then the tth iteration result of the extended step parameter vector is Construct an iterative objective function of the expanded step parameter vector, where the iterative objective function takes the expanded step parameter vector and fingerprint information as input, and takes the objective function value corresponding to the expanded step parameter vector as output, and takes the expanded step parameter vector θ as output. t The iterative objective function with input is:

[0131]

[0132]

[0133] Where: F(θ t ) represents the expanded step size parameter vector θ t is the input iterative objective function, that is, the expanded step parameter vector θ t The corresponding objective function value; L n represents the fingerprint information at the nth fixed position, Represents the expanded step size parameter vector θ t Fingerprint information at the nth fixed position obtained by interpolation calculation; ||·|| 2 represents the L2 norm; L near (n) represents the fingerprint information at the fixed position closest to the nth fixed position; loc n Indicates the position coordinates of the nth fixed position, loc n (near) represents the position coordinates of the fixed position closest to the nth fixed position among the N fixed positions, dis(loc n ,loc n (near)) represents the position coordinate loc n ,loc n (near) the distance between; represents the iterative objective function F(θ t ) is an intermediate parameter variable in the calculation process; based on the improved grey wolf algorithm, the step size parameter vector θ is expanded t Iterate and get θ t+1 ; Let t = t + 1, until the maximum number of iterations is reached, and get θ Max ; Select the location in the building where no fingerprint information is generated, combine the fingerprint information of all currently known locations, and expand the step size parameter vector θ Max Perform interpolation calculation to obtain the fingerprint information of the selected position; calculate and obtain the fingerprint information of H positions in the building to form a building fingerprint library, wherein the H positions include N fixed positions and the HN positions selected in step S16, and the representation form of the building fingerprint library is:

[0134] {L h =(r 1 (h),r 2(h),...,r m (h),...,r M (h))|h∈[1,H]}

[0135] Where: L h represents the fingerprint information of the hth position in the building, (r 1 (h),r 2 (h),...,r m (h),...,r M (h)) represents the signal strength of the WI-FI signal emitted by M signal collectors at the hth position in the building, where the position coordinates of the hth position in the building are loc(h). This solution uses the signal collector to collect fingerprint information at fixed positions in the building, and combines the gray wolf optimization algorithm to implement iterative optimization of interpolation parameters. The interpolation parameters are used to interpolate the fingerprint information of the building position that lacks fingerprint information, and the fingerprint information in the building is expanded. The signal strength information of the mobile device in the internal area of ​​the building is collected, and the fingerprint information in the building fingerprint library is matched and searched in combination with the neighboring search algorithm to realize the positioning processing of the mobile device in the current internal area of ​​the building, and obtain the location information of the people in the building.

[0136] At the same time, this scheme proposes an adaptive density clustering method to cluster the personnel location information and obtain the personnel location distribution clusters inside the building. The personnel location information clustering process is as follows: extract the current G personnel location information in the building:

[0137] {loc g |g∈[1,G]}

[0138] Where: loc g represents the g-th person location information; calculates the Euclidean distance between any two different person location information; calculates the local distance density of each person location information, where the person location information loc g The local distance density is:

[0139]

[0140] Where: g Indicates the location information of the person loc g The local distance density of the g ,loc j ) indicates the personnel location information loc g ,loc j The Euclidean distance between them; calculate the gravity value of each person's location information, where the person's location information loc g The gravitational value of is:

[0141]

[0142] Among them: g Indicates the location information of the person loc g The gravity value of each person’s location information is calculated to obtain the cluster center weight, where the person’s location information loc g The cluster center weight is:

[0143]

[0144] Among them; weight g Indicates the location information of the person loc g The cluster center weight is selected; the personnel location information with the highest weight of U cluster centers is selected as the cluster center of the personnel location distribution cluster, where U represents the number of personnel location distribution clusters, and the personnel location information of non-cluster centers is merged into the personnel location distribution cluster where the cluster center with the closest Euclidean distance is located, and the cluster center location information and cluster radius of each personnel location distribution cluster are extracted, where the cluster center location information of the u-th personnel location distribution cluster extracted is δ(u), and the cluster radius is R(u), the cluster radius is the maximum Euclidean distance from the non-cluster center to the cluster center in the personnel location distribution cluster, and the cluster center location information is the personnel location information as the cluster center in the personnel location distribution cluster. This scheme adopts an adaptive density clustering method to calculate the local distance density and gravity value of different personnel location information, realize adaptive screening and clustering processing of cluster centers, and construct an air-conditioning adaptive control model. The personnel location distribution cluster information, the number of people in the cluster, and the air-conditioning position are mapped to high-dimensional space respectively, and the personnel location distribution cluster vector and the air-conditioning position vector are fused to obtain the corresponding relationship between the air-conditioning position vector and the personnel location distribution cluster vector in space, that is, the personnel location distribution characteristics of different air-conditioning locations. The air-conditioning adjustment parameters are adjusted in combination with the personnel location distribution characteristics. The denser the personnel, the higher the energy consumption of the corresponding air-conditioning after adjustment, thereby realizing air-conditioning energy-saving control based on personnel location distribution. BRIEF DESCRIPTION OF THE DRAWINGS

