Multi-dimensional data driven smart park energy saving system and method

Abstract

The invention discloses a multi-dimensional data-driven smart park energy saving system and method, and relates to the technical field of park energy saving, and the energy saving method comprises the following steps: collecting historical data of any energy consumption device in a park, determining the energy consumption type of each energy consumption device, and carrying out the type division; the comprehensive energy-saving efficiency of any energy consumption type is obtained; performing difference comparison on any energy consumption equipment in the same energy consumption type to obtain correlation factors of the same energy consumption type, and generating a reference energy-saving range of any energy consumption type; comparing the energy-saving condition of any energy consumption equipment with the reference energy-saving range of the corresponding energy consumption type, identifying the energy-saving deviation, and formulating an expected energy-saving strategy; energy consumption data of any energy consumption equipment is collected in real time, and based on the comparison situation with any expected energy saving strategy, the reasonability of the actual energy saving situation is judged; the energy-saving potential of various types of equipment can be accurately identified, and the pertinence and operability of an energy-saving strategy are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of energy-saving technology in industrial parks, specifically a multi-dimensional data-driven smart park energy-saving system and method. Background Technology

[0002] With the deepening of smart park construction, IoT-based energy-saving systems have been widely used. These systems collect multi-dimensional data by deploying a large number of sensors to achieve automated energy-saving control of various devices. However, existing technologies generally suffer from insufficient rationality judgment when evaluating energy-saving effects. Current methods rely heavily on simple comparisons of energy consumption data, failing to adequately consider various dynamic variables affecting energy consumption. These include changes in park occupancy rates, load fluctuations under different seasons and weather conditions, and the specific nature of production or operational activities, all of which significantly impact energy consumption benchmarks. Ignoring these key variables can easily lead to misjudgments of energy-saving effects. For example, attributing a natural decrease in energy consumption due to business contraction to energy-saving measures, or mistakenly rejecting effective energy-saving strategies due to increased energy consumption during peak operating periods. Summary of the Invention

[0003] The purpose of this invention is to provide a multi-dimensional data-driven smart park energy-saving system and method to solve the problems raised in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a multi-dimensional data-driven smart park energy-saving method, the energy-saving method comprising the following steps: Step S100: Collect historical data of any energy-consuming equipment in the park, determine the energy consumption type of each energy-consuming equipment and classify the types; analyze the historical data of each energy-consuming equipment in any energy consumption type to obtain the comprehensive energy-saving efficiency of any energy consumption type. Step S200: Compare the historical data of any energy-consuming device of the same energy consumption type to obtain the relevant factors affecting energy-saving efficiency of the same energy consumption type; Step S300: Analyze the impact of various relevant factors on energy efficiency and generate a baseline energy-saving range for any energy consumption type; Step S400: Compare the energy-saving status of any energy-consuming device with the benchmark energy-saving range of the corresponding energy consumption type to identify energy-saving deviations; based on the energy-saving deviations, identify relevant factors with energy-saving potential and formulate desired energy-saving strategies; Step S500: Collect energy consumption data of any energy-consuming device in real time, analyze the real-time energy efficiency of any energy-consuming device, and extract abnormal related factors; based on the comparison with any desired energy-saving strategy, make a reasonable judgment on the actual energy saving situation.

[0005] Furthermore, step S100 includes the following steps: Step S101: First, build an equipment monitoring system to store the equipment information of various energy-consuming devices in the park. At the same time, pre-divide several environmental dimensions and energy consumption indicators in the equipment monitoring system. Collect environmental data and energy consumption data of any energy-consuming device at every unit time point during operation through the equipment monitoring system. Match all environmental data with each environmental dimension to obtain the environmental data group of the energy-consuming device. At the same time, distribute each energy consumption data to each energy consumption indicator to obtain the various energy consumption indicators involved in the energy-consuming device. Generate the energy consumption indicator group of the energy-consuming device, and generate the corresponding energy consumption dataset for any energy consumption indicator in the energy consumption indicator group. Environmental dimensions include indoor and outdoor temperature and humidity, lighting, foot traffic, equipment operating status and scheduling, production plans, etc., while energy consumption indicators include total power consumption, power per unit time, running time, load rate, energy efficiency ratio, average energy consumption, carbon emissions, etc. Step S102: Randomly select two energy-consuming devices, extract the device information and energy consumption index groups of the two energy-consuming devices respectively, compare the two energy consumption index groups respectively, if the energy consumption indexes contained in the two energy consumption index groups are exactly the same, then use natural language processing technology to compare the similarity of the device information of the two energy-consuming devices, preset a similarity threshold, if the similarity of the two energy-consuming devices exceeds the similarity threshold, then set the two selected energy-consuming devices as devices of the same energy consumption type; Step S103: Divide all energy-consuming devices in the park into several sets of energy-consuming devices of different types. Randomly select one energy-consuming device from one set of energy-consuming devices of a certain type, and select the a-th energy consumption index from the selected energy-consuming device. Obtain the energy consumption data E collected at the first time point of the a-th energy consumption index. s And the energy consumption data E collected at the last time point e The energy efficiency C of the a-th energy consumption index is calculated. a =(E e -E s ) / E s The average energy-saving efficiency of the selected energy-consuming equipment is obtained by averaging the energy-saving efficiency of each energy-consuming index among the selected energy-consuming equipment. Step S104: Obtain the average energy-saving efficiency of each energy-consuming device in the selected energy-consuming type device set, and calculate the average value to obtain the comprehensive energy-saving efficiency of the selected energy-consuming type device set.

