Equipment energy efficiency evaluation method for smart station

Abstract

The invention relates to the technical field of smart station energy efficiency evaluation, and discloses a smart station equipment energy efficiency evaluation method, which comprises the steps of constructing a change relation curve of historical monitoring data and historical operation evaluation values in historical monitoring logs of each piece of equipment, and determining historical energy efficiency data of a smart station and concerned equipment; classifying the historical energy efficiency data according to the working condition type, and constructing a plurality of energy efficiency data combinations corresponding to the concerned equipment according to a classification result; randomly screening historical energy efficiency data in a plurality of energy efficiency data combinations of each concerned device and combining the historical energy efficiency data to obtain a plurality of concerned energy efficiency data sets, and constructing a comprehensive energy efficiency evaluation model; the real-time concerned energy efficiency data set is obtained, the comprehensive energy efficiency evaluation value of the real-time concerned energy efficiency data set is generated based on the comprehensive energy efficiency evaluation model, whether the energy efficiency optimization instruction of the concerned equipment is generated or not is judged according to the comprehensive energy efficiency evaluation value, the judgment accuracy of energy consumption related data is improved, and efficient and stable operation of the equipment of the intelligent station is guaranteed.

Description

Technical Field

[0001] This application relates to the technical field of energy efficiency evaluation of intelligent stations, and particularly to a method for evaluating the energy efficiency of equipment in an intelligent station. Background Art

[0002] Intelligent stations combine video monitoring, AI algorithms, and front-end devices to improve the safety and management efficiency of stations. In practical applications, improving energy efficiency is also one of the important goals of intelligent stations.

[0003] Currently, due to the excessive number of monitoring devices and the large amount of data analysis, the energy efficiency of equipment cannot be evaluated and analyzed in a timely manner, and it cannot be ensured that the equipment can provide services in an efficient and economical operation mode. Therefore, there is an urgent need for a method for evaluating the energy efficiency of equipment in an intelligent station to accurately monitor and analyze the energy efficiency-related data of the equipment and improve the accuracy of overall energy efficiency evaluation. Summary of the Invention

[0004] To solve the above technical problems, this application provides a method for evaluating the energy efficiency of equipment in an intelligent station. By determining the equipment of concern in the intelligent station and historical energy consumption data, constructing multiple groups of energy efficiency data of concern, and constructing a comprehensive energy efficiency evaluation model, obtaining a comprehensive energy efficiency evaluation value according to the comprehensive energy efficiency evaluation model and the real-time energy efficiency data group of concern, and generating an energy efficiency optimization instruction for the corresponding equipment of concern according to the comprehensive energy efficiency evaluation value, the judgment accuracy of energy consumption-related data is improved, and the equipment in the intelligent station is ensured to operate efficiently and stably.

[0005] In some embodiments of this application, a method for evaluating the energy efficiency of equipment in an intelligent station is provided, including: Obtain the historical monitoring logs of each device, construct a curve of the change relationship between the historical monitoring data and the historical operation evaluation value in the historical monitoring logs of each device, and determine the historical energy efficiency data and the equipment of concern in the intelligent station; Classify the historical energy efficiency data of each device of concern according to the working condition type, and construct several energy efficiency data combinations corresponding to the device of concern according to the classification result; Randomly select and combine the historical energy efficiency data in several energy efficiency data combinations of each device of concern to obtain multiple groups of energy efficiency data of concern, and construct a comprehensive energy efficiency evaluation model; Obtain real-time energy efficiency data and construct a real-time energy efficiency data group of concern, generate a comprehensive energy efficiency evaluation value of the real-time energy efficiency data group of concern based on the comprehensive energy efficiency evaluation model, and judge whether to generate an energy efficiency optimization instruction for the device of concern according to the comprehensive energy efficiency evaluation value.

[0006] In some embodiments of this application, determining the historical energy efficiency data and the equipment of concern in the intelligent station includes: Preset multiple evaluation indicators for an intelligent station, where the evaluation indicators include energy efficiency indicators and other indicators, and determine the associated indicators of the energy efficiency indicators according to the degree of association between the other indicators and the energy efficiency indicators; Determine the comprehensive evaluation indicator according to the energy efficiency indicator and the corresponding associated indicator; Establish a time reference line based on the historical monitoring duration of the historical monitoring log, and set data collection nodes based on a preset time interval; Collect the historical monitoring data of multiple devices in the corresponding historical monitoring log and the historical operation evaluation value of the comprehensive evaluation indicator according to the data collection nodes, and map them onto the corresponding time reference line to obtain multiple change relationship curves of the historical monitoring data and the historical operation evaluation value of multiple historical monitoring logs; Screen out several mutation nodes of the historical operation evaluation value in each change relationship curve and the first mutation characteristics of the corresponding mutation nodes, where the first mutation characteristics include the first mutation variable value and the first mutation rate; Set the preset influence period of several mutation nodes, and screen out the second mutation characteristics of the historical monitoring data in each preset influence period, where the second mutation characteristics include the second mutation variable value and the second mutation rate; Construct a first mutation characteristic sequence and a second mutation characteristic sequence in chronological order, and determine the correlation and degree of correlation between each historical monitoring data and the historical operation evaluation value in the same change relationship curve according to the first mutation characteristic sequence and the second mutation characteristic sequence; Generate the comprehensive correlation degree between the corresponding historical monitoring data and the comprehensive evaluation indicator according to the multiple correlation degrees of the same historical monitoring data in multiple change relationship curves; If the comprehensive correlation degree is greater than the preset comprehensive correlation degree threshold, set the corresponding historical monitoring data as historical energy efficiency data, and set the weight coefficient of each historical energy efficiency data; Generate the attention coefficient of the corresponding device according to the number of historical energy efficiency data involved in the device and the weight coefficient of the corresponding historical energy efficiency data; Set the device with an attention coefficient greater than the preset attention coefficient threshold as the attention device.

