A power distribution network operation optimization control method based on a multi-light storage integrated machine

Abstract

The application discloses a power distribution network operation optimization control method based on a multi-light storage integrated machine, relates to the technical field of energy scheduling, and improves the dynamic adaptive capability in the power distribution network control process.The application generates an expected power generation curve and a predicted power consumption curve according to historical power generation data and historical load fluctuation data, maps each item of historical environmental data and historical power generation data in a multi-dimensional coordinate system, then generates a power production causal chain from the historical environmental data and the historical power generation data through an environmental factor node and a power production node, and further obtains a power production causal network under each energy scheduling time point; at the beginning of each energy scheduling period, each item of real-time environmental data is matched with a corresponding power production causal dynamic network, the predicted power consumption of each power consumption area is obtained according to a predicted power consumption curve, then an energy deployment decision is generated through the predicted power consumption and the estimated power production of each photovoltaic power station, and the energy deployment decision is executed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of energy scheduling, in particular to a power distribution network operation optimization control method based on a multi-light storage integrated machine. BACKGROUND

[0002] With the rapid development and wide application of renewable energy, the power system is undergoing profound changes. Solar photovoltaic power generation has gradually become an important part of the power system due to its clean, renewable and distributed characteristics. However, the intermittency and instability of photovoltaic power generation have brought new challenges to the operation of the power distribution network. Traditional power distribution networks mainly rely on centralized power generation and fixed load characteristics for operation, while the connection of photovoltaic power generation makes the operation state of the power distribution network more complex and dynamic.

[0003] The existing power distribution network operation control technology has the following defects:

[0004] Single control strategy: Most of the existing multi-light storage integrated machine control methods use a single control strategy, such as local information-based or centralized control, which lacks comprehensive consideration of the overall operation state. This leads to unstable control effect in complex and variable power distribution network environment, making it difficult to achieve global optimization.

[0005] Lack of dynamic adaptability: Traditional control methods are often based on static models and fixed parameters, making it difficult to adapt to the dynamic changes of photovoltaic power generation and load in the power distribution network. The variability of the power distribution network operation environment requires a high degree of dynamic adaptability in the control method, but existing technologies are insufficient in this regard.

[0006] Therefore, how to improve the dynamic adaptability of the power distribution network control process while achieving comprehensive consideration of the overall operation state is a difficult point in the existing technology, and a power distribution network operation optimization control method based on a multi-light storage integrated machine is provided. SUMMARY

[0007] In order to solve the above technical problems, the purpose of the present application is to provide a power distribution network operation optimization control method based on a multi-light storage integrated machine.

[0008] In order to achieve the above purpose, the present application provides the following technical solutions:

[0009] A power distribution network operation optimization control method based on a multi-light storage integrated machine, comprising the following steps:

[0010] Step S1, set a number of energy scheduling periods with a data collection period of one year, and then collect historical environmental data and historical power generation data of photovoltaic power stations in each region in each energy scheduling period, and collect historical load fluctuation data of each power consumption area;

[0011] Step S2, a two-dimensional coordinate system is established, historical power generation data and historical load fluctuation data are respectively mapped in the two-dimensional coordinate system, and an energy scheduling period is divided into a plurality of energy scheduling time points, historical power generation data and historical load fluctuation data are split into a plurality of curve segments through the energy scheduling time points, expected power generation curves and expected power consumption curves are generated according to the plane area of each curve segment and the x-axis, and then expected power generation reference lines are generated according to the expected power generation curves;

[0012] Step S3, a multi-dimensional coordinate system, an environmental factor node and an electric quantity production node are set, historical environmental data and historical power generation data are mapped in the multi-dimensional coordinate system, and then data points at each energy scheduling time point are intercepted from the historical environmental data and the historical power generation data through the environmental factor node and the electric quantity production node, corresponding electric quantity production causal chains are generated, the environmental factor nodes and the electric quantity production nodes of the same type of data in the electric quantity production causal chains are compared, each electric quantity production causal chain is locally connected according to the comparison result, and then an electric quantity production causal network at each energy scheduling time point is obtained.