[0145] Figure 1 A flowchart of an intelligent AI energy-saving method based on integrated energy management provided by one embodiment of the present invention;

[0146] Figure 2 A schematic diagram of the structure of an electronic device for implementing an intelligent AI energy-saving method based on integrated energy management provided by one embodiment of the present invention.

[0147] In the figure: 1 electronic device, 10 processor, 11 memory, 12 program, 13 communication interface.

[0148] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0149] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0150] The embodiment of the present application provides an intelligent AI energy-saving method based on integrated energy management. The execution subject of the intelligent AI energy-saving method based on integrated energy management includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided in the embodiment of the present application. In other words, the intelligent AI energy-saving method based on integrated energy management can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0151] Embodiment 1:

[0152] S1: deploy signal collectors in the building, and use the signal collectors to collect signal strengths at different fixed positions to form fingerprint information, and perform fingerprint expansion on the collected fingerprint information to obtain a building fingerprint library.

[0153] In the step S1, a signal collector is deployed in the building, and the signal collector is used to collect signal strengths at different fixed positions to form fingerprint information, including:

[0154] M signal collectors are deployed in the building, and N fixed positions are set in the building. The signal collectors are used to collect the signal strength at different fixed positions to form fingerprint information. The signal collector can send out WI-FI signals. The fingerprint information collection process at different fixed positions is as follows:

[0155] Place the mobile device with WI-FI turned on in a fixed position to receive the WI-FI signal from the signal collector;

[0156] The mobile device records the wireless signal addresses and corresponding signal strengths of different WI-FI signals received, and sends the recorded content to the data processing center for aggregation;

[0157] The data center matches the wireless signal address of the WI-FI signal with the wireless signal address of the signal collector, and summarizes the signal strengths of the WI-FI signals emitted by the M signal collectors recorded by different mobile devices;

[0158] The signal strength of the Wi-Fi signals emitted by M signal collectors recorded by mobile devices at N fixed locations constitutes fingerprint information, where the fingerprint information is expressed in the form of:

[0159]

[0160] in:

[0161] L n represents the fingerprint information at the nth fixed position, Indicates the signal strength of the Wi-Fi signal emitted by M signal collectors at the nth fixed position, Indicates the signal strength of the Wi-Fi signal emitted by the mth signal collector at the nth fixed position;

[0162] The collected fingerprint information is expanded to obtain a building fingerprint library.

[0163] The step of performing fingerprint expansion on the collected fingerprint information to obtain a building fingerprint library includes:

[0164] The fingerprint expansion process is as follows:

[0165] S11: Initialize and generate the expansion step parameter vector in is the expanded step size parameter vector θ 0 The three parameters in;

[0166] S12: Set the current iteration number of the extended step parameter vector to t, the initial value of t is 0, and the maximum value is Max. Then the tth iteration result of the extended step parameter vector is

[0167] S13: construct an iterative objective function of the expanded step parameter vector, wherein the iterative objective function takes the expanded step parameter vector and the fingerprint information as input, takes the objective function value corresponding to the expanded step parameter vector as output, and takes the expanded step parameter vector θ as output. t The iterative objective function with input is:

[0168]

[0169] in:

[0170] F(θ t ) represents the expanded step size parameter vector θ t is the input iterative objective function, that is, the expanded step parameter vector θ t The corresponding objective function value;

[0171] L n represents the fingerprint information at the nth fixed position, Represents the expanded step size parameter vector θ t Fingerprint information at the nth fixed position obtained by interpolation calculation;

[0172] ||·|| 2 represents the L2 norm;

[0173] L near (n) represents the fingerprint information at the fixed position closest to the nth fixed position;

[0174] loc n Indicates the position coordinates of the nth fixed position, loc n (near) represents the position coordinates of the fixed position closest to the nth fixed position among the N fixed positions, dis(loc n ,loc n (near)) represents the position coordinate loc n ,loc n (near) the distance between;

[0175] represents the iterative objective function F(θ t ) Intermediate parameter variables in the calculation process;

[0176] S14: Based on the improved grey wolf algorithm, the step size parameter vector θ is expanded t Iterate and get θ t+1 , where the iterative process is:

