AIOT energy consumption data analysis system and method for energy-saving intelligent control

Abstract

The invention relates to an AIOT energy consumption data analysis system and method for energy-saving intelligent control, and belongs to the technical field of energy consumption analys.The method specifically comprises the steps that according to floor numbers and device priorities, energy consumption devices in a building are numbered to obtain an energy consumption device set, historical energy consumption data are divided according to a preset period, and the number of the energy consumption devices in the building is calculated; obtaining an energy consumption standard value according to the data size of the historical energy consumption data and the energy consumption category, predicting the energy consumption of the next period according to the current energy consumption data to obtain energy consumption prediction data, and obtaining energy consumption abnormal information according to the energy consumption standard value and the energy consumption prediction data; the equipment operation sequence is obtained based on the energy consumption abnormal information, the energy consumption equipment set and the energy consumption standard value, the regulation and control instruction is generated according to the equipment operation sequence, the energy consumption equipment in the building is controlled through the regulation and control instruction, the real-time performance and accuracy of building energy consumption analysis are improved, intelligent control over the energy consumption equipment of the building is achieved, and the practicability is high. And the purposes of energy conservation and consumption reduction of buildings are achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of energy consumption analysis, and specifically relates to an AIOT energy consumption data analysis system and method for energy-saving intelligent control. Background Art

[0003] The traditional public building energy management method is usually based on manual experience summary, setting time to adjust the operating time of energy-consuming equipment to achieve energy saving and consumption reduction of the building. This method cannot collect building energy consumption information in real time, and cannot analyze the building's energy consumption data in real time. It cannot achieve refined, personalized and intelligent energy saving and consumption reduction based on energy consumption data, resulting in inefficient building energy management and energy waste. Summary of the invention

[0004] In order to solve the above-mentioned problems existing in the prior art, the present invention provides an AIOT energy consumption data analysis system and method for energy-saving intelligent control. According to the floor number and equipment priority, the energy consumption equipment in the building is numbered to obtain an energy consumption equipment set, and the historical energy consumption data is divided according to a preset period to obtain the energy consumption category, and the energy consumption standard value is obtained according to the data volume and energy consumption category of the historical energy consumption data. The energy consumption of the next period is predicted according to the current energy consumption data to obtain energy consumption forecast data, and energy consumption abnormality information is obtained according to the energy consumption standard value and the energy consumption forecast data. Based on the energy consumption abnormality information, the energy consumption equipment set and the energy consumption standard value, the equipment operation sequence is obtained, and the control instructions are generated according to the equipment operation sequence. The energy consumption equipment in the building is controlled by the control instructions, which improves the real-time and accuracy of the building energy consumption analysis, realizes the intelligent control of the building's energy consumption equipment, and achieves the goal of energy saving and consumption reduction of the building.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] An AIOT energy consumption data analysis method for energy-saving intelligent control includes the following steps:

[0007] S1: According to the floor number and equipment priority, the energy-consuming equipment in the building is uniformly numbered to obtain the energy-consuming equipment set;

[0008] S2: obtaining current energy consumption data and historical energy consumption data of the building, dividing the historical energy consumption data according to a preset period, analyzing the divided historical energy consumption data to obtain energy consumption categories, and obtaining energy consumption standard values ​​by processing the data volume and energy consumption categories of the historical energy consumption data;

[0009] S3: predicting the energy consumption of the next cycle according to the current energy consumption data to obtain energy consumption prediction data, and obtaining energy consumption abnormality information based on the energy consumption standard value and the energy consumption prediction data;

[0010] S4: Based on the abnormal energy consumption information, the energy consumption equipment set and the energy consumption standard value are analyzed and processed to obtain an equipment operation sequence, and a control instruction is generated according to the equipment operation sequence, and the energy consumption equipment in the building is controlled by the control instruction.

[0011] Preferably, the calculation steps of the energy consumption categories are specifically as follows:

[0012] Step 1: normalize the historical energy consumption data to obtain first historical energy consumption data, obtain a historical energy consumption data set based on the first historical energy consumption data, and divide the historical energy consumption data set according to a preset period. The historical energy consumption data set is expressed as: And d and h are both positive integers, where d represents the day index and h represents the hour index. Indicates the first historical energy consumption data at h time on day d, represents the first historical energy consumption data set of 1 to 24 hours with day index d, and

[0013] Step 2: The number of energy consumption categories is preset to k. In the historical energy consumption data set, a first historical energy consumption data is randomly selected as the initial class representative element. According to the historical energy consumption data set and the initial class representative element, the distance between the first class representative element and the historical energy consumption data set is calculated. in, represents the distance of the first class representative element, Ni represents the initial class representative element, and the class representative element probability is calculated based on the distance of the first class representative element.

[0014] Step 3: Take the first historical energy consumption data corresponding to the maximum probability of the class representative element as the new class representative element, take the new class representative element as the new initial class representative element, repeat step 2, and in the process of repeating step 2, perform deduplication processing on the selected class representative elements until k class representative elements are selected to obtain the class representative element set C j Indicates the representative element of the jth class;

[0015] Step 4: Extract features from the historical energy consumption data set and the class representative element set to obtain the historical energy consumption feature vector and class represents element feature vector in, express The coordinate value of the corresponding historical energy consumption feature vector in the zth dimension, Represents C j The coordinate value of the corresponding class representative element feature vector in the zth dimension; according to the historical energy consumption feature vector and the class representative element feature vector, the distance of the second class representative element is calculated, The second type of representative element distance represents the distance from the first historical energy consumption data to the representative elements of each class, D d，h，j Represents the distance of the second category representative element. By comparing the distance between the historical energy consumption data and the second category representative element of each class representative element, the first historical energy consumption data is divided into the class where the class representative element closest to it is located. After the first historical energy consumption data is divided, k energy consumption categories and corresponding energy consumption category data sets are obtained.

