Method, system and air conditioner for intelligently reducing consumption of air conditioning system

Abstract

The present invention provides a method, system and air conditioner for intelligent energy saving of an air conditioning system. The method utilizes an existing BAS system and additional sensors, and realizes intelligent energy saving of the air conditioning system through machine learning and online learning methods, combined with the operating characteristics of the air conditioning system, automatic and intelligent control. The method comprises: collecting operation strategies, environmental impact indicators, comfort index values ​​and energy efficiency ratios corresponding to operation actions; forming a group of corresponding operation strategy record cards with the above four types of data; screening out record cards with qualified comfort indexes and assembling them into an operation database; collecting real-time environmental impact indicators, and finding the operation strategy with the highest energy efficiency ratio under the same or similar environmental impact indicators from the operation database; recommending the above operation strategy to the operator or directly issuing operation instructions to the equipment; after adopting a new operation strategy, repeatedly learning the new operation actions according to the above steps, so that the operation strategy record cards in the operation database are continuously optimized.

Description

Technical Field

[0001] The present invention relates to the field of energy saving and consumption reduction, and in particular to a method, system and air conditioner for intelligent consumption reduction of an air conditioning system. Background Art

[0002] With the continuous acceleration of my country's urbanization process, many places are equipped with air-conditioning systems.

[0003] Air conditioning provides people with a suitable temperature environment, making people more comfortable, but the energy consumed during operation is very huge, which has increased the problem of energy shortage in my country, especially the air conditioning system in super large spaces, such as the lobby of star-rated hotels, large industrial buildings, shopping malls, etc. The capacity of the chiller, refrigeration water pump, cooling water pump and cooling fan of the traditional super large space air conditioning system is selected according to the maximum design heat load of the building, and a certain margin is left, but most of the current equipment systems are operated at full load under the maximum heat load conditions, which consumes a lot of energy. Because the outside temperature changes with the day and night of the four seasons and the heat load of the working operation period and the non-working operation period in the building are quite different, the heat load of the air conditioning system is much lower than the design load most of the time, that is, the air conditioning system can actually operate at a low load most of the time.

[0004] At present, most air-conditioning systems are equipped with an automation system (Building Automation System, BAS) or an environment and equipment monitoring system (Equipment Monitoring & Control System, EMCS). By collecting and processing the data of the air-conditioning system, unified monitoring and control of the equipment in the air-conditioning system can be achieved, achieving automatic control and energy saving and consumption reduction.

[0005] However, the traditional BAS system can only achieve simple inching start-stop control and interlock protection for air conditioning, and display the operating status of equipment such as cooling towers, cooling water systems, chillers, fans, and other equipment, as well as parameters such as ambient temperature and humidity. The traditional BAS system cannot further coordinate and optimize the control of these devices and systems based on environmental parameters, and cannot achieve intelligent energy saving in ultra-large space air-conditioning systems. Summary of the invention

[0006] The present invention provides a method for intelligent consumption reduction of an air-conditioning system. From the perspective of sustainable development, the existing BAS system is utilized, and multiple types of sensors required for data collection are additionally added. Through machine learning and online learning methods, combined with the operating characteristics of the air-conditioning system, automatic and intelligent control is performed to achieve intelligent consumption reduction of the air-conditioning system, especially the air-conditioning system for ultra-large spaces.

[0007] The method comprises:

[0008] Collect the operation strategies, environmental impact indicators, comfort index values, and energy efficiency ratios corresponding to the historical or real-time operation actions;

[0009] The combination of the above four types of data at the same time forms a record card;

[0010] Screen the record cards that meet the comfort index requirements and compile them into an operation database;

[0011] Carry out multi-dimensional and multi-level classification of the combination of various environmental impact indicators contained in the above record cards;

[0012] Sort the operation strategies corresponding to each category from high to low according to the energy efficiency ratio, and retain several of them to form the optimal operation strategy;

[0013] Collect real-time environmental impact indicators and match them to the same or similar categories of environmental impact indicators in the above operation database;

[0014] Recommend the best operation strategy in the same or similar categories to the operator or directly issue operation instructions to the equipment;

[0015] After adopting a new operating strategy, follow the above steps to conduct machine learning;

[0016] The energy efficiency ratio of the new operation strategy is compared with the optimal operation recommendation corresponding to the same category, and the operation strategy with high energy efficiency ratio is retained, so that the optimal operation strategy of the same or similar category in the operation database can be continuously optimized.

