Photovoltaic system regulation and control method of micro-grid

Abstract

The invention discloses a photovoltaic system regulation and control method of a micro-grid, and belongs to the technical field of photovoltaic systems. Comprising the following steps: S1, acquiring predicted power of all photovoltaic systems in a micro-grid at the current moment; S2, setting a power fluctuation threshold value of each photovoltaic system; S3, calculating a charge state of each energy storage device; S4, dividing the energy storage devices in the micro-grid into quick response type energy storage devices and long-period adjustment type energy storage devices according to a response time standard; s6, enabling the quick response type energy storage equipment to discharge to each photovoltaic system, or enabling the photovoltaic system to charge each quick response type energy storage equipment, or maintaining the operation of the quick response type energy storage equipment; and S7, enabling the long-period adjusting type energy storage equipment to discharge to each photovoltaic system, or enabling the photovoltaic system to charge each long-period adjusting type energy storage equipment, or maintaining operation. According to the method, the power fluctuation of the micro-grid can be more reasonably stabilized.

Description

Technical Field

[0001] The present invention belongs to the technical field of photovoltaic systems, and particularly relates to a method for regulating and controlling a photovoltaic system of a microgrid. Background Art

[0002] A microgrid refers to a small power network composed of distributed power sources (such as solar panels, wind turbines), energy storage systems, loads, and monitoring and protection devices.

[0003] Since the output power of a photovoltaic power generation system highly depends on weather conditions (such as light intensity, cloud cover, etc.), its output has significant intermittency and unpredictability. Therefore, when a large number of photovoltaic systems are connected to the grid, it will cause severe fluctuations in the power flow in the grid, and further pose a threat to the grid frequency and voltage stability.

[0004] Among them, the configuration of energy storage devices can suppress the power fluctuations of the photovoltaic system and balance the power of the microgrid. The current difficulties are as follows: 1. There is a lack of a suitable model for predicting photovoltaic power; 2. A suitable power fluctuation threshold is not established in combination with the parameters of the photovoltaic system. As a result, the effect of suppressing the power fluctuations of the photovoltaic system may not be obvious. Summary of the Invention

[0005] In order to solve the above problems, the present invention proposes a method for regulating and controlling a photovoltaic system of a microgrid.

[0006] In order to achieve the above object, the present invention proposes the following technical solutions:

[0007] A method for regulating and controlling a photovoltaic system of a microgrid, comprising the following steps:

[0008] S1: Obtain the predicted power of all photovoltaic systems in the microgrid at the current moment;

[0009] S2: Set the power fluctuation threshold for each photovoltaic system; the power fluctuation threshold includes a maximum fluctuation threshold, a minimum fluctuation threshold, and a standard fluctuation threshold;

[0010] S3: Obtain the number of energy storage devices in the microgrid, number the energy storage devices; calculate the state of charge of each energy storage device;

[0011] S4: Classify the energy storage devices in the microgrid into fast-response energy storage devices and long-cycle regulation energy storage devices according to the response time standard;

[0012] S5: Calculate the net load demand of each photovoltaic system at the current moment t;

[0013] S6: Monitor the real-time power fluctuations of the PV system, compare the real-time power fluctuations of the PV system with the set standard fluctuation threshold, and according to the comparison result, let the fast-response energy storage device discharge to each PV system, or let the PV system charge each fast-response energy storage device, or maintain the operation.

[0014] S7: Compare the static load demand of each PV system with the set load threshold, and according to the comparison result, let the long-cycle regulation energy storage device discharge to each PV system, or let the PV system charge each long-cycle regulation energy storage device, or maintain the operation.

[0015] Further, step S1 includes:

[0016] S1.1: Number all the PV systems in the microgrid; obtain the historical multi-characteristics of each PV system, and convert the historical multi-characteristics of each PV system into the corresponding feature matrix.

[0017] Assume that there are N PV systems in the microgrid, the numbers of the PV systems are 1 to N, and the expression of the historical multi-characteristics of the PV system numbered i (i ∈ [1, N]) is:

[0018]

[0019] X i =[ESD i , AQI i , TH i , DI i T (1)

[0020] In formula (1), ESD i represents the effective sunshine duration at the location of the PV system numbered i; Tdh i represents the total sunshine hours at the location of the PV system numbered i; CCF i represents the cloud cover factor at the location of the PV system numbered i; AQI i represents the optimized air quality index at the location of the PV system numbered i; AQ i represents the original air quality index at the location of the PV system numbered i; Hu i represents the air humidity percentage at the location of the PV system numbered i; TH i represents the interaction effect of temperature and humidity at the location of the PV system numbered i; Th i represents the ambient temperature at the location of the PV system numbered i; DI i represents the dust deposition effect at the location of the PV system numbered i; W i represents the wind speed at the location of the PV system numbered i; D i ​Indicates the dust concentration at the location of the photovoltaic system numbered i; X i Indicates the feature matrix of the photovoltaic system numbered i;