[0007] In some embodiments of the present application, the calculation formula for the historical operation evaluation value of the comprehensive evaluation indicator is: ; where P is the comprehensive evaluation indicator, a1 is the initial weight of the energy efficiency indicator, n is the number of associated indicators, is the weight corresponding to the i-th type of associated indicator, is the historical operation sub-evaluation value of the i-th type of associated indicator, Y is the preset operation sub-evaluation value threshold, k is a preset constant, is the historical operation sub-evaluation value of the energy efficiency indicator.

[0008] In some embodiments of the present application, several energy efficiency data combinations corresponding to the concerned devices are constructed according to the classification results, including: The working condition types of each concerned device and the corresponding standard energy efficiency data intervals for each working condition type are preset; The historical energy efficiency data of each concerned device is compared with the corresponding standard energy efficiency data intervals for each working condition type, and the working condition type corresponding to each historical energy efficiency data is determined according to the comparison results; The historical energy efficiency data of the same working condition type is used to construct the corresponding pending energy efficiency data combination; Calculate the historical operation sub-evaluation values of the historical energy efficiency data corresponding to the concerned devices in each pending energy efficiency data combination, and generate the historical fluctuation curve of the historical operation sub-evaluation values of the same pending energy efficiency data combination; Obtain the number of nodes with a fluctuation degree greater than the preset fluctuation degree threshold in each historical fluctuation curve and the difference in the fluctuation degree greater than the preset fluctuation degree threshold, and generate the energy efficiency influence coefficient of the corresponding pending energy efficiency data combination for the concerned device; If the energy efficiency influence coefficient is greater than the preset energy efficiency influence coefficient threshold, calculate the difference in the energy efficiency influence coefficient, set the collection number of the corresponding pending energy efficiency data combination according to the difference in the energy efficiency influence coefficient, collect the historical energy efficiency data in the corresponding pending energy efficiency data group according to the collection number, and construct the energy efficiency data combination of the corresponding working condition type; If the energy efficiency influence coefficient is less than the preset energy efficiency influence coefficient threshold, screen out the historical energy efficiency data at the corresponding nodes, use the screened historical energy efficiency data as the division nodes, perform mean processing on the remaining historical energy efficiency data in the pending energy efficiency data combination according to the division nodes, and construct the energy efficiency data combination of the corresponding working condition type according to the mean-processed historical energy efficiency data and the screened historical energy efficiency data.

[0009] In some embodiments of the present application, the calculation formula of the energy efficiency influence coefficient is: ; where H is the energy efficiency influence coefficient, h1 is the energy efficiency influence conversion coefficient, n0 is the number of nodes with a fluctuation degree greater than the preset fluctuation degree threshold, N is the total number of nodes, is the difference in the fluctuation degree at the s-th node.

[0010] In some embodiments of the present application, a comprehensive energy efficiency evaluation model is constructed, including: Randomly combine the historical energy efficiency data in several energy efficiency data combinations of different concerned devices to obtain a plurality of concerned energy efficiency data combinations, where each concerned energy efficiency data combination includes one historical energy efficiency data of all concerned devices; Randomly screen multiple historical concerned energy efficiency data combinations, use the screened historical concerned energy efficiency data combinations as training input data, and use the corresponding historical comprehensive energy efficiency evaluation values of the historical concerned energy efficiency data combinations as training output data to perform neural network training to obtain an initial comprehensive energy efficiency evaluation model; Perform iterative training on the initial comprehensive energy efficiency evaluation model based on multiple concerned energy efficiency data combinations, and calculate the credibility of the initial comprehensive energy efficiency evaluation model; If the credibility is greater than the preset credibility threshold, set the corresponding initial comprehensive energy efficiency evaluation model as the comprehensive energy efficiency evaluation model.

[0011] In some embodiments of the present application, generating a comprehensive energy efficiency evaluation value of the real-time concerned energy efficiency data group based on the comprehensive energy efficiency evaluation model includes: Obtain the real-time energy efficiency data of each concerned device at the current monitoring time node, and construct a real-time concerned energy efficiency data group according to the real-time energy efficiency data of each concerned device; Input the real-time concerned energy efficiency data group into the comprehensive energy efficiency evaluation model to obtain the comprehensive energy efficiency evaluation value at the current monitoring time node.