[0017] Furthermore, the comfort index includes an air distribution characteristic index (ADPI), which is the ratio of the number of measuring points that meet the specified wind speed and temperature requirements to the total number of measuring points, and the calculation formula is:

[0018]

[0019] Where ΔET is the effective temperature difference,

[0020] ΔET=(t i -t n )-7.66(u i -0.15)

[0021] t i and t n are the air temperature at a certain point in the working area and the given indoor design temperature respectively;

[0022] u i is the air velocity at a certain point in the working area.

[0023] Furthermore, the environmental impact index is a real-time variable, including influencing factors such as the system's external environmental conditions and the system's internal environmental conditions, such as external temperature, external humidity, regional set temperature, air-conditioning inlet water temperature, indoor CO2 concentration, indoor PM2.5 concentration, heat recovery water tank temperature, etc.

[0024] Furthermore, the operation strategy includes a combination of multiple operation factors, such as the status of the fan equipment of the fresh air system, the status of the fan coil equipment, and the cooling or heating working set temperature of the air conditioning system.

[0025] Furthermore, the energy efficiency ratio is the ratio of energy conversion efficiency. The larger the energy efficiency ratio is, the higher the energy conversion efficiency is when the central air conditioner is currently running, and the more electric energy is saved.

[0026] The energy efficiency ratio of air conditioners includes cooling energy efficiency ratio and heating energy efficiency ratio.

[0027] The cooling energy efficiency ratio EER is the cooling performance coefficient of the air conditioner, which is called the cooling energy efficiency ratio. It indicates the cooling capacity of the air conditioner under unit power. The calculation formula is as follows:

[0028]

[0029] Where Qc refers to the nominal cooling capacity per unit time, in kW, and W refers to the power consumed by the air conditioner per unit time, in kW;

[0030] The heating efficiency ratio COP is the heating performance coefficient of the air conditioner, which is called the heating efficiency ratio. It indicates the heating capacity of the air conditioner per unit power. The calculation formula is as follows:

[0031]

[0032] Qh refers to the nominal heating capacity per unit time, in kW, and W refers to the power consumed by the air conditioner per unit time, in kW.

[0033] Furthermore, the environmental impact indicators are classified in multiple dimensions and at multiple levels, and the operating strategies for each environmental impact indicator level are ranked from high to low with the energy efficiency ratio as the goal; during use, each environmental impact indicator or environmental impact indicator classification is improved in turn, and each environmental impact indicator or environmental impact indicator classification corresponds to an optimal operating strategy, forming an optimal operating strategy table; when the real-time environmental impact indicator matches the corresponding environmental impact indicator classification in the operating strategy database, the operating strategy ranked first (i.e., with the highest energy efficiency ratio) in the environmental impact indicator classification is obtained.

[0034] Furthermore, the environmental impact indicator classification is matched with the operation strategy through coding.

[0035] Furthermore, the coding includes variable coding, and different variable coding corresponds to different variables in the environmental impact index.

[0036] Furthermore, the encoding adopts adaptive encoding, and the calculation formula is: variable encoding = rounding function ((variable - variable minimum value) / variable step length);

[0037] The environmental impact index classification is matched with the operation strategy through coding.

[0038] The minimum value of the variable depends on the normal distribution of the variable data:

[0039] A. If the data is standard normal distribution, the lowest value of the variable = mean - 3*standard deviation;

[0040] B. If the data is left-skewed (skewness coefficient SK<-0.1) (there is a long tail on the left side of the distribution), when the data is concentrated (kurtosis coefficient KT>0), the lowest value of the variable = mean-3*standard deviation; when the data is dispersed (kurtosis coefficient KT<0), the lowest value of the variable = mean-standard deviation;

[0041] C. If the data is right-skewed (skewness coefficient SK>0.1) (there is a long tail on the right side of the distribution), when the data is concentrated (kurtosis coefficient KT>0), the lowest value of the variable = mean-3*standard deviation; when the data is dispersed (kurtosis coefficient KT<0), the lowest value of the variable = mean-standard deviation;