[0021] S1.2: Obtain the observed power generation of each photovoltaic system at the current moment, and then combine the feature matrices of each photovoltaic system to construct a power prediction model and calculate the predicted power of each photovoltaic system at the current moment;

[0022] The power prediction model is:

[0023] P i (t) = W p ·RNN(h N-1 , [CNN(X i ) ; S i (t)]) + b p (2)

[0024] In formula (2), P i (t) represents the predicted power of the photovoltaic system numbered i at the current moment t; W p is the weight matrix; RNN represents the recurrent neural network; h N-1 represents the state of the recurrent neural network at time step N - 1; CNN(X i ) represents processing the feature matrix X numbered i i using a convolutional neural network; S i (t) represents the observed power generation of the photovoltaic system numbered i at the current moment t; b p is the bias vector.

[0025] Furthermore, in step S2, the expressions for the maximum fluctuation threshold, minimum fluctuation threshold, and standard fluctuation threshold are respectively:

[0026] The formula for the minimum power fluctuation threshold of the photovoltaic system numbered i is:

[0027]

[0028] The formula for the maximum power fluctuation threshold of the photovoltaic system numbered i is:

[0029]

[0030] The formula for the standard power fluctuation threshold of the photovoltaic system numbered i is:

[0031]

[0032] In formulas (3), (4), and (5), α represents the coefficient of the geographical feature vector; β represents the coefficient of the device type feature vector; γ represents the coefficient of the historical data feature vector; Represents the minimum geographical feature vector of the photovoltaic system numbered i; Represents the minimum equipment type feature vector of the photovoltaic system numbered i; Represents the minimum historical data feature vector of the photovoltaic system numbered i; D i Represents the dust concentration at the location of the photovoltaic system numbered i; D max Represents the maximum dust concentration at the location of the photovoltaic system numbered i; P i max Represents the maximum charging power of the photovoltaic system numbered i; Represents the historical minimum load of the photovoltaic system numbered i; Represents the historical negative fluctuation standard deviation of the photovoltaic system numbered i; Represents the charging efficiency of the photovoltaic system numbered i; Represents the maximum geographical feature vector of the photovoltaic system numbered i; Represents the maximum historical data feature vector of the photovoltaic system numbered i; Represents the maximum equipment type feature vector of the photovoltaic system numbered i; W i Represents the wind speed at the location of the photovoltaic system numbered i; Wmax represents the maximum wind speed at the location of the photovoltaic system numbered i; Represents the maximum discharge power of the photovoltaic system numbered i; Represents the discharge efficiency of the photovoltaic system numbered i; Th max Represents the maximum ambient temperature at the location of the photovoltaic system numbered i; Th min Represents the minimum ambient temperature at the location of the photovoltaic system numbered i; Represents the historical highest load of the photovoltaic system numbered i; Represents the historical positive fluctuation standard deviation of the photovoltaic system numbered i; Represents the standard geographical feature vector of the photovoltaic system numbered i; Represents the standard equipment type feature vector of the photovoltaic system numbered i; Represents the standard historical data feature vector of the photovoltaic system numbered i; Represents the historical average load of the photovoltaic system numbered i; Represents the historical average fluctuation standard deviation of the photovoltaic system numbered i.

[0033] Furthermore, in steps S3 and S4, it is assumed that there are M energy storage devices in the microgrid, numbered from 1 to M, including X fast-response energy storage devices and Y long-cycle regulation energy storage devices;

[0034] The state of charge formula for the energy storage device numbered k (k ∈ [1, M]) at the next moment t + 1 is:

[0035]

[0036] In formula (6), SOC k (t + 1) represents the state of charge of the energy storage device numbered k at the next moment t + 1;

[0037] SOC k (t) represents the state of charge of the energy storage device numbered k at the current moment t; represents the charging efficiency of the energy storage device numbered k at moment t; represents the charging power of the energy storage device numbered k at moment t; Δt represents the time interval; represents the discharging efficiency of the energy storage device numbered k at moment t; represents the discharging power of the energy storage device numbered k at moment t.

[0038] 5. A photovoltaic system regulation method for a microgrid according to claim 4, wherein the formula for the net load demand of the photovoltaic system numbered i is:

[0039] P i v (t) = P l (t) - P i (t)(7)

[0040] In formula (7), P l (t) represents the power demand of the user for the photovoltaic system numbered i at moment t; P i v (t) represents the net load demand of the user for the photovoltaic system numbered i at moment t; P i (t) represents the predicted power of the photovoltaic system numbered i at moment t.