[0016] Preferably, the specific calculation method of the energy consumption standard value includes: obtaining the energy consumption category data amount according to the historical energy consumption data amount and the energy consumption category data set; arranging the first historical energy consumption data in the energy consumption category data set in order from small to large, and calculating the median position of the energy consumption category, and the calculation formula is: Among them, P j represents the median position of the energy consumption category data set corresponding to the jth energy consumption category, m j The energy consumption category data volume of the j-th energy consumption category is represented, and the first historical energy consumption data corresponding to the median position is used as the energy consumption standard value of the j-th energy consumption category.

[0017] Preferably, the step P3 specifically comprises the following steps:

[0018] S301: adding current energy consumption data to a historical energy consumption data set, extracting features from the historical energy consumption data set to obtain historical time features, constructing an energy consumption prediction model based on the historical time features combined with the historical energy consumption data set, and obtaining energy consumption prediction data through calculation using the energy consumption prediction model;

[0019] S302: According to the time characteristics of the current energy consumption data, the time characteristics to be predicted of the energy consumption prediction data are obtained. According to the time characteristics to be predicted, the energy consumption prediction data NH=α+β×wi is calculated by the energy consumption prediction model, wherein NH represents the energy consumption prediction data, α and β represent parameter factors, and wi represents the time characteristics to be predicted. The parameter factors are obtained by processing the historical energy consumption data set by the least square method.

[0020] S302: According to the characteristics of the time to be predicted, the energy consumption category and energy consumption standard value corresponding to the energy consumption prediction data are obtained. When the energy consumption prediction data is greater than or equal to the energy consumption standard value, it indicates that the energy consumption is abnormal, and energy consumption abnormality information is generated. When the energy consumption prediction data is less than the energy consumption standard value, it indicates that the energy consumption is normal.

[0021] Preferably, the step P4 specifically comprises the following steps:

[0022] S401: Obtaining the energy consumption equipment priority and the energy consumption per unit time of the energy consumption equipment, constraining the energy consumption equipment set according to the energy consumption standard value by presetting the equipment operation constraint condition, and generating random equipment operation sequence information, wherein the energy consumption equipment set is represented as: R, f and Th are all positive integers, where G represents the set of energy-consuming devices, R represents the priority of energy-consuming devices, f represents the floor number, and Th represents the device number. represents an energy-consuming device with a priority of R, a floor number of f, and a device number of Th. The device operation constraint condition is expressed as: in, Indicates the energy consumption per unit time of energy-consuming equipment, J max Indicates the standard value of energy consumption, Indicates the number of energy-consuming devices with priority 1 running on the same floor. Indicates the minimum standard for the number of energy-consuming devices with priority 1 running on the same floor;

[0023] S402: The random device operation sequence information includes the number of energy-consuming devices in operation on each floor and the number of energy-consuming devices. According to the number of energy-consuming devices in operation on each floor, the formula Calculate the sequence fluctuation value of the random device operation sequence information, where x i represents the number of energy-consuming devices in operation at the i-th layer, and BD represents the sequence fluctuation value;

[0024] S403: when the sequence fluctuation value is greater than or equal to the preset threshold, repeat steps P401-S402 until the sequence fluctuation value is less than the preset threshold, and obtain the equipment operation sequence according to the random equipment operation sequence information corresponding to the sequence fluctuation value;

[0025] S404: Generate control instructions according to the equipment operation sequence, and control energy-consuming equipment in the building through the control instructions.

[0026] Furthermore, the present application also provides an AIOT energy consumption data analysis system for energy-saving intelligent control, including the following modules:

[0027] The data acquisition module is used to uniformly number the energy-consuming equipment in the building according to the floor number and equipment priority, obtain the energy-consuming equipment set, and obtain the current energy consumption data and historical energy consumption data of the building;

[0028] An energy consumption category determination module is used to classify the historical energy consumption data according to a preset period, analyze the classified historical energy consumption data to obtain energy consumption categories, and obtain energy consumption standard values ​​by processing the data volume and energy consumption categories of the historical energy consumption data;

[0029] An energy consumption prediction module is used to predict the energy consumption of the next cycle according to the current energy consumption data to obtain energy consumption prediction data, and obtain energy consumption abnormality information based on the energy consumption standard value and the energy consumption prediction data;

[0030] The energy-consuming equipment control module analyzes and processes the energy-consuming equipment set and the energy consumption standard value based on the energy consumption abnormality information to obtain an equipment operation sequence, generates a control instruction according to the equipment operation sequence, and controls the energy-consuming equipment in the building through the control instruction.

[0031] Preferably, the method for calculating the energy consumption category specifically includes the following steps:

[0032] Step P1: normalizing the historical energy consumption data to obtain first historical energy consumption data, obtaining a historical energy consumption data set based on the first historical energy consumption data, and dividing the historical energy consumption data set according to a preset period.