[0006] Furthermore, step S200 includes the following steps: Step S201: Arbitrarily select a set of energy-consuming devices of a certain type. From this set, arbitrarily select one energy-consuming device. Obtain the average energy-saving efficiency of the selected device during its operation over a specified time interval. Define the average energy-saving efficiency at the i-th time point within the selected time interval as C. i The average energy-saving efficiency at the j-th time point is C. j A preset efficiency change threshold ΔC is defined. th If |C i -C j |≥ΔC th Then, the i-th time point and the j-th time point are set as the first target time point group. If |C i -C j |<ΔC th Then, the i-th time point and the j-th time point are set as the second target time point group; the first target time point group indicates that the energy efficiency change between the two time points is relatively large, and the second target time point group indicates that the energy efficiency change between the two time points is relatively small; Step S202: Acquire the environmental data sets for the two time points in the target time point group respectively. Randomly select one environmental dimension and extract the environmental data for that dimension from the two environmental data sets. Let the environmental data for the i-th time point be D. i The environmental data at time point j is D. j A preset data change threshold ΔD is set for the selected environmental dimension. th If |D i -D j |≥ΔD th Then the environmental dimension will be set as the first target dimension. If |D i -D j |<ΔD th If the environmental dimension is selected, it will be set as the second target dimension; the first target dimension indicates that the data variation of the environmental dimension is too large, and the second target dimension indicates that the data variation of the environmental dimension is too small. Step S203: Calculate the average energy efficiency difference between any two time points to generate several first target time point groups and several second target time point groups. Divide each environmental dimension in any target time point group into a first target dimension set and a second target dimension set. Perform a union operation on several first target dimension sets in several first target time point groups to obtain a relevant dimension set. Perform a union operation on several first target dimension sets in several second target time point groups to obtain an irrelevant dimension set. Simultaneously, perform a union operation on several second target dimension sets in several first target time point groups to obtain another irrelevant dimension set. Perform a union operation on the two irrelevant dimension sets to obtain a comprehensive irrelevant dimension set. If a target dimension in the relevant dimension set is the same as a target dimension in the comprehensive irrelevant dimension set, remove the target dimension from the relevant dimension set to obtain a corrected relevant dimension set. To determine which dimensions affect energy efficiency, we first identify which dimensions have the largest changes and also the largest changes in energy efficiency. These are the most direct influencing dimensions. Meanwhile, dimensions with large changes but small changes in energy efficiency are defined as irrelevant dimensions. Next, we summarize the influencing dimensions by taking the union of different devices and unify the influencing dimensions by comparing similar devices. This will give us accurate relevant factors. Step S204: Obtain the relevant dimension set and the comprehensive irrelevant dimension set for each energy-consuming device in the selected energy-consuming device set during operation. Perform an intersection operation on all relevant dimension sets to obtain a common dimension set and a difference dimension set. Randomly select a difference dimension from the difference dimension set. If the selected difference dimension does not have the same dimension in any of the comprehensive irrelevant dimension sets, add the selected difference dimension to the common dimension set to obtain a corrected common dimension set. Set each environmental dimension in the corrected common dimension set as a relevant factor affecting the energy-saving efficiency of the selected energy-consuming device set.

[0007] Furthermore, step S300 includes the following steps: Step S301: Arbitrarily select one energy-consuming device from the set of energy-consuming devices of a certain type, extract the k-th environmental dimension from the selected energy-consuming device, and obtain the environmental data (D) of the k-th environmental dimension at the first time point during the operation of the selected energy-consuming device. s ) k and environmental data at the last time point (D e ) k The change magnitude Δf of the kth environmental dimension is calculated. k =[(D e ) k -(Ds ) k ] / (D s ) k ; Step S302: Set the overall energy-saving efficiency of the selected energy-consuming equipment to C. com According to the formula: ; Where p is a positive integer and p∈[1,u], u is the total number of environment dimensions, and Δf p Let p be the magnitude of change in the p-th environmental dimension; calculate the degree of influence y of the k-th environmental dimension. k ; Step S303: Extract several relevant factors from the set of energy-consuming devices of the selected energy-consuming type, and obtain the influence degree of each relevant factor. The influence degree of the q-th relevant factor is set as y. ’ q And the change range is Δf ’ q According to the formula: ; Where v represents the number of relevant factors for selecting energy-consuming equipment; the baseline energy-saving efficiency C of the selected energy-consuming equipment is calculated. ’ com The baseline energy-saving efficiency of each energy-consuming device in the selected energy-consuming device set is obtained, and the baseline energy-saving range of the selected energy-consuming device set is generated. The baseline energy-saving range is used to determine whether the energy-consuming device meets the most basic energy-saving efficiency, and at the same time, it can provide a basis for comparison to determine whether there is further energy-saving space.