[0041] Furthermore, step S6 includes the following steps:

[0042] S6.1: Let the real-time power fluctuation of the monitored photovoltaic system numbered i be Compare the real-time power fluctuation of the photovoltaic system numbered i with its standard power fluctuation threshold According to the comparison result, let the fast-response energy storage device discharge to each photovoltaic system, or the photovoltaic system charge each fast-response energy storage device, or maintain operation;

[0043] S6.2: If then it is case one, and the fast-response energy storage device needs to discharge to the photovoltaic system;

[0044] Step S6.2 further includes the following steps:

[0045] S6.2.1: Calculate the SOC values of all fast-response energy storage devices, sort and number them in descending order of the SOC values, and discharge them to the photovoltaic system in turn in descending order of the SOC values;

[0046] S6.2.2: The discharge power of the fast-response energy storage device numbered j (j ∈ [1, X]) to the photovoltaic system numbered i is:

[0047]

[0048] In the formula, represents the discharge power of the fast-response energy storage device numbered j to the photovoltaic system numbered i; is the maximum power fluctuation threshold of the photovoltaic system numbered i; P md,j is the maximum discharge power of the fast-response energy storage device numbered j when discharging to the photovoltaic system numbered i; min{*, *} represents taking the minimum value of the two;

[0049] During the discharge process of the fast-response energy storage device numbered j to the photovoltaic system numbered i, the SOC value of the fast-response energy storage device numbered j is calculated in real time. If the SOC value of this fast-response energy storage device is lower than 20% of its total state of charge, it stops discharging to the photovoltaic system numbered i, and the fast-response energy storage device numbered (j + 1) continues to discharge to the photovoltaic system numbered i; or at the same time, during the discharge process of the fast-response energy storage device numbered j to the photovoltaic system numbered i, the net load demand P i v (t) of the photovoltaic system numbered i is calculated in real time. When the net load demand of the photovoltaic system numbered i satisfies it also stops discharging to the photovoltaic system numbered i, and the fast-response energy storage device numbered (j + 1) continues to discharge to the photovoltaic system numbered i;

[0050] S6.3: If then it is Case 2, and the photovoltaic system needs to charge the fast-response energy storage device;

[0051] Step S6.3 further includes the following steps:

[0052] S6.3.1: Calculate the SOC values of all fast-response energy storage devices, sort and number them in ascending order of the SOC values, and the photovoltaic system charges each fast-response energy storage device in turn in ascending order of the SOC values;

[0053] S6.3.2: The charging power of the photovoltaic system numbered i to the fast-response energy storage device numbered g is:

[0054]

[0055] Wherein, represents the charging power of the photovoltaic system numbered i to the fast-response energy storage device numbered g; P mc,g is the maximum charging power of the photovoltaic system numbered i to the fast-response energy storage device numbered g;

[0056] During the charging process of the photovoltaic system numbered i to the fast-response energy storage device numbered g, calculate the SOC value of the fast-response energy storage device numbered g; when the SOC value of the fast-response energy storage device is higher than 90% of its total state of charge, the photovoltaic system numbered i stops charging the fast-response energy storage device numbered g, and the photovoltaic system numbered i continues to charge the fast-response energy storage device numbered (g + 1); or at the same time, during the charging process of the photovoltaic system numbered i to the fast-response energy storage device numbered g, calculate the net load demand P i v (t) of the photovoltaic system numbered i. When the net load demand of the photovoltaic system numbered i meets the photovoltaic system numbered i stops charging the fast-response energy storage device numbered g, and the photovoltaic system numbered i continues to charge the fast-response energy storage device numbered (g + 1);

[0057] S6.4: When maintain the current operation.

[0058] Furthermore, step S7 includes:

[0059] S7.1: Set the load threshold to 0; compare the net load demand P i v (t) of the photovoltaic system numbered i with the load threshold, and according to the result, let the long-cycle regulation energy storage device discharge to each photovoltaic system, or the photovoltaic system charge each long-cycle regulation energy storage device, or maintain the operation;

[0060] S7.2: If P i v (t) < 0, which is case one, the photovoltaic system needs to charge the long-cycle regulation energy storage device;

[0061] S7.2 further includes the following steps:

[0062] S7.2.1: Calculate the SOC values of all the long-cycle regulation energy storage devices, sort them in ascending order of the SOC values and number them, and the photovoltaic system charges each long-cycle regulation energy storage device in ascending order of the SOC values;

[0063] The charging power of the photovoltaic system numbered i to the long-cycle regulation energy storage device numbered r is:

[0064]