[0033] Step P2: The number of energy consumption categories is preset to k. In the historical energy consumption data set, a first historical energy consumption data is randomly selected as the initial class representative element. According to the historical energy consumption data set and the initial class representative element, the first class representative element distance is calculated. According to the first class representative element distance, the class representative element probability is calculated;

[0034] Step P3: taking the first historical energy consumption data corresponding to the maximum probability of the class representative element as a new class representative element, taking the new class representative element as a new initial class representative element, repeating step 2, and in the process of repeating step P2, performing deduplication processing on the selected class representative elements until k class representative elements are selected to obtain a set of class representative elements;

[0035] Step P4: Feature extraction is performed on the historical energy consumption data set and the class representative element set respectively to obtain historical energy consumption feature vectors and class representative element feature vectors. The distance of the second class representative element is calculated based on the historical energy consumption feature vectors and the class representative element feature vectors. By comparing the distance of the second class representative element between the historical energy consumption data and each class representative element, the first historical energy consumption data is divided into the class where the class representative element closest to the historical energy consumption data is located. After the first historical energy consumption data is divided, k energy consumption categories and corresponding energy consumption category data sets are obtained.

[0036] Preferably, the calculation method of the energy consumption standard value is specifically as follows: obtain the energy consumption category data amount based on the historical energy consumption data amount combined with the energy consumption category data set; arrange the first historical energy consumption data in the energy consumption category data set in order from small to large, calculate the median position of the energy consumption category, and use the first historical energy consumption data corresponding to the median position as the energy consumption standard value of the energy consumption category.

[0037] Preferably, the specific calculation method of the equipment operation sequence includes the following steps:

[0038] Step A1: Obtain the energy consumption equipment priority and the energy consumption per unit time of the energy consumption equipment, constrain the energy consumption equipment set according to the energy consumption standard value and preset equipment operation constraint conditions, and generate random equipment operation sequence information.

[0039] Step A2: The random device operation sequence information includes the number of energy-consuming devices in operation on each floor and the number of energy-consuming devices. According to the number of energy-consuming devices in operation on each floor, the sequence fluctuation value of the random device operation sequence information is obtained by calculation;

[0040] Step A3: when the sequence fluctuation value is greater than or equal to the preset threshold, repeat steps A1-A2 until the sequence fluctuation value is less than the preset threshold, and obtain the device operation sequence according to the random device operation sequence information corresponding to the sequence fluctuation value;

[0041] Step A4: Generate control instructions according to the equipment operation sequence, and control the energy-consuming equipment in the building through the control instructions.

[0042] Furthermore, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. Through computer executable instructions, when the processor executes the computer program, the above-mentioned AIOT energy consumption data analysis method for energy-saving intelligent control is implemented.

[0043] The beneficial effects of the present invention are:

[0044] 1. By setting an energy consumption category identification module, the historical energy consumption data is divided according to a preset period, and the divided historical energy consumption data is analyzed to obtain energy consumption categories. By processing the data volume and energy consumption categories of the historical energy consumption data, the energy consumption standard value is obtained. By analyzing and processing the historical energy consumption data, the energy consumption category and the energy consumption standard value are obtained, and the energy consumption pattern classification of the historical energy consumption data is realized to obtain the energy consumption category. Each energy consumption category has a corresponding energy consumption standard value, which is conducive to analyzing the energy consumption category corresponding to the future energy consumption data, judging whether the energy consumption forecast data has abnormalities, and improving the data analysis efficiency;

[0045] 2. By setting up an energy-consuming equipment control module, based on the energy consumption abnormality information, the energy-consuming equipment set and the energy consumption standard value are analyzed and processed to obtain the equipment operation sequence, and a control instruction is generated according to the equipment operation sequence. The energy-consuming equipment in the building is controlled by the control instruction, and the equipment operation sequence is obtained based on the energy consumption standard value, thereby realizing rational and intelligent control of the building's energy-consuming equipment, improving the energy-saving and consumption-reduction efficiency of the building, and improving the control accuracy of the energy-consuming equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0047] Figure 1 It is a flow chart of an AIOT energy consumption data analysis method for energy-saving intelligent control according to the present invention. DETAILED DESCRIPTION

[0048] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0049] See also Figure 1 , an AIOT energy consumption data analysis method for energy-saving intelligent control, comprising the following steps:

[0050] S1: According to the floor number and equipment priority, the energy-consuming equipment in the building is uniformly numbered to obtain the energy-consuming equipment set;

[0051] S2: obtaining current energy consumption data and historical energy consumption data of the building, dividing the historical energy consumption data according to a preset period, analyzing the divided historical energy consumption data to obtain energy consumption categories, and obtaining energy consumption standard values ​​by processing the data volume and energy consumption categories of the historical energy consumption data;

[0052] S3: predicting the energy consumption of the next cycle according to the current energy consumption data to obtain energy consumption prediction data, and obtaining energy consumption abnormality information based on the energy consumption standard value and the energy consumption prediction data;

[0053] S4: Based on the abnormal energy consumption information, the energy consumption equipment set and the energy consumption standard value are analyzed and processed to obtain an equipment operation sequence, and a control instruction is generated according to the equipment operation sequence, and the energy consumption equipment in the building is controlled by the control instruction.

