A Smart Control Method for an Integrated Source-Grid-Load-Storage System

Abstract

This invention discloses an intelligent control method for an integrated power generation, grid, load, and energy storage system, relating to the field of power generation, grid, load, and energy storage technology. The method includes: retrieving monitoring records from the integrated system, extracting several time windows, and establishing a power change function for each power source; calculating the deviation between time windows, merging the time windows to obtain several sub-time periods within a day; obtaining the date to be set, and obtaining the marker window for each sub-time period within that date; retrieving fault records from the integrated system, and obtaining the power disturbance value for each sub-time period; obtaining the minimum data acquisition interval based on server resource values, and obtaining the data acquisition interval for each sub-time period within the date to be set based on the power disturbance value. This invention, by analyzing historical records of the power generation, grid, load, and energy storage system, obtains a reasonable acquisition interval, which helps to balance data monitoring intensity and server resources, meeting the needs of dynamic changes in power grid operation.

Description

Technical Field

[0001] This invention relates to the field of power generation, grid, load and storage technology, specifically an intelligent control method for an integrated power generation, grid, load and storage system. Background Technology

[0002] With the explosive development of new energy in my country, the installed capacity and power generation share of the power system have continued to rise. Due to the volatility of new energy, the power system faces the problem of balancing supply and demand. The emergence of the integrated source-grid-load-storage system has played a key role in building a clean, low-carbon, safe, controllable, flexible and efficient power system, which can ensure the coordinated optimization and dynamic regulation of the entire power generation, transmission, consumption and storage chain.

[0003] To balance the power system, ensure reliable grid operation, and achieve efficient absorption of new energy sources and dynamic supply-demand balance, it is necessary to monitor the integrated source-grid-load-storage system in real time and dynamically collect and analyze data from the power source side, grid side, load side, and energy storage side. However, current monitoring is usually carried out at fixed collection intervals. When the collection interval is long, there will be data lag and the problem of missing key disturbances. When the collection interval is short, it will lead to data redundancy and increase the computational pressure. Therefore, the fixed collection interval monitoring mode is no longer suitable for the needs of dynamic changes in grid operation. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent control method for an integrated power generation, grid, load, and storage system to solve the problems raised in the prior art.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] A smart control method for an integrated power generation, grid, load, and storage system includes the following steps:

[0007] Retrieve historical monitoring records from the integrated power generation, grid, load, and energy storage system, and extract monitoring records from different power sides, including the power source side, grid side, load side, and energy storage side; extract several time windows within a day, set monitoring values ​​for the power side according to the type of power side, and establish a power change function corresponding to each power side based on the monitoring values ​​of the power side in the time windows;

[0008] Extract any two adjacent time windows, obtain the degree of deviation between the two adjacent time windows according to the power change function, determine whether to merge the time windows according to the degree of deviation, and then obtain several sub-time periods within a day according to the merged time windows;

[0009] The current date is used as the date to be set. Based on the predicted temperature value of the date to be set and the corresponding power consumption area on the load side, the marking window corresponding to each sub-time period within the date to be set is obtained.

[0010] Retrieve historical fault records from the integrated system, and obtain the power change threshold based on the total power value at several moments before the fault occurred; obtain the power disturbance value for each sub-period based on the total power and power change threshold at each moment within the marked window;

[0011] Obtain the server's historical resource usage values ​​at different data collection intervals. Based on the server's available resource usage values, determine the minimum data collection interval. Preset the maximum data collection interval and, based on the minimum interval and the power disturbance value for each sub-period, determine the data collection interval for each sub-period within the set date.

[0012] One possible implementation involves establishing a power change function for each power source, including the following steps:

[0013] Set the duration of the time window to H, extract several time windows within a day, and obtain a certain time window W from them;

[0014] If the power side is the power source side, set the monitoring value to the total power generation and extract the total power generation of the power source side within the time window W.

[0015] If the power side is the load side, set the monitoring value to the total power consumption and extract the total power consumption of the load side within the time window W.