[0042] The variable step size depends on the amount of change in data precision:

[0043] DIFF = (mean + 3 * standard deviation) - (mean - 3 * standard deviation)

[0044] IF DIFF<=5AND DIFF>=0.5:

[0045] Variable step size = 0.1

[0046] ELIF DIFF>5AND DIFF<=50:

[0047] Variable step size = 1

[0048] ELIF DIFF<0.5:

[0049] Variable step size = 0.01

[0050] ELIF DIFF>50:

[0051] Variable step size = 10

[0052] Adaptive coding is used to better locate the operation strategy. It is the mapping of each variable to the operation strategy, that is, the corresponding operation strategy can be quickly found according to each variable. This variable coding can not only greatly reduce the memory space of the sample, increase the model training speed, but also greatly improve the learning accuracy.

[0053] A system for intelligently reducing the energy consumption of an air-conditioning system adopts the above-mentioned method for intelligently reducing the energy consumption of an air-conditioning system and is applied to the air-conditioning system, comprising a data acquisition module, a data transmission module and a data analysis module; the data acquisition module collects environmental impact indicators and operation strategies through sensors, and transmits them to the data analysis module through the data transmission module to analyze the comfort index value and energy efficiency ratio.

[0054] An air conditioner adopts the above-mentioned method for intelligent energy saving of air conditioner system, comprising a sensor and a processor. The sensor is used to collect environmental impact indicators, the processor obtains environmental impact indicators and collection operation strategies, analyzes and calculates the comfort index value and energy efficiency ratio, and pushes the optimal operation strategy to the execution component of the air conditioner.

[0055] It can be seen from the above description of the present invention that, compared with the prior art, the method, system and air conditioner for intelligently reducing energy consumption of an air conditioning system proposed by the present invention have the following advantages:

[0056] 1. This method continuously learns the operation data online through online updates, learns and iterates to generate the optimal operation strategy. It does not need to be calculated, but only needs to judge the current environmental impact index, and then match the same or similar environmental impact index from the database to obtain the optimal operation strategy, so as to achieve rapid and efficient reduction of the energy consumption of the air-conditioning system.

[0057] 2. Record historical operation strategies to facilitate viewing and comparison of operation strategies;

[0058] 3. The recommended best operation strategy must exist in the historical operation records, that is, the reproduction of history, safe and reliable, and there is no need to worry about errors due to wrong selection of operation strategy;

[0059] 4. Adaptive coding is adopted, which is free from fixed subjective experience and is not mixed with any subjective views or experience after observing data statistics. It first adaptively learns and adjusts the screening rules to extract safe and normal valuable data, and then adjusts the coding formula parameters through adaptive learning, so as to achieve the goal of mining potential value in historical data. It avoids the need for manual data analysis and parameter adjustment when new data comes in later, which ensures the feasibility of the data and greatly improves the flexibility of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] The drawings described herein are used to provide further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0061] in:

[0062] Figure 1 This is a diagram of the practical application steps of a method for intelligently reducing energy consumption in an air-conditioning system according to the present invention. DETAILED DESCRIPTION

[0063] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention clearer and more understandable, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0064] Embodiment 1:

[0065] A method for intelligent consumption reduction of air-conditioning systems, starting from the perspective of sustainable development, utilizes the existing BAS system and adds multiple types of sensors required for data collection. Through machine learning and online learning methods, combined with the operating characteristics of the air-conditioning system, automatic and intelligent control is implemented to achieve intelligent consumption reduction of the air-conditioning system, especially the air-conditioning system for ultra-large spaces.

[0066] The method comprises:

[0067] S10: Collecting the operation strategy, environmental impact index, comfort index value, and energy efficiency ratio corresponding to each operation action. The operation action can be a historical operation or a current operation;

[0068] S20: The above four types of data form a set of corresponding operation strategy record cards;

[0069] S30: collecting the record cards whose comfort indexes meet the conditions into an operation database;

[0070] S40: Collecting real-time environmental impact indicators, and finding the operation strategy with the highest energy efficiency ratio under the same or similar environmental impact indicators from the above operation database;

[0071] S50: recommending the above operation strategy to the operator or directly issuing an operation instruction to the device;

[0072] S60: After adopting the new operation strategy, the new operation action is repeatedly learned according to the above steps, so that the operation strategy record card in the operation database is continuously optimized.