[0012] In some embodiments of the present application, before judging whether to generate an energy efficiency optimization instruction for a concerned device according to the comprehensive energy efficiency evaluation value, it further includes: Generate the difference between the concerned coefficient of each concerned device and the preset concerned coefficient threshold; Preset a first preset concerned coefficient difference interval, a second preset concerned coefficient difference interval, and a third preset concerned coefficient difference interval; When the concerned coefficient difference of a concerned device is within the first preset concerned coefficient difference interval, set the corresponding concerned device as a low-concern device; When the concerned coefficient difference of a concerned device is within the second preset concerned coefficient difference interval, set the corresponding concerned device as a medium-concern device; When the concerned coefficient difference of a concerned device is within the third preset concerned coefficient difference interval, set the corresponding concerned device as a high-concern device.

[0013] In some embodiments of the present application, judging whether to generate an energy efficiency optimization instruction for a concerned device according to the comprehensive energy efficiency evaluation value includes: Preset a comprehensive energy efficiency evaluation value threshold; If the comprehensive energy efficiency evaluation value is greater than the comprehensive energy efficiency evaluation value threshold, do not generate an energy efficiency optimization instruction for the concerned device; If the comprehensive energy efficiency evaluation value is less than the comprehensive energy efficiency evaluation value threshold, generate a comprehensive energy efficiency evaluation value difference; Preset a first preset comprehensive energy efficiency evaluation value difference interval, a second preset comprehensive energy efficiency evaluation value difference interval, and a third preset comprehensive energy efficiency evaluation value difference interval; When the difference value of the comprehensive energy efficiency evaluation value is within the first preset comprehensive energy efficiency evaluation value difference range, a first energy efficiency optimization instruction for low - attention devices is generated; The first energy efficiency optimization instruction includes screening out the low - attention devices to be optimized and the corresponding energy efficiency data to be optimized, and analyzing the low - attention devices to be optimized and the corresponding energy efficiency data to be optimized based on the objective function of the lowest energy efficiency optimization cost and the objective function of the shortest optimization duration to obtain a first energy efficiency optimization strategy; When the difference value of the comprehensive energy efficiency evaluation value is within the second preset comprehensive energy efficiency evaluation value difference range, a second energy efficiency optimization instruction for medium - attention devices is generated; The second energy efficiency optimization instruction includes screening out the medium - attention devices to be optimized and the corresponding energy efficiency data to be optimized, and analyzing the medium - attention devices to be optimized and the corresponding energy efficiency data to be optimized based on the objective function of the lowest energy efficiency optimization cost and the objective function of the shortest optimization duration to obtain a second energy efficiency optimization strategy; When the difference value of the comprehensive energy efficiency evaluation value is within the third preset comprehensive energy efficiency evaluation value difference range, a third energy efficiency optimization instruction for high - attention devices is generated; The third energy efficiency optimization instruction includes screening out the high - attention devices to be optimized and the corresponding energy efficiency data to be optimized, and analyzing the high - attention devices to be optimized and the corresponding energy efficiency data to be optimized based on the objective function of the lowest energy efficiency optimization cost and the objective function of the shortest optimization duration to obtain a third energy efficiency optimization strategy.

[0014] An equipment energy efficiency evaluation method for a smart station in an embodiment of the present application, compared with the prior art, has the beneficial effects that: By determining the attention devices and historical energy consumption data of the smart station, constructing multiple attention energy efficiency data groups, and constructing a comprehensive energy efficiency evaluation model, obtaining a comprehensive energy efficiency evaluation value according to the comprehensive energy efficiency evaluation model and the real - time attention energy efficiency data group, and generating an energy efficiency optimization instruction for the corresponding attention devices according to the comprehensive energy efficiency evaluation value, the judgment accuracy of energy - consumption - related data is improved, and the efficient and stable operation of the equipment in the smart station is ensured. Brief Description of the Drawings

[0015] Figure 1 is a flowchart of an equipment energy efficiency evaluation method for a smart station in an embodiment of the present application. Detailed Description of the Embodiment

[0016] The following combines the drawings and embodiments to further describe the detailed implementation of the present application in detail. The following embodiments are used to illustrate the present application but not to limit the scope of the present application.

[0017] In the description of the present application, it should be understood that the orientation or positional relationships indicated by the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. are based on the orientation or positional relationships shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present application.

[0018] The terms "first" and "second" are only used for descriptive purposes and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present application, unless otherwise specified, the meaning of "a plurality" is two or more.

[0019] In the description of the present application, it should be noted that unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.

[0020] As Figure 1 shown, a method for evaluating the energy efficiency of devices in an intelligent station of an embodiment of the present application includes: Step S101: Obtain the historical monitoring logs of each device, construct a curve of the change relationship between the historical monitoring data and the historical operation evaluation values in the historical monitoring logs of each device, and determine the historical energy efficiency data of the intelligent station and the devices to be concerned. Step S102: Classify the historical energy efficiency data of each device to be concerned according to the working condition types, and construct several energy efficiency data combinations corresponding to the devices to be concerned according to the classification results. Step S103: Randomly select and combine the historical energy efficiency data in several energy efficiency data combinations of each device to be concerned to obtain multiple concerned energy efficiency data groups, and construct a comprehensive energy efficiency evaluation model. Step S104: Obtain the real-time energy efficiency data and construct a real-time concerned energy efficiency data group, generate a comprehensive energy efficiency evaluation value of the real-time concerned energy efficiency data group based on the comprehensive energy efficiency evaluation model, and determine whether to generate an energy efficiency optimization instruction for the device to be concerned according to the comprehensive energy efficiency evaluation value.