[0065] In the formula, represents the charging power of the photovoltaic system numbered i to the long-term regulation energy storage device numbered r; P mch,r represents the maximum charging power of the photovoltaic system numbered i to the long-term regulation energy storage device numbered r;

[0066] During the charging process of the photovoltaic system numbered i to the long-term regulation energy storage device numbered r, calculate the SOC value of the long-term regulation energy storage device. When its SOC value is greater than 90% of its total state of charge, the photovoltaic system numbered i stops charging the long-term regulation energy storage device numbered r and continues to charge the long-term regulation energy storage device numbered (r + 1); or at the same time, calculate the net load demand P i v (t) of the photovoltaic system numbered i in real time. When the net load demand of the photovoltaic system numbered i satisfies the photovoltaic system numbered i stops charging the long-term regulation energy storage device and continues to charge the long-term regulation energy storage device numbered (r + 1);

[0067] S7.3: If P i v (t) > 0, it is Case 2; the long-term regulation energy storage device needs to discharge to the photovoltaic system;

[0068] S7.3 further includes the following steps:

[0069] S7.3.1: Calculate the SOC values of all long-term regulation energy storage devices, sort them in descending order and number them. Each long-term regulation energy storage device discharges to the photovoltaic system in order;

[0070] S7.3.2: The discharge power of the long-term regulation energy storage device numbered o to the photovoltaic system numbered i is:

[0071]

[0072] In the formula, represents the discharge power of the long-term regulation energy storage device numbered o to the photovoltaic system numbered i; P mdh,o represents the maximum discharge power of the long-term regulation energy storage device numbered o to the photovoltaic system numbered i;

[0073] During the discharge process of the long - cycle regulation energy storage device numbered o to the photovoltaic system numbered i, the SOC value of the long - cycle regulation energy storage device is calculated in real time. If the SOC value is less than 20% of its total state of charge, the long - cycle regulation energy storage device numbered o stops discharging to the photovoltaic system numbered i, and the long - cycle regulation energy storage device numbered (o + 1) discharges to the photovoltaic system numbered i; or at the same time, the net load demand P of the photovoltaic system numbered i is calculated in real time i v (t). When the net load demand of the photovoltaic system numbered i is reached, the long - cycle regulation energy storage device numbered o stops discharging to the photovoltaic system numbered i, and the long - cycle regulation energy storage device numbered (o + 1) discharges to the photovoltaic system numbered i;

[0074] S7.4: When P i v (t)=0, maintain the current operation.

[0075] Adopting the above technical solutions, the beneficial effects that can be achieved are:

[0076] 1. This scheme proposes a photovoltaic power prediction model, which can predict photovoltaic power more accurately.

[0077] 2. Three types of power fluctuation thresholds are adopted, namely the maximum power fluctuation threshold, the minimum power fluctuation threshold, and the standard power fluctuation threshold. With the joint cooperation of the three parameters, the charging or discharging of the energy storage device is realized, and the strategy of suppressing fluctuations is more reasonable. Brief Description of the Drawings

[0078] Figure 1 is the flowchart of the method. Detailed Embodiment

[0079] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0080] A photovoltaic system regulation method for a micro - grid includes the following steps:

[0081] S1: Obtain the predicted power of all photovoltaic systems in the micro - grid at the current moment.

[0082] S1 specifically includes the following steps:

[0083] S1.1: Number all the photovoltaic systems in the microgrid, obtain the historical multi - feature of each photovoltaic system, and convert the historical multi - feature of each photovoltaic system into the corresponding feature matrix.

[0084] S1.1 further includes the following steps:

[0085] S1.1.1: Number all the photovoltaic systems in the microgrid. Assume that the number of photovoltaic systems in the microgrid is N, and number each photovoltaic system, with the numbers ranging from 1 to N.

[0086] S1.1.2: Obtain the historical multi - feature of each photovoltaic system. The historical multi - feature includes: effective sunshine duration (ESD), original air quality index (AQI), interaction effect of temperature and humidity (TH), and dust deposition impact (DI).

[0087] Among them, the expression of the historical multi - feature of the photovoltaic system numbered i (i ∈ [1, N]) is:

[0088]

[0089] In formula (1), i ∈ [1, N] and i is an integer, representing the photovoltaic system numbered i; ESD i represents the effective sunshine duration at the location of the photovoltaic system numbered i; Tdh i represents the total sunshine hours at the location of the photovoltaic system numbered i; CCF i represents the cloud cover factor at the location of the photovoltaic system numbered i;

[0090] AQI i represents the optimized air quality index at the location of the photovoltaic system numbered i; AQ i represents the original air quality index at the location of the photovoltaic system numbered i; Hu i represents the air humidity percentage at the location of the photovoltaic system numbered i;

[0091] TH i represents the interaction effect of temperature and humidity at the location of the photovoltaic system numbered i; Th i represents the ambient temperature at the location of the photovoltaic system numbered i;

[0092] DI i represents the dust deposition impact at the location of the photovoltaic system numbered i; W i represents the wind speed at the location of the photovoltaic system numbered i; D i represents the dust concentration at the location of the photovoltaic system numbered i.

