A smart city energy management system for the Internet of Things
By dividing electricity consumption into time periods and conducting stability analysis in the smart city energy management system, the problem of inaccurate stability analysis of electricity consumption data has been solved, enabling more accurate monitoring and management of electricity consumption.
Patent Information
- Application Number
- CN202410478193.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-19
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-04-19
AI Technical Summary
Existing technologies do not provide accurate stability analysis results for electricity consumption data under different conditions, and their adaptability to electricity management is weak.
By acquiring electricity consumption data over a preset historical time period, dividing the electricity consumption status into time periods, and monitoring and managing the data based on time period difference indicators and electricity consumption stability indicators, including merging and matching electricity consumption status time periods, and using normalized indicators to analyze the stability of electricity consumption data.
It improves the accuracy of judging the stability of electricity consumption data and the adaptability of electricity management, enabling more accurate monitoring of changes in electricity consumption at any given moment and timely issuance of early warning notices.
Smart Images

Figure CN118368301B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of Internet of Things control technology, and in particular to a smart city energy management system for the Internet of Things. Background Art
[0002] Smart city energy management systems can be used to analyze and manage the power consumption of urban public equipment. The Internet of Things can deploy sensors such as energy meters to monitor the overall power consumption of urban public equipment in real time. Through comprehensive analysis of historical data, real-time data and other relevant data, it can discover power consumption trends, peaks and valleys, perceive changes in power consumption, ensure the balance between power supply and demand of urban public equipment, and provide intelligent power regulation and power consumption management through energy management systems.
[0003] There is a certain regularity in the electricity consumption of urban public equipment, such as day and night patterns. Based on this feature, the electricity consumption can be divided into time periods for stability analysis. However, due to seasonal or weather changes, the regularity of data changes over a period of time will change. The judgment of regularity alone will make the judgment of the stability of the current electricity consumption data unreliable. The stability analysis results of electricity consumption data in different situations are not accurate, and the adaptability of electricity management is weak. Summary of the invention
[0004] In order to solve the technical problems in the prior art that the stability analysis results of power consumption data in different situations are inaccurate and the adaptability of power management is weak, the purpose of the present invention is to provide a smart city energy management system for the Internet of Things, and the technical solutions adopted are as follows:
[0005] The present invention provides a smart city energy management system for the Internet of Things, the system comprising:
[0006] The power consumption data acquisition module is used to obtain the power consumption data of urban public equipment at each sampling time in a preset historical time period;
[0007] The power consumption status time period division module is used to divide the preset historical time period into power consumption status time periods according to the fluctuation stability and trend approximation of the power consumption data at the continuous sampling moments in the preset historical time period;
[0008] The current power consumption state time period matching module is used to obtain the time period difference index between the current power consumption state time period and each other power consumption state time period according to the distribution deviation of the power consumption data and the fluctuation difference degree of the power consumption data between the current power consumption state time period and each other power consumption state time period; determine the stable matching segment of the current power consumption state time period based on the time period difference index;
[0009] The current power consumption status management module is used to obtain the current power consumption stability index based on the time period difference index and time distribution deviation between the current power consumption status time period and each stable matching segment, as well as the distribution stability of power consumption data; and monitor and manage the current power consumption based on the power consumption stability index.
[0010] Furthermore, the method for acquiring the power usage status time period includes:
[0011] In a preset historical time period, the power consumption data at the first sampling moment is taken as the starting moment, and the interval consisting of a preset number of consecutive sampling moments after the starting moment is taken as the initial division interval; the average value of the power consumption data at all sampling moments in the initial division interval is taken as the interval distribution value of the initial division interval; the variance of the power consumption data at all sampling moments in the initial division interval is calculated to obtain the interval fluctuation value of the initial division interval; the difference in the power consumption data between each two adjacent sampling times in the initial division interval is calculated, and the average value is obtained to obtain the interval change value of the initial division interval;
[0012] The next sampling moment of the initial division interval is used as the time to be divided; the difference between the power consumption data at the time to be divided and the last sampling moment in the initial division interval is calculated to obtain the power consumption difference value at the time to be divided; the difference between the power consumption difference value at the time to be divided and the interval change value of the initial division interval is used as the change stability index at the time to be divided;
[0013] Calculate the variance of the power consumption data corresponding to all sampling moments in the initial partition interval and the moment to be partitioned as the data fluctuation value at the moment to be partitioned; take the difference between the data fluctuation value at the moment to be partitioned and the interval fluctuation value of the initial partition interval as the distribution stability index at the moment to be partitioned;
[0014] The difference between the power consumption data at the time to be divided and the interval distribution value of the initial divided interval is used as the numerical stability index at the time to be divided;
[0015] Obtain the slope of each sampling moment; obtain the combined index of the time to be divided according to the change stability index, distribution stability index and numerical stability index of the time to be divided, and the deviation between the slope of the sampling moment in the initial division interval and the slope of the time to be divided;
[0016] When the merge index of the time to be divided meets the preset merge condition, the time to be divided is merged with the initial division interval to obtain a new merged initial division interval, and the merging is stopped when the merge index of the time to be divided does not meet the preset merging condition, and the initial division interval at this time is used as a power consumption state time period; the first sampling moment of the remaining unmerged sampling moments is used as the new starting moment to iteratively merge and divide to obtain the power consumption state time period until all sampling moments are merged;
[0017] If the number of all remaining uncombined sampling moments after the start moment is less than or equal to a preset number, all remaining uncombined sampling moments are taken as a power usage state time period.