[0013] Step S4, at the beginning of each energy scheduling period, real-time environmental data of each photovoltaic power station is collected, the real-time environmental data is matched with the corresponding electric quantity production causal dynamic network, estimated production electric quantity is obtained according to the matching result, and expected power consumption is obtained according to the expected power consumption curve of each power consumption area, and then energy deployment decisions are generated through the expected power consumption and the estimated production electric quantity of each photovoltaic power station and executed.

[0014] Further, the collection process comprises:

[0015] A plurality of energy scheduling periods are set with years as a data collection period, a voltage sensor and i kinds of environmental perception sensors are set in each photovoltaic power station, and a power sensor is installed in each power consumption area, each sensor in the photovoltaic power station and the power consumption area is reset when each energy scheduling period starts, and then historical environmental data, historical power generation data and historical load fluctuation data in the corresponding energy scheduling period are collected.

[0016] Further, the acquisition process of the plane area of each curve segment and the x-axis comprises:

[0017] A two-dimensional coordinate system is established, historical power generation data and historical load fluctuation data of the same data type but different data collection periods in the same energy scheduling period are respectively mapped in the same two-dimensional coordinate system, a plurality of energy scheduling time points with equal time intervals are divided in each energy scheduling period, and curves on each two-dimensional coordinate system are split into a plurality of curve segments according to each energy scheduling time point.

[0018] The plane area values between each curve segment and the x-axis between each pair of energy scheduling time points are obtained, and then the plane area mean values of each pair of energy scheduling time points under different data collection periods are obtained in sequence according to the time sequence.

[0019] Further, the expected power generation curve and the expected power consumption curve establishment process comprises:

[0020] The plane area mean values between each curve segment and the x-axis between each pair of energy scheduling time points under different numbers of data collection periods are mapped on the same two-dimensional coordinate system, and a smooth curve is set, so that the smooth curve passes through each plane area mean value as much as possible in the corresponding interval between each energy scheduling time point.

[0021] The smooth curves of different data types are respectively mapped on the two-dimensional coordinate system where the historical power generation data or the historical load fluctuation data are located, and then all the curve segments below the smooth curve are removed, and the smooth curve corresponding to the retained curve segments is obtained according to the generation process of the smooth curve, and is respectively recorded as the expected power generation curve and the expected power consumption curve.

[0022] Further, the process of generating the expected power generation reference line according to the expected power generation curve comprises:

[0023] According to the historical power generation data corresponding to the retained curve segments, the associated historical environmental data are retrieved, a multi-dimensional coordinate system is established, and the historical power generation data corresponding to the retained curve segments and the associated historical environmental data are mapped in the same multi-dimensional coordinate system, and then a multi-dimensional linear regression equation is established with power generation as the dependent variable and various types of environmental data as the independent variable.

[0024] The process of establishing the expected power generation curve is collected to obtain the ideal state environmental curve corresponding to each environmental data, the ideal state environmental curve corresponding to each environmental data and the expected power generation curve are substituted into the multi-dimensional linear regression equation, the relationship parameters in the multi-dimensional linear regression equation are modified according to the substitution result, the straight line corresponding to the modified multi-dimensional linear regression equation is mapped in the same multi-dimensional coordinate system, and then the expected power generation reference line of each photovoltaic power station at different energy scheduling time periods under the ideal environmental state is obtained.

[0025] Further, the generation process of the power production causal chain comprises:

[0026] i environmental factor nodes and power production nodes are set, and the data names of various types of environmental data are respectively set for each environmental factor node.

[0027] Further, all data is extracted from the historical power station data set and the historical power consumption data set from each data collection period, and each item of historical environmental data and historical power generation data corresponding to the same power substation but different data collection periods and the same energy collection period is mapped in the same multi-dimensional coordinate system, and several energy scheduling time points are marked on the multi-dimensional coordinate system, and the expected power generation reference line is mapped in the multi-dimensional coordinate system.