[0016] If the power supply side is the grid side, the monitoring value is set as the loss rate. Based on the total power generation F1 and total power consumption F2 within the time window W, the loss rate of the grid side within the time window W is obtained. ;

[0017] If the power side is the energy storage side, the monitoring value is set to the change in power. When the power on the energy storage side increases, the change in power is positive, and when the power on the energy storage side decreases, the change in power is negative, thus obtaining the change in power on the energy storage side within the time window W.

[0018] Total power generation, total power consumption, loss rate, and power change are parameters used to characterize the power situation on the power source side, grid side, load side, and energy storage side, respectively. Since power generation, consumption, and storage are regular, such as low power consumption in the early morning and high power consumption during the day, the target monitoring value can be obtained by averaging the monitoring values ​​at the same time over multiple days. The following calculations can be performed using the target monitoring value, which can make the analysis process and judgment results more reliable.

[0019] The monitoring values ​​for a certain power sector over a certain number of days are obtained within a time window W. The average value is calculated to obtain the target monitoring value for that power sector within the time window W. Then, the target monitoring value for that power sector in each time window is obtained, and a power change function is established to show the change of the target monitoring value with the time window.

[0020] One possible implementation involves obtaining several sub-time periods within a day, including the following steps:

[0021] Extract any two adjacent time windows, win1 and win2. Based on the power change function, obtain all target monitoring values ​​within time windows win1 and win2, and determine the degree of deviation between time windows win1 and win2. Among them, W n Let V1 be the weight of the nth type of power supply side, e be the natural constant, and V1 be the weight of the nth type of power supply side. n For the nth type of power side, V2 represents the target monitoring value within time window win1. n This represents the target monitoring value for the nth type of power supply within the time window win2.

[0022] If the deviation level X is less than the preset threshold, then time windows win1 and win2 are merged, and all time windows that meet the merging conditions are merged. Then, based on the merged time windows, several sub-time periods within a day are obtained.

[0023] Here, time windows with similar power conditions are merged, and based on these merged time windows, the day is divided into multiple sub-time periods. Then, by analyzing and judging each sub-time period separately, the computational resource consumption can be reduced. Formula: y=1-e -x It is a function of x, where x>0, and y is a function of x, where x>0. y increases as x increases. Since the larger the deviation between two time windows on a certain type of power side, the greater the degree of deviation between the two time windows, this formula can be used in the design of this scheme.

[0024] One possible implementation involves obtaining the marker window corresponding to each sub-time period within the date to be set, including the following steps:

[0025] Since electricity usage is also seasonal and regional, meaning that electricity consumption follows patterns in different regions and temperatures—for example, higher electricity consumption in winter and summer, and lower consumption in autumn and winter—we can extract a marker window based on this pattern. The specific implementation steps are as follows:

[0026] The current date is used as the date to be set. The load side includes supplying power to several power consumption areas. Based on the weather forecast platform, the predicted temperature value of each power consumption area for each sub-period within the date to be set is obtained. Several time period windows corresponding to a certain sub-period D are extracted from historical dates. The actual temperature value TPW of each power consumption area in a certain time period window is obtained, as well as the predicted temperature value of each power consumption area in sub-period D. If the difference between the actual temperature value and the predicted temperature value corresponding to each power consumption area is less than a preset difference threshold, the time period window TPW is marked to obtain all marked windows corresponding to sub-period D. Thus, the marked windows corresponding to each sub-period within the date to be set are obtained.