[0018] Furthermore, the expression of the combined index is:
[0019] Where P n It is expressed as the combined index of the nth time to be divided, ΔE n Expressed as the change stability index at the nth time to be divided, ΔS n Expressed as the distribution stability index of the nth time to be divided, ΔF n Expressed as the numerical stability index of the nth time to be divided, K n It is expressed as the total number of sampling moments in the initial partition interval before the nth time to be partitioned, k i It is expressed as the slope of the i-th sampling time in the initial partition interval before the n-th time to be partitioned, k n It is represented as the slope before the nth time to be divided, || is represented as the absolute value extraction function, and norm[] is represented as the normalization processing function.
[0020] Furthermore, the preset merging condition is that the merging index of the time to be divided is greater than or equal to a preset merging threshold.
[0021] Furthermore, the method for obtaining the time period difference index includes:
[0022] For any power consumption time period, calculate the average value of the power consumption data at all sampling times within the power consumption time period to obtain the time period mean value of the power consumption time period; calculate the variance of the power consumption data at all sampling times within the power consumption time period to obtain the time period fluctuation value of the power consumption time period; use the range of the power consumption data at the sampling times within the power consumption time period as the time period data distribution value of the power consumption time period;
[0023] The difference between the time period mean value of the current power consumption time period and each other power consumption time period is used as the time period value difference index; the difference between the time period data distribution value of the current power consumption time period and each other power consumption time period is used as the time period distribution difference index; the difference between the time period fluctuation value of the current power consumption time period and each other power consumption time period is used as the time period fluctuation difference index;
[0024] According to the time period value difference index, time period distribution difference index and time period fluctuation difference index between the current power consumption time period and each other power consumption time period, the time period difference index between the current power consumption time period and each other power consumption time period is obtained; the time period value difference index, time period distribution difference index and time period fluctuation difference index are all positively correlated with the time period difference index, and the time period difference index is the normalized value.
[0025] Furthermore, the step of determining the stable matching segment of the power usage state time period at the current moment based on the time period difference index includes:
[0026] When the time period difference index is less than or equal to the preset difference threshold, the corresponding other power usage state time periods are used as stable matching segments of the power usage state time period at the current moment.
[0027] Furthermore, the method for obtaining the power consumption stability index includes:
[0028] Get the maximum value of the power consumption data in each stable matching segment, and calculate the average value to obtain the stable maximum value at the current moment; get the minimum value of the power consumption data in each stable matching segment, and calculate the average value to obtain the stable minimum value at the current moment; calculate the average value of the time period averages of all stable matching segments to obtain the stable average value at the current moment;
[0029] Based on the stable maximum value, stable minimum value and stable mean value at the current moment, and according to the distribution deviation of the power consumption data in the power consumption state time period at the current moment, the numerical stability at the current moment is obtained;
[0030] Obtain the timing length of the power consumption state time period and the stable matching section at the current moment; obtain the timing stability at the current moment according to the deviation of the timing length of the current moment and each stable matching section and the time period difference index;
[0031] When there is a negative number between the numerical stability and the temporal stability, the power consumption stability index at the current moment is set to the preset unstable value; otherwise, the power consumption stability index at the current moment is obtained based on the numerical stability and the temporal stability at the current moment, and both the numerical stability and the temporal stability are positively correlated with the power consumption stability index, and the power consumption stability index is the normalized value; the preset unstable value is a negative number.
[0032] Furthermore, the expression of the numerical stability is:
[0033] In the formula, R represents the numerical stability at the current moment, M a It is represented by the total number of stable matching segments in the power consumption state time period a at the current moment, F m,max It is expressed as the maximum value of the power consumption data in the mth stable matching segment, F a,max It is represented by the maximum value of the power consumption data in the current power consumption state time period a, F a,min It is represented by the minimum value of the power consumption data in the current power consumption state time period a.
[0034] F m,min It is expressed as the minimum value of the power consumption data in the mth stable matching segment, Y a It is represented by the mean value of the current power consumption time period a, Y m It is expressed as the time period mean of the mth stable matching segment, || is expressed as the absolute value extraction function, exp() is expressed as the exponential function with the natural constant as the base, and α is expressed as the preset adjustment coefficient.
[0035] Furthermore, the expression of the timing stability is:
[0036] In the formula, H represents the timing stability at the current moment, M a It is represented by the total number of stable matching segments in the power consumption state time period a at the current moment, T a,m It is expressed as the time difference index between the current power consumption time period a and the mth stable matching period, D a It is expressed as the time length of the power consumption state time period a at the current moment, D m It is represented as the timing length of the mth stable matching segment, and β is represented as the preset adjustment coefficient.
[0037] Furthermore, the monitoring and management of the current power consumption based on the power consumption stability index includes:
[0038] When the power consumption stability index is less than the preset stability threshold, an early warning notification is issued through the energy management system; the preset stability threshold is a positive number.