[0008] Furthermore, step S400 includes the following steps: Step S401: Randomly select a set of energy-consuming devices of a certain type to obtain the baseline energy-saving range of the selected set of energy-consuming devices. Randomly select one energy-consuming device from the selected set of energy-consuming devices to obtain the comprehensive energy-saving efficiency C of the selected energy-consuming device. com The maximum energy efficiency within the benchmark energy-saving range is set as (C). com ) max The energy-saving deviation value ΔC of the selected energy-consuming equipment is obtained. com =(C com ) max -C com ; Step S402: Obtain the influence degree of each relevant factor in the selected energy-consuming equipment, and set the influence degree of the q-th relevant factor as y. ’ q According to the formula: ; Where r is a positive integer and r∈[1,v], v is the number of relevant factors selected for energy-consuming devices, and y ’ r The influence degree of the r-th related factor is given; the expected energy saving deviation (Δf) of the q-th related factor is calculated. ex ) q ;Preset the expected deviation range of the relevant factors, if the expected energy saving deviation (Δf) of the q-th relevant factor ex ) q If the value is within the expected deviation range, then the qth relevant factor is set as the energy-saving potential factor; Step S403: Summarize each energy-saving potential factor and its corresponding expected energy-saving deviation to generate a corresponding expected energy-saving strategy.

[0009] Furthermore, step S500 includes the following steps: Step S501: Randomly select an energy-consuming device, collect the environmental data set and energy consumption data set of the selected energy-consuming device in real time, and obtain the real-time energy-saving efficiency of the selected energy-consuming device; extract several relevant factors of the set of energy-consuming devices of the selected energy-consuming device type, obtain the influence degree of each relevant factor, and obtain the expected change range of each relevant factor. Step S502: Obtain the real-time change range of each relevant factor. If the real-time change range of a certain relevant factor is less than the expected change range, then set the certain relevant factor as an abnormal relevant factor to obtain several abnormal relevant factors for selecting energy consumption equipment. Step S503: Randomly select an abnormal correlation factor and obtain the expected energy saving deviation of the selected abnormal factor; arbitrarily select an expected energy saving strategy. If all the abnormal correlation factors of the selected energy-consuming equipment and the corresponding expected energy saving deviations are within the selected expected energy saving strategy, then the expected energy saving strategy is invoked for energy saving adjustment. If they are not within the selected expected energy saving strategy, then an abnormal reminder is sent to the real-time energy saving status.

[0010] To better implement the above methods, a smart park energy-saving system is also proposed. The energy-saving system includes an energy consumption data analysis module, a relevant factor extraction module, a benchmark range generation module, an energy-saving potential realization module, and a reasonable energy-saving judgment module. The energy consumption data analysis module is used to collect historical data of any energy-consuming equipment in the park, determine the energy consumption type of each energy-consuming equipment and classify the types; analyze the historical data of each energy-consuming equipment in any energy consumption type to obtain the comprehensive energy-saving efficiency of any energy consumption type. The relevant factor extraction module is used to compare the historical data of any energy-consuming device of the same energy consumption type to obtain the relevant factors affecting energy-saving efficiency of the same energy consumption type. The benchmark range generation module is used to analyze the impact of various relevant factors on energy efficiency and generate benchmark energy-saving ranges for any energy consumption type. The energy-saving potential realization module is used to compare the energy-saving status of any energy-consuming device with the benchmark energy-saving range of the corresponding energy consumption type to identify energy-saving deviations; based on the energy-saving deviations, it identifies relevant factors with energy-saving potential and formulates the desired energy-saving strategy. The reasonable energy-saving judgment module is used to collect energy consumption data of any energy-consuming device in real time, analyze the real-time energy efficiency of any energy-consuming device, and extract abnormal related factors; based on the comparison with any expected energy-saving strategy, it makes a reasonable judgment on the actual energy saving situation.

[0011] Furthermore, the energy consumption data analysis module includes an energy consumption type classification unit and an energy efficiency assessment unit; The energy consumption type classification unit is used to collect historical data of any energy-consuming equipment in the park, determine the energy consumption type of each energy-consuming equipment, and classify the types; the energy efficiency evaluation unit is used to analyze the historical data of each energy-consuming equipment in any energy consumption type to obtain the comprehensive energy efficiency of any energy consumption type.

[0012] Furthermore, the energy-saving potential realization module includes an energy-saving deviation analysis unit and an expected potential setting unit; The energy-saving deviation analysis unit is used to compare the energy-saving status of any energy-consuming device with the benchmark energy-saving range of the corresponding energy consumption type to identify energy-saving deviations; the expected potential setting unit is used to identify relevant factors with energy-saving potential based on energy-saving deviations and formulate expected energy-saving strategies.

[0013] Furthermore, the reasonable energy-saving judgment module includes a real-time energy efficiency analysis unit and an abnormal energy-saving identification unit; The real-time energy efficiency analysis unit is used to collect energy consumption data of any energy-consuming device in real time, analyze the real-time energy efficiency of any energy-consuming device, and extract abnormal related factors; the abnormal energy saving identification unit is used to make a reasonable judgment on the actual energy saving based on the comparison with any expected energy saving strategy.