[0073] Specifically, in this embodiment,

[0074] The comfort index includes an air distribution characteristic index (ADPI), which is the ratio of the number of measuring points that meet the specified wind speed and temperature requirements to the total number of measuring points. The calculation formula is:

[0075]

[0076] Where ΔET is the effective temperature difference,

[0077] ΔET=(t i -t n )-7.66(u i -0.15)

[0078] t i and t n are the air temperature at a certain point in the working area and the given indoor design temperature, in °C;

[0079] u i It is the air velocity at a certain point in the working area, in m / s.

[0080] The environmental impact index is a real-time variable, including influencing factors such as the system's external environmental conditions and the system's internal environmental conditions, such as external temperature, external humidity, regional set temperature, air-conditioning inlet water temperature, indoor CO2 concentration, indoor PM2.5 concentration, heat recovery water tank temperature, etc.; the operation strategy includes a combination of multiple operation factors, and the operation factors include the fresh air system fan equipment status, fan coil equipment status, and the air-conditioning system cooling or heating working set temperature.

[0081] The energy efficiency ratio is the ratio of energy conversion efficiency. The larger the energy efficiency ratio is, the higher the energy conversion efficiency is when the air conditioner is currently running, and the more electric energy is saved.

[0082] The energy efficiency ratio of air conditioners includes cooling energy efficiency ratio and heating energy efficiency ratio.

[0083] The cooling energy efficiency ratio EER is the cooling performance coefficient of the air conditioner, which is called the cooling energy efficiency ratio. It indicates the cooling capacity of the air conditioner per unit power. The calculation formula is as follows:

[0084]

[0085] Where Q c It refers to the nominal cooling capacity per unit time, in kW; W refers to the power consumed by the air conditioner per unit time, in kW.

[0086] The heating efficiency ratio COP is the heating performance coefficient of the air conditioner, which is called the heating efficiency ratio. It indicates the heating capacity of the air conditioner per unit power. The calculation formula is as follows:

[0087]

[0088] Where Q h It refers to the nominal heating capacity per unit time, in kW; W refers to the power consumed by the air conditioner per unit time, in kW.

[0089] Embodiment 2:

[0090] On the basis of Example 1, the environmental impact indicators are classified, and further graded in the classification. Specifically, the graded classification means that each influencing factor in the environmental impact indicator is first graded, and then combined into different environmental impact indicator grades, and the operation strategies of each environmental impact indicator grade are sorted from high to low with the energy efficiency ratio as the goal.

[0091] During use, each environmental impact indicator or environmental impact indicator level is improved in turn. Each environmental impact indicator or environmental impact indicator level corresponds to an optimal operation strategy, forming an optimal operation strategy table.

[0092] When the real-time environmental impact indicator matches the corresponding environmental impact indicator level in the operation strategy database, the operation strategy that ranks first (ie, has the highest energy efficiency ratio) in the environmental impact indicator level is obtained.

[0093] By classifying the environmental impact indicators, the practicability of the method can be improved and the memory space of the samples can be reduced.

[0094] Embodiment three:

[0095] On the basis of the first or second embodiment, the environmental impact index or the environmental impact index level is matched with the operation strategy one by one through coding, and the coding includes variable coding, and different variable codes correspond to different variables in the environmental impact index.

[0096] In this embodiment, the encoding adopts adaptive encoding.