[0093] The numerical values of all the above - mentioned parameters are provided by NASA or commercial service providers.

[0094] S1.1.3: Obtain the characteristic matrix corresponding to each photovoltaic system according to the historical multivariate characteristics of each photovoltaic system.

[0095] X i =[ESD i , AQI i , TH i , DI i T (2)

[0096] In formula (2), X i represents the characteristic matrix of the photovoltaic system numbered i.

[0097] S1.2: Obtain the observed power generation of each photovoltaic system at the current moment, and then combine the characteristic matrix of each photovoltaic system to construct a power prediction model to calculate the predicted power of each photovoltaic system at the current moment.

[0098] The power prediction model is:

[0099] P i (t) = W p · RNN(h N-1 , [CNN(X i );S i (t)]) + b p (3)

[0100] In formula (3), t represents the current moment; P i (t) represents the predicted power of the photovoltaic system numbered i at the current moment t; W p is the weight matrix; RNN represents the recurrent neural network; h N-1 represents the state of the recurrent neural network at time step N - 1; CNN(X i ) represents processing the characteristic matrix X i numbered i using a convolutional neural network; S i (t) represents the observed power generation of the photovoltaic system numbered i at the current moment t; b p is the bias vector.

[0101] In formula (3), the expression for the observed power generation S i (t) of the photovoltaic system numbered i is:

[0102] S i (t) = T i (t) + M i (t) + R i (t) (4)

[0103] In formula (4), T i ​(t) represents the power generation trend component observed by the photovoltaic system labeled i at the current time t; M i (t) represents the seasonal component corresponding to the power generation observed by the photovoltaic system labeled i at the current time t; R i (t) represents the remaining error corresponding to the power generation observed by the photovoltaic system labeled i at the current time t.

[0104] According to Equation (4), the predicted power at the current time t of each photovoltaic system is calculated.

[0105] S2: Set the power fluctuation threshold for each photovoltaic system.

[0106] The power fluctuation threshold includes: the maximum power fluctuation threshold, the minimum power fluctuation threshold, and the standard power fluctuation threshold.

[0107] Among them, the minimum power fluctuation threshold of the photovoltaic system numbered i The formula is:

[0108]

[0109] In Equation (5), α represents the geographical feature vector coefficient; β represents the device type feature vector coefficient; γ represents the historical data feature vector coefficient; represents the minimum geographical feature vector of the photovoltaic system numbered i; CCF i represents the cloud cover factor at the location of the photovoltaic system numbered i; D i represents the dust concentration at the location of the photovoltaic system numbered i; D max represents the maximum dust concentration at the location of the photovoltaic system numbered i. In this embodiment, D max is set to 500 μg / m 3 ; Hu i represents the air humidity percentage at the location of the photovoltaic system numbered i; represents the minimum device type feature vector of the photovoltaic system numbered i; P i max represents the maximum charging power of the photovoltaic system numbered i; represents the charging efficiency of the photovoltaic system numbered i; represents the minimum historical data feature vector of the photovoltaic system numbered i; represents the historical minimum load of the photovoltaic system numbered i; represents the historical negative fluctuation standard deviation of the photovoltaic system numbered i.

[0110] The maximum power fluctuation threshold of the photovoltaic system numbered i The formula is:

[0111]

[0112] In formula (6), represents the maximum geographical feature vector of the photovoltaic system numbered i; represents the maximum equipment type feature vector of the photovoltaic system numbered i; represents the maximum historical data feature vector of the photovoltaic system numbered i; ESD i represents the effective sunshine duration at the location of the photovoltaic system numbered i; W i represents the wind speed at the location of the photovoltaic system numbered i; Wmax represents the maximum wind speed at the location of the photovoltaic system numbered i; Th i represents the ambient temperature at the location of the photovoltaic system numbered i; Th max represents the maximum ambient temperature at the location of the photovoltaic system numbered i; Th min represents the minimum ambient temperature at the location of the photovoltaic system numbered i; represents the maximum discharge power of the photovoltaic system numbered i; represents the discharge efficiency of the photovoltaic system numbered i; represents the historical highest load of the photovoltaic system numbered i; represents the historical positive fluctuation standard deviation of the photovoltaic system numbered i.