[0014] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention achieves refined classification and management of energy-consuming equipment in the park through multi-dimensional data fusion analysis and intelligent classification of equipment types. It can accurately identify the energy-saving potential of various types of equipment, effectively improve the pertinence and operability of energy-saving strategies, and significantly enhance the scientific and systematic nature of energy-saving management. 2. By constructing a benchmark energy-saving range based on relevant factors, this invention can dynamically identify energy-saving deviations during equipment operation and automatically generate desired energy-saving strategies, realizing closed-loop management from data monitoring to strategy formulation, and significantly improving the real-time performance and accuracy of energy-saving decisions. 3. This invention has real-time energy efficiency analysis and anomaly identification capabilities, which can quickly capture energy efficiency anomalies and related factor changes during equipment operation, realize dynamic evaluation and feedback on the implementation effect of energy-saving strategies, effectively avoid misjudgment of energy saving caused by changes in external environment or operating status, and improve the adaptability and reliability of the system in practical applications. Attached Figure Description

[0015] Figure 1 A schematic diagram illustrating the steps of a multi-dimensional data-driven smart park energy-saving method; Figure 2 This is a schematic diagram of a multi-dimensional data-driven smart park energy-saving system. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Example: Figures 1 to 2 As shown, this invention provides a multi-dimensional data-driven smart park energy-saving method, which includes the following steps: Step S100: Collect historical data of any energy-consuming equipment in the park, determine the energy consumption type of each energy-consuming equipment and classify the types; analyze the historical data of each energy-consuming equipment in any energy consumption type to obtain the comprehensive energy-saving efficiency of any energy consumption type. Step S100 includes the following steps: Step S101: First, build an equipment monitoring system to store the equipment information of various energy-consuming devices in the park. At the same time, pre-divide several environmental dimensions and energy consumption indicators in the equipment monitoring system. Collect environmental data and energy consumption data of any energy-consuming device at every unit time point during operation through the equipment monitoring system. Match all environmental data with each environmental dimension to obtain the environmental data group of the energy-consuming device. At the same time, distribute each energy consumption data to each energy consumption indicator to obtain the various energy consumption indicators involved in the energy-consuming device. Generate the energy consumption indicator group of the energy-consuming device, and generate the corresponding energy consumption dataset for any energy consumption indicator in the energy consumption indicator group. Step S102: Randomly select two energy-consuming devices, extract the device information and energy consumption index groups of the two energy-consuming devices respectively, compare the two energy consumption index groups respectively, if the energy consumption indexes contained in the two energy consumption index groups are exactly the same, then use natural language processing technology to compare the similarity of the device information of the two energy-consuming devices, preset a similarity threshold, if the similarity of the two energy-consuming devices exceeds the similarity threshold, then set the two selected energy-consuming devices as devices of the same energy consumption type; Step S103: Divide all energy-consuming devices in the park into several sets of energy-consuming devices of different types. Randomly select one energy-consuming device from one set of energy-consuming devices of a certain type, and select the a-th energy consumption index from the selected energy-consuming device. Obtain the energy consumption data E collected at the first time point of the a-th energy consumption index. s And the energy consumption data E collected at the last time point e The energy efficiency C of the a-th energy consumption index is calculated. a =(E e -E s ) / E s The average energy-saving efficiency of the selected energy-consuming equipment is obtained by averaging the energy-saving efficiency of each energy-consuming index among the selected energy-consuming equipment. Step S104: Obtain the average energy-saving efficiency of each energy-consuming device in the selected energy-consuming type device set, and calculate the average value to obtain the comprehensive energy-saving efficiency of the selected energy-consuming type device set.

[0018] Step S200: Compare the historical data of any energy-consuming device of the same energy consumption type to obtain the relevant factors affecting energy-saving efficiency of the same energy consumption type; Step S200 includes the following steps: Step S201: Arbitrarily select a set of energy-consuming devices of a certain type. From this set, arbitrarily select one energy-consuming device. Obtain the average energy-saving efficiency of the selected device during its operation over a specified time interval. Define the average energy-saving efficiency at the i-th time point within the selected time interval as C. i The average energy-saving efficiency at the j-th time point is C. j A preset efficiency change threshold ΔC is defined. th If |C i -C j |≥ΔC th Then, the i-th time point and the j-th time point are set as the first target time point group. If |C i -C j |<ΔC th Then the i-th time point and the j-th time point are set as the second target time point group; Step S202: Acquire the environmental data sets for the two time points in the target time point group respectively. Randomly select one environmental dimension and extract the environmental data for that dimension from the two environmental data sets. Let the environmental data for the i-th time point be D. i The environmental data at time point j is D. j A preset data change threshold ΔD is set for the selected environmental dimension. th If |D i -D j |≥ΔD th Then the environmental dimension will be set as the first target dimension. If |D i -D j |<ΔD th Then the selected environment dimension will be set as the second target dimension; Step S203: Calculate the average energy efficiency difference between any two time points to generate several first target time point groups and several second target time point groups. Divide each environmental dimension in any target time point group into a first target dimension set and a second target dimension set. Perform a union operation on several first target dimension sets in several first target time point groups to obtain a relevant dimension set. Perform a union operation on several first target dimension sets in several second target time point groups to obtain an irrelevant dimension set. Simultaneously, perform a union operation on several second target dimension sets in several first target time point groups to obtain another irrelevant dimension set. Perform a union operation on the two irrelevant dimension sets to obtain a comprehensive irrelevant dimension set. If a target dimension in the relevant dimension set is the same as a target dimension in the comprehensive irrelevant dimension set, remove the target dimension from the relevant dimension set to obtain a corrected relevant dimension set. Step S204: Obtain the relevant dimension set and the comprehensive irrelevant dimension set for each energy-consuming device in the selected energy-consuming device set during operation. Perform an intersection operation on all relevant dimension sets to obtain a common dimension set and a difference dimension set. Randomly select a difference dimension from the difference dimension set. If the selected difference dimension does not have the same dimension in any of the comprehensive irrelevant dimension sets, add the selected difference dimension to the common dimension set to obtain a corrected common dimension set. Set each environmental dimension in the corrected common dimension set as a relevant factor affecting the energy-saving efficiency of the selected energy-consuming device set. Example 1: Select an air conditioning unit and set its efficiency C at the i-th time point. i =20%, efficiency C at time point j j=15%, and a preset efficiency change threshold of 5%. Since |20%-15%|≥5%, the i-th and j-th time points are set as the first target time point group. For the first target time point group, the outdoor temperature dimension is extracted from the environmental data group, and the temperature D at time point i is... i =28℃, temperature D at time point j j =32℃, preset data change threshold ΔD th =3℃. Since |28-32|=4℃≥3℃, "outdoor temperature" is set as the first target dimension. Repeat the above operation to generate multiple first target time point groups and second target time point groups. By taking the union and intersection operations, the relevant dimension set including outdoor temperature and operating mode is obtained, and the irrelevant dimension set including humidity is integrated. Take the intersection of the relevant dimension sets of all devices in the air conditioning equipment set to obtain the common dimension set as outdoor temperature, and add the other differential dimensions as operating mode to the common set. Finally, the relevant factors are determined to be outdoor temperature and operating mode.