[0039] The present invention has the following beneficial effects:
[0040] The present invention divides the power consumption state into time periods according to the continuous fluctuation stability of the power consumption data in the historical time period, reflects the stable time period of the change of the power consumption data in the time period through the power consumption state time period, and reflects the local stability of different power consumption in different time periods through the power consumption state time period. Based on the fluctuation situation and the data value distribution situation, the stable matching segment is selected for the power consumption state time period at the current moment, and the fluctuation time period that is relatively close to the power consumption state at the current moment is reflected, so as to facilitate the subsequent specific analysis of the stability at the current moment, and obtain the power consumption stability index at the current moment according to the stability situation between the power consumption state time period at the current moment and the stable matching segment in the time series and the value distribution, and perform the power consumption monitoring and management at the current moment. The present invention analyzes the stability of the power consumption data at the current moment by segmenting the different power consumption states in history, so as to make the judgment of the stability of the power consumption data more accurate, and the power consumption monitoring and management has strong adaptability. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0042] Figure 1 A smart city energy management system for the Internet of Things provided by an embodiment of the present invention;
[0043] Figure 2 A schematic diagram of electricity consumption data of urban public facilities over a four-day period provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0044] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation method, structure, features and effects of a smart city energy management system for the Internet of Things proposed by the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.
[0045] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
[0046] The following is a detailed description of a specific solution of a smart city energy management system for the Internet of Things provided by the present invention in conjunction with the accompanying drawings.
[0047] See also Figure 1 , which shows a structure diagram of a smart city energy management system for the Internet of Things provided by an embodiment of the present invention, the system includes: an electricity consumption data acquisition module 101, an electricity consumption status time period division module 102, a current electricity consumption status time period matching module 103, and a current electricity consumption status management module 104.
[0048] The power consumption data acquisition module 101 is used to acquire the power consumption data of urban public equipment at each sampling time in a preset historical time period.
[0049] The public equipment in a city requires a huge amount of electricity, and has different levels of consumption at different time periods every day. In order to ensure the power supply balance of the city's public equipment, it is necessary to monitor, analyze and adjust the electricity consumption. In the embodiment of the present invention, when managing the electricity consumption of the city's public equipment in the Internet of Things system, it is first necessary to deploy various sensors in the city's public equipment, such as electricity meters, etc., to detect key data such as the equipment's electricity consumption, energy consumption and power consumption status. In the preset historical time period, the electricity consumption data of the city's public equipment at each sampling moment is obtained. The preset historical time period is the time period of 7 days before the current moment. The sampling moment is the moment of each monitoring and analysis, and it is set to collect and analyze every half hour. The specific data implementers can adjust according to the specific implementation situation, and there is no restriction here. Please refer to Figure 2 , which shows a schematic diagram of electricity consumption data of urban public facilities in the past four days provided by an embodiment of the present invention.
[0050] The power usage status time period division module 102 is used to divide the preset historical time period into power usage status time periods according to the fluctuation stability and trend approximation of the power usage data at consecutive sampling moments in the preset historical time period.
[0051] In general, urban public equipment has certain stable electricity consumption trends in different time periods, such as urban traffic management during the day and street lighting at night. Therefore, different time periods will have different electricity consumption trends. Due to the influence of different seasons and weather, the startup and duration of many electrical equipment will be adjusted differently. Therefore, there will be certain differences in the time of their stable data segments. Under normal circumstances, the electricity consumption will show a relatively stable change, that is, a slow rise, fall or stability, to ensure the stability of the power supply. Therefore, the time period can be divided based on the different stability conditions in the time period, that is, according to the fluctuation stability and trend approximation of the electricity consumption data at the continuous sampling moments in the preset historical time period, the preset historical time period is divided into electricity consumption status time periods.
[0052] Preferably, in a preset historical time period, the electricity data at the first sampling moment is taken as the starting moment, and the interval consisting of a preset number of consecutive sampling moments after the starting moment is taken as the initial division interval, and the sampling moments are traversed backwards with the initial division interval for merging. The average value of the electricity consumption data at all sampling moments in the initial division interval is taken as the interval distribution value of the initial division interval, reflecting the numerical distribution trend in the interval. The variance of the electricity consumption data at all sampling moments in the initial division interval is calculated to obtain the interval fluctuation value of the initial division interval, reflecting the degree of data confusion in the interval. The difference in electricity consumption data between each two adjacent sampling times in the initial division interval is calculated, and the average value is obtained to obtain the interval change value of the initial division interval, reflecting the trend of the degree of change between adjacent numerical values in the interval.
[0053] Combine the data characteristics in the interval to make a merge judgment for the next sampling moment, take the next sampling moment of the initial partition interval as the moment to be partitioned, and analyze the impact of merging the moment to be partitioned. Calculate the difference between the power consumption data at the moment to be partitioned and the last sampling moment in the initial partition interval to obtain the power consumption difference value at the moment to be partitioned. Take the difference between the power consumption difference value at the moment to be partitioned and the interval change value of the initial partition interval as the change stability index at the moment to be partitioned, and use the similarity between the change difference at the moment to be partitioned and the overall change difference of the interval as the judgment index on the change characteristics.
[0054] The variance of the power consumption data corresponding to all sampling moments in the initial partition interval and the time to be partitioned is calculated as the data fluctuation value of the time to be partitioned, reflecting the degree of data fluctuation after the time to be partitioned is added to the interval. The difference between the data fluctuation value of the time to be partitioned and the interval fluctuation value of the initial partition interval is used as the distribution stability index of the time to be partitioned, and the deviation of the impact of the chaos of the data after the time to be partitioned is used as the judgment index on the fluctuation characteristics.