[0113] The formula for the standard power fluctuation threshold of the photovoltaic system numbered i is:

[0114]

[0115] In formula (7), represents the standard geographical feature vector of the photovoltaic system numbered i; represents the standard equipment type feature vector of the photovoltaic system numbered i; represents the standard historical data feature vector of the photovoltaic system numbered i; represents the historical average load of the photovoltaic system numbered i; represents the historical average fluctuation standard deviation of the photovoltaic system numbered i.

[0116] S3: Obtain the number of energy storage devices in the microgrid, number the energy storage devices; calculate the state of charge of each energy storage device.

[0117] Assume that there are M energy storage devices in the microgrid, number the M energy storage devices, and the numbers are 1 to M;

[0118] The formula for the state of charge of the energy storage device numbered k (k ∈ [1, M]) at the next moment t + 1 is:

[0119]

[0120] In formula (8), SOC k (t + 1) represents the state of charge of the energy storage device numbered k at the next moment t + 1;

[0121] SOC k (t) represents the state of charge of the energy storage device numbered k at the current moment t;

[0122] represents the charging efficiency of the energy storage device numbered k at moment t; represents the charging power of the energy storage device numbered k at moment t; Δt represents the time interval;

[0123] represents the discharging efficiency of the energy storage device numbered k at moment t; represents the discharging power of the energy storage device numbered k at moment t;

[0124] S4: Classify the energy storage devices in the microgrid into fast - response energy storage devices and long - cycle regulation energy storage devices according to the response time standard.

[0125] Classify the M energy storage devices into fast - response energy storage devices and long - cycle regulation energy storage devices according to the response time standard. Based on the fact that the response time of fast - response energy storage devices is in the millisecond level, while the response time of long - cycle regulation energy storage devices is in the minute level, so by setting the response time threshold, such as the threshold is 5 seconds, group the M energy storage devices. Assume there are X fast - response energy storage devices and Y long - cycle regulation energy storage devices in total;

[0126] S5: Calculate the net load demand of each photovoltaic system at the current moment t.

[0127] The formula for the net load demand of the photovoltaic system numbered i is:

[0128] P i v (t) = P l (t) - P i (t)(9)

[0129] In formula (9), P l (t) represents the power demand of the user for the photovoltaic system numbered i at moment t, which is obtained by predicting with the time - series model LSTM.

[0130] P i v (t) represents the net load demand of the user for the photovoltaic system numbered i at moment t;

[0131] P i (t) represents the predicted power of the photovoltaic system numbered i at moment t, which is obtained by calculating with formula (2);

[0132] S6: Monitor the real-time power fluctuation of the photovoltaic system, compare the real-time power fluctuation of the photovoltaic system with the set standard fluctuation threshold, and according to the comparison result, let the fast-response energy storage device discharge to each photovoltaic system, or let the photovoltaic system charge each fast-response energy storage device, or maintain the operation.

[0133] S6.1: Let the real-time power fluctuation of the monitored photovoltaic system numbered i be

[0134] Compare the real-time power fluctuation of the photovoltaic system numbered i with its standard power fluctuation threshold According to the comparison result, let the fast-response energy storage device discharge to each photovoltaic system, or let the photovoltaic system charge each fast-response energy storage device, or maintain the operation.

[0135] S6.2: If Then it is case one, and the fast-response energy storage device needs to discharge to the photovoltaic system.

[0136] Step S6.2 further includes the following steps:

[0137] S6.2.1: Calculate the SOC values of all fast-response energy storage devices according to Equation (8), sort and number them in descending order of the SOC values, and discharge to the photovoltaic system in descending order of the SOC values.

[0138] S6.2.2: Assume that the fast-response energy storage devices numbered j (j ∈ [1, X]) are discharging to the i-th photovoltaic system, then the discharge power of the fast-response energy storage device numbered j to the i-th photovoltaic system is:

[0139]

[0140] In Equation (10), represents the discharge power of the fast-response energy storage device numbered j to the i-th photovoltaic system; is the maximum power fluctuation threshold of the i-th photovoltaic system, obtained by calculating according to Equation (6). min{*, *} represents taking the minimum of the two; P md,j is the maximum discharge power of the fast-response energy storage device numbered j discharging to the i-th photovoltaic system;

[0141] During the discharge process of the fast-response energy storage device numbered j to the photovoltaic system numbered i, the SOC value of the fast-response energy storage device numbered j is calculated in real time according to Equation (8). If the SOC value of this fast-response energy storage device is lower than 20% of its total state of charge (total SOC value), it stops discharging to the photovoltaic system numbered i, and the fast-response energy storage device numbered (j + 1) continues to discharge to the photovoltaic system numbered i. Or at the same time, during the discharge process of the fast-response energy storage device numbered j to the photovoltaic system numbered i, the net load demand P i v (t) of the photovoltaic system numbered i is calculated in real time according to Equation (9). When the net load demand of the photovoltaic system numbered i satisfies , it also stops discharging to the photovoltaic system numbered i, and the fast-response energy storage device numbered (j + 1) continues to discharge to the photovoltaic system numbered i.