[0019] Step S300: Analyze the impact of various relevant factors on energy efficiency and generate a baseline energy-saving range for any energy consumption type; Step S300 includes the following steps: Step S301: Arbitrarily select one energy-consuming device from the set of energy-consuming devices of a certain type, extract the k-th environmental dimension from the selected energy-consuming device, and obtain the environmental data (D) of the k-th environmental dimension at the first time point during the operation of the selected energy-consuming device. s ) k and environmental data at the last time point (D e ) k The change magnitude Δf of the kth environmental dimension is calculated. k =[(D e ) k -(D s ) k ] / (D s ) k ; Step S302: Set the overall energy-saving efficiency of the selected energy-consuming equipment to C. com According to the formula: ; Where p is a positive integer and p∈[1,u], u is the total number of environment dimensions, and Δf p Let p be the magnitude of change in the p-th environmental dimension; calculate the degree of influence y of the k-th environmental dimension. k ; Example 2: Select an air conditioning unit, extract the outdoor temperature dimension, and obtain the environmental data (D) at the first time point. s ) k=25℃, the last time point (D) e ) k =30℃, calculate the change range Δf k =(30-25) / 25=0.2; Set the overall energy efficiency C of the equipment. com =18%, there are two environmental dimensions, the variation of the other environmental dimension is 0.3, and the influence of the outdoor temperature dimension y is calculated. k =18%×0.2 / 0.5=7.2%; Step S303: Extract several relevant factors from the set of energy-consuming devices of the selected energy-consuming type, and obtain the influence degree of each relevant factor. The influence degree of the q-th relevant factor is set as y. ’ q And the change range is Δf ’ q According to the formula: ; Where v represents the number of relevant factors for selecting energy-consuming equipment; the baseline energy-saving efficiency C of the selected energy-consuming equipment is calculated. ’ com The baseline energy-saving efficiency of each energy-consuming device in the selected energy-consuming type device set is obtained, and the baseline energy-saving range of the selected energy-consuming type device set is generated. Example 3: Based on Example 2, the influence and variation of all relevant factors, including outdoor temperature and operating mode, were extracted, where it was assumed that y ’ 1 = 7.2%, Δf ’ 1 = 0.2, y ’ 2=5%, Δf ’ If 2 = 0.1, then C ’ com =7.2%×0.2+5%×0.1=1.44%+0.5%=1.94%; The baseline energy efficiency of all equipment in the air conditioning equipment set is 1.94% for equipment 1, 2.1% for equipment 2, and 1.8% for equipment 3. Taking the range, the baseline energy efficiency range is [1.8%, 2.1%].

[0020] Step S400: Compare the energy-saving status of any energy-consuming device with the benchmark energy-saving range of the corresponding energy consumption type to identify energy-saving deviations; based on the energy-saving deviations, identify relevant factors with energy-saving potential and formulate desired energy-saving strategies; Step S400 includes the following steps: Step S401: Randomly select a set of energy-consuming devices of a certain type to obtain the baseline energy-saving range of the selected set of energy-consuming devices. Randomly select one energy-consuming device from the selected set of energy-consuming devices to obtain the comprehensive energy-saving efficiency C of the selected energy-consuming device.com The maximum energy efficiency within the benchmark energy-saving range is set as (C). com ) max The energy-saving deviation value ΔC of the selected energy-consuming equipment is obtained. com =(C com ) max -C com ; Step S402: Obtain the influence degree of each relevant factor in the selected energy-consuming equipment, and set the influence degree of the q-th relevant factor as y. ’ q According to the formula: ; Where r is a positive integer and r∈[1,v], v is the number of relevant factors selected for energy-consuming devices, and y ’ r The influence degree of the r-th related factor is given; the expected energy saving deviation (Δf) of the q-th related factor is calculated. ex ) q ;Preset the expected deviation range of the relevant factors, if the expected energy saving deviation (Δf) of the q-th relevant factor ex ) q If the value is within the expected deviation range, then the qth relevant factor is set as the energy-saving potential factor; Step S403: Summarize each energy-saving potential factor and its corresponding expected energy-saving deviation to generate a corresponding expected energy-saving strategy.