[0027] One possible implementation involves obtaining the power change threshold, including the following steps:

[0028] Retrieve historical fault records of the integrated system. These records are those from when the server detects and alarms when a fault occurs in the integrated system. Extract the total power value of the load at each moment within the time period TP prior to the fault occurrence. Extract any two adjacent moments t1 and t2 within the time period TP as a time combination C1. 2 Based on the total power values ​​pv1 and pv2 at times t1 and t2, the time combination C1 is obtained. 2 The power change value |pv1-pv2| is obtained, the power change value of several time moments is combined, and the average value is calculated as the power change threshold pv0;

[0029] Since power outages or alarms are usually caused by irregular power usage, which is represented here by power, for example, if the power at a certain moment changes significantly compared to the previous moment, this sudden power change will disrupt the original power balance and stable operation of the power grid, leading to increased equipment stress and reduced system stability. In such cases, the probability of a power grid failure increases significantly. Therefore, this solution analyzes power to obtain power disturbance values. A larger power disturbance value indicates a significantly increased probability of a failure during that sub-period, and the data collection interval for monitoring that sub-period should be relatively short to better observe the power grid situation, detect anomalies promptly, and prevent larger power failures. Conversely, a smaller power disturbance value indicates a relatively lower probability of a failure during that sub-period, and the data collection interval for monitoring that sub-period should be relatively long to better conserve resources and avoid resource waste due to over-monitoring.

[0030] One possible implementation involves obtaining the power disturbance value for each sub-time period, including the following steps:

[0031] Obtain all marked windows corresponding to a certain sub-time period D, and extract any two adjacent times t3 and t4 within the marked windows as time combination C3.4 If the total power values ​​pv3 and pv4 at times t3 and t4 satisfy |pv3-pv4|>pv0, then the time combination C3 4 The time slots are marked; then, from the N time slot combinations corresponding to sub-time slot D, the number of marked time slot combinations N0 is obtained, and the power disturbance value M of sub-time slot D is also obtained. D =1-e -N0 Then the power disturbance value for each sub-period is obtained.

[0032] One possible implementation involves obtaining the minimum time interval for data acquisition, including the following steps:

[0033] The integrated system deploys all power-side components on the same server and simultaneously collects and analyzes data. It obtains the server's resource usage values ​​when collecting data from the integrated system at different historical collection intervals, and fits a linear function to the change in resource usage values ​​over the collection intervals to obtain the variation function. The available resource usage value Z0 of the server is then obtained and substituted into the variation function to obtain the minimum data collection interval K. min .

[0034] One possible implementation involves obtaining the time interval for data collection for each sub-period within the set date, including the following steps:

[0035] Obtain the power disturbance value for each sub-time period, and obtain the maximum disturbance value M. max and minimum disturbance value M min ; Use the current preset maximum data acquisition time interval as K max According to the minimum time interval K min And the power disturbance value M of a certain sub-period D D According to the formula The sampling time interval K of sub-time period D is obtained. D This allows us to obtain the time interval for data collection for each sub-period within the set date.

[0036] Here, the data collection interval is set reasonably according to the rule that the smaller the power disturbance value, the longer the data collection interval, and the larger the power disturbance value, the shorter the data collection interval. This balances the monitoring intensity with the resources occupied, and meets the needs of dynamic changes in power grid operation.

[0037] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention provides an intelligent control method for an integrated power generation, grid, load, and energy storage system, comprising: retrieving monitoring records from the integrated power generation, grid, load, and energy storage system; extracting several time windows; establishing a power change function for each power side; calculating the degree of deviation between time windows; merging the time windows to obtain several sub-time periods within a day; obtaining the date to be set; obtaining the marked window for each sub-time period within the date to be set; retrieving fault records from the integrated system; obtaining the power disturbance value for each sub-time period; obtaining the minimum data acquisition time interval based on server resource values; and obtaining the data acquisition time interval for each sub-time period within the date to be set based on the power disturbance value. This invention, by analyzing historical records of power generation, grid, load, and energy storage, obtains a reasonable acquisition time interval, which helps to balance data monitoring intensity and server resources, and meets the needs of dynamic changes in power grid operation. Attached Figure Description

[0038] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a flowchart illustrating an intelligent control method for an integrated source-grid-load-storage system according to the present invention. Detailed Implementation

[0040] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] Example: Figure 1 As shown, this invention provides a technical solution for an intelligent control method of an integrated source-grid-load-storage system, comprising the following steps:

[0042] (1) Retrieve the historical monitoring records of the integrated power generation, grid, load and energy storage system, and extract the monitoring records of different power sides. The power side includes the power source side, grid side, load side and energy storage side. Extract several time windows within a day, set the monitoring values ​​for the power side according to the type of power side, and establish the power change function corresponding to each power side according to the monitoring values ​​of the power side in the time window.