[0054] Specifically, the calculation steps of the energy consumption category are as follows:

[0055] Step 1: normalize the historical energy consumption data to obtain first historical energy consumption data, obtain a historical energy consumption data set based on the first historical energy consumption data, and divide the historical energy consumption data set according to a preset period. The historical energy consumption data set is expressed as: And d and h are both positive integers, where d represents the day index and h represents the hour index. Indicates the first historical energy consumption data at h time on day d, represents the first historical energy consumption data set of 1 to 24 hours with day index d, and

[0056] Step 2: The number of energy consumption categories is preset to k. In the historical energy consumption data set, a first historical energy consumption data is randomly selected as the initial class representative element. According to the historical energy consumption data set and the initial class representative element, the distance between the first class representative element and the historical energy consumption data set is calculated. in, represents the distance of the first class representative element, Ni represents the initial class representative element, and the class representative element probability is calculated based on the distance of the first class representative element.

[0057] Step 3: Take the first historical energy consumption data corresponding to the maximum probability of the class representative element as the new class representative element, take the new class representative element as the new initial class representative element, repeat step 2, and in the process of repeating step 2, perform deduplication processing on the selected class representative elements until k class representative elements are selected to obtain the class representative element set C j Indicates the representative element of the jth class;

[0058] Step 4: Extract features from the historical energy consumption data set and the class representative element set to obtain the historical energy consumption feature vector and class represents element feature vector in, express The coordinate value of the corresponding historical energy consumption feature vector in the zth dimension, Represents C j The coordinate value of the corresponding class representative element feature vector in the zth dimension; according to the historical energy consumption feature vector and the class representative element feature vector, the distance of the second class representative element is calculated, The second type of representative element distance represents the distance from the first historical energy consumption data to the representative elements of each class, D d，h，j Represents the distance of the second category representative element. By comparing the distance between the historical energy consumption data and the second category representative element of each class representative element, the first historical energy consumption data is divided into the class where the class representative element closest to it is located. After the first historical energy consumption data is divided, k energy consumption categories and corresponding energy consumption category data sets are obtained.

[0059] It should be further explained that the historical energy consumption data includes but is not limited to electricity energy consumption data, air conditioning energy consumption data and water energy consumption data. Different energy consumption data have inconsistent units. In this embodiment, a normalization method is adopted to unify the dimensions for easy calculation. In addition, by presetting the number of energy consumption categories, the initial class representative element is selected, and the distance from the historical energy consumption data to the first class representative element of the class representative element is calculated. The next class representative element is selected according to the distance of the first class representative element. The calculation is terminated after k class representative elements are selected. The distance from the historical energy consumption data to each class representative element is calculated, and the first historical energy consumption data is divided into the class where the class representative element with the smallest distance is located. After processing all the first historical energy consumption data, k energy consumption categories and corresponding energy consumption category data sets are obtained, and the energy consumption mode classification for the historical energy consumption data is realized. The energy consumption category is obtained, which is conducive to analyzing the energy consumption category corresponding to the future energy consumption data, judging whether the energy consumption forecast data has abnormalities, and improving the data analysis efficiency.

[0060] For example, the number of preset energy consumption categories is 3, the initial category representative element Ni=4, and after processing according to the above step 1, there is a first historical energy consumption data set of 3 to 8 hours with a day index of 2. By formula The calculated distance of the first representative element corresponding to the first historical energy consumption data set is Calculate the probability of the class representative element based on the distance of the first class representative element The new initial class representative element is obtained as follows: Repeat the above steps to get the first type of representative element distance The probability of class representative element is This results in a new initial class representative element Get the class representative element set C = {C1 = 4, C2 = 9, C3 = 1};

[0061] The historical energy consumption feature vectors are (3, 3), (4, 2), (5, 1), (6, 4), and (7, 8); the class representative element feature vectors are (6, 4), (8, 9), and (5, 1). It should be noted that the above two types of feature vectors are composed of time index features and energy consumption features. For example, (4, 2) represents a historical energy consumption feature vector composed of a time index feature of 4 and an energy consumption feature of 2. According to the above data, the energy consumption category data set H3 of the third energy consumption category is {3, 2, 1}.

[0062] Specifically, the specific calculation method of the energy consumption standard value includes: obtaining the energy consumption category data amount according to the historical energy consumption data amount and the energy consumption category data set; arranging the first historical energy consumption data in the energy consumption category data set in order from small to large, and calculating the median position of the energy consumption category, and the calculation formula is: Among them, P jrepresents the median position of the energy consumption category data set corresponding to the jth energy consumption category, m j The energy consumption category data volume of the j-th energy consumption category is represented, and the first historical energy consumption data corresponding to the median position is used as the energy consumption standard value of the j-th energy consumption category.

[0063] It should be further explained that the purpose of normalization processing is to divide the historical energy consumption data according to a preset period, wherein the period includes but is not limited to year, month, week and day. The energy consumption standard values ​​of buildings of different categories are different. However, due to changes in actual usage conditions, the category to which the building belongs is no longer the initial category, so it is inaccurate to manage the building energy consumption with the energy consumption standard value of the initial category of the building. In this embodiment, the preset period is the daily hourly historical energy consumption data, and the energy consumption category and energy consumption standard value are obtained from the historical energy consumption data, wherein the day index represents the date of the historical energy consumption data, and the hour index represents 1 to 24 hours per purpose. By analyzing each energy consumption category to obtain the corresponding energy consumption standard value, it is judged whether the energy consumption forecast data is abnormal based on the energy consumption standard value, thereby improving the data analysis efficiency;

[0064] For example, the energy consumption category data set of the third energy consumption category is arranged from large to small as H3={1, 2, 3}, then P3=(3+1) / 2=2, and the energy consumption standard value of the third energy consumption category is 2.