[0021] Step S500: Collect energy consumption data of any energy-consuming device in real time, analyze the real-time energy efficiency of any energy-consuming device, and extract abnormal related factors; make a reasonable judgment on the actual energy saving based on the comparison with any expected energy-saving strategy. Step S500 includes the following steps: Step S501: Randomly select an energy-consuming device, collect the environmental data set and energy consumption data set of the selected energy-consuming device in real time, and obtain the real-time energy-saving efficiency of the selected energy-consuming device; extract several relevant factors of the set of energy-consuming devices of the selected energy-consuming device type, obtain the influence degree of each relevant factor, and obtain the expected change range of each relevant factor. Step S502: Obtain the real-time change range of each relevant factor. If the real-time change range of a certain relevant factor is less than the expected change range, then set the certain relevant factor as an abnormal relevant factor to obtain several abnormal relevant factors for selecting energy consumption equipment. Step S503: Randomly select an abnormal correlation factor and obtain the expected energy saving deviation of the selected abnormal factor; arbitrarily select an expected energy saving strategy. If all the abnormal correlation factors of the selected energy-consuming equipment and the corresponding expected energy saving deviations are within the selected expected energy saving strategy, then the expected energy saving strategy is invoked for energy saving adjustment. If they are not within the selected expected energy saving strategy, then an abnormal reminder is sent to the real-time energy saving status.

[0022] A smart park energy-saving system includes an energy consumption data analysis module, a relevant factor extraction module, a benchmark range generation module, an energy-saving potential realization module, and a reasonable energy-saving judgment module. The energy consumption data analysis module is used to collect historical data of any energy-consuming equipment in the park, determine the energy consumption type of each energy-consuming equipment and classify the types; analyze the historical data of each energy-consuming equipment in any energy consumption type to obtain the comprehensive energy-saving efficiency of any energy consumption type. The relevant factor extraction module is used to compare the historical data of any energy-consuming device of the same energy consumption type to obtain the relevant factors affecting energy-saving efficiency of the same energy consumption type. The benchmark range generation module is used to analyze the impact of various relevant factors on energy efficiency and generate benchmark energy-saving ranges for any energy consumption type. The energy-saving potential realization module is used to compare the energy-saving status of any energy-consuming device with the benchmark energy-saving range of the corresponding energy consumption type to identify energy-saving deviations; based on the energy-saving deviations, it identifies relevant factors with energy-saving potential and formulates the desired energy-saving strategy. The reasonable energy-saving judgment module is used to collect energy consumption data of any energy-consuming device in real time, analyze the real-time energy efficiency of any energy-consuming device, and extract abnormal related factors; based on the comparison with any expected energy-saving strategy, it makes a reasonable judgment on the actual energy saving situation.

[0023] The energy consumption data analysis module includes an energy consumption type classification unit and an energy efficiency evaluation unit. The energy consumption type classification unit is used to collect historical data of any energy-consuming equipment in the park, determine the energy consumption type of each energy-consuming equipment, and classify the types; the energy efficiency evaluation unit is used to analyze the historical data of each energy-consuming equipment in any energy consumption type to obtain the comprehensive energy efficiency of any energy consumption type.

[0024] The energy-saving potential realization module includes an energy-saving deviation analysis unit and an expected potential setting unit. The energy-saving deviation analysis unit is used to compare the energy-saving status of any energy-consuming device with the benchmark energy-saving range of the corresponding energy consumption type to identify energy-saving deviations; the expected potential setting unit is used to identify relevant factors with energy-saving potential based on energy-saving deviations and formulate expected energy-saving strategies.

[0025] The reasonable energy-saving judgment module includes a real-time energy efficiency analysis unit and an abnormal energy-saving identification unit. The real-time energy efficiency analysis unit is used to collect energy consumption data of any energy-consuming device in real time, analyze the real-time energy efficiency of any energy-consuming device, and extract abnormal related factors; the abnormal energy saving identification unit is used to make a reasonable judgment on the actual energy saving based on the comparison with any expected energy saving strategy.