[0043] Set the duration of the time window to H, extract several time windows within a day, and obtain a certain time window W from them;

[0044] If the power side is the power source side, set the monitoring value to the total power generation and extract the total power generation of the power source side within the time window W.

[0045] If the power side is the load side, set the monitoring value to the total power consumption and extract the total power consumption of the load side within the time window W.

[0046] If the power supply side is the grid side, the monitoring value is set as the loss rate. Based on the total power generation F1 and total power consumption F2 within the time window W, the loss rate of the grid side within the time window W is obtained. ;

[0047] If the power side is the energy storage side, the monitoring value is set to the change in power. When the power on the energy storage side increases, the change in power is positive, and when the power on the energy storage side decreases, the change in power is negative, thus obtaining the change in power on the energy storage side within the time window W.

[0048] Total power generation, total power consumption, loss rate, and power change are parameters used to characterize the power situation on the power source side, grid side, load side, and energy storage side, respectively. Since power generation, consumption, and storage are regular, such as low power consumption in the early morning and high power consumption during the day, the target monitoring value can be obtained by averaging the monitoring values ​​at the same time over multiple days. The following calculations can be performed using the target monitoring value, which can make the analysis process and judgment results more reliable.

[0049] The monitoring values ​​for a certain power sector over a certain number of days are obtained within a time window W. The average value is calculated to obtain the target monitoring value for that power sector within the time window W. Then, the target monitoring value for that power sector in each time window is obtained, and a power change function is established to show the change of the target monitoring value with the time window.

[0050] (2) Extract any two adjacent time windows, obtain the degree of deviation between the two adjacent time windows according to the power change function, determine whether to merge the time windows according to the degree of deviation, and then obtain several sub-time periods within a day according to the merged time windows.

[0051] Extract any two adjacent time windows, win1 and win2. Based on the power change function, obtain all target monitoring values ​​within time windows win1 and win2, and determine the degree of deviation between time windows win1 and win2. Among them, W n Let V1 be the weight of the nth type of power supply side, e be the natural constant, and V1 be the weight of the nth type of power supply side. n For the nth type of power side, V2 represents the target monitoring value within time window win1. nThe target monitoring value for the nth type of power supply within the time window win2;

[0052] If the deviation level X is less than the preset threshold, then time windows win1 and win2 are merged, and all time windows that meet the merging conditions are merged. Then, based on the merged time windows, several sub-time periods within a day are obtained.

[0053] Here, time windows with similar power conditions are merged, and based on these merged time windows, the day is divided into multiple sub-time periods. Then, by analyzing and judging each sub-time period separately, the computational resource consumption can be reduced. Formula: y=1-e -x When x > 0, y is a function of 0 to 1, and y increases as x increases. Since the larger the deviation between two time windows on a certain type of power side, the greater the degree of deviation between the two time windows, this formula can be used in the design of this scheme. Similarly, the design method of the power disturbance value below also refers to this principle. In this embodiment, the sum of the weights of all power sides is 1, that is, W1 + W2 + W3 + W4 = 1. The final deviation degree X is between 0 and 1. Here, the degree threshold is set to 0.3, that is, when the deviation degree X between time windows win1 and win2 is less than 0.3, time windows win1 and win2 are merged.

[0054] (3) Take the current date as the date to be set, and obtain the marking window corresponding to each sub-period within the date to be set based on the predicted temperature value of the date to be set and the corresponding power consumption area on the load side.