[0177] S141: Input the t expanded step length parameter vectors obtained in the current iteration into the iterative objective function, and use the expanded step length parameter vector with the smallest objective function value as the prey coordinate vector of this iteration

[0178] S142: Based on prey coordinate vector The expansion step parameter vector is iterated, where the iteration formula is:

[0179]

[0180] in:

[0181] rand(0,1) represents a random number between 0 and 1;

[0182] a t Represents the proportional coefficient of the objective function value used for iteration, a max Represents the preset maximum objective function value proportional coefficient, a min Indicates the preset minimum objective function value proportional coefficient;

[0183] ε represents the convergence factor;

[0184] b t represents the dynamic weight factor used for iteration, b max Indicates the preset maximum weight factor, b min Indicates the preset minimum weight factor;

[0185] S15: Let t = t + 1, return to step S14, until the maximum number of iterations is reached, and get θ Max ;

[0186] S16: Select the location in the building where no fingerprint information is generated, combine the fingerprint information of all currently known locations, and expand the step length parameter vector θ Max And the calculation formula in step S13 is used for interpolation calculation to obtain the fingerprint information of the selected position;

[0187] S17: Repeat step S16 to calculate the fingerprint information of H positions in the building to form a building fingerprint library, where the H positions include N fixed positions and HN positions selected in step S16. The representation form of the building fingerprint library is:

[0188] {L h =(r 1 (h),r 2 (h),...,r m (h),...,r M (h))|h∈[1,H]}

[0189] in:

[0190] L h represents the fingerprint information of the hth position in the building, (r 1 (h),r 2 (h),...,r m (h),...,r M (h)) represents the signal strength of the WI-FI signal emitted by M signal collectors at the h-th position in the building, where the position coordinate of the h-th position in the building is loc(h).

[0191] S2: Collect signal strength information of mobile devices in the internal area of ​​the building, perform fingerprint matching in combination with the building fingerprint library, obtain location information of mobile devices in the building, and use the location information of mobile devices as personnel location information.

[0192] The step S2 includes collecting signal strength information of mobile devices in the internal area of ​​the building, including:

[0193] Collect the signal strength information of mobile devices in the internal area of ​​the building. The collection process of the mobile device signal strength information is as follows:

[0194] The mobile device receives the WI-FI signals from the M signal collectors and records the signal strength of each WI-FI signal to form the signal strength information of the mobile device;

[0195] The mobile device sends the signal strength information to the data processing center in real time;

[0196] The data processing center forms a collection of signal strength information of all mobile devices currently in the building:

[0197] {L(g)|g∈[1,G]}

[0198] in:

[0199] L(g) represents the signal strength information of the gth mobile device in the internal area of ​​the building, and G represents the total number of all mobile devices currently in the building;

[0200] Combined with the building fingerprint database, the signal strength information of the mobile device is fingerprint matched to obtain the location information of the mobile device inside the building, and the location information of the mobile device is used as the personnel location information. The location information of the mobile device obtained by matching the signal strength information L(g) is loc g ,loc g is the position coordinate of the g-th mobile device in the internal area of ​​the building corresponding to the signal strength information L(g).

[0201] The fingerprint matching of the signal strength information of the mobile device in combination with the building fingerprint library to obtain the location information of the mobile device inside the building includes:

[0202] Combined with the building fingerprint library, the signal strength information of the mobile device is fingerprint matched to obtain the location information of the mobile device inside the building. The fingerprint matching process of the signal strength information L(g) is as follows:

[0203] Calculate the similarity between the signal strength signal L(g) and the fingerprint information corresponding to H positions in the building fingerprint database, where the signal strength signal L(g) and the fingerprint information L h The similarity calculation formula is:

[0204]

[0205] in:

[0206] Sim(L(g),L h ) represents the signal strength signal L(g) and the fingerprint information L h similarity;

[0207] ||·|| 2 represents the L2 norm;

[0208] Select the K position coordinates corresponding to the K fingerprint information with the highest similarity as the neighboring position coordinates of the signal strength signal L(g):

[0209] (loc g (1),loc g (2),...,loc g (k),...,locg (K))

[0210] in:

[0211] loc g (k) represents the kth neighbor position coordinate of the signal strength signal L(g), k∈[1,K];

[0212] Perform weighted calculation on the K nearest neighbor position coordinates to obtain the mobile device location information loc corresponding to the signal strength information L(g) g :

[0213]

[0214] in:

[0215] w F Represents the weighted weight of the kth neighbor position coordinate;

[0216] Represents the kth neighbor position coordinates to the mobile device location information loc g The intensity weight of

[0217] Represents the kth neighbor position coordinates to the mobile device location information loc g The distance weight of

[0218] e represents a natural constant;

[0219] d 0 Indicates the reference distance of signal propagation, d 1 Represents the attenuation coefficient of signal propagation;

[0220] L g (k) represents the kth neighbor position coordinate loc g (k) Fingerprint information in the building fingerprint database.