[0028] Each item of historical environmental data and historical power generation data is framed by the environmental factor node and the power production node, and the environmental factor node and the power production node at each energy scheduling node are sequentially connected at the data point position at each energy scheduling time point, and then several power production causal chains are obtained.

[0029] Further, the establishment process of the power production causal network comprises:

[0030] A deviation threshold is set, and the values of each item of environmental data on the expected power generation reference line at each energy scheduling time point are set as the interval center to set the corresponding deviation detection interval;

[0031] The size relationship between the difference value between the values contained in the power production node in each power production causal chain and the values on the expected power generation reference line at the corresponding energy scheduling time point and the deviation threshold is judged;

[0032] According to the judgment result, it is further judged whether the difference value between the values contained in each environmental factor node in the corresponding power production causal chain and the values on the expected power generation reference line at the corresponding energy scheduling time point is within the deviation detection interval, and according to the judgment result, the corresponding environmental factor node is marked as an environmental impact node;

[0033] A value approximation threshold is set for each type of environmental data, and it is judged whether the absolute value of the difference between the data values recorded by the environmental impact nodes of the same data type in each power production causal chain at the same energy scheduling time point of the same photovoltaic power station is less than or equal to the corresponding value approximation threshold, and according to the judgment result, the consistent environmental impact nodes in each power production causal chain at the same energy scheduling time point of the same photovoltaic power station are locally connected, wherein the environmental impact nodes at the connection are directly overlapped, and then the power production causal network at each energy scheduling time point of each photovoltaic power station is obtained.

[0034] The power production causal networks at the same energy scheduling period are sequentially overlapped and spliced according to time, and then the power production causal dynamic network at the corresponding energy scheduling period is obtained.

[0035] Further, the generation process of the energy deployment decision comprises:

[0036] acquire the energy transmission efficiency between each photovoltaic power station and each power consumption area, and then acquire the expected power consumption of each power consumption area between each energy scheduling time point according to the expected power consumption curve of each power consumption area when each energy scheduling time period starts;

[0037] Meanwhile, each time an energy scheduling time point starts, collect each real-time environmental data of each photovoltaic power generation area, and according to the current energy scheduling time period and energy scheduling time point, separate the corresponding power production causal network from the power production causal dynamic network;

[0038] First, judge whether each real-time environmental data is within the deviation detection interval of the corresponding type of environmental data, and according to the judgment result, match each real-time environmental data with the environmental factor node to acquire the value fit degree of the real-time environmental data and the data contained in the environmental factor node;

[0039] According to the matching result of each real-time environmental data and the environmental factor node, select the actual power production chain from the power production causal network, and record the value of the power production node in the actual power production chain as the corresponding photovoltaic power station, and the estimated production power before the next energy scheduling time point;

[0040] According to the estimated production power and the energy transmission efficiency between each photovoltaic power station and each power consumption area, collect the energy allocation decision generated by the optimization algorithm and execute it.

[0041] Compared with the prior art, the beneficial effects of the present application are:

[0042] The present application generates the expected power generation curve and the expected power consumption curve according to the historical power generation data and the historical load fluctuation data, and maps each historical environmental data and historical power generation data in a multi-dimensional coordinate system, and then generates the power production causal chain from the historical environmental data and the historical power generation data through the environmental factor node and the power production node, and then obtains the power production causal network at each energy scheduling time point. At the beginning of each energy scheduling time period, match each real-time environmental data with the corresponding power production causal dynamic network, and obtain the expected power consumption according to the expected power consumption curve of each power consumption area, and then generate the energy allocation decision through the expected power consumption and the estimated production power of each photovoltaic power station and execute it, which realizes improving the dynamic adaptability in the power distribution network control process while comprehensively considering the global operation state. BRIEF DESCRIPTION OF DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments or the prior art, the following will briefly introduce the drawings needed in the embodiments. Obviously, the drawings described below are only some embodiments of the present application.