[0055] Since electricity usage is also seasonal and regional, meaning that electricity consumption follows patterns in different regions and temperatures—for example, higher electricity consumption in winter and summer, and lower consumption in autumn and winter—we can extract a marker window based on this pattern. The specific implementation steps are as follows:

[0056] The current date is used as the date to be set. The load side includes supplying power to several power consumption areas. Based on the weather forecast platform, the predicted temperature value of each power consumption area for each sub-period within the date to be set is obtained. Several time period windows corresponding to a certain sub-period D are extracted from historical dates. The actual temperature value TPW of each power consumption area in a certain time period window is obtained, as well as the predicted temperature value of each power consumption area in sub-period D. If the difference between the actual temperature value and the predicted temperature value corresponding to each power consumption area is less than a preset difference threshold, the time period window TPW is marked to obtain all marked windows corresponding to sub-period D. Thus, the marked windows corresponding to each sub-period within the date to be set are obtained.

[0057] (4) Retrieve the historical fault records of the integrated system and obtain the power change threshold based on the total power value at several moments before the fault occurred.

[0058] Retrieve historical fault records of the integrated system. These records are those from when the server detects and alarms when a fault occurs in the integrated system. Extract the total power value of the load at each moment within the time period TP prior to the fault occurrence. Extract any two adjacent moments t1 and t2 within the time period TP as a time combination C1. 2 Based on the total power values ​​pv1 and pv2 at times t1 and t2, the time combination C1 is obtained. 2 The power change value |pv1-pv2| is obtained, and the power change value of several time periods is combined. The average value is then used as the power change threshold pv0.

[0059] Since power outages or alarms are usually caused by irregular power usage, which is represented here by power, for example, if the power at a certain moment changes significantly compared to the previous moment, this sudden power change will disrupt the original power balance and stable operation of the power grid, leading to increased equipment stress and reduced system stability. In such cases, the probability of a power grid failure increases significantly. Therefore, this solution analyzes power to obtain power disturbance values. A larger power disturbance value indicates a significantly increased probability of a failure during that sub-period, and the data collection interval for monitoring that sub-period should be relatively short to better observe the power grid situation, detect anomalies promptly, and prevent larger power failures. Conversely, a smaller power disturbance value indicates a relatively lower probability of a failure during that sub-period, and the data collection interval for monitoring that sub-period should be relatively long to better conserve resources and avoid resource waste due to over-monitoring.

[0060] (5) Based on the total power and power change threshold at each time point within the marked window, the power disturbance value for each sub-period is obtained.

[0061] Obtain all marked windows corresponding to a certain sub-time period D, and extract any two adjacent times t3 and t4 within the marked windows as time combination C3. 4 If the total power values ​​pv3 and pv4 at times t3 and t4 satisfy |pv3-pv4|>pv0, then the time combination C3 4 The time slots are marked; then, from the N time slot combinations corresponding to sub-time slot D, the number of marked time slot combinations N0 is obtained, and the power disturbance value M of sub-time slot D is also obtained. D =1-e -N0 Then the power disturbance value for each sub-period is obtained.

[0062] Here, N time-time combinations are extracted from each sub-time period to make the calculation of power disturbance values ​​more reasonable.

[0063] (6) Obtain the resource usage values ​​of the server during historical operation at different data collection intervals, and obtain the minimum data collection interval based on the available resource usage values ​​of the server.

[0064] The integrated system deploys all power-side components on the same server and simultaneously collects and analyzes data. It obtains the server's resource usage values ​​when collecting data from the integrated system at different historical collection intervals, and fits a linear function to the change in resource usage values ​​over the collection intervals to obtain the variation function. The available resource usage value Z0 of the server is then obtained and substituted into the variation function to obtain the minimum data collection interval K. min .

[0065] (7) Preset the maximum time interval for data collection, and obtain the time interval for data collection for each sub-period within the date to be set based on the minimum time interval and the power disturbance value of each sub-period.

[0066] Obtain the power disturbance value for each sub-time period, and obtain the maximum disturbance value M. max and minimum disturbance value M min ; Use the current preset maximum data acquisition time interval as K max According to the minimum time interval K min And the power disturbance value M of a certain sub-period D D According to the formula The sampling time interval K of sub-time period D is obtained. D This allows us to obtain the time interval for data collection for each sub-period within the set date.