[0055] The difference between the power consumption data at the time to be divided and the interval distribution value of the initial division interval is used as the numerical stability index of the time to be divided, and the difference between the value of the power consumption data at the time to be divided and the overall numerical distribution in the interval is used as the judgment index of the numerical characteristics.
[0056] The slope of each sampling moment is obtained to reflect the change of each sampling moment. According to the change stability index, distribution stability index and numerical stability index of the time to be divided, and the deviation between the slope of the sampling moment in the initial division interval and the slope of the time to be divided, the combined index of the time to be divided is obtained. In one embodiment of the present invention, the expression of the combined index is:
[0057]
[0058] Where Pn It is expressed as the combined index of the nth time to be divided, ΔE n Expressed as the change stability index at the nth time to be divided, ΔS n Expressed as the distribution stability index of the nth time to be divided, ΔF n Expressed as the numerical stability index of the nth time to be divided, K n It is expressed as the total number of sampling moments in the initial partition interval before the nth time to be partitioned, k i It is expressed as the slope of the i-th sampling time in the initial partition interval before the n-th time to be partitioned, k n It is represented as the slope before the nth moment to be divided, || is represented as the absolute value extraction function, norm[] is represented as the normalization processing function. It should be noted that normalization is a technical means well known to those skilled in the art. The normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.
[0059] Where, ΔE n The smaller it is, the closer the change degree of the time to be analyzed is to the overall change degree in the interval, and the better the effect of merging the time to be divided. n The smaller the value, the smaller the value, which reflects that the addition of the time to be divided has little effect on the chaos of the value distribution in the interval, and the higher the possibility of merging the time to be divided. n The smaller it is, the closer the value distribution of the time to be divided is to the overall value distribution in the interval, and the higher the possibility of merging. It reflects the approximation degree of the change trend between the time to be divided and the initial division interval. The smaller it is, the closer the change trend of the time to be divided is to the interval, and the higher the possibility of merging the time to be divided. Therefore, through normalized negative correlation, the higher the possibility of merging the time to be divided, the larger the merging index of the time to be divided.
[0060] When the merge index of the time to be divided meets the preset merge condition, the time to be divided is merged with the initial division interval to obtain a new merged initial division interval. In the embodiment of the present invention, the preset merge condition is that the merge index of the time to be divided is greater than or equal to the preset merge threshold, that is, when the merge index of the time to be divided is greater than the preset merge threshold, it means that the necessity of merging the time to be divided is relatively large, and the time to be divided is merged with the initial division interval to update the initial division interval. The preset merge threshold is set to 0.8. The specific value is adjusted by the implementer according to the specific implementation situation and is not limited here. Iterate the merging until the merge index of the time to be divided does not meet the preset merging condition, then stop the merging, and take the initial division interval at this time as a power consumption state time period, that is, when there is a merge index of the time to be divided that is less than the preset merge threshold, stop merging the time to be divided, and divide the initial division interval at this time as a power consumption state time period.
[0061] The first sampling moment among the remaining unmerged sampling moments is used as the new starting moment for iterative merging and division to obtain the power usage state time period until all sampling moments are merged, that is, each sampling moment has a corresponding power usage state time period. If the number of all remaining unmerged sampling moments after the starting moment is less than or equal to the preset number, it means that the remaining moments cannot form the initial division interval for judgment, so all the remaining unmerged sampling moments are used as a power usage state time period.
[0062] At this point, the sampling moments in the historical time period have been divided, and each power consumption state time period reflects the local stable change interval of the power consumption data, which is used for subsequent reference analysis of power consumption data changes in similar situations.
[0063] The current power usage status time period matching module 103 is used to obtain the time period difference index between the current power usage status time period and each other power usage status time period according to the distribution deviation of the power usage data and the degree of fluctuation difference of the power usage data between the current power usage status time period and each other power usage status time period; and determine the stable matching segment of the current power usage status time period based on the time period difference index.
[0064] The changes in electricity consumption in the same city over a period of time generally do not change greatly. Even if there are influences from weather and seasonal changes, the degree of influence in a short period of time is relatively similar, that is, the fluctuations in electricity consumption data in similar electricity consumption status time periods of each day within the same week or month will not be very different. Therefore, the stability of the current electricity consumption can be judged based on the time periods with similar fluctuations obtained in the historical time periods. First, the time period difference index between the current electricity consumption status time period and each other electricity consumption status time period is obtained based on the distribution deviation of the electricity consumption data and the degree of fluctuation difference in the electricity consumption data between the current electricity consumption status time period and each other electricity consumption status time period.
[0065] Preferably, for any power usage time period, the average value of the power usage data at all sampling moments in the power usage time period is calculated to obtain the time period mean of the power usage time period, reflecting the overall distribution of data in the power usage time period. The variance of the power usage data at all sampling moments in the power usage time period is calculated to obtain the time period fluctuation value of the power usage time period, reflecting the distribution stability of data in the power usage time period. The range of the power usage data at the sampling moments in the power usage time period is used as the time period data distribution value of the power usage time period, reflecting the degree of fluctuation of data in the power usage time period.
[0066] The difference between the current power consumption time period is calculated by the difference in data distribution and data fluctuation. The difference between the time period mean value between the current power consumption time period and each other power consumption time period is used as the time period value difference index to reflect the degree of difference in data value distribution. The difference in the time period data distribution value between the current power consumption time period and each other power consumption time period is used as the time period distribution difference index to reflect the degree of difference in data stability. The difference in the time period fluctuation value between the current power consumption time period and each other power consumption time period is used as the time period fluctuation difference index to reflect the degree of difference in data fluctuation.