[0065] Specifically, the step P3 comprises the following steps:

[0066] S301: adding current energy consumption data to a historical energy consumption data set, extracting features from the historical energy consumption data set to obtain historical time features, constructing an energy consumption prediction model based on the historical time features combined with the historical energy consumption data set, and obtaining energy consumption prediction data through calculation using the energy consumption prediction model;

[0067] S302: According to the time characteristics of the current energy consumption data, the time characteristics to be predicted of the energy consumption prediction data are obtained. According to the time characteristics to be predicted, the energy consumption prediction data NH=α+β×wi is calculated by the energy consumption prediction model, wherein NH represents the energy consumption prediction data, α and β represent parameter factors, and wi represents the time characteristics to be predicted. The parameter factors are obtained by processing the historical energy consumption data set by the least square method.

[0068] S302: According to the characteristics of the time to be predicted, the energy consumption category and energy consumption standard value corresponding to the energy consumption prediction data are obtained. When the energy consumption prediction data is greater than or equal to the energy consumption standard value, it indicates that the energy consumption is abnormal, and energy consumption abnormality information is generated. When the energy consumption prediction data is less than the energy consumption standard value, it indicates that the energy consumption is normal.

[0069] It should be further explained that an energy consumption prediction model is established by analyzing the relationship between the time characteristics of historical energy consumption data and the energy consumption amount, and the energy consumption prediction model is used to predict the energy consumption prediction data of the next period. The energy consumption category and the energy consumption standard value are obtained according to the time characteristics to be predicted of the energy consumption prediction data. In this embodiment, since the first historical energy consumption data in the historical energy consumption data set carries a day index and a time index, the energy consumption category and the first historical energy consumption data in the energy consumption category data set also carry a day index and a time index. Therefore, the corresponding energy consumption category and the energy consumption standard value are found through the time characteristics to be predicted of the energy consumption prediction data, and whether the energy consumption of the building is abnormal is determined by comparing the energy consumption standard value and the energy consumption prediction data.

[0070] Specifically, the step P4 comprises the following steps:

[0071] S401: Obtaining the energy consumption equipment priority and the energy consumption per unit time of the energy consumption equipment, constraining the energy consumption equipment set according to the energy consumption standard value by presetting the equipment operation constraint condition, and generating random equipment operation sequence information, wherein the energy consumption equipment set is represented as: R, f and Th are all positive integers, where G represents the set of energy-consuming devices, R represents the priority of energy-consuming devices, f represents the floor number, and Th represents the device number. represents an energy-consuming device with a priority of R, a floor number of f, and a device number of Th. The device operation constraint condition is expressed as: in, Indicates the energy consumption per unit time of energy-consuming equipment, J max Indicates the standard value of energy consumption, Indicates the number of energy-consuming devices with priority 1 running on the same floor. Indicates the minimum standard for the number of energy-consuming devices with priority 1 running on the same floor;

[0072] S402: The random device operation sequence information includes the number of energy-consuming devices in operation on each floor and the number of energy-consuming devices. According to the number of energy-consuming devices in operation on each floor, the formula The sequence fluctuation value of the random equipment operation sequence information is calculated, where xi represents the number of energy-consuming equipment in operation at the i-th layer, and BD represents the sequence fluctuation value;

[0073] S403: when the sequence fluctuation value is greater than or equal to the preset threshold, repeat steps P401-S402 until the sequence fluctuation value is less than the preset threshold, and obtain the equipment operation sequence according to the random equipment operation sequence information corresponding to the sequence fluctuation value;

[0074] S404: Generate control instructions according to the equipment operation sequence, and control energy-consuming equipment in the building through the control instructions.

[0075] It should be further explained that the equipment priority is set in advance, and it is necessary to ensure that the equipment with the highest priority can be operated first at any time. The smaller the priority number of the energy-consuming equipment, the higher the priority. In addition, the energy consumption per unit time represents the energy consumption of the energy-consuming equipment in one hour. The sequence fluctuation value of the random equipment operation sequence is calculated. The purpose is to ensure that the number of energy-consuming equipment running on each floor is kept as balanced as possible while ensuring that the energy-consuming equipment with high priority meets the operating conditions, so as to achieve rational and intelligent control of the building's energy-consuming equipment, improve the energy-saving and consumption-reducing efficiency of the building, and improve the control accuracy of the energy-consuming equipment;

[0076] For example, there are 5 floors in a building. In the random equipment operation sequence, the number of energy-consuming equipment running on floors 1 to 5 is (2, 4, 6, 5, 3) respectively. The preset threshold is 0.25. If BD>0.25, steps P401-S402 are repeated until the sequence fluctuation value is less than a preset threshold.