[0026] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A multi-dimensional data-driven energy-saving method for a smart park, characterized in that: The energy-saving method comprises the following steps: Step S100: collecting historical data of any energy-consuming device in the park, determining the energy consumption type of each energy-consuming device and performing type classification; analyzing the historical data of each energy-consuming device in any energy consumption type to obtain the comprehensive energy-saving efficiency of any energy consumption type; Step S200: comparing the historical data of any energy-consuming device in the same energy consumption type to obtain the relevant factors affecting the energy-saving efficiency in the same energy consumption type; Step S300: analyzing the influence of each relevant factor on the energy-saving efficiency to generate a benchmark energy-saving range for any energy consumption type; Step S400: comparing the energy-saving situation of any energy-consuming device with the benchmark energy-saving range of the corresponding energy consumption type to identify the energy-saving deviation; identifying the relevant factors with energy-saving potential based on the energy-saving deviation and formulating an expected energy-saving strategy; Step S500: collecting energy consumption data of any energy-consuming device in real time, analyzing the real-time energy efficiency of any energy-consuming device, and extracting abnormal relevant factors; based on the comparison with any expected energy-saving strategy, reasonably judging the actual energy-saving situation. 2.The multi-dimensional data-driven wisdom park energy-saving method of claim 1, wherein: The step S100 comprises the following steps: Step S101: a device monitoring system is pre-built to store device information of each energy-consuming device in the park, and a plurality of environmental dimensions and energy consumption indicators are pre-classified in the device monitoring system; the environmental data and energy consumption data of any energy-consuming device during operation are collected by the device monitoring system every unit time; all environmental data are matched with each environmental dimension to obtain an environmental data set of the energy-consuming device, and each energy consumption data is allocated to each energy consumption indicator to obtain each energy consumption indicator involved in the energy-consuming device, an energy consumption indicator set of the energy-consuming device is generated, and an energy consumption data set of any energy consumption indicator in the energy consumption indicator set is generated; Step S102: any two energy-consuming devices are selected, the device information and energy consumption indicator set of the two energy-consuming devices are extracted, the two energy consumption indicator sets are compared, if the energy consumption indicators contained in the two energy consumption indicator sets are completely same, the similarity of the device information of the two energy-consuming devices is compared by using natural language processing technology, a similarity threshold is preset, if the similarity of the two energy-consuming devices exceeds the similarity threshold, the selected two energy-consuming devices are set as the same energy consumption type device; Step S103: dividing all energy consumption devices in the park into several energy consumption type device sets, selecting an energy consumption device in an energy consumption type device set, and selecting an a-th energy consumption index from the selected energy consumption device, to obtain energy consumption data E s collected at a first time point and energy consumption data E e collected at a last time point, respectively, to calculate an energy saving efficiency C a of the a-th energy consumption index as (E e -E s ) / E s ; and performing average value calculation on the energy saving efficiencies of the energy consumption indexes of the selected energy consumption device to obtain an average energy saving efficiency of the selected energy consumption device; Step S104: the average energy-saving efficiency of each energy-consuming device in the selected energy consumption type device set is obtained, and the average value is calculated to obtain the comprehensive energy-saving efficiency of the selected energy consumption type device set. 3.The multi-dimensional data-driven wisdom zone energy-saving method of claim 2, wherein: The step S200 comprises the following steps: Step S201: arbitrarily select a set of energy consumption type devices, arbitrarily select an energy consumption device from the selected set of energy consumption type devices, acquire the time interval of the selected energy consumption device in the running process, obtain the average energy saving efficiency of the selected energy consumption device at each time point in the selected time interval, set the average energy saving efficiency of the i th time point in the selected time interval as C i And the average energy saving efficiency of the j th time point is C j ; preset an efficiency change threshold ΔC th , if |C i -C j |≥ΔC th , then set the i th time point and the j th time point as the first target time point group, if |C i -C j |<ΔC th , then set the i th time point and the j th time point as the second target time point group; Step S202: Obtain the environment data groups of the two time points in the target time point group respectively, select an environment dimension at random, extract the environment data of the selected environment dimension in the two environment data groups, set the environment data of the i th time point as D i and the environment data of the j th time point as D j , preset a data change threshold ΔD th for the selected environment dimension, if |D i -D j | ≥ ΔD th , set the selected environment dimension as the first target dimension, if |D i -D j | < ΔD th , set the selected environment dimension as the second target dimension; Step S203: Calculate the average energy saving efficiency difference between any two time points, generate a plurality of first target time point groups and a plurality of second target time point groups, divide each environmental dimension in any target time point group into a first target dimension set and a second target dimension set; perform a set union operation on a plurality of first target dimension sets in a plurality of first target time point groups to obtain a related dimension set, perform a set union operation on a plurality of first target dimension sets in a plurality of second target time groups to obtain an unrelated dimension set, and perform a set union operation on a plurality of second target dimension sets in a plurality of first target time groups to obtain another unrelated dimension set; perform a set union operation on the two unrelated dimension sets to obtain a comprehensive unrelated dimension set; if there is a target dimension in the related dimension set that is the same as the target dimension in the comprehensive unrelated dimension set, remove the target dimension from the related dimension set to obtain a modified related dimension set; Step S204: Obtain the related dimension set and the comprehensive unrelated dimension set of each energy consumption device in the selected energy consumption type device set during the operation process, perform a set intersection operation on all related dimension sets to obtain a public dimension set and a difference dimension set respectively, and select a difference dimension from the difference dimension set; if the selected difference dimension does not exist in each comprehensive unrelated dimension set, add the selected difference dimension to the public dimension set to obtain a modified public dimension set, and set each environmental dimension in the modified public dimension set as a related factor affecting the energy saving efficiency of the selected energy consumption type device set.

4. The multi-dimensional data-driven wisdom park energy-saving method of claim 3, wherein: The step S300 includes the following steps: Step S301: Select an energy consumption device from a set of energy consumption devices, extract the kth environmental dimension from the selected energy consumption device, and obtain the environmental data of the kth environmental dimension at the first time point (D s ) k and the last time point (D e ) k , and calculate the change amplitude Δf k =[(D e ) k -(D s ) k ] / (D s ) k of the kth environmental dimension. Step S302: Set the comprehensive energy-saving efficiency of the selected energy-consuming device as C com According to the formula: ； wherein p is a positive integer and p∈[1, u], u is the total number of environmental dimensions, Δf p is the change range of the pth environmental dimension; the influence degree y k of the kth environmental dimension is calculated. Step S303: extracting several relevant factors in the energy consumption type device set where the energy consumption device is selected, obtaining the influence degree of each relevant factor, wherein the influence degree of the qth relevant factor is y ’ q and the change range is Δf ’ q According to the formula: ； Wherein, v is the number of selected energy consumption equipment related factors; the reference energy-saving efficiency C of the selected energy consumption equipment is calculated ’ com ; the reference energy-saving efficiency of each energy consumption equipment in the selected energy consumption type equipment set is obtained, and the reference energy-saving range of the selected energy consumption type equipment set is generated.