[0097] Calculation formula: variable code = rounding function ((variable – variable minimum value) / variable step size);

[0098] in,

[0099] The minimum value of the variable depends on the normal distribution of the variable data:

[0100] A. If the data is standard normal distribution, the lowest value of the variable = mean - 3*standard deviation;

[0101] B. If the data is left-skewed (skewness coefficient SK<-0.1) (there is a long tail on the left side of the distribution), when the data is concentrated (kurtosis coefficient KT>0), the lowest value of the variable = mean-3*standard deviation; when the data is dispersed (kurtosis coefficient KT<0), the lowest value of the variable = mean-standard deviation;

[0102] C. If the data is right-skewed (skewness coefficient SK>0.1) (there is a long tail on the right side of the distribution), when the data is concentrated (kurtosis coefficient KT>0), the lowest value of the variable = mean-3*standard deviation; when the data is dispersed (kurtosis coefficient KT<0), the lowest value of the variable = mean-standard deviation;

[0103] The variable step size depends on the amount of change in data precision:

[0104] DIFF = (mean + 3 * standard deviation) - (mean - 3 * standard deviation)

[0105] IF DIFF<=5AND DIFF>=0.5:

[0106] Variable step size = 0.1

[0107] ELIF DIFF>5AND DIFF<=50:

[0108] Variable step size = 1

[0109] ELIF DIFF<0.5:

[0110] Variable step size = 0.01

[0111] ELIF DIFF>50:

[0112] Variable step size = 10

[0113] Adaptive coding is used to better locate the operation strategy. It is the mapping of each variable to the operation strategy, that is, the corresponding operation strategy can be quickly found according to each variable. This variable coding can not only greatly reduce the memory space of the sample, increase the model training speed, but also greatly improve the learning accuracy.

[0114] The encoding method of the present invention is variable and replaceable, and this embodiment is only a specific implementation method. The present invention is not limited to this specific encoding method.

[0115] Embodiment 4:

[0116] See also Figure 1 , the above embodiment is applied to an actual air conditioning system. When the system starts working, its working steps include:

[0117] Initialize the system;

[0118] The sensor collects data, and the system receives the data and analyzes the working conditions, including the operation strategy, environmental impact index, and comfort index value corresponding to each operation action;

[0119] Then compare the target set information of the best operation strategy corresponding to the real-time environmental impact index with the same or similar environmental impact index in the database, the target set information includes comfort index, energy efficiency ratio, and energy consumption, to realize strategy condition judgment, when the historical target set information is better than the real-time environmental impact index, it is judged that a strategy is needed, and the database launches the strategy, when the historical target set information is not better than the real-time environmental impact index, it is judged that no strategy is needed;

[0120] Regardless of whether the strategy introduced by the database or the original operation strategy is adopted, it is necessary to make a learning condition judgment, that is, to learn the target set information corresponding to the current operation strategy, so as to further update the database;

[0121] Wait for data and enter the next round of data receiving.

[0122] Embodiment five:

[0123] A system for intelligent energy saving of an air conditioning system, which adopts the method for intelligent energy saving of an air conditioning system of the above embodiment and is applied to an air conditioner, comprises a data acquisition module, a data transmission module and a data analysis module; the data acquisition module collects environmental impact indicators and operation strategies through sensors, and transmits them to the data analysis module through the data transmission module to analyze the comfort index value and energy efficiency ratio.

[0124] Embodiment six:

[0125] An air conditioner adopts the method and system for intelligent energy saving of an air conditioner system of the above embodiment, comprising a sensor and a processor. The sensor is used to collect environmental impact indicators, the processor obtains environmental impact indicators and collection operation strategies, analyzes and calculates the comfort index value and energy efficiency ratio, and pushes the optimal operation strategy to the execution component of the air conditioner.

[0126] It can be seen from the above description of the present invention that, compared with the prior art, the method, system and air conditioner for intelligently reducing energy consumption of an air conditioning system proposed by the present invention have the following advantages:

[0127] 1. This method continuously learns the operation data online through online updates, learns and iterates to generate the optimal operation strategy. It does not need to be calculated, but only needs to judge the current environmental impact index, and then match the same or similar environmental impact index from the database to obtain the optimal operation strategy, so as to achieve rapid and efficient reduction of the energy consumption of the air-conditioning system.

[0128] 2. Record historical operation strategies to facilitate viewing and comparison of operation strategies;

[0129] 3. The recommended best operation strategy must exist in the historical operation records, that is, the reproduction of history, safe and reliable, and there is no need to worry about errors due to wrong selection of operation strategy;

[0130] 4. Adaptive coding is adopted, which is free from fixed subjective experience and is not mixed with any subjective views or experience after observing data statistics. It first adaptively learns and adjusts the screening rules to extract safe and normal valuable data, and then adjusts the coding formula parameters through adaptive learning, so as to achieve the goal of mining potential value in historical data. It avoids the need for manual data analysis and parameter adjustment when new data comes in later, which ensures the feasibility of the data and greatly improves the flexibility of the model.