[0044] Figure 1A method flowchart of the present application. DETAILED DESCRIPTION

[0045] In order to make the objectives, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described in detail below. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0046] As shown in Figure 1 A power distribution network operation optimization control method based on a multi-optical storage all-in-one machine, comprising the following steps:

[0047] Step S1, set a plurality of energy scheduling periods with a year as the data collection period, and then collect the historical environmental data and historical power generation data of each regional photovoltaic power station in each energy scheduling period, and collect the historical load fluctuation data of each power consumption area;

[0048] The specific implementation process of step S1 includes:

[0049] Set m energy scheduling periods with a year as the data collection period, it should be noted that the time span of each energy scheduling period is not completely the same, that is, there is an energy scheduling period whose time length is one month, and there is an energy scheduling period whose time length is two months, wherein m is a natural number greater than 0;

[0050] The photovoltaic power station is composed of a power generation unit and an energy storage unit, a voltage sensor, a GPS positioning device and i kinds of environmental perception sensors are arranged at each photovoltaic power station, and a power sensor is installed at each power consumption area, wherein the types of environmental sensors include temperature sensors, light sensors, humidity sensors, etc., and i is a natural number greater than 0;

[0051] Each time an energy scheduling period starts, all sensors in the photovoltaic power station and the power consumption area are reset, and then the historical environmental data, historical power generation data and historical load fluctuation data in the corresponding energy scheduling period are collected, wherein the historical environmental data includes historical humidity change data, historical light change data, etc. It should be noted that the above data is collected under the condition that there is no hardware device damage in the photovoltaic power station and the power consumption area;

[0052] When the energy scheduling period ends, all the data generated by each photovoltaic power station and power consumption area is summarized to obtain the historical power station data set and the historical power consumption data set in the corresponding energy scheduling period.

[0053] Step S2, a two-dimensional coordinate system is established, historical power generation data and historical load fluctuation data are respectively mapped in the two-dimensional coordinate system, and an energy scheduling period is divided into a plurality of energy scheduling time points, the historical power generation data and the historical load fluctuation data are split into a plurality of curve segments through the energy scheduling time points, expected power generation curves and expected power consumption curves are generated according to the plane area of each curve segment and the x-axis, and then expected power generation reference lines are generated according to the expected power generation curves;

[0054] The specific implementation process of the step S2 includes:

[0055] A two-dimensional coordinate system is established, and historical power generation data and historical load fluctuation data of the same data type in historical power station data sets and historical power consumption data sets of different data collection periods but the same energy scheduling period are respectively mapped in the same two-dimensional coordinate system, wherein the x-axis of the two-dimensional coordinate system is associated with time, and the y-axis is associated with the numerical unit of the corresponding type of data.

[0056] A plurality of energy scheduling time points with equal time intervals are divided in each energy scheduling period, and the curves on each two-dimensional coordinate system are split into a plurality of curve segments according to each energy scheduling time point.

[0057] The plane area values between each curve segment and the x-axis within each pair of energy scheduling time points are obtained, and then the plane area mean values of each pair of energy scheduling time points under different numbers of data collection periods are obtained in time sequence, the specific process includes: first, the plane area mean values between each curve segment and the x-axis within each pair of energy scheduling time points under the first data collection period and the second data collection period are obtained, then the plane area mean values between each curve segment and the x-axis within each pair of energy scheduling time points under the first data collection period, the second data collection period and the third data collection period are obtained, and so on, until the plane area mean values between each curve segment and the x-axis within each pair of energy scheduling time points under all existing data collection periods are obtained.

[0058] Further, the plane area mean values between each curve segment and the x-axis within each pair of energy scheduling time points under different numbers of data collection periods are mapped on the same two-dimensional coordinate system, and a smooth curve is set, so that the smooth curve passes through as many plane area mean values as possible in the corresponding interval between each energy scheduling time point.