[0221] S3: Cluster the personnel location information to obtain the personnel location distribution clusters inside the building, and extract the cluster center location information and cluster radius of each personnel location distribution cluster.

[0222] In the step S3, the personnel location information is clustered to obtain the personnel location distribution cluster inside the building, including:

[0223] The personnel location information is clustered to obtain the personnel location distribution clusters inside the building. The personnel location information clustering process is as follows:

[0224] S31: Extract the current location information of G persons in the building:

[0225] {loc g |g∈[1,G]}

[0226] in:

[0227] loc g Indicates the location information of the g-th person;

[0228] S32: Calculate the Euclidean distance between any two different personnel location information;

[0229] S33: Calculate the local distance density of each person's location information, where the person's location information loc g The local distance density of is;

[0230]

[0231] in:

[0232] ρ g Indicates the location information of the person loc g The local distance density of

[0233] dist represents the preset cutoff distance;

[0234] dis(loc g ,loc j ) indicates the personnel location information loc g ,loc j The Euclidean distance between

[0235] S34: Calculate the gravity value of each person's location information, where the person's location information loc g The gravitational value of is:

[0236]

[0237] in:

[0238] f g Indicates the location information of the person loc g The gravitational value of

[0239] S35: Calculate the cluster center weight of each person's location information, where the person's location information loc g The cluster center weight is:

[0240]

[0241] in:

[0242] weight g Indicates the location information of the person loc g The cluster center weights of

[0243] S36: Select the personnel location information with the highest weights among U cluster centers as the cluster centers of the personnel location distribution cluster, where U represents the number of personnel location distribution clusters, merge the personnel location information of non-cluster centers into the personnel location distribution cluster where the cluster center with the closest Euclidean distance is located, and extract the cluster center location information and cluster radius of each personnel location distribution cluster, where the extracted cluster center location information of the u-th personnel location distribution cluster is δ(u), and the cluster radius is R(u), the cluster radius is the maximum Euclidean distance from the non-cluster center to the cluster center in the personnel location distribution cluster, and the cluster center location information is the personnel location information as the cluster center in the personnel location distribution cluster.

[0244] S4: Construct an air conditioning adaptive control model, and adaptively adjust the air conditioning parameters in combination with the personnel location distribution cluster information and the air conditioning location information. The air conditioning adaptive control model takes the personnel location distribution cluster information, the number of people in the cluster and the air conditioning location as input, and takes the current air conditioning adjustment parameters as output, wherein the personnel location distribution cluster information includes the cluster center location information and radius of the personnel location distribution cluster.

[0245] The air conditioning adaptive control model is constructed in step S4, including:

[0246] Construct an air conditioning adaptive control model, and use the air conditioning adaptive control model to perform adaptive air conditioning parameter adjustment in combination with personnel position distribution cluster information and air conditioning position information. The air conditioning adaptive control model takes personnel position distribution cluster information, the number of personnel in the cluster, and the air conditioning position as input, and takes the current air conditioning adjustment parameters as output. The personnel position distribution cluster information includes the cluster center position information and radius of the personnel position distribution cluster, wherein the air conditioning adaptive control model includes an input layer, a feature mapping layer, a feature information fusion layer, and an output layer;

[0247] The input layer is used to receive the personnel position distribution cluster information, the number of people in the cluster and the air-conditioning position. The feature mapping layer is used to map the personnel position distribution cluster information, the number of people in the cluster and the air-conditioning position to the high-dimensional space respectively to obtain the personnel position distribution cluster vector and the air-conditioning position vector. The feature information fusion layer is used to fuse the personnel position distribution cluster vector and the air-conditioning position vector to obtain the corresponding relationship between the air-conditioning position vector and the personnel position distribution cluster vector in space, that is, the personnel position distribution characteristics of different air-conditioning positions. The output layer calculates the air-conditioning adjustment parameters based on the personnel position distribution characteristics of different air-conditioning positions to obtain the current air-conditioning adjustment parameters of different air-conditioning positions.