[0142] S6.3: If Then it is Case 2. The photovoltaic system needs to charge the fast-response energy storage device.

[0143] Step S6.3 further includes the following steps:

[0144] S6.3.1: Calculate the SOC values of all fast-response energy storage devices according to Equation (8), sort them in ascending order of SOC value and number them, and the photovoltaic system charges each fast-response energy storage device in ascending order of SOC value.

[0145] S6.3.2: Assume that the photovoltaic system numbered i is charging the fast-response energy storage device numbered g (g ∈ [1, X]). The charging power of the photovoltaic system numbered i to the fast-response energy storage device numbered g is:

[0146]

[0147] In Equation (11), represents the charging power of the photovoltaic system numbered i to the fast-response energy storage device numbered g; P mc,g is the maximum charging power of the photovoltaic system numbered i to the fast-response energy storage device numbered g;

[0148] During the charging process of the fast-response energy storage device numbered g by the photovoltaic system numbered i, the SOC value of the fast-response energy storage device numbered g is calculated according to Equation (8); when the SOC value of this fast-response energy storage device is higher than 90% of its total state of charge (total SOC value), the photovoltaic system numbered i stops charging the fast-response energy storage device numbered g, and the photovoltaic system numbered i continues to charge the fast-response energy storage device numbered (g + 1). Or simultaneously, during the charging process of the fast-response energy storage device numbered g by the photovoltaic system numbered i, the net load demand P i v (t) of the photovoltaic system numbered i is calculated according to Equation (9). When the net load demand of the photovoltaic system numbered i satisfies , the photovoltaic system numbered i stops charging the fast-response energy storage device numbered g, and the photovoltaic system numbered i continues to charge the fast-response energy storage device numbered (g + 1).

[0149] S6.4: When , maintain the current operation.

[0150] S7: Compare the static load demand of each photovoltaic system with the set load threshold. According to the comparison result, let the long-cycle regulation energy storage device discharge to each photovoltaic system, or the photovoltaic system charge each long-cycle regulation energy storage device, or maintain the operation.

[0151] S7 specifically includes the following steps:

[0152] S7.1: Set the load threshold to 0. Compare the net load demand P i v (t) of the photovoltaic system numbered i with the load threshold, and according to the result, let the long-cycle regulation energy storage device discharge to each photovoltaic system, or the photovoltaic system charge each long-cycle regulation energy storage device, or maintain the operation.

[0153] S7.2: If P i v (t) < 0, which is Case 1, the photovoltaic system needs to charge the long-cycle regulation energy storage device.

[0154] S7.2 further includes the following steps:

[0155] S7.2.1: Calculate the SOC values of all long-cycle regulation energy storage devices according to Equation (8), sort them in ascending order of SOC value and number them, and the photovoltaic system charges each long-cycle regulation energy storage device in ascending order of SOC value.

[0156] S7.2.2: Assume that the photovoltaic system numbered i is charging the long-cycle regulated energy storage device numbered r (r ∈ [1, Y]). Then, the charging power of the photovoltaic system numbered i to the long-cycle regulated energy storage device numbered r is as follows:

[0157]

[0158] In Equation (12), represents the charging power of the photovoltaic system numbered i to the long-cycle regulated energy storage device numbered r;

[0159] P mch,r represents the maximum charging power of the photovoltaic system numbered i to the long-cycle regulated energy storage device numbered r.

[0160] During the charging process of the photovoltaic system numbered i to the long-cycle regulated energy storage device numbered r, calculate the SOC value of this long-cycle regulated energy storage device according to Equation (8). When its SOC value is greater than 90% of its total state of charge (total SOC value), the photovoltaic system numbered i stops charging the long-cycle regulated energy storage device numbered r and continues to charge the long-cycle regulated energy storage device numbered (r + 1). Or simultaneously, calculate the net load demand P i v (t) of the photovoltaic system numbered i in real time. When the net load demand of the photovoltaic system numbered i satisfies the photovoltaic system numbered i stops charging the long-cycle regulated energy storage device and continues to charge the long-cycle regulated energy storage device numbered (r + 1).

[0161] S7.3: If P i v (t) > 0, it is Case 2; the long-cycle regulated energy storage device needs to discharge to the photovoltaic system.