[0021] In this embodiment, the historical energy efficiency data refers to the data that has a greater impact on the energy efficiency evaluation, and the devices to be concerned refer to the devices that have a greater impact on the energy efficiency evaluation.

[0022] In this embodiment, by determining the devices of interest and historical energy efficiency data, the subsequent data processing volume and analysis volume are reduced, the overall energy efficiency evaluation accuracy is improved, and the efficient and stable operation of the devices is ensured.

[0023] In some embodiments of the present application, determining the historical energy efficiency data and the devices of interest of the intelligent station includes: Preset multiple evaluation indicators for the intelligent station, where the evaluation indicators include energy efficiency indicators and other indicators, and determine the associated indicators of the energy efficiency indicators according to the degree of association between the other indicators and the energy efficiency indicators; Determine the comprehensive evaluation indicators according to the energy efficiency indicators and the corresponding associated indicators; Establish a time reference line based on the historical monitoring duration of the historical monitoring logs, and set data collection nodes based on a preset time interval; Collect the historical monitoring data of multiple devices in the corresponding historical monitoring logs and the historical operation evaluation values of the comprehensive evaluation indicators according to the data collection nodes, and map them onto the corresponding time reference lines to obtain multiple change relationship curves of the historical monitoring data and the historical operation evaluation values of the multiple historical monitoring logs; Screen out several mutation nodes of the historical operation evaluation values in each change relationship curve and the first mutation characteristics of the corresponding mutation nodes, where the first mutation characteristics include the first mutation amount value and the first mutation rate; Set the preset influence periods of several mutation nodes, and screen out the second mutation characteristics of the historical monitoring data in each preset influence period, where the second mutation characteristics include the second mutation amount value and the second mutation rate; Construct a first mutation characteristic sequence and a second mutation characteristic sequence in chronological order, and determine the correlation relationship and the degree of correlation between each historical monitoring data and the historical operation evaluation value in the same change relationship curve according to the first mutation characteristic sequence and the second mutation characteristic sequence; Generate the comprehensive correlation degree between the corresponding historical monitoring data and the comprehensive evaluation indicators according to the multiple degrees of correlation of the same historical monitoring data in multiple change relationship curves; If the comprehensive correlation degree is greater than the preset comprehensive correlation degree threshold, set the corresponding historical monitoring data as historical energy efficiency data, and set the weight coefficient of each historical energy efficiency data; Generate the attention coefficient of the corresponding device according to the number of historical energy efficiency data involved in the device and the weight coefficient of the corresponding historical energy efficiency data; Set the device with the attention coefficient greater than the preset attention coefficient threshold as the device of interest.

[0024] In this embodiment, other indicators refer to the evaluation indicators describing the operating status of the intelligent station, and the associated indicators refer to the evaluation indicators that affect the energy efficiency indicators. The comprehensive evaluation indicator is generated based on the energy efficiency indicator and the associated indicator, laying a foundation for subsequent screening of energy efficiency-related data and attention to equipment, and improving the accuracy of energy efficiency-related data.

[0025] In this embodiment, the correlation includes a linear correlation and a non-linear correlation. It is determined whether the first mutation variable value and the first mutation rate in the first mutation feature sequence change with the second mutation variable value and the second mutation rate in the second mutation feature sequence. If so, there is a correlation, and the degree of correlation is set according to the influence degree of the second mutation variable value on the first mutation variable value. That is, when the second mutation variable value is smaller but the first mutation variable value is larger, the corresponding degree of correlation is larger, and vice versa.

[0026] In this embodiment, when the degree of correlation is larger, the weight coefficient of the corresponding historical energy consumption data is larger. When the number of historical energy efficiency data involved in each device is larger and the weight coefficient of the corresponding historical energy efficiency data is larger, the attention coefficient is larger, and vice versa.

[0027] In some embodiments of the present application, the calculation formula for the historical operation evaluation value of the comprehensive evaluation indicator is: ; where P is the comprehensive evaluation indicator, a1 is the initial weight of the energy efficiency indicator, n is the number of associated indicators, is the weight corresponding to the i-th type of associated indicator, is the historical operation sub-evaluation value of the i-th type of associated indicator, Y is the preset operation sub-evaluation value threshold, k is a preset constant, is the historical operation sub-evaluation value of the energy efficiency indicator.

[0028] In this embodiment, by taking the energy efficiency indicator as the main evaluation center and correcting the historical operation sub-evaluation value of the energy efficiency indicator according to the historical operation sub-evaluation value of the associated indicator, the historical operation evaluation value of the comprehensive evaluation indicator is obtained, improving the accuracy of the historical operation evaluation value, and thus ensuring the accuracy of the historical energy consumption data.