[0248] The method of using the air conditioning adaptive control model to perform adaptive air conditioning parameter adjustment in combination with the personnel location distribution cluster information and the air conditioning location information includes:

[0249] The air conditioning adaptive control model is used to combine the personnel location distribution cluster information and the air conditioning location information to perform adaptive air conditioning parameter adjustment. The adaptive air conditioning parameter adjustment process is as follows:

[0250] S41: The input layer receives the personnel position distribution cluster information, the number of personnel in the cluster and the air conditioner position, where the number of personnel in the u-th personnel position distribution cluster is num(u), and the air conditioner position is:

[0251] {AC q |q∈[1,Q]}

[0252] in:

[0253] AC q represents the position coordinates of the qth air conditioner, and Q represents the total number of air conditioners;

[0254] S42: The feature mapping layer maps the personnel position distribution cluster information, the number of people in the cluster, and the air conditioner position to the high-dimensional space to obtain the personnel position distribution cluster vector Y 1 and the air conditioner position vector Y 2 :

[0255] Y 1 =[(num(1),δ(1),R(1)),...,(num(u),δ(u),R(u)),...,(num(U),δ(U),R(U))]

[0256] Y 2 =(AC 1 ,...,AC q ,...,AC Q )

[0257] in:

[0258] δ(u) is the cluster center location information of the u-th personnel location distribution cluster, and R(u) is the cluster radius;

[0259] S43: The feature information fusion layer fuses the personnel position distribution cluster vector and the air conditioner position vector to obtain the spatial correspondence between the air conditioner position vector and the personnel position distribution cluster vector, that is, the personnel position distribution characteristics of different air conditioner positions, where the qth air conditioner position AC q The personnel location distribution characteristics are:

[0260] X q =[X q (1),X q (2),...,X q (u),...,X q (U)]

[0261]

[0262] in:

[0263] X q Indicates the qth air conditioner position AC q The personnel location distribution features include the mapping fusion information of U personnel location distribution clusters, X q (u) represents the mapping fusion information of the u-th personnel location distribution cluster;

[0264] dis(δ(u),AC q ) represents δ(u),AC q The Euclidean distance between them, exp(·) represents the exponential function with the natural constant as the base;

[0265] S44: The output layer calculates the adjustment parameters of the air conditioner based on the personnel position distribution characteristics of different air conditioner positions, and obtains the current air conditioner adjustment parameters of different air conditioner positions, where the air conditioner position AC q The current air conditioning adjustment parameters of the air conditioner are:

[0266]

[0267] in:

[0268] Indicates the air conditioner position AC q The current air conditioning adjustment parameters of the air conditioner, W represents the adjustment parameter mapping matrix, and T represents transpose.

[0269] Embodiment 2:

[0270] like Figure 2 , is a schematic diagram of the structure of an electronic device for implementing an intelligent AI energy-saving method based on integrated energy management provided by an embodiment of the present invention.

[0271] The electronic device 1 may include a processor 10 , a memory 11 , a communication interface 13 and a bus, and may also include a computer program stored in the memory 11 and executable on the processor 10 , such as a program 12 .

[0272] The memory 11 includes at least one type of readable storage medium, and the readable storage medium includes a flash memory, a mobile hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), a magnetic memory, a disk, an optical disk, etc. The memory 11 may be an internal storage unit of the electronic device 1 in some embodiments, such as a mobile hard disk of the electronic device 1. The memory 11 may also be an external storage device of the electronic device 1 in other embodiments, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 1. Further, the memory 11 may also include both an internal storage unit of the electronic device 1 and an external storage device. The memory 11 may be used not only to store application software and various types of data installed in the electronic device 1, such as the code of the program 12, etc., but also to temporarily store data that has been output or is to be output.

[0273] The processor 10 may be composed of an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or a plurality of integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips. The processor 10 is the control core (Control Unit) of the electronic device, and uses various interfaces and lines to connect the various components of the entire electronic device, and executes or executes the programs or modules stored in the memory 11 (such as the program 12 for realizing intelligent AI energy saving based on integrated energy management), and calls the data stored in the memory 11 to execute various functions of the electronic device 1 and process data.

[0274] The communication interface 13 may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices, and to achieve connection and communication between internal components of the electronic device.

[0275] The bus may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is configured to realize connection and communication between the memory 11 and at least one processor 10, etc.

[0276] Figure 2 Only an electronic device with components is shown, and those skilled in the art will understand that Figure 2 The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0277] For example, although not shown, the electronic device 1 may also include a power source (such as a battery) for supplying power to each component. Preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that the power management device can realize functions such as charging management, discharging management, and power consumption management. The power source may also include any components such as one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, etc. The electronic device 1 may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0278] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode) touch device. The display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device 1 and to display a visual user interface.