5. The multi-dimensional data-driven smart park energy-saving method according to claim 4, characterized in that: The step S400 includes the following steps: Step S401: randomly select a set of energy consumption type devices, obtain a benchmark energy saving range of the selected set of energy consumption type devices, randomly select an energy consumption device from the selected set of energy consumption type devices, obtain a comprehensive energy saving efficiency C of the selected energy consumption device com ; set the maximum energy saving efficiency of the benchmark energy saving range as (C com ) max , obtain an energy saving deviation value AC of the selected energy consumption device com =(C com ) max -C com ; Step S402: Obtain the influence degree of each related factor in the selected energy consumption device, and set the influence degree of the qth related factor as y ’ q According to the formula: ； wherein r is a positive integer and r ∈ [1, v], v is a number of selected relevant factors of energy consumption equipment, y ’ r is an influence degree of the rth relevant factor; the expected energy-saving deviation (Δf ex ) q of the qth relevant factor is calculated; an expected deviation range of the relevant factor is preset, if the expected energy-saving deviation (Δf ex ) q of the qth relevant factor is in the expected deviation range, the qth relevant factor is set as an energy-saving potential factor. Step S403: Summarize each energy saving potential factor and the corresponding expected energy saving deviation to generate a corresponding expected energy saving strategy.

6. The multi-dimensional data-driven wisdom zone energy-saving method according to claim 5, characterized in that: The step S500 includes the following steps: Step S501: Arbitrarily select an energy consumption device, collect the environmental data group and the energy consumption data set of the selected energy consumption device in real time to obtain the real-time energy saving efficiency of the selected energy consumption device; extract a plurality of related factors of the energy consumption type device set to which the selected energy consumption device belongs, respectively obtain the influence degree of each related factor, and obtain the expected change amplitude of each related factor; Step S502: Obtain the real-time change amplitude of each related factor, and if the real-time change amplitude of a certain related factor is less than the expected change amplitude, set the certain related factor as an abnormal related factor to obtain a plurality of abnormal related factors of the selected energy consumption device; Step S503: Arbitrarily select an abnormal related factor to obtain the expected energy saving deviation of the selected abnormal factor; arbitrarily select an expected energy saving strategy, if each abnormal related factor of the selected energy consumption device and the corresponding expected energy saving deviation is in the selected expected energy saving strategy, call the expected energy saving strategy to perform energy saving adjustment, if not in the selected expected energy saving strategy, send an abnormal reminder to the real-time energy saving situation.

7. A smart park energy saving system for performing a multi-dimensional data-driven smart park energy saving method according to any one of claims 1-6, characterized in that: The energy-saving system comprises an energy consumption data analysis module, a correlation factor extraction module, a benchmark range generation module, an energy-saving potential realization module and a reasonable energy-saving judgment module. The energy consumption data analysis module is configured to collect historical data of any energy consumption device in the park, determine the energy consumption type of each energy consumption device and perform type division, analyze the historical data of each energy consumption device in any energy consumption type, and obtain the comprehensive energy-saving efficiency of any energy consumption type. The correlation factor extraction module is configured to perform difference comparison on the historical data of any energy consumption device in the same energy consumption type, and obtain the correlation factor affecting the energy-saving efficiency in the same energy consumption type. The benchmark range generation module is configured to analyze the influence of each correlation factor on the energy-saving efficiency, and generate the benchmark energy-saving range of any energy consumption type. The energy-saving potential realization module is configured to compare the energy-saving condition of any energy consumption device with the benchmark energy-saving range of the corresponding energy consumption type, identify the energy-saving deviation, identify the correlation factor with energy-saving potential based on the energy-saving deviation, and formulate an expected energy-saving strategy. The reasonable energy-saving judgment module is configured to collect the energy consumption data of any energy consumption device in real time, analyze the real-time energy efficiency of any energy consumption device, extract abnormal correlation factors, and judge the rationality of the actual energy-saving condition based on the comparison with any expected energy-saving strategy. 8.The energy saving system of a smart park of claim 7, wherein: The energy consumption data analysis module comprises an energy consumption type division unit and an energy-saving efficiency evaluation unit. The energy consumption type division unit is configured to collect the historical data of any energy consumption device in the park, determine the energy consumption type of each energy consumption device and perform type division. 9.The energy saving system of a smart park of claim 7, wherein: The energy-saving potential realization module comprises an energy-saving deviation analysis unit and an expected potential setting unit. The energy-saving deviation analysis unit is configured to compare the energy-saving condition of any energy consumption device with the benchmark energy-saving range of the corresponding energy consumption type, identify the energy-saving deviation, and the expected potential setting unit is configured to identify the correlation factor with energy-saving potential based on the energy-saving deviation, and formulate an expected energy-saving strategy.

10. The energy saving system of a smart park according to claim 7, characterized in that: The reasonable energy-saving judgment module comprises a real-time energy efficiency analysis unit and an abnormal energy-saving identification unit. The real-time energy efficiency analysis unit is configured to collect the energy consumption data of any energy consumption device in real time, analyze the real-time energy efficiency of any energy consumption device, and extract abnormal correlation factors. The abnormal energy-saving identification unit is configured to judge the rationality of the actual energy-saving condition based on the comparison with any expected energy-saving strategy.

Images