[0029] In some embodiments of the present application, a number of energy efficiency data combinations corresponding to the attention devices are constructed according to the classification results, including: The working condition type of each attention device and the corresponding standard energy efficiency data interval for each working condition type are preset in advance; The historical energy efficiency data of each attention device is compared with the standard energy efficiency data interval corresponding to each working condition type, and the working condition type corresponding to each historical energy efficiency data is determined according to the comparison result; Construct corresponding to-be-determined energy efficiency data combinations for historical energy efficiency data of the same working condition type; Calculate the historical operation sub-evaluation values of the historical energy efficiency data in each to-be-determined energy efficiency data combination for the concerned device, and generate a historical fluctuation curve of the historical operation sub-evaluation values of the same to-be-determined energy efficiency data combination; Obtain the number of nodes in each historical fluctuation curve with a fluctuation degree greater than the preset fluctuation degree threshold and the difference in fluctuation degree greater than the preset fluctuation degree threshold, and generate the energy efficiency influence coefficient of the to-be-determined energy efficiency data combination corresponding to the working condition type for the concerned device; If the energy efficiency influence coefficient is greater than the preset energy efficiency influence coefficient threshold, calculate the difference in energy efficiency influence coefficients, set the acquisition number of the to-be-determined energy efficiency data combination corresponding to the working condition type according to the difference in energy efficiency influence coefficients, collect the historical energy efficiency data in the corresponding to-be-determined energy efficiency data group according to the acquisition number, and construct the energy efficiency data combination corresponding to the working condition type; If the energy efficiency influence coefficient is less than the preset energy efficiency influence coefficient threshold, screen out the historical energy efficiency data at the corresponding nodes, use the screened historical energy efficiency data as the division nodes, perform mean processing on the remaining historical energy efficiency data in the to-be-determined energy efficiency data combination according to the division nodes, and construct the energy efficiency data combination corresponding to the working condition type based on the mean-processed historical energy efficiency data and the screened historical energy efficiency data.

[0030] In this embodiment, taking the size order of multiple historical energy efficiency data in the corresponding to-be-determined energy efficiency data combination as the time reference line, map the historical operation sub-evaluation values of the corresponding historical energy efficiency data to the time reference line, and construct the historical fluctuation curve of the to-be-determined energy efficiency data combination of each working condition type.

[0031] In this embodiment, the energy efficiency influence coefficient is an evaluation of the energy efficiency influence degree of multiple historical energy consumption data in the to-be-determined energy efficiency data combination of the same working condition type on the historical operation sub-evaluation value of the corresponding concerned device. When the energy efficiency influence coefficient is larger, it indicates that the historical energy consumption data in the to-be-determined energy efficiency data combination of the corresponding working condition type has a greater impact on the energy efficiency of the corresponding concerned device. Then, set the acquisition number of the to-be-determined energy efficiency data combination in the corresponding working condition type according to the difference in energy efficiency influence coefficients. When the difference in energy efficiency influence coefficients is larger, the acquisition number should be more, thereby laying a foundation for subsequent construction of multiple concerned energy efficiency data combinations and comprehensive energy efficiency evaluation models, improving the evaluation accuracy of historical energy efficiency data for the energy efficiency of each concerned device, and thus obtaining the overall energy efficiency evaluation result of the intelligent station.

[0032] In some embodiments of the present application, the calculation formula of the energy efficiency influence coefficient is: ; Among them, H is the energy efficiency impact coefficient, h1 is the energy efficiency impact conversion coefficient, n0 is the number of nodes with a fluctuation degree greater than the preset fluctuation degree threshold, and N is the total number of nodes. is the difference in the degree of fluctuation at the s-th node.

[0033] In some embodiments of the present application, a comprehensive energy efficiency evaluation model is constructed, including: Randomly combine the historical energy efficiency data in several energy efficiency data combinations of different concerned devices to obtain multiple concerned energy efficiency data combinations. Among them, a historical energy efficiency data of all concerned devices is included in the concerned energy efficiency data combination. Randomly select multiple historical concerned energy efficiency data combinations. Use the selected historical concerned energy efficiency data combinations as training input data, and use the corresponding historical comprehensive energy efficiency evaluation values of the historical concerned energy efficiency data combinations as training output data to perform neural network training to obtain an initial comprehensive energy efficiency evaluation model. Based on multiple concerned energy efficiency data combinations, perform iterative training on the initial comprehensive energy efficiency evaluation model, and calculate the credibility of the initial comprehensive energy efficiency evaluation model. If the credibility is greater than the preset credibility threshold, set the corresponding initial comprehensive energy efficiency evaluation model as the comprehensive energy efficiency evaluation model.

[0034] In this embodiment, based on multiple concerned energy efficiency data combinations, perform iterative training on the initial comprehensive energy efficiency evaluation model, and re-screen multiple historical concerned energy efficiency data combinations to judge the credibility of the iterated initial comprehensive energy efficiency evaluation model, thereby improving the accuracy of the comprehensive energy efficiency evaluation model.

[0035] In some embodiments of the present application, generating a comprehensive energy efficiency evaluation value of the real-time concerned energy efficiency data group based on the comprehensive energy efficiency evaluation model includes: Obtain the real-time energy efficiency data of each concerned device at the current monitoring time node, and construct a real-time concerned energy efficiency data group according to the real-time energy efficiency data of each concerned device. Input the real-time concerned energy efficiency data group into the comprehensive energy efficiency evaluation model to obtain the comprehensive energy efficiency evaluation value at the current monitoring time node.