[0059] The smooth curves of different data types are respectively mapped on the two-dimensional coordinate systems where the historical power generation data or the historical load fluctuation data are located, and then all the curve segments below the smooth curves are removed, and then the smooth curves corresponding to the retained curve segments are obtained according to the generation process of the smooth curves, and are respectively recorded as expected power generation curves and expected power consumption curves.

[0060] Meanwhile, according to the historical power generation data corresponding to the reserved curve segment, the associated historical environmental data is called, a multi-dimensional coordinate system is established, and the historical power generation data corresponding to the reserved curve segment and the historical environmental data corresponding to the called are mapped in the same multi-dimensional coordinate system, and then a multi-dimensional linear regression equation is established with power generation as the dependent variable and various environmental data as the independent variable, wherein the multi-dimensional linear regression equation can be expressed as: U = a1x1 + … + anxn + e n x n , wherein U represents power generation, x n and a n represent the nth environmental data and the corresponding relationship parameter, respectively, and n is a natural number greater than 0;

[0061] In the process of establishing the expected power generation curve, the ideal state environmental curve corresponding to each environmental data is obtained, and the ideal state environmental curve corresponding to each environmental data and the expected power generation curve are substituted into the multi-dimensional linear regression equation, and each relationship parameter in the multi-dimensional linear regression equation is modified according to the substitution result.

[0062] The straight line corresponding to the modified multi-dimensional linear regression equation is mapped in the same multi-dimensional coordinate system, and then the expected power generation reference line of each photovoltaic power station at different energy scheduling time periods under the ideal environmental state is obtained.

[0063] Step S3, set up a multi-dimensional coordinate system, an environmental factor node and a power production node, map each historical environmental data and historical power generation data in the multi-dimensional coordinate system, and then cut the data point position at each energy scheduling time point from the historical environmental data and the historical power generation data through the environmental factor node and the power production node, generate the corresponding power production causal chain, compare the environmental factor node and the power production node of the same type data in the power production causal chain, modify each power production causal chain according to the comparison result, and then obtain the power production causal network at each energy scheduling time point;

[0064] The specific implementation process of step S3 includes:

[0065] Set up i environmental factor nodes and power production nodes, and set up corresponding data names for each environmental factor node according to the data names of various types of environmental data;

[0066] Then all the data are extracted from the historical power station data set and the historical power consumption data set in each data collection period, the historical environmental data and the historical power generation data corresponding to the same power substation but different data collection periods and the same energy collection period are mapped in the same multi-dimensional coordinate system, and the expected power generation reference line is mapped in the multi-dimensional coordinate system.

[0067] Each item of historical environmental data and historical power generation data is framed by the environmental factor node and the power generation node, and the data points are connected in sequence at each energy scheduling time point, so as to obtain a plurality of power generation causal chains;

[0068] A deviation threshold is set, and the values of each item of environmental data on the expected power generation reference line at each energy scheduling time point are set as the interval center to set the corresponding deviation detection interval;

[0069] Then, the difference between the values contained in the power generation node in each power generation causal chain and the values on the expected power generation reference line at the corresponding energy scheduling time point is judged in relation to the size of the deviation threshold;

[0070] If the difference is greater than or equal to the deviation threshold, the corresponding power generation causal chain is deleted;

[0071] If the difference is less than the deviation threshold, it is further judged whether the difference between the values contained in each environmental factor node in the corresponding power generation causal chain and the values on the expected power generation reference line at the corresponding energy scheduling time point is within the corresponding deviation detection interval;

[0072] It should be noted that when the power generation of a photovoltaic power station in a period of time is greater than or equal to the power generation given by the expected power generation reference line at the corresponding time, each item of environmental data at the corresponding time is within the corresponding deviation detection interval;

[0073] Then, if the difference is within the corresponding deviation detection interval, no operation is performed on the corresponding environmental factor node;