[0067] Maximum time interval K max This is the preset maximum time for collecting monitoring data; the following example is given: Here, if the maximum time interval K is set... max =10 seconds, minimum time interval K min =3 seconds, maximum disturbance value M max =0.9, minimum disturbance value M min =0.2, then if the power disturbance value M in sub-period D is 0.2, then... D =0.6, and based on the above formula, the data acquisition time interval K can be obtained. D =6 seconds, if the power disturbance value M of sub-period D D =0.4, and based on the above formula, the data acquisition time interval K can be obtained. D =8 seconds.

[0068] The integrated power generation, grid, load, and energy storage system is a new type of power system architecture that takes the safe and stable operation of the power system as its core. It achieves efficient operation of the entire power production-transmission-consumption-storage chain through deep collaboration and data interoperability among the power source side (generation), grid side (transmission), load side (consumption), and energy storage side (energy storage). Real-time monitoring of the integrated power generation, grid, load, and energy storage system is the core support for ensuring the safe, efficient, and low-carbon operation of the integrated system. Here, the sampling time interval for each sub-period is reasonably set according to the rule that the smaller the power disturbance value, the longer the sampling time interval, and the larger the power disturbance value, the shorter the sampling time interval. This balances the monitoring intensity and the resources occupied, and meets the needs of dynamic changes in grid operation.