[0077] Furthermore, the present application also provides an AIOT energy consumption data analysis system for energy-saving intelligent control, including the following modules:

[0078] The data acquisition module is used to uniformly number the energy-consuming equipment in the building according to the floor number and equipment priority, obtain the energy-consuming equipment set, and obtain the current energy consumption data and historical energy consumption data of the building;

[0079] An energy consumption category determination module is used to classify the historical energy consumption data according to a preset period, analyze the classified historical energy consumption data to obtain energy consumption categories, and obtain energy consumption standard values ​​by processing the data volume and energy consumption categories of the historical energy consumption data;

[0080] An energy consumption prediction module is used to predict the energy consumption of the next cycle according to the current energy consumption data to obtain energy consumption prediction data, and obtain energy consumption abnormality information based on the energy consumption standard value and the energy consumption prediction data;

[0081] The energy-consuming equipment control module analyzes and processes the energy-consuming equipment set and the energy consumption standard value based on the energy consumption abnormality information to obtain an equipment operation sequence, generates a control instruction according to the equipment operation sequence, and controls the energy-consuming equipment in the building through the control instruction.

[0082] Specifically, the energy consumption category calculation method comprises the following steps:

[0083] Step P1: normalizing the historical energy consumption data to obtain first historical energy consumption data, obtaining a historical energy consumption data set based on the first historical energy consumption data, and dividing the historical energy consumption data set according to a preset period.

[0084] Step P2: The number of energy consumption categories is preset to k. In the historical energy consumption data set, a first historical energy consumption data is randomly selected as the initial class representative element. According to the historical energy consumption data set and the initial class representative element, the first class representative element distance is calculated. According to the first class representative element distance, the class representative element probability is calculated;

[0085] Step P3: taking the first historical energy consumption data corresponding to the maximum probability of the class representative element as a new class representative element, taking the new class representative element as a new initial class representative element, repeating step 2, and in the process of repeating step P2, performing deduplication processing on the selected class representative elements until k class representative elements are selected to obtain a set of class representative elements;

[0086] Step P4: Feature extraction is performed on the historical energy consumption data set and the class representative element set respectively to obtain historical energy consumption feature vectors and class representative element feature vectors. The distance of the second class representative element is calculated based on the historical energy consumption feature vectors and the class representative element feature vectors. By comparing the distance of the second class representative element between the historical energy consumption data and each class representative element, the first historical energy consumption data is divided into the class where the class representative element closest to the historical energy consumption data is located. After the first historical energy consumption data is divided, k energy consumption categories and corresponding energy consumption category data sets are obtained.

[0087] Specifically, the method for calculating the energy consumption standard value is as follows: obtain the energy consumption category data amount based on the historical energy consumption data amount and the energy consumption category data set; arrange the first historical energy consumption data in the energy consumption category data set in order from small to large, calculate the median position of the energy consumption category, and use the first historical energy consumption data corresponding to the median position as the energy consumption standard value of the energy consumption category.

[0088] Specifically, the specific calculation method of the equipment operation sequence includes the following steps:

[0089] Step A1: Obtain the energy consumption equipment priority and the energy consumption per unit time of the energy consumption equipment, constrain the energy consumption equipment set according to the energy consumption standard value and preset equipment operation constraint conditions, and generate random equipment operation sequence information.

[0090] Step A2: The random device operation sequence information includes the number of energy-consuming devices in operation on each floor and the number of energy-consuming devices. According to the number of energy-consuming devices in operation on each floor, the sequence fluctuation value of the random device operation sequence information is obtained by calculation;

[0091] Step A3: when the sequence fluctuation value is greater than or equal to the preset threshold, repeat steps A1-A2 until the sequence fluctuation value is less than the preset threshold, and obtain the device operation sequence according to the random device operation sequence information corresponding to the sequence fluctuation value;

[0092] Step A4: Generate control instructions according to the equipment operation sequence, and control the energy-consuming equipment in the building through the control instructions.

[0093] Furthermore, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. Through computer executable instructions, when the processor executes the computer program, the above-mentioned AIOT energy consumption data analysis method for energy-saving intelligent control is implemented.

[0094] The working principle and use process of the present invention:

[0095] Through the data acquisition module, the equipment in the building is uniformly numbered according to the floor number and equipment priority to obtain an energy-consuming equipment set, the current energy consumption data and historical energy consumption data of the building are obtained, the historical energy consumption data are divided according to a preset period, and the divided historical energy consumption data are analyzed to obtain energy consumption categories. By processing the data volume and energy consumption categories of the historical energy consumption data, the energy consumption standard value is obtained, and the energy consumption of the next period is predicted according to the current energy consumption data to obtain energy consumption prediction data. Based on the energy consumption standard value and the energy consumption prediction data, energy consumption abnormality information is obtained. Based on the energy consumption abnormality information, the energy consumption equipment set and the energy consumption standard value are analyzed and processed to obtain the equipment operation sequence, and control instructions are generated according to the equipment operation sequence. The energy-consuming equipment in the building is controlled by the control instructions.

[0096] The computer storage medium of the embodiment of the present invention may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device.