[0067] The judgment is made by comprehensively considering the difference indicators in different aspects. According to the time period value difference indicator, time period distribution difference indicator and time period fluctuation difference indicator between the current power consumption state time period and each other power consumption state time period, the time period difference indicator between the current power consumption state time period and each other power consumption state time period is obtained, reflecting the degree of characteristic difference between the current power consumption state time period and other time periods. The time period value difference indicator, time period distribution difference indicator and time period fluctuation difference indicator are all positively correlated with the time period difference indicator. The time period difference indicator is the value after normalization. In the embodiment of the present invention, the expression of the time period difference indicator is:
[0068] Ta,v =softmax[|S v -S a |×|(F v,max -F v,min )-(F a,max -F a,min )|×|Y a -Y v |]
[0069] Where, T a,v It is expressed as the time difference index between the current power consumption state time period a and the vth other power consumption state time period, S v It is expressed as the period fluctuation value of the vth other power consumption state time period, S a It is represented by the fluctuation value of the power consumption time period a at the current moment, F v,max It is expressed as the maximum value of the power consumption data in the vth other power consumption state time period, F v,min It is expressed as the minimum value of the power consumption data in the vth other power consumption state time period, F a,max It is represented by the maximum value of the power consumption data in the current power consumption state time period a, F a,min It is represented by the minimum value of the power consumption data in the current power consumption state time period a, Y a It is represented by the mean value of the current power consumption time period a, Y v It is represented as the time period mean of the vth other power consumption status time period, || is represented as the absolute value extraction function, and softmax[] is represented as the normalized exponential function.
[0070] Among them, (F v,max -F v,min ) represents the time period data distribution value of the vth other power consumption state time period,
[0071] (F a,max -F a,min ) represents the data distribution value of the time period a of the current power consumption state, |(F v,max -F v,min )-(F a,max -F a,min )| represents the time period distribution difference index between the current power consumption state time period a and the vth other power consumption state time period, |S v -S a | represents the time period fluctuation difference index between the current power consumption time period a and the vth other power consumption time period, |Y a -Y v| is represented by the time period numerical difference index between the power consumption status time period a at the current moment and the vth other power consumption status time period, and reflects in the form of product that the time period numerical difference index, the time period distribution difference index and the time period fluctuation difference index are all positively correlated with the time period difference index. In other embodiments of the present invention, other basic mathematical operations can also be used to reflect that the time period numerical difference index, the time period distribution difference index and the time period fluctuation difference index are all positively correlated with the time period difference index, such as addition or power operation, etc., which are not limited here.
[0072] The larger the time period distribution difference index, the time period fluctuation difference index and the time period value difference index, the greater the difference between the two power consumption time periods.
[0073] Based on the time period difference index, a relatively matching time period with little difference can be screened out for the current power usage state time period, which is convenient for subsequent stable analysis of the power usage state at the current moment, and a stable matching segment of the current power usage state time period is determined based on the time period difference index. Preferably, when the time period difference index is less than or equal to the preset difference threshold, it means that the difference between the power usage state time periods is small, and the corresponding other power usage state time periods are used as stable matching segments of the current power usage state time period. In the embodiment of the present invention, the preset difference threshold is set to 0.2, and the implementer can adjust it according to the specific implementation scenario, which is not limited here.
[0074] At this point, based on the distribution and stability of the power consumption data, a stable matching segment with similar power consumption status is selected for the power consumption status time period at the current moment.
[0075] The current power consumption status management module 104 is used to obtain the current power consumption stability index based on the time period difference index and time distribution deviation between the current power consumption status time period and each stable matching segment, as well as the distribution stability of the power consumption data; and monitor and manage the current power consumption based on the power consumption stability index.
[0076] The current stability is judged by the approximate stable time periods in history. The more stable the current electricity consumption data is in multiple stable matching segments and the more credible it is in the stable time period, the more reliable the current electricity consumption is in maintaining stability. Therefore, the current electricity consumption stability index is obtained based on the time period difference index and time distribution deviation between the current electricity consumption status time period and each stable matching segment, as well as the distribution stability of the electricity consumption data.
[0077] Preferably, the maximum value of the power consumption data in each stable matching segment is obtained, and the average value is calculated to obtain the stable maximum value at the current moment, reflecting the average maximum situation of the power consumption data in the stable time period. The minimum value of the power consumption data in each stable matching segment is obtained, and the average value is calculated to obtain the stable minimum value at the current moment, reflecting the average minimum situation of the power consumption data in the stable time period. The average value of the time period averages of all stable matching segments is calculated to obtain the stable average value at the current moment, reflecting the average situation of the power consumption data in the stable time period.