[0279] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0280] The program 12 stored in the memory 11 in the electronic device 1 is a combination of multiple instructions, and when executed in the processor 10, it can achieve:

[0281] Deploy signal collectors in the building, and use the signal collectors to collect signal strengths at different fixed positions to form fingerprint information, and perform fingerprint expansion on the collected fingerprint information to obtain a building fingerprint library;

[0282] Collect signal strength information of mobile devices in the internal area of ​​the building, perform fingerprint matching in combination with the building fingerprint database, obtain location information of mobile devices in the building, and use the location information of mobile devices as personnel location information;

[0283] Cluster the personnel location information to obtain the personnel location distribution clusters inside the building, and extract the cluster center location information and cluster radius of each personnel location distribution cluster;

[0284] An air conditioning adaptive control model is constructed, and adaptive air conditioning parameters are adjusted based on the personnel location distribution cluster information and air conditioning location information.

[0285] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to Figure 1 to Figure 2 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0286] It should be noted that the serial numbers of the above-mentioned embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. And the terms "including", "comprising" or any other variants thereof in this article are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the term "including a ..." does not exclude the existence of other identical elements in the process, device, article or method including the element.

[0287] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0288] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. An intelligent AI energy-saving method based on comprehensive energy management, characterized in that: The method comprises: S1: deploy signal collectors in the building, and use the signal collectors to collect signal strengths at different fixed positions to form fingerprint information, and perform fingerprint expansion on the collected fingerprint information to obtain a building fingerprint library; The fingerprint expansion process is as follows: S11: Initialize and generate the expansion step parameter vector in are the three parameters in the expanded step parameter vector θ0; S12: Set the current iteration number of the extended step parameter vector to t, the initial value of t is 0, and the maximum value is Max. Then the tth iteration result of the extended step parameter vector is S13: construct an iterative objective function of the expanded step parameter vector, wherein the iterative objective function takes the expanded step parameter vector and the fingerprint information as input, takes the objective function value corresponding to the expanded step parameter vector as output, and takes the expanded step parameter vector θ as output. t The iterative objective function with input is: in: N represents the number of fixed positions; F(θ t ) represents the expanded step size parameter vector θ t is the input iterative objective function, that is, the expanded step parameter vector θ t The corresponding objective function value; L n represents the fingerprint information at the nth fixed position, Represents the expanded step size parameter vector θ t Fingerprint information at the nth fixed position obtained by interpolation calculation; ||·||2 represents the L2 norm; L near (n) represents the fingerprint information at the fixed position closest to the nth fixed position; loc n Indicates the position coordinates of the nth fixed position, loc n (near) represents the position coordinates of the fixed position closest to the nth fixed position among the N fixed positions, dis(loc n ,loc n (near)) represents the position coordinate loc n ,loc n (near) the distance between; represents the iterative objective function F(θ t ) Intermediate parameter variables in the calculation process; S14: Based on the improved grey wolf algorithm, the step size parameter vector θ is expanded t Iterate and get θ t+1 ; S15: Let t = t + 1, return to step S14, until the maximum number of iterations is reached, and get θ Max ; S16: Select the location in the building where no fingerprint information is generated, combine the fingerprint information of all currently known locations, and expand the step length parameter vector θ Max And the calculation formula in step S13 is used for interpolation calculation to obtain the fingerprint information of the selected position; S17: Repeat step S16 to calculate the fingerprint information of H positions in the building to form a building fingerprint library, where the H positions include N fixed positions and HN positions selected in step S16. The representation form of the building fingerprint library is: {L h =(r 1 (h),r 2 (h),...,r m (h),...,r M (h))|h∈[1,H]} in: L h represents the fingerprint information of the hth position in the building, (r 1 (h),r 2 (h),...,r m (h),...,r M (h)) represents the signal strength of the WI-FI signal emitted by M signal collectors at the h-th position in the building, where the position coordinate of the h-th position in the building is loc(h); S2: Collect signal strength information of mobile devices in the internal area of ​​the building, perform fingerprint matching in combination with the building fingerprint database, obtain location information of mobile devices in the building, and use the location information of mobile devices as personnel location information; S3: Clustering personnel location information to obtain personnel location distribution clusters inside the building, and extracting cluster center location information and cluster radius of each personnel location distribution cluster; S4: construct an air conditioning adaptive control model, and perform adaptive air conditioning parameter adjustment based on the personnel position distribution cluster information and the air conditioning position information, wherein the air conditioning adaptive control model takes the personnel position distribution cluster information, the number of personnel in the cluster and the air conditioning position as input, and takes the current air conditioning adjustment parameters as output, wherein the personnel position distribution cluster information includes the cluster center position information and radius of the personnel position distribution cluster; The adaptive air conditioning parameter adjustment process is as follows: S41: The input layer receives the personnel position distribution cluster information, the number of personnel in the cluster and the air conditioner position, where the number of personnel in the u-th personnel position distribution cluster is num(u), and the air conditioner position is: {AC q |q∈[1,Q]} in: AC q represents the position coordinates of the qth air conditioner, and Q represents the total number of air conditioners; S42: The feature mapping layer maps the personnel position distribution cluster information, the number of personnel in the cluster, and the air conditioner position to the high-dimensional space to obtain the personnel position distribution cluster vector Y1 and the air conditioner position vector Y2: Y1=[(num(1),δ(1),R(1)),...,(num(u),δ(u),R(u)),...,(num(U),δ(U),R(U))] Y2=(AC1,...,AC q ,..., AND Q ) in: δ(u) is the cluster center location information of the u-th personnel location distribution cluster, and R(u) is the cluster radius; S43: The feature information fusion layer fuses the personnel position distribution cluster vector and the air conditioner position vector to obtain the spatial correspondence between the air conditioner position vector and the personnel position distribution cluster vector, that is, the personnel position distribution characteristics of different air conditioner positions, where the qth air conditioner position AC q The personnel location distribution characteristics are: X q =[X q (1),X q (2),...,X q (u),...,X q (U)] in: x q Indicates the qth air conditioner position AC q The personnel location distribution feature contains the mapping fusion information of U personnel location distribution clusters, x q (u) represents the mapping fusion information of the u-th personnel location distribution cluster; dis(δ(u),AC q ) represents δ(u),AC q The Euclidean distance between them, exp(·) represents the exponential function with the natural constant as the base; S44: The output layer calculates the adjustment parameters of the air conditioner based on the personnel position distribution characteristics of different air conditioner positions, and obtains the current air conditioner adjustment parameters of different air conditioner positions, where the air conditioner position AC A The current air conditioning adjustment parameters of the air conditioner are: in: Indicates the air conditioner position AC q The current air conditioning adjustment parameters of the air conditioner, W represents the adjustment parameter mapping matrix, and T represents transpose.