[0097] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0098] The program code included in the computer-readable medium can be transmitted with any appropriate medium, including but not limited to wireless, wire, optical cable, etc., or any suitable combination of the above. The computer program code for performing the operation of the present invention can be written in one or more programming languages ​​or their combinations, and the programming language includes object-oriented programming languages-such as Java, Smalltalk, C++, and also includes conventional procedural programming languages-such as "C" language or similar programming languages. The program code can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0099] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. An AIOT energy consumption data analysis method for energy-saving intelligent control, characterized in that: The following steps are involved: S1: According to the floor number and equipment priority, the energy-consuming equipment in the building is uniformly numbered to obtain the energy-consuming equipment set; S2: obtaining current energy consumption data and historical energy consumption data of the building, dividing the historical energy consumption data according to a preset period, analyzing the divided historical energy consumption data to obtain energy consumption categories, and obtaining energy consumption standard values ​​by processing the data volume and energy consumption categories of the historical energy consumption data; S3: predicting the energy consumption of the next cycle according to the current energy consumption data to obtain energy consumption prediction data, and obtaining energy consumption abnormality information based on the energy consumption standard value and the energy consumption prediction data; S4: Based on the abnormal energy consumption information, the energy consumption equipment set and the energy consumption standard value are analyzed and processed to obtain an equipment operation sequence, and a control instruction is generated according to the equipment operation sequence, and the energy consumption equipment in the building is controlled by the control instruction.

2. According to claim 1, the AIOT energy consumption data analysis method for energy-saving intelligent control is characterized in that: The calculation steps of the energy consumption category are specifically as follows: Step 1: normalize the historical energy consumption data to obtain first historical energy consumption data, obtain a historical energy consumption data set based on the first historical energy consumption data, and divide the historical energy consumption data set according to a preset period. The historical energy consumption data set is expressed as: And d and h are both positive integers, where d represents the day index and h represents the hour index. Indicates the first historical energy consumption data at h time on day d, represents the first historical energy consumption data set of 1 to 24 hours with day index d, and Step 2: The number of energy consumption categories is preset to k. In the historical energy consumption data set, a first historical energy consumption data is randomly selected as the initial class representative element. According to the historical energy consumption data set and the initial class representative element, the distance between the first class representative element and the historical energy consumption data set is calculated. in, represents the distance of the first class representative element, Ni represents the initial class representative element, and the class representative element probability is calculated based on the distance of the first class representative element. Step 3: Take the first historical energy consumption data corresponding to the maximum probability of the class representative element as the new class representative element, take the new class representative element as the new initial class representative element, repeat step 2, and in the process of repeating step 2, perform deduplication processing on the selected class representative elements until k class representative elements are selected to obtain the class representative element set C j Indicates the representative element of the jth class; Step 4: Extract features from the historical energy consumption data set and the class representative element set to obtain the historical energy consumption feature vector and class represents element feature vector in, express The coordinate value of the corresponding historical energy consumption feature vector in the zth dimension, Represents C j The coordinate value of the corresponding class representative element feature vector in the zth dimension; according to the historical energy consumption feature vector and the class representative element feature vector, the distance of the second class representative element is calculated, The second type of representative element distance represents the distance from the first historical energy consumption data to the representative elements of each class, D d，h，j Represents the distance of the second category representative element. By comparing the distance between the historical energy consumption data and the second category representative element of each class representative element, the first historical energy consumption data is divided into the class where the class representative element closest to it is located. After the first historical energy consumption data is divided, k energy consumption categories and corresponding energy consumption category data sets are obtained.

3. According to claim 1, the AIOT energy consumption data analysis method for energy-saving intelligent control is characterized in that: The specific calculation method of the energy consumption standard value includes: obtaining the energy consumption category data volume according to the historical energy consumption data volume and the energy consumption category data volume; arranging the first historical energy consumption data in the energy consumption category data volume in the order of small to large, and calculating the median position of the energy consumption category, and the calculation formula is: Among them, P j represents the median position of the energy consumption category data set corresponding to the jth energy consumption category, m j The energy consumption category data volume of the j-th energy consumption category is represented, and the first historical energy consumption data corresponding to the median position is used as the energy consumption standard value of the j-th energy consumption category.

4. According to claim 1, the AIOT energy consumption data analysis method for energy-saving intelligent control is characterized in that: The step P3 specifically comprises the following steps: S301: adding current energy consumption data to a historical energy consumption data set, extracting features from the historical energy consumption data set to obtain historical time features, constructing an energy consumption prediction model based on the historical time features combined with the historical energy consumption data set, and obtaining energy consumption prediction data through calculation using the energy consumption prediction model; S302: According to the time characteristics of the current energy consumption data, the time characteristics to be predicted of the energy consumption prediction data are obtained. According to the time characteristics to be predicted, the energy consumption prediction data NH=α+β×wi is calculated by the energy consumption prediction model, wherein NH represents the energy consumption prediction data, α and β represent parameter factors, and wi represents the time characteristics to be predicted. The parameter factors are obtained by processing the historical energy consumption data set by the least square method. S302: According to the characteristics of the time to be predicted, the energy consumption category and energy consumption standard value corresponding to the energy consumption prediction data are obtained. When the energy consumption prediction data is greater than or equal to the energy consumption standard value, it indicates that the energy consumption is abnormal, and energy consumption abnormality information is generated. When the energy consumption prediction data is less than the energy consumption standard value, it indicates that the energy consumption is normal.