[0078] Based on the stable maximum value, stable minimum value and stable mean value at the current moment, the numerical stability at the current moment is obtained according to the distribution deviation of the power consumption data in the power consumption state time period at the current moment. The smaller the deviation of the power consumption data in the power consumption state time period at the current moment is, the more stable the data distribution is. In one embodiment of the present invention, the expression of the numerical stability is:
[0079]
[0080] In the formula, R represents the numerical stability at the current moment, M a It is represented by the total number of stable matching segments in the power consumption state time period a at the current moment, F m,max It is expressed as the maximum value of the power consumption data in the mth stable matching segment, F a,max It is represented by the maximum value of the power consumption data in the current power consumption state time period a, F a,min It is represented by the minimum value of the power consumption data in the current power consumption state time period a, F m,min It is expressed as the minimum value of the power consumption data in the mth stable matching segment, Y a It is represented by the mean value of the current power consumption time period a, Y m It is represented as the time period mean of the mth stable matching segment, || represents the absolute value extraction function, exp() represents the exponential function with the natural constant as the base, α represents the preset adjustment coefficient. In the embodiment of the present invention, the preset adjustment coefficient is set to 0.0001. Its purpose is to prevent the denominator from being zero, making the formula meaningless.
[0081] in, It is represented as the stable mean value of the power consumption state time period a at the current moment, It reflects the difference between the current power consumption time period a and the stable mean. The smaller the difference, the more likely the power consumption state at the current moment is to be stable. It is expressed as the stable maximum value of the power consumption state time period a at the current moment, It is represented by the stable minimum value of the power consumption state time period a at the current moment, It is expressed as the difference between the stable maximum value and the maximum value of the current power consumption time period a, reflecting the distance between the current power consumption time period and the stable maximum value. It is expressed as the difference between the minimum value of the power consumption time period a at the current moment and the stable minimum value, reflecting the distance between the power consumption time period at the current moment and the stable minimum value.
[0082] Therefore Reflected in the analysis of numerical stability adjustment by deviation from the limit value, when When it is negative, it means that the maximum value in the current power consumption time period is greater than the stable maximum value or the minimum value is less than the stable minimum value, which means that there is a problem with the stability of the value. Therefore, the value stability is made negative to facilitate subsequent analysis and judgment. It should be noted that when the maximum value is greater than the stable maximum value and the minimum value is less than the stable minimum value, The positive situation is impossible because the stable matching section is similar to the data fluctuation of the power consumption state time period at the current moment, so there will not be a situation where the deviation is large at the same time. When the impact judgment is positive, it is credible and reliable. When is positive, if The larger the value is, the closer the power consumption data is to the stable minimum value in the current power consumption period, and the overall power consumption data is small. At this time, more data differences are considered to facilitate subsequent judgment and improvement of power supply capacity. The smaller it is, the closer the power consumption data in the current power consumption time period is to the stable maximum value, and the overall power consumption data is closer to the power consumption limit. At this time, there are fewer data differences to consider, and more analysis of duration factors is needed to facilitate the control of the overall power supply situation and balance the power supply pressure.
[0083] When R is larger, it means that when the data difference is larger, the data difference is smaller, that is, The smaller, The larger it is, the more stable the data is. Otherwise, it means that the data difference is large or the data difference is small. The smaller the value reflects the stability, the smaller R is.
[0084] Therefore, the continuous timing is further analyzed to obtain the timing length of the power consumption state time period and the stable matching segment at the current moment, reflecting the continuous state length of the power consumption state in the timing. According to the deviation of the timing length of the current moment and each stable matching segment and the time period difference index, the timing stability at the current moment is obtained, and the power consumption stability is judged by the continuity of the timing. In one embodiment of the present invention, the expression of the timing stability is:
[0085]
[0086] In the formula, H represents the timing stability at the current moment, M a It is represented by the total number of stable matching segments in the power consumption state time period a at the current moment, T a,m It is expressed as the time difference index between the current power consumption time period a and the mth stable matching period, D a It is expressed as the time length of the power consumption state time period a at the current moment, D m It represents the timing length of the mth stable matching segment, and β represents the preset adjustment coefficient. In the embodiment of the present invention, the preset adjustment coefficient is set to 0.0001, and its purpose is to prevent the denominator from being zero, which makes the formula meaningless.
[0087] in, Reflects the deviation between the current power consumption time period a and the timing length of the mth stable matching segment. When it is greater than 1, it indicates that the timing length of the stable matching segment is greater than the power consumption time period at the previous moment. It is expressed as the inverse proportion of the time period difference index between the current power consumption state time period a and the mth stable matching segment, reflecting the credibility of the timing deviation. The smaller the time period difference index is, the more similar the stable matching segment is to the current power consumption state time period, and the more credible the judgment of the timing deviation is.
[0088] It is expressed as the comprehensive deviation obtained by integrating the timing deviation of all stable matching segments after adjusting the confidence weight. When the overall value is less than 1, it means that the timing length of the integrated stable matching segment is less than the current power consumption time period. The current power consumption time period is less reliable in terms of timing stability. Therefore, through the difference with 1, when When it is negative, it means that the current power consumption status is too long in time sequence and its credibility is low, and power supply monitoring and adjustment are needed. For positive time, analyze the timing continuity. The larger H is, the shorter the current power consumption period is than the integrated stable matching period, and the more reliably the timing reflects stability. Therefore, the larger H is, the greater the timing stability is.