[0036] In some embodiments of the present application, before judging whether to generate an energy efficiency optimization instruction for the concerned device according to the comprehensive energy efficiency evaluation value, it further includes: Generate the difference between the concern coefficient of each concerned device and the preset concern coefficient threshold. Pre-set a first preset concern coefficient difference interval, a second preset concern coefficient difference interval, and a third preset concern coefficient difference interval. When the concern coefficient difference of the concerned device is within the first preset concern coefficient difference interval, set the corresponding concerned device as a low-concern device. When the difference in the attention coefficients of the devices to be monitored is within the second preset attention coefficient difference range, the corresponding devices to be monitored are set as medium - attention devices; When the difference in the attention coefficients of the devices to be monitored is within the third preset attention coefficient difference range, the corresponding devices to be monitored are set as high - attention devices.

[0037] In this embodiment, the first preset attention coefficient difference range < the second preset attention coefficient difference range < the third preset attention coefficient difference range.

[0038] In this embodiment, by determining low - attention devices, medium - attention devices, and high - attention devices according to the attention coefficients, it lays a foundation for formulating subsequent energy - efficiency optimization instructions, improves the application effect of the energy - efficiency optimization instructions, ensures the overall energy efficiency of the intelligent station, and the efficient and stable operation of the devices.

[0039] In some embodiments of the present application, judging whether to generate an energy - efficiency optimization instruction for the devices to be monitored according to the comprehensive energy - efficiency evaluation value includes: Presetting a comprehensive energy - efficiency evaluation value threshold in advance; If the comprehensive energy - efficiency evaluation value is greater than the comprehensive energy - efficiency evaluation value threshold, no energy - efficiency optimization instruction for the devices to be monitored is generated; If the comprehensive energy - efficiency evaluation value is less than the comprehensive energy - efficiency evaluation value threshold, a comprehensive energy - efficiency evaluation value difference is generated; Presetting a first preset comprehensive energy - efficiency evaluation value difference range, a second preset comprehensive energy - efficiency evaluation value difference range, and a third preset comprehensive energy - efficiency evaluation value difference range in advance; When the comprehensive energy - efficiency evaluation value difference is within the first preset comprehensive energy - efficiency evaluation value difference range, a first energy - efficiency optimization instruction for low - attention devices is generated; The first energy - efficiency optimization instruction includes screening out the low - attention devices to be optimized and the corresponding energy - efficiency data to be optimized, and analyzing the low - attention devices to be optimized and the corresponding energy - efficiency data to be optimized based on the objective function of the lowest energy - efficiency optimization cost and the objective function of the shortest optimization duration to obtain a first energy - efficiency optimization strategy; When the comprehensive energy - efficiency evaluation value difference is within the second preset comprehensive energy - efficiency evaluation value difference range, a second energy - efficiency optimization instruction for medium - attention devices is generated; The second energy - efficiency optimization instruction includes screening out the medium - attention devices to be optimized and the corresponding energy - efficiency data to be optimized, and analyzing the medium - attention devices to be optimized and the corresponding energy - efficiency data to be optimized based on the objective function of the lowest energy - efficiency optimization cost and the objective function of the shortest optimization duration to obtain a second energy - efficiency optimization strategy; When the comprehensive energy - efficiency evaluation value difference is within the third preset comprehensive energy - efficiency evaluation value difference range, a third energy - efficiency optimization instruction for high - attention devices is generated; The third energy efficiency optimization instruction includes screening out high-concern devices to be optimized and corresponding energy efficiency data to be optimized, and analyzing the high-concern devices to be optimized and the corresponding energy efficiency data to be optimized based on the objective function of the lowest energy efficiency optimization cost and the objective function of the shortest optimization duration, so as to obtain the third energy efficiency optimization strategy.

[0040] In this embodiment, an energy efficiency optimization model is constructed through the historical energy efficiency optimization strategies in the historical energy efficiency optimization log. The low-concern devices, medium-concern devices or high-concern devices to be optimized and the corresponding data to be optimized are input into the energy efficiency optimization model to obtain multiple pending energy efficiency optimization strategies. Based on the objective function of the lowest energy efficiency optimization cost and the objective function of the shortest optimization duration, the multiple pending energy efficiency optimization strategies are analyzed to obtain the energy efficiency optimization strategy with the lowest optimization cost and the longest optimization duration, thereby improving the energy efficiency optimization efficiency.

[0041] The above are only the preferred embodiments of the present application. It should be noted that for those of ordinary skill in the art, without departing from the technical principle of the present application, several improvements and replacements can be made, and these improvements and replacements should also be regarded as the protection scope of the present application.

Claims

1. A method for evaluating equipment energy efficiency of a smart station, characterized in that: include: Obtain the historical monitoring log of each device, construct the change relationship curve between the historical monitoring data and the historical operation evaluation value in the historical monitoring log of each device, and determine the historical energy efficiency data of the smart station and the focus equipment; Classify the historical energy efficiency data of each device of interest according to the operating condition type, and construct several energy efficiency data combinations corresponding to the device of interest based on the classification results; Randomly select and combine historical energy efficiency data from several energy efficiency data combinations of each concerned device to obtain multiple concerned energy efficiency data groups, and construct a comprehensive energy efficiency evaluation model; Acquire real-time energy efficiency data and construct a real-time energy efficiency data group, generate a comprehensive energy efficiency evaluation value of the real-time energy efficiency data group based on the comprehensive energy efficiency evaluation model, and determine whether to generate energy efficiency optimization instructions for the concerned equipment based on the comprehensive energy efficiency evaluation value.