[0074] If the difference is not within the corresponding deviation detection interval, the corresponding environmental factor node in the power generation causal chain is marked as an environmental impact node;

[0075] A value approximation threshold is set for each type of environmental data, and it is judged whether the absolute value of the difference between the data values recorded by the environmental impact nodes of the same data type in each power generation causal chain at the same energy scheduling time point of the same photovoltaic power station is less than or equal to the corresponding value approximation threshold;

[0076] If the absolute value of the difference is greater than the corresponding value approximation threshold, no operation is performed;

[0077] If the absolute value of the difference is less than or equal to the corresponding value approximation threshold, the corresponding environmental impact nodes are consistent;

[0078] Further, according to the judgment result, the same photovoltaic power station is connected with each other in the same energy scheduling time point, and the same environmental impact node in each power production causal chain is connected, and the environmental impact node at the connection is directly overlapped, and then the power production causal network of each photovoltaic power station in each energy scheduling time point is obtained.

[0079] The power production causal network in the same energy scheduling period is sequentially overlapped and spliced according to time, and then the power production causal dynamic network corresponding to the energy scheduling period is obtained.

[0080] Step S4, at the beginning of each energy scheduling period, collecting each real-time environmental data of each photovoltaic power station, matching each real-time environmental data with the corresponding power production causal dynamic network, obtaining the estimated production power according to the matching result, and obtaining the estimated power consumption according to the estimated power consumption curve of each power consumption area, and then generating an energy allocation decision according to the estimated power consumption of each photovoltaic power station and the estimated power consumption, and executing the energy allocation decision.

[0081] The specific implementation process of the step S4 includes:

[0082] Obtaining the energy transmission efficiency between each photovoltaic power station and each power consumption area, and then when an energy scheduling period starts, obtaining the estimated power consumption of each power consumption area between each energy scheduling time point according to the estimated power consumption curve of each power consumption area;

[0083] At the same time, when an energy scheduling time point starts, collecting each real-time environmental data of each photovoltaic power area, and according to the current energy scheduling period and the energy scheduling time point, the corresponding power production causal network is separated from the power production causal dynamic network;

[0084] First, it is judged whether the real-time environmental data is in the deviation detection interval of the corresponding type of environmental data. If it is, the corresponding real-time environmental data is matched with the environmental factor node in the power production causal network without environmental impact node mark and with the same type of data;

[0085] If not, the corresponding real-time environmental data is matched with the environmental factor node in the power production causal network with the environmental impact node mark and with the same type of data;

[0086] In the process of matching each real-time environmental data with the environmental factor node, the numerical fitness degree of the real-time environmental data and the data contained in the environmental factor node is obtained, wherein the numerical fitness degree is the numerical value obtained by dividing the numerical value of the real-time environmental data by the numerical value of the data contained in the environmental factor node;

[0087] According to the matching results of the real-time environmental data and the environmental factor nodes, a number of preliminary power generation chains are traversed from the power production causal network, and a preliminary power generation chain with the largest numerical agreement degree sum is selected and recorded as an actual power generation chain;

[0088] Further, the numerical values recorded by the power generation nodes in the actual power generation chain are recorded as the estimated production power of the corresponding photovoltaic power station before the next energy scheduling time point;

[0089] Further, according to the estimated production power of the photovoltaic power station before the next energy scheduling time point and the energy transmission efficiency between each photovoltaic power station and each power consumption area, an energy allocation decision is generated by using an optimization algorithm, so that the energy allocation decision meets the condition that the allocated power of each power consumption area is greater than or equal to the expected power consumption of each power consumption area before the next energy scheduling time point.

[0090] Each photovoltaic power station executes the energy allocation decision, and the energy allocation decision is repeatedly generated and executed until the next energy allocation time point starts.

[0091] The above embodiments are only used to illustrate the technical method of the present application and not to limit it. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical method of the present application.