[0069] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0070] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0071] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An intelligent control method of a source-network-load-storage integrated system, characterized in that, The method comprises the following steps: The historical monitoring records of the source network load storage integrated system are called, and the monitoring records of different power sides are extracted, including the power source side, the power grid side, the load side and the energy storage side; a plurality of time windows within a day are extracted, the monitoring values of the power sides are set according to the types of the power sides, the monitoring values of the power sides in the time windows are obtained, and the power change functions corresponding to each power side are established; Any two adjacent time windows are extracted, the deviation degree between the two adjacent time windows is obtained according to the power change functions, whether the time windows are merged is judged according to the deviation degree, and then a plurality of sub-periods within a day are obtained according to the merged time windows. The current date is taken as a to-be-set date, the corresponding marking window of each sub-period in the to-be-set date is obtained according to the predicted temperature value of the to-be-set date and the corresponding power consumption area of the load side; The historical fault records of the integrated system are called, the power change threshold value is obtained according to the total power value at a plurality of moments before the fault occurrence time, the total power at the moments in the marking window and the power change threshold value are obtained, and the power disturbance value of each sub-period is obtained. The resource occupation values of the server when running at different collection time intervals are obtained, the minimum time interval of data collection is obtained according to the available resource occupation value of the server; The maximum time interval of data collection is preset, and the time interval of data collection for each sub-period in the to-be-set date is obtained according to the minimum time interval and the power disturbance value of each sub-period. The power change function corresponding to each power side is established, comprising the following steps: The length of the time window is set as H, a plurality of time windows within a day are extracted, and a certain time window W is obtained; If the power side is the power source side, the monitoring value is set as the total power generation, and the total power generation of the power source side in the time window W is extracted; If the power side is the load side, the monitoring value is set as the total power consumption, and the total power consumption of the load side in the time window W is extracted; If the power side is the grid side, the monitoring value is set as the loss rate, and according to the total power generation F1 and the total power consumption F2 in the time window W, the loss rate of the grid side in the time window W is obtained as ; If the power side is the energy storage side, the monitoring value is set as the power change amount, the power change amount is positive when the power of the energy storage side increases, and the power change amount is negative when the power of the energy storage side decreases, and the power change amount of the energy storage side in the time window W is obtained; The monitoring values of a plurality of days corresponding to a certain power side in the time window W are obtained, the average value is obtained, the target monitoring value corresponding to the certain power side in the time window W is obtained, and then the target monitoring value corresponding to each time window of the certain power side is obtained, and the power change function of the target monitoring value changing with the time window is established. 2.The intelligent regulation method of the source-network-load integrated system according to claim 1, wherein, A plurality of sub-periods within a day are obtained, comprising the following steps: Extracting any two adjacent time windows win1 and win2, according to the power change function, all target monitoring values in the time windows win1 and win2 are obtained, and the deviation degree between the time windows win1 and win2 is obtained: Wherein, W n is the weight of the nth power side, e is a natural constant, V1 n is the target monitoring value of the nth power side in the time window win1, V2 n is the target monitoring value of the nth power side in the time window win2; If the deviation degree X is less than the preset degree threshold value, the time windows win1 and win2 are merged, all the time windows meeting the merging condition are merged, and then a plurality of sub-periods within a day are obtained according to the merged time windows. 3.The intelligent regulation method of the source-network-load integrated system according to claim 1, wherein, The marking window corresponding to each sub-period in the to-be-set date is obtained, comprising the following steps: The current date is taken as a to-be-set date, the load side includes power supply to several power consumption areas, according to a weather prediction platform, a predicted temperature value of each power consumption area in each sub-period in the to-be-set date is obtained; a plurality of time period windows corresponding to a certain sub-period D are extracted from historical dates, an actual temperature value of each power consumption area in a certain time period window TPW and a predicted temperature value of each power consumption area in the sub-period D are obtained, if the difference between the actual temperature value and the predicted temperature value corresponding to each power consumption area is less than a preset difference threshold, the time period window TPW is marked to obtain all marked windows corresponding to the sub-period D; and then the marked window corresponding to each sub-period in the to-be-set date is obtained. 4.The intelligent regulation method of the source-network-load integrated system according to claim 1, wherein, The power change threshold is obtained, including the following steps: The history fault record of the integrated system is called, the fault record is the record that the server monitors the fault of the integrated system and alarms, the total power value of the load end at each time in the period TP before the fault occurrence time is extracted; any two adjacent times t1 and t2 in the period TP are extracted as time combination C1 2 According to the total power values pv1 and pv2 of the times t1 and t2, the power change value |pv1-pv2| of the time combination C1 2 is obtained, the power change values of several time combinations are obtained, and the average value is taken as the power change threshold pv0. 5.The intelligent regulation method of the source-network-load integrated system according to claim 4, wherein, The power disturbance value of each sub-period is obtained, including the following steps: Obtain all the marked windows corresponding to a certain sub-period D, extract any two adjacent time points t3 and t4 in the marked window as a time combination C3 4 If the total power values pv3 and pv4 of the time points t3 and t4 satisfy |pv3-pv4|>pv0, the time combination C3 4 is marked; then from the N time combinations corresponding to the sub-period D, obtain the number of marked time combinations N0, and at the same time obtain the power disturbance value M D =1-e -N0 of the sub-period D; then obtain the power disturbance value of each sub-period. 6.The intelligent regulation method of the source-network-load integrated system according to claim 1, wherein, The minimum time interval of data acquisition is obtained, including the following steps: The integrated system is to deploy all power sides in the same server, and collect data for analysis at the same time; when acquiring history, the resource occupation value of the server is obtained when the integrated system is data collected according to different collection time intervals, and a first function fitting of the resource occupation value with the change of the collection time interval is carried out to obtain a change function; the available resource occupation value Z0 of the server is obtained, and substituted into the change function to obtain the minimum time interval K of data collection min . 7.The intelligent regulation method of the source-network-load integrated system according to claim 6, wherein, The time interval of data acquisition for each sub-period in the to-be-set date is obtained, including the following steps: acquiring the power disturbance value of each sub-period and obtaining the maximum disturbance value M max and the minimum disturbance value M min ; taking the maximum time interval of the preset data acquisition as K max , according to the minimum time interval K min and the power disturbance value M D of a certain sub-period D, obtaining the acquisition time interval K of the sub-period D according to the formula D , and further obtaining the time interval of data acquisition for each sub-period in the to-be-set date.

Images