2. The method for evaluating the equipment energy efficiency of a smart station according to claim 1, characterized in that: Determine the historical energy efficiency data of the smart site and focus on the equipment, including: Preset multiple evaluation indicators of the smart station, the evaluation indicators include energy efficiency indicators and other indicators, and determine the correlation indicators of the energy efficiency indicators according to the correlation degree between other indicators and the energy efficiency indicators; Determine comprehensive evaluation indicators based on energy efficiency indicators and corresponding related indicators; Establish a time reference line based on the historical monitoring duration of the historical monitoring log, and set data collection nodes based on preset time intervals; According to the data collection node, historical monitoring data of multiple devices in the corresponding historical monitoring log and historical operation evaluation values ​​of the comprehensive evaluation index are collected, and mapped to the corresponding time reference line to obtain multiple change relationship curves between the historical monitoring data and the historical operation evaluation values ​​of the multiple historical monitoring logs; Screening out a number of mutation nodes of the historical operation evaluation value in each change relationship curve and first mutation features of the corresponding mutation nodes, wherein the first mutation features include a first mutation value and a first mutation rate; Setting a preset impact period of several mutation nodes, and screening out a second mutation feature of the historical monitoring data in each preset impact period, wherein the second mutation feature includes a second mutation value and a second mutation rate; Constructing a first mutation feature sequence and a second mutation feature sequence in chronological order, and determining the correlation and correlation degree between each historical monitoring data and the historical operation evaluation value in the same change relationship curve according to the first mutation feature sequence and the second mutation feature sequence; Generate a comprehensive correlation degree between the corresponding historical monitoring data and the comprehensive evaluation index according to multiple correlation degrees of the same historical monitoring data in multiple change relationship curves; If the comprehensive correlation degree is greater than the preset comprehensive correlation degree threshold, the corresponding historical monitoring data is set as historical energy efficiency data, and a weight coefficient of each historical energy efficiency data is set; Generate a concern coefficient of the corresponding device according to the number of historical energy efficiency data involved in the device and the weight coefficient of the corresponding historical energy efficiency data; A device whose attention coefficient is greater than a preset attention coefficient threshold is set as a attention device.

3. The method for evaluating the equipment energy efficiency of a smart station according to claim 2, characterized in that: The calculation formula of the historical operation evaluation value of the comprehensive evaluation index is: ； Among them, P is the comprehensive evaluation index, a1 is the initial weight of the energy efficiency index, n is the number of related indicators, is the weight corresponding to the i-th correlation index, is the historical running sub-evaluation value of the i-th related indicator, Y is the preset running sub-evaluation value threshold, k is the preset constant, It is the historical operation sub-evaluation value of the energy efficiency indicator.

4. The method for evaluating the equipment energy efficiency of a smart station according to claim 3, characterized in that: According to the classification results, several energy efficiency data combinations of the corresponding equipment of interest are constructed, including: Pre-set the working condition type of each device of concern and the standard energy efficiency data range corresponding to each working condition type; Compare the historical energy efficiency data of each device of interest with the standard energy efficiency data range corresponding to each operating condition type, and determine the operating condition type corresponding to each historical energy efficiency data based on the comparison results; The historical energy efficiency data of the same operating condition type are used to construct a corresponding combination of pending energy efficiency data; Calculate the historical operation sub-evaluation value of the equipment of interest corresponding to the historical energy efficiency data in each pending energy efficiency data combination, and generate a historical fluctuation curve of the historical operation sub-evaluation value of the same pending energy efficiency data combination; Obtain the number of nodes in each historical fluctuation curve whose fluctuation degree is greater than a preset fluctuation degree threshold and the fluctuation degree difference whose fluctuation degree is greater than the preset fluctuation degree threshold to generate the energy efficiency influence coefficient of the pending energy efficiency data combination of the corresponding working condition type on the concerned equipment; If the energy efficiency impact coefficient is greater than the preset energy efficiency impact coefficient threshold, the energy efficiency impact coefficient difference is calculated, and the number of collections of pending energy efficiency data combinations of the corresponding working condition type is set according to the energy efficiency impact coefficient difference, and the historical energy efficiency data in the corresponding pending energy efficiency data group is collected according to the number of collections, and the energy efficiency data combination of the corresponding working condition type is constructed; If the energy efficiency impact coefficient is less than the preset energy efficiency impact coefficient threshold, the historical energy efficiency data at the corresponding node is screened out, and the screened out historical energy efficiency data is used as the division node. The remaining historical energy efficiency data in the determined energy efficiency data combination are averaged according to the division node, and the energy efficiency data combination of the corresponding operating condition type is constructed based on the historical energy efficiency data after average processing and the screened out historical energy efficiency data.