Claims

1. A power distribution network operation optimization control method based on a multi-optical storage all-in-one machine, characterized in that, The method comprises the following steps: Step S1, setting a plurality of energy scheduling periods with a data collection period of years, and then collecting historical environmental data and historical power generation data of each regional photovoltaic power station in each energy scheduling period, and collecting historical load fluctuation data of each power consumption region; Step S2, establishing a two-dimensional coordinate system, mapping the historical power generation data and the historical load fluctuation data in the two-dimensional coordinate system, and dividing the energy scheduling period into a plurality of energy scheduling time points, and splitting the historical power generation data and the historical load fluctuation data into a plurality of curve segments through the energy scheduling time points, generating an expected power generation curve and a predicted power consumption curve according to the plane area of each curve segment and the x-axis, and then generating an expected power generation reference line according to the expected power generation curve; Step S3, setting a multi-dimensional coordinate system, an environmental factor node and an electric quantity production node, mapping each item of historical environmental data and historical power generation data in the multi-dimensional coordinate system, and then intercepting data points at each energy scheduling time point from the historical environmental data and the historical power generation data through the environmental factor node and the electric quantity production node, generating a corresponding electric quantity production causal chain, comparing the environmental factor node and the electric quantity production node of the same type of data in the electric quantity production causal chain, and locally connecting each electric quantity production causal chain according to the comparison result, and then obtaining an electric quantity production causal network at each energy scheduling time point; Step S4, at the beginning of each energy scheduling period, collecting each item of real-time environmental data of each photovoltaic power station, matching each item of real-time environmental data with the corresponding electric quantity production causal dynamic network, obtaining an estimated production electric quantity according to the matching result, and obtaining a predicted power consumption according to the predicted power consumption curve of each power consumption region, and then generating an energy allocation decision according to the predicted power consumption and the estimated production electric quantity of each photovoltaic power station and executing the energy allocation decision; The generation process of the electric quantity production causal chain comprises: Setting i environmental factor nodes and electric quantity production nodes, mapping each item of historical environmental data and historical power generation data corresponding to the same substation but different data collection periods and the same energy scheduling period in the same multi-dimensional coordinate system, and labeling a plurality of energy scheduling time points on the multi-dimensional coordinate system, and mapping the expected power generation reference line in the multi-dimensional coordinate system; Through the environmental factor node and the electric quantity production node, the data points of each item of historical environmental data and historical power generation data at each energy scheduling time point are framed out, the environmental factor nodes and the electric quantity production nodes at each energy scheduling node are connected in turn, and then a plurality of electric quantity production causal chains are obtained.

2. The power distribution network operation optimization control method based on the multi-optical storage all-in-one machine according to claim 1, characterized in that, The collection process comprises: Setting a plurality of energy scheduling periods with a data collection period of years, setting voltage sensors and i kinds of environmental perception sensors at each photovoltaic power station, and installing power sensors at each power consumption region, and resetting all sensors in the photovoltaic power station and the power consumption region when each energy scheduling period starts, and then collecting each item of historical environmental data, historical power generation data and historical load fluctuation data in the corresponding energy scheduling period, and i is a natural number greater than 0.

3. The power distribution network operation optimization control method based on the multi-optical storage all-in-one machine according to claim 2, characterized in that, The acquisition process of the plane area of each curve segment and the x-axis comprises: A two-dimensional coordinate system is established, and historical power generation data and historical load fluctuation data of the same data type in the same energy scheduling period but different data collection periods are respectively mapped in the same two-dimensional coordinate system. A plurality of energy scheduling time points are divided in each energy scheduling period, and the curves in the two-dimensional coordinate system are divided into a plurality of curve segments according to the energy scheduling time points. The plane area values between each curve segment and the x-axis between each pair of energy scheduling time points are obtained, and then the plane area mean values of each pair of energy scheduling time points under different numbers of data collection periods are obtained in time sequence.