2. The intelligent AI energy-saving method based on comprehensive energy management as claimed in claim 1, characterized in that: In the step S1, a signal collector is deployed in the building, and the signal collector is used to collect signal strengths at different fixed positions to form fingerprint information, including: M signal collectors are deployed in the building, and N fixed positions are set in the building. The signal collectors are used to collect the signal strength at different fixed positions to form fingerprint information. The signal collector can send out WI-FI signals. The fingerprint information collection process at different fixed positions is as follows: Place the mobile device with WI-FI turned on in a fixed position to receive the WI-FI signal from the signal collector; The mobile device records the wireless signal addresses and corresponding signal strengths of different WI-FI signals received, and sends the recorded content to the data processing center for aggregation; The data center matches the wireless signal address of the WI-FI signal with the wireless signal address of the signal collector, and summarizes the signal strengths of the WI-FI signals emitted by the M signal collectors recorded by different mobile devices; The signal strength of the Wi-Fi signals emitted by M signal collectors recorded by mobile devices at N fixed locations constitutes fingerprint information, where the fingerprint information is expressed in the form of: in: L n represents the fingerprint information at the nth fixed position, Indicates the signal strength of the Wi-Fi signal emitted by M signal collectors at the nth fixed position, Indicates the signal strength of the Wi-Fi signal emitted by the mth signal collector at the nth fixed position; The collected fingerprint information is expanded to obtain a building fingerprint library.

3. The intelligent AI energy-saving method based on comprehensive energy management according to claim 1, characterized in that: The step S2 includes collecting signal strength information of mobile devices in the internal area of ​​the building, including: Collect the signal strength information of mobile devices in the internal area of ​​the building. The collection process of the mobile device signal strength information is as follows: The mobile device receives the WI-FI signals from the M signal collectors and records the signal strength of each WI-FI signal to form the signal strength information of the mobile device; The mobile device sends the signal strength information to the data processing center in real time; The data processing center forms a collection of signal strength information of all mobile devices currently in the building: {L(g)|g∈[1,G]} in: L(g) represents the signal strength information of the gth mobile device in the internal area of ​​the building, and G represents the total number of all mobile devices currently in the building; Combined with the building fingerprint database, the signal strength information of the mobile device is fingerprint matched to obtain the location information of the mobile device inside the building, and the location information of the mobile device is used as the personnel location information. The location information of the mobile device obtained by matching the signal strength information L(g) is loc g ,loc g is the position coordinate of the g-th mobile device in the internal area of ​​the building corresponding to the signal strength information L(g).