5. The method for evaluating the equipment energy efficiency of a smart station according to claim 4, characterized in that: The calculation formula of the energy efficiency impact coefficient is: ； Among them, H is the energy efficiency impact coefficient, h1 is the energy efficiency impact conversion coefficient, n0 is the number of nodes whose fluctuation degree is greater than the preset fluctuation degree threshold, N is the total number of nodes, is the fluctuation degree difference at the sth node.

6. The method for evaluating the equipment energy efficiency of a smart station according to claim 5, characterized in that: Construct a comprehensive energy efficiency assessment model, including: Randomly combining historical energy efficiency data in a plurality of energy efficiency data combinations of different concerned devices to obtain a plurality of concerned energy efficiency data combinations, wherein the concerned energy efficiency data combination includes a historical energy efficiency data of all concerned devices; Randomly select multiple historical energy efficiency data combinations, use the selected historical energy efficiency data combinations as training input data, use the historical comprehensive energy efficiency evaluation values ​​corresponding to the historical energy efficiency data combinations as training output data, perform neural network training, and obtain an initial comprehensive energy efficiency evaluation model; Iteratively train the initial comprehensive energy efficiency evaluation model based on multiple energy efficiency data combinations, and calculate the credibility of the initial comprehensive energy efficiency evaluation model; If the credibility is greater than the preset credibility threshold, the corresponding initial comprehensive energy efficiency evaluation model is set as the comprehensive energy efficiency evaluation model.

7. The method for evaluating the equipment energy efficiency of a smart station according to claim 6, characterized in that: Generate a comprehensive energy efficiency evaluation value of the real-time energy efficiency data group based on the comprehensive energy efficiency evaluation model, including: Obtain the real-time energy efficiency data of each concerned device at the current monitoring time node, and construct a real-time concerned energy efficiency data group according to the real-time energy efficiency data of each concerned device; The real-time energy efficiency data group is input into the comprehensive energy efficiency evaluation model to obtain the comprehensive energy efficiency evaluation value of the current monitoring time node.

8. The method for evaluating the equipment energy efficiency of a smart station according to claim 7, characterized in that: Before determining whether to generate an energy efficiency optimization instruction for the device of interest according to the comprehensive energy efficiency evaluation value, the method further includes: Generate an attention coefficient difference between an attention coefficient of each attention device and a preset attention coefficient threshold; Presetting a first preset attention coefficient difference interval, a second preset attention coefficient difference interval and a third preset attention coefficient difference interval; When the attention coefficient difference of the attention device is within the first preset attention coefficient difference interval, setting the corresponding attention device as a low attention device; When the attention coefficient difference of the attention device is within the second preset attention coefficient difference interval, setting the corresponding attention device as a medium attention device; When the attention coefficient difference of the attention device is within the third preset attention coefficient difference interval, the corresponding attention device is set as a high attention device.

9. The method for evaluating the equipment energy efficiency of a smart station according to claim 8, characterized in that: Determine whether to generate energy efficiency optimization instructions for the equipment of interest based on the comprehensive energy efficiency evaluation value, including: Pre-set comprehensive energy efficiency evaluation value threshold; If the comprehensive energy efficiency evaluation value is greater than the comprehensive energy efficiency evaluation value threshold, no energy efficiency optimization instruction for the concerned device is generated; If the comprehensive energy efficiency evaluation value is less than the comprehensive energy efficiency evaluation value threshold, a comprehensive energy efficiency evaluation value difference is generated; Presetting a first preset comprehensive energy efficiency evaluation value difference interval, a second preset comprehensive energy efficiency evaluation value difference interval and a third preset comprehensive energy efficiency evaluation value difference interval; When the comprehensive energy efficiency evaluation value difference is within a first preset comprehensive energy efficiency evaluation value difference interval, generating a first energy efficiency optimization instruction for the low-concern device; The first energy efficiency optimization instruction includes screening out low-concern devices to be optimized and corresponding energy efficiency data to be optimized, analyzing the low-concern devices to be optimized and corresponding energy efficiency data to be optimized based on the energy efficiency optimization minimum cost objective function and the optimization shortest time objective function to obtain a first energy efficiency optimization strategy; When the comprehensive energy efficiency evaluation value difference is within a second preset comprehensive energy efficiency evaluation value difference interval, generating a second energy efficiency optimization instruction for the device under consideration; The second energy efficiency optimization instruction includes screening out the medium-focus equipment to be optimized and the corresponding energy efficiency data to be optimized, analyzing the medium-focus equipment to be optimized and the corresponding energy efficiency data to be optimized based on the energy efficiency optimization minimum cost objective function and the optimization shortest time objective function to obtain a second energy efficiency optimization strategy; When the comprehensive energy efficiency evaluation value difference is within a third preset comprehensive energy efficiency evaluation value difference interval, generating a third energy efficiency optimization instruction for the high-concern equipment; The third energy efficiency optimization instruction includes screening out high-attention devices to be optimized and corresponding energy efficiency data to be optimized, analyzing the high-attention devices to be optimized and corresponding energy efficiency data to be optimized based on the energy efficiency optimization minimum cost objective function and the shortest optimization time objective function to obtain the third energy efficiency optimization strategy.

Images