[0089] Comprehensive analysis is performed on numerical stability and timing stability. At this time, the case where the numerical stability and timing stability are negative numbers is considered. When there are negative numbers in the numerical stability and timing stability, it means that there is an extreme instability in the numerical value or timing, or both the numerical value and the timing are unstable. Therefore, the power consumption stability index at the current moment is set to a preset unstable value. In the embodiment of the present invention, the preset unstable value is a negative number, which is convenient for subsequent analysis. The preset unstable value is set to -1, and its numerical value can be adjusted by the implementer according to the specific implementation situation. Otherwise, that is, when both the numerical stability and the timing stability are positive numbers, the power consumption stability index at the current moment is obtained according to the numerical stability and the timing stability at the current moment, and both the numerical stability and the timing stability are positively correlated with the power consumption stability index. The power consumption stability index is a normalized value. In one embodiment of the present invention, the product of the numerical stability and the timing stability is normalized to obtain the power consumption stability index. In other embodiments of the present invention, other basic mathematical operations can also be used to reflect that both the numerical stability and the timing stability are positively correlated with the power consumption stability index, such as addition or power operation, etc., which are not limited here.
[0090] After obtaining the electricity consumption stability index reflecting the current moment, the electricity consumption at the current moment is monitored and managed based on the electricity consumption stability index. In one embodiment of the present invention, when the electricity consumption stability index is less than the preset stability threshold, it means that the electricity consumption state at the current moment deviates greatly from the historical stability state. An early warning notification is issued through the energy management system so that the staff can further make adjustments according to the specific abnormal power supply demand, and realize the monitoring and management of electricity consumption data based on the electricity consumption stability index at the current moment. In the embodiment of the present invention, the preset stability threshold is a positive number, and the preset stability threshold is set to 0.6. The implementer can adjust its value according to the specific implementation situation, and no limitation is made here. In other embodiments of the present invention, when the electricity consumption stability index is less than the preset stability threshold, the electricity consumption stability index can be input into the control module for numerical analysis and control to realize the control of unstable electricity consumption resources. No limitation is made here on the specific monitoring and management method.
[0091] In summary, the present invention divides the continuous fluctuation stability of electricity consumption data in the historical time period into preliminary electricity consumption time periods, reflects the stable time period of the change of electricity consumption data in the time period through the electricity consumption time period, and reflects the local stability of different electricity consumption in different time periods through the electricity consumption time period. Based on the fluctuation situation and data value distribution, the stable matching segment is selected for the electricity consumption time period at the current moment, reflecting the fluctuation time period that is relatively close to the electricity consumption state at the current moment, so as to facilitate the subsequent specific analysis of the stability at the current moment, and obtain the current moment electricity consumption stability index according to the stability of the current moment electricity consumption time period and the stable matching segment in time series and value distribution, and perform electricity consumption monitoring and management at the current moment. The present invention analyzes the stability of electricity consumption data at the current moment by segmenting different historical electricity consumption states, so as to make the judgment of the stability of electricity consumption data more accurate and the adaptability of electricity monitoring and management stronger.
[0092] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0093] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.
Claims
1. A smart city energy management system for the Internet of Things, characterized in that: The system comprises: The power consumption data acquisition module is used to obtain the power consumption data of urban public equipment at each sampling time in a preset historical time period; The power consumption status time period division module is used to divide the preset historical time period into power consumption status time periods according to the fluctuation stability and trend approximation of the power consumption data at the continuous sampling moments in the preset historical time period; The current power consumption state time period matching module is used to obtain the time period difference index between the current power consumption state time period and each other power consumption state time period according to the distribution deviation of the power consumption data and the fluctuation difference degree of the power consumption data between the current power consumption state time period and each other power consumption state time period; determine the stable matching segment of the current power consumption state time period based on the time period difference index; The current power consumption status management module is used to obtain the current power consumption stability index based on the time period difference index and time distribution deviation between the current power consumption status time period and each stable matching section, as well as the distribution stability of power consumption data; monitor and manage the current power consumption based on the power consumption stability index; The method for obtaining the power usage status time period includes: In a preset historical time period, the power consumption data at the first sampling moment is taken as the starting moment, and the interval consisting of a preset number of consecutive sampling moments after the starting moment is taken as the initial division interval; the average value of the power consumption data at all sampling moments in the initial division interval is taken as the interval distribution value of the initial division interval; the variance of the power consumption data at all sampling moments in the initial division interval is calculated to obtain the interval fluctuation value of the initial division interval; the difference in the power consumption data between each two adjacent sampling times in the initial division interval is calculated, and the average value is obtained to obtain the interval change value of the initial division interval; The next sampling moment of the initial division interval is used as the time to be divided; the difference between the power consumption data at the time to be divided and the last sampling moment in the initial division interval is calculated to obtain the power consumption difference value at the time to be divided; the difference between the power consumption difference value at the time to be divided and the interval change value of the initial division interval is used as the change stability index at the time to be divided; Calculate the variance of the power consumption data corresponding to all sampling moments in the initial partition interval and the moment to be partitioned as the data fluctuation value at the moment to be partitioned; take the difference between the data fluctuation value at the moment to be partitioned and the interval fluctuation value of the initial partition interval as the distribution stability index at the moment to be partitioned; The difference between the power consumption data at the time to be divided and the interval distribution value of the initial divided interval is used as the numerical stability index at the time to be divided; Obtain the slope of each sampling moment; obtain the combined index of the time to be divided according to the change stability index, distribution