4. The power distribution network operation optimization control method based on the multi-optical storage all-in-one machine according to claim 3, characterized in that, The expected power generation curve and the predicted power consumption curve establishment process comprises: The plane area mean values between each curve segment and the x-axis between each pair of energy scheduling time points under different numbers of data collection periods are mapped in the same two-dimensional coordinate system, a smooth curve is set, the smooth curves of different data types are respectively mapped in the two-dimensional coordinate system of the historical power generation data or the historical load fluctuation data, and then all the curve segments below the smooth curve are removed. According to the generation process of the smooth curve, the smooth curve corresponding to the retained curve segments is obtained, and is respectively recorded as the expected power generation curve and the predicted power consumption curve.

5. The power distribution network operation optimization control method based on the multi-optical storage all-in-one machine according to claim 4, characterized in that, The process of generating the expected power generation reference line according to the expected power generation curve comprises: A multi-dimensional coordinate system is established, and the historical power generation data corresponding to the retained curve segments and the historical environment data corresponding to the historical environment data are mapped in the same multi-dimensional coordinate system. A multi-dimensional linear regression equation is established with power generation as the dependent variable and various types of environment data as the independent variable. The ideal state environment curve corresponding to each item of environment data is obtained, and the relationship parameters in the multi-dimensional linear regression equation are modified according to the ideal state environment curve corresponding to each item of environment data and the expected power generation curve. The straight line corresponding to the modified multi-dimensional linear regression equation is mapped in the same multi-dimensional coordinate system, and the expected power generation reference line of each photovoltaic power station in different energy scheduling periods under the ideal environment state is obtained.

6. The power distribution network operation optimization control method based on the multi-optical storage all-in-one machine according to claim 5, characterized in that, The establishment process of the power production causal network comprises: A deviation threshold is set, and the values of each item of environment data on the expected power generation reference line at each energy scheduling time point are set as the interval center, and the corresponding deviation detection interval is set. The relationship between the difference between the values contained in each power production node in each power production causal chain and the values on the expected power generation reference line at the corresponding energy scheduling time point and the deviation threshold is determined. According to the judgment result, further judge whether the difference between the values contained in each environmental factor node in the corresponding power generation causal chain and the values of the expected power generation reference line at the corresponding energy scheduling time point is located in the related deviation detection interval, according to the judgment result, mark the corresponding environmental factor node as an environmental impact node, set a value approximation threshold for various types of environmental data, and judge whether the absolute value of the difference between the data values recorded by the environmental impact nodes of the same data type in each power generation causal chain at the same energy scheduling time point of the same photovoltaic power station is less than or equal to the corresponding value approximation threshold, according to the judgment result, locally connect each power generation causal chain, wherein the environmental impact nodes at the connection place directly overlap, and then obtain the power generation causal network of each photovoltaic power station at each energy scheduling time point.

7. The power distribution network operation optimization control method based on the multi-optical storage all-in-one machine according to claim 6, characterized in that, The generation process of the energy allocation decision includes: Obtain the energy transmission efficiency between each photovoltaic power station and each power consumption area, and then when a energy scheduling period starts, obtain the expected power consumption of each power consumption area between each energy scheduling time point according to the expected power consumption curve of each power consumption area; At the same time, when an energy scheduling time point starts, collect various real-time environmental data of each photovoltaic power generation area, and according to the current energy scheduling period and energy scheduling time point, extract the corresponding power generation causal network from the power generation causal dynamic network; Judge whether each real-time environmental data is within the deviation detection interval of the corresponding type of environmental data, and according to the judgment result, match each real-time environmental data with the environmental factor node to obtain the value fit degree of the real-time environmental data and the data contained in the environmental factor node; According to the matching result of each real-time environmental data and the environmental factor node, select the actual power generation chain from the power generation causal network, and record the value recorded by the power generation node in the actual power generation chain as the estimated power generation of the corresponding photovoltaic power station before the next energy scheduling time point; According to the estimated power generation and the energy transmission efficiency between each photovoltaic power station and each power consumption area, an optimization algorithm is used to generate an energy allocation decision and execute it.

Images