5. The AIOT energy consumption data analysis method for energy-saving intelligent control according to claim 1 is characterized in that: The step P4 specifically comprises the following steps: S401: Obtaining the energy consumption equipment priority and the energy consumption per unit time of the energy consumption equipment, constraining the energy consumption equipment set according to the energy consumption standard value by presetting the equipment operation constraint condition, and generating random equipment operation sequence information, wherein the energy consumption equipment set is represented as: R, f and Th are all positive integers, where G represents the set of energy-consuming devices, R represents the priority of energy-consuming devices, f represents the floor number, and Th represents the device number. represents an energy-consuming device with a priority of R, a floor number of f, and a device number of Th. The device operation constraint condition is expressed as: in, Indicates the energy consumption per unit time of energy-consuming equipment, J max Indicates the standard value of energy consumption, Indicates the number of energy-consuming devices with priority 1 running on the same floor. Indicates the minimum standard for the number of energy-consuming devices with priority 1 running on the same floor; S402: The random device operation sequence information includes the number of energy-consuming devices in operation on each floor and the number of energy-consuming devices. According to the number of energy-consuming devices in operation on each floor, the formula Calculate the sequence fluctuation value of the random device operation sequence information, where x i represents the number of energy-consuming devices in operation at the i-th layer, and BD represents the sequence fluctuation value; S403: when the sequence fluctuation value is greater than or equal to the preset threshold, repeat steps P401-S402 until the sequence fluctuation value is less than the preset threshold, and obtain the device operation sequence according to the random device operation sequence information corresponding to the sequence fluctuation value; S404: Generate control instructions according to the equipment operation sequence, and control energy-consuming equipment in the building through the control instructions.

6. An AIOT energy consumption data analysis system for energy-saving intelligent control, the system adopts an AIOT energy consumption data analysis method for energy-saving intelligent control as described in any one of claims 1-5, characterized in that: Includes the following modules: The data acquisition module is used to uniformly number the energy-consuming equipment in the building according to the floor number and equipment priority, obtain the energy-consuming equipment set, and obtain the current energy consumption data and historical energy consumption data of the building; An energy consumption category determination module is used to classify the historical energy consumption data according to a preset period, analyze the classified historical energy consumption data to obtain energy consumption categories, and obtain energy consumption standard values ​​by processing the data volume and energy consumption categories of the historical energy consumption data; An energy consumption prediction module is used to predict the energy consumption of the next cycle according to the current energy consumption data to obtain energy consumption prediction data, and obtain energy consumption abnormality information based on the energy consumption standard value and the energy consumption prediction data; The energy-consuming equipment control module analyzes and processes the energy-consuming equipment set and the energy consumption standard value based on the energy consumption abnormality information to obtain an equipment operation sequence, generates a control instruction according to the equipment operation sequence, and controls the energy-consuming equipment in the building through the control instruction.

7. The AIOT energy consumption data analysis system for energy-saving intelligent control according to claim 6 is characterized in that: The method for calculating the energy consumption category specifically comprises the following steps: Step P1: normalizing the historical energy consumption data to obtain first historical energy consumption data, obtaining a historical energy consumption data set based on the first historical energy consumption data, and dividing the historical energy consumption data set according to a preset period. Step P2: The number of energy consumption categories is preset to k. In the historical energy consumption data set, a first historical energy consumption data is randomly selected as the initial class representative element. According to the historical energy consumption data set and the initial class representative element, the first class representative element distance is calculated. According to the first class representative element distance, the class representative element probability is calculated; Step P3: taking the first historical energy consumption data corresponding to the maximum probability of the class representative element as a new class representative element, taking the new class representative element as a new initial class representative element, and repeating step 2. In the process of repeating step 2, performing deduplication processing on the selected class representative elements until k class representative elements are selected to obtain a set of class representative elements; Step P4: Feature extraction is performed on the historical energy consumption data set and the class representative element set respectively to obtain historical energy consumption feature vectors and class representative element feature vectors. The distance of the second class representative element is calculated based on the historical energy consumption feature vectors and the class representative element feature vectors. By comparing the distance of the second class representative element between the historical energy consumption data and each class representative element, the first historical energy consumption data is divided into the class where the class representative element closest to the historical energy consumption data is located. After the first historical energy consumption data is divided, k energy consumption categories and corresponding energy consumption category data sets are obtained.

8. The AIOT energy consumption data analysis system for energy-saving intelligent control according to claim 6 is characterized in that: The specific method for calculating the energy consumption standard value is: obtain the energy consumption category data amount based on the historical energy consumption data amount and the energy consumption category data set; arrange the first historical energy consumption data in the energy consumption category data set in order from small to large, calculate the median position of the energy consumption category, and use the first historical energy consumption data corresponding to the median position as the energy consumption standard value of the energy consumption category.

9. The AIOT energy consumption data analysis system for energy-saving intelligent control according to claim 6 is characterized in that: The specific calculation method of the equipment operation sequence includes the following steps: Step A1: Obtain the energy consumption equipment priority and the energy consumption per unit time of the energy consumption equipment, constrain the energy consumption equipment set according to the energy consumption standard value and preset equipment operation constraint conditions, and generate random equipment operation sequence information. Step A2: The random device operation sequence information includes the number of energy-consuming devices in operation on each floor and the number of energy-consuming devices. According to the number of energy-consuming devices in operation on each floor, the sequence fluctuation value of the random device operation sequence information is obtained by calculation; Step A3: when the sequence fluctuation value is greater than or equal to the preset threshold, repeat steps A1-A2 until the sequence fluctuation value is less than the preset threshold, and obtain the device operation sequence according to the random device operation sequence information corresponding to the sequence fluctuation value; Step A4: Generate control instructions according to the equipment operation sequence, and control the energy-consuming equipment in the building through the control instructions.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: By means of computer executable instructions, when the processor executes the computer program, an AIOT energy consumption data analysis method for energy-saving intelligent control as described in any one of claims 1-5 is implemented.

Images