stability index and numerical stability index of the time to be divided, and the deviation between the slope of the sampling moment in the initial division interval and the slope of the time to be divided; When the merge index of the time to be divided meets the preset merge condition, the time to be divided is merged with the initial division interval to obtain a new merged initial division interval, and the merging is stopped when the merge index of the time to be divided does not meet the preset merging condition, and the initial division interval at this time is used as a power consumption state time period; the first sampling moment of the remaining unmerged sampling moments is used as the new starting moment to iteratively merge and divide to obtain the power consumption state time period until all sampling moments are merged; If the number of all remaining unmerged sampling moments after the start time is less than or equal to the preset number, all remaining unmerged sampling moments are taken as a power usage state time period; The expression of the combined index is: Where P n It is expressed as the combined index of the nth time to be divided, ΔE n Expressed as the change stability index at the nth time to be divided, ΔS n Expressed as the distribution stability index of the nth time to be divided, ΔF n Expressed as the numerical stability index of the nth time to be divided, K n It is expressed as the total number of sampling moments in the initial partition interval before the nth time to be partitioned, k i It is expressed as the slope of the i-th sampling time in the initial partition interval before the n-th time to be partitioned, k n It is represented as the slope before the nth time to be divided, || is represented as the absolute value extraction function, and norm[] is represented as the normalization processing function; The method for obtaining the time period difference index includes: For any power consumption time period, calculate the average value of the power consumption data at all sampling times within the power consumption time period to obtain the time period mean value of the power consumption time period; calculate the variance of the power consumption data at all sampling times within the power consumption time period to obtain the time period fluctuation value of the power consumption time period; use the range of the power consumption data at the sampling times within the power consumption time period as the time period data distribution value of the power consumption time period; The difference between the time period mean value of the current power consumption time period and each other power consumption time period is used as the time period value difference index; the difference between the time period data distribution value of the current power consumption time period and each other power consumption time period is used as the time period distribution difference index; the difference between the time period fluctuation value of the current power consumption time period and each other power consumption time period is used as the time period fluctuation difference index; According to the time period value difference index, time period distribution difference index and time period fluctuation difference index between the current power consumption state time period and each other power consumption state time period, the time period difference index between the current power consumption state time period and each other power consumption state time period is obtained; the time period value difference index, the time period distribution difference index and the time period fluctuation difference index are all positively correlated with the time period difference index, and the time period difference index is the normalized value; The step of determining the stable matching segment of the power usage state time period at the current moment based on the time period difference index includes: When the time period difference index is less than or equal to the preset difference threshold, the corresponding other power usage state time periods are used as stable matching segments of the power usage state time period at the current moment.
2. According to claim 1, a smart city energy management system for the Internet of Things is characterized in that: The preset merging condition is that the merging index of the time to be divided is greater than or equal to the preset merging threshold.
3. According to claim 1, a smart city energy management system for the Internet of Things is characterized in that: The method for obtaining the power consumption stability index includes: Get the maximum value of the power consumption data in each stable matching segment, and calculate the average value to obtain the stable maximum value at the current moment; get the minimum value of the power consumption data in each stable matching segment, and calculate the average value to obtain the stable minimum value at the current moment; calculate the average value of the time period averages of all stable matching segments to obtain the stable average value at the current moment; Based on the stable maximum value, stable minimum value and stable mean value at the current moment, and according to the distribution deviation of the power consumption data in the power consumption state time period at the current moment, the numerical stability at the current moment is obtained; Obtain the timing length of the power consumption state time period and the stable matching segment at the current moment; obtain the timing stability at the current moment according to the deviation of the timing length of the current moment and each stable matching segment and the time period difference index; When there is a negative number between the numerical stability and the temporal stability, the power consumption stability index at the current moment is set to the preset unstable value; otherwise, the power consumption stability index at the current moment is obtained based on the numerical stability and the temporal stability at the current moment, and both the numerical stability and the temporal stability are positively correlated with the power consumption stability index, and the power consumption stability index is the normalized value; the preset unstable value is a negative number.
4. A smart city energy management system for the Internet of Things according to claim 3, characterized in that: The expression of numerical stability is: In the formula, R represents the numerical stability at the current moment, M a It is represented by the total number of stable matching segments in the power consumption state time period a at the current moment, F m,max It is expressed as the maximum value of the power consumption data in the mth stable matching segment, F a,max It is represented by the maximum value of the power consumption data in the current power consumption state time period a, F a,min It is represented by the minimum value of the power consumption data in the current power consumption state time period a, F m,min It is expressed as the minimum value of the power consumption data in the mth stable matching segment, Y a It is represented by the mean value of the current power consumption time period a, Y m It is expressed as the time period mean of the mth stable matching segment, || is expressed as the absolute value extraction function, exp() is expressed as the exponential function with the natural constant as the base, and α is expressed as the preset adjustment coefficient.
5. According to claim 3, a smart city energy management system for the Internet of Things is characterized in that: The expression of the timing stability is: In the formula, H represents the timing stability at the current moment, M a It is represented by the total number of stable matching segments in the power consumption state time period a at the current moment, T a,m It is expressed as the time difference index between the current power consumption time period a and the mth stable matching period, D a It is expressed as the time length of the power consumption state time period a at the current moment, D m It is represented as the timing length of the mth stable matching segment, and β is represented as the preset adjustment coefficient.
6. According to claim 3, a smart city energy management system for the Internet of Things is characterized in that: The monitoring and management of the current power consumption based on the power consumption stability index includes: When the power consumption stability index is less than the preset stability threshold, an early warning notification is issued through the energy management system; the preset stability threshold is a positive number.