[0131] The present invention is described above by way of example in conjunction with the accompanying drawings. It is obvious that the specific implementation of the present invention is not limited to the above-mentioned method. As long as various non-substantial improvements are made using the method concept and technical solution of the present invention, or the concept and technical solution of the present invention are directly applied to other occasions without improvement, they are all within the protection scope of the present invention.

Claims

1. A method for intelligently reducing consumption of an air conditioning system, characterized in that: The method comprises: Collect the operation strategies, environmental impact indicators, comfort index values, and energy efficiency ratios corresponding to the operation actions that occurred historically or in real time; combine the above four types of data at the same time to form a record card; screen the record cards with qualified comfort indicators to form an operation database; conduct multi-dimensional and multi-level classification of the combinations of various environmental impact indicators contained in the above record cards; sort the operation strategies corresponding to each category from high to low according to the energy efficiency ratio, and retain several of them to form the optimal operation strategy; collect real-time environmental impact indicators and match them to the same or similar categories of environmental impact indicators in the above operation database; recommend the optimal operation strategy in the above same or similar categories to the operator or directly issue operation instructions to the equipment; after adopting a new operation strategy, perform machine learning according to the above steps; compare the energy efficiency ratio of the new operation strategy with the optimal operation suggestion corresponding to the same category, and retain the operation strategy with high energy efficiency ratio, so that the optimal operation strategy of the same or similar category in the operation database is continuously optimized.

2. The method for intelligently reducing energy consumption of an air conditioning system according to claim 1, characterized in that: The comfort index includes an air distribution characteristic index (ADPI), which is the ratio of the number of measuring points that meet the specified wind speed and temperature requirements to the total number of measuring points. The calculation formula is: Where ΔET is the effective temperature difference, ΔET=(t i -t n )-7.66(u i -0.15) t i and t n are the air temperature at a certain point in the working area and the given indoor design temperature respectively; u i is the air velocity at a certain point in the working area.

3. The method for intelligently reducing consumption of an air conditioning system according to claim 1, characterized in that: The environmental impact indicators are real-time variables, including outside temperature, outside humidity, comfort value, air conditioning water inlet temperature, and heat recovery water tank temperature; the comfort value includes the temperature of each indoor area, the humidity of each indoor area, the CO2 concentration of each indoor area, and the PM2.5 concentration of each indoor area.

4. The method for intelligently reducing consumption of an air conditioning system according to claim 1, characterized in that: The data of the operation strategy includes the setting values ​​of various parameters of the fresh air system fan equipment, the setting values ​​of various parameters of the fan coil equipment, the cooling or heating working set temperature of the air-conditioning system, and the target temperature of each area.

5. The method for intelligently reducing consumption of an air conditioning system according to claim 1, characterized in that: The environmental impact indicators are classified in multiple dimensions and at multiple levels, and the operation strategies corresponding to each category are sorted from high to low with the energy efficiency ratio as the goal; when the real-time environmental impact indicator is the same as or similar to the environmental impact indicator in a category in the operation strategy database, several top-ranked operation strategies in the operation strategies corresponding to the category are obtained.

6. A system for intelligently reducing consumption of an air conditioning system, characterized in that: The method for intelligent energy saving of an air-conditioning system according to any one of claims 1 to 5 is applied to an air conditioner, and comprises a data acquisition module, a data transmission module and a data analysis module; the data acquisition module collects environmental impact indicators and operation strategies through sensors, and transmits them to the data analysis module through the data transmission module to analyze the comfort index value and energy efficiency ratio.

7. An air conditioner, characterized in that: The method for intelligent energy saving of an air-conditioning system according to any one of claims 1 to 5 comprises a sensor and a processor; the processor obtains environmental impact indicators and acquisition operation strategies, analyzes and calculates the comfort index value and energy efficiency ratio, and pushes the optimal operation strategy to the execution component of the air conditioner.

Images