4. The intelligent AI energy-saving method based on integrated energy management as claimed in claim 3, characterized in that: The fingerprint matching of the signal strength information of the mobile device in combination with the building fingerprint library to obtain the location information of the mobile device inside the building includes: Combined with the building fingerprint library, the signal strength information of the mobile device is fingerprint matched to obtain the location information of the mobile device inside the building. The fingerprint matching process of the signal strength information L(g) is as follows: Calculate the similarity between the signal strength signal L(g) and the fingerprint information corresponding to H positions in the building fingerprint database, where the signal strength signal L(g) and the fingerprint information L h The similarity calculation formula is: in: Sim(L(g),L h ) represents the signal strength signal L(g) and the fingerprint information L h similarity; ||·||2 represents the L2 norm; Select the K position coordinates corresponding to the K fingerprint information with the highest similarity as the neighboring position coordinates of the signal strength signal L(g): (place g (1),place g (2),...,place g (k),...,place g (K)) in: loc g (k) represents the kth neighbor position coordinate of the signal strength signal L(g), k∈[1,K]; Perform weighted calculation on the K nearest neighbor position coordinates to obtain the mobile device location information loc corresponding to the signal strength information L(g) Z : in: w k Represents the weighted weight of the kth neighbor position coordinate; Represents the kth neighbor position coordinates to the mobile device location information loc g The intensity weight of Represents the kth neighbor position coordinates to the mobile device location information loc g The distance weight of e represents a natural constant; d0 represents the reference distance of signal propagation, and d1 represents the attenuation coefficient of signal propagation; L g (k) represents the kth neighbor position coordinate loc g (k) Fingerprint information in the building fingerprint database.

5. The intelligent AI energy-saving method based on comprehensive energy management as claimed in claim 1, characterized in that: In the step S3, the personnel location information is clustered to obtain the personnel location distribution cluster inside the building, including: The personnel location information is clustered to obtain the personnel location distribution clusters inside the building. The personnel location information clustering process is as follows: S31: Extract the current location information of G persons in the building: {place g |g∈[1,G]} in: loc g Indicates the location information of the g-th person; S32: Calculate the Euclidean distance between any two different personnel location information; S33: Calculate the local distance density of each person's location information, where the person's location information loc g The local distance density is: in: ρ g Indicates the location information of the person loc g The local distance density of dist represents the preset cutoff distance; dis(loc g ,loc j ) indicates the personnel location information loc g ,loc j The Euclidean distance between S34: Calculate the gravity value of each person's location information, where the person's location information loc g The gravitational value of is: in: f Z Indicates the location information of the person loc g The gravitational value of S35: Calculate the cluster center weight of each person's location information, where the person's location information loc g The cluster center weight is: in: weight g Indicates the location information of the person loc g The cluster center weights of S36: Select the personnel location information with the highest weights among U cluster centers as the cluster centers of the personnel location distribution cluster, where U represents the number of personnel location distribution clusters, merge the personnel location information of non-cluster centers into the personnel location distribution cluster where the cluster center with the closest Euclidean distance is located, and extract the cluster center location information and cluster radius of each personnel location distribution cluster, where the extracted cluster center location information of the u-th personnel location distribution cluster is δ(u), and the cluster radius is R(u), the cluster radius is the maximum Euclidean distance from the non-cluster center to the cluster center in the personnel location distribution cluster, and the cluster center location information is the personnel location information as the cluster center in the personnel location distribution cluster.

6. The intelligent AI energy-saving method based on comprehensive energy management according to claim 1, characterized in that: The air conditioning adaptive control model is constructed in step S4, including: Construct an air conditioning adaptive control model, and use the air conditioning adaptive control model to perform adaptive air conditioning parameter adjustment in combination with personnel position distribution cluster information and air conditioning position information. The air conditioning adaptive control model takes personnel position distribution cluster information, the number of personnel in the cluster, and the air conditioning position as input, and takes the current air conditioning adjustment parameters as output. The personnel position distribution cluster information includes the cluster center position information and radius of the personnel position distribution cluster, wherein the air conditioning adaptive control model includes an input layer, a feature mapping layer, a feature information fusion layer, and an output layer; The input layer is used to receive the personnel position distribution cluster information, the number of people in the cluster and the air-conditioning position. The feature mapping layer is used to map the personnel position distribution cluster information, the number of people in the cluster and the air-conditioning position to the high-dimensional space respectively to obtain the personnel position distribution cluster vector and the air-conditioning position vector. The feature information fusion layer is used to fuse the personnel position distribution cluster vector and the air-conditioning position vector to obtain the corresponding relationship between the air-conditioning position vector and the personnel position distribution cluster vector in space, that is, the personnel position distribution characteristics of different air-conditioning positions. The output layer calculates the air-conditioning adjustment parameters based on the personnel position distribution characteristics of different air-conditioning positions to obtain the current air-conditioning adjustment parameters of different air-conditioning positions.

Images