[0162] S7.3 further includes the following steps:

[0163] S7.3.1: Calculate the SOC values of all long-cycle regulated energy storage devices according to Equation (8), sort them in descending order and number them. Each long-cycle regulated energy storage device discharges to the photovoltaic system in order.

[0164] S7.3.2: Assume that the number of the long-cycle regulated energy storage device that is discharging is o (o ∈ [1, Y]). The discharging power of the long-cycle regulated energy storage device numbered o to the photovoltaic system numbered i is as follows:

[0165]

[0166] In Equation (13), Indicates the discharge power of the long - term regulation type energy storage device numbered o to the photovoltaic system numbered i; P mdh,o Indicates the maximum discharge power of the long - term regulation type energy storage device numbered o to the photovoltaic system numbered i.

[0167] During the discharge process of the long - term regulation type energy storage device numbered o to the photovoltaic system numbered i, the SOC value of this long - term regulation type energy storage device is calculated in real - time according to Equation (8). If its SOC value is less than 20% of its total state of charge (total SOC value), the long - term regulation type energy storage device numbered o stops discharging to the photovoltaic system numbered i, and the long - term regulation type energy storage device numbered (o + 1) discharges to the photovoltaic system numbered i. Or simultaneously, the net load demand P i v (t) of the photovoltaic system numbered i is calculated in real - time. When the net load demand of the photovoltaic system numbered i is met, the long - term regulation type energy storage device numbered o stops discharging to the photovoltaic system numbered i, and the long - term regulation type energy storage device numbered (o + 1) discharges to the photovoltaic system numbered i.

[0168] S7.4: When P i v (t)=0, maintain the current operation.

[0169] Enlightened by the ideal embodiments of the present invention described above, through the above - mentioned description content, relevant staff can completely make various changes and modifications without departing from the technical idea of this invention. The technical scope of this invention is not limited to the content in the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A photovoltaic system control method for a microgrid, characterized in that: The following steps are involved: S1: Obtain the predicted power of all photovoltaic systems in the microgrid at the current moment; S2: Setting the power fluctuation threshold of each photovoltaic system; the power fluctuation threshold includes the maximum fluctuation threshold, the minimum fluctuation threshold and the standard fluctuation threshold; S3: Obtain the number of energy storage devices in the microgrid, number the energy storage devices, and calculate the state of charge of each energy storage device; S4: The energy storage devices in the microgrid are divided into fast response type energy storage devices and long cycle regulation type energy storage devices according to the response time standard; S5: Calculate the net load demand of each photovoltaic system at the current time t; S6: monitor the real-time power fluctuation of the photovoltaic system, compare the real-time power fluctuation of the photovoltaic system with the set standard fluctuation threshold, and according to the comparison result, let the fast response energy storage device discharge to each photovoltaic system, or the photovoltaic system charge each fast response energy storage device, or maintain operation; S7: The static load demand of each photovoltaic system is compared with the set load threshold. According to the comparison result, the long-cycle regulating energy storage device is discharged to each photovoltaic system, or the photovoltaic system charges each long-cycle regulating energy storage device, or maintains operation.

2. A photovoltaic system control method for a microgrid according to claim 1, characterized in that: Step S1 includes: S1.1: Number all photovoltaic systems in the microgrid; obtain the historical multivariate features of each photovoltaic system, and convert the historical multivariate features of each photovoltaic system into a corresponding feature matrix; Assuming that there are N photovoltaic systems in the microgrid, the photovoltaic systems are numbered from 1 to N, and the historical multivariate characteristic expression of the photovoltaic system numbered i (i∈[1,N]) is: X i =[ESD i ,AQI i ,TH i ,IN i ] T (1) In formula (1), ESD i Indicates the effective sunshine duration at the location of the photovoltaic system numbered i; Tdh i Indicates the total sunshine hours at the location of the photovoltaic system numbered i; CCF i Indicates the cloud coverage factor at the location of the photovoltaic system numbered i; AQI i Represents the air quality index after optimization of the location of the photovoltaic system numbered i; AQ i Represents the original air quality index of the location of the photovoltaic system numbered i; Hu i Indicates the air humidity percentage at the location of the photovoltaic system numbered i; TH i represents the interaction effect of temperature and humidity at the location of the photovoltaic system numbered i; Th i Indicates the ambient temperature of the location of the photovoltaic system numbered i; DI i Indicates the dust deposition effect at the location of the photovoltaic system numbered i; W i Indicates the wind speed at the location of the photovoltaic system numbered i; D i Indicates the dust concentration at the location of the photovoltaic system numbered i; X i represents the characteristic matrix of the photovoltaic system numbered i; S1.2: Obtain the observed power generation of each photovoltaic system at the current moment, and then build a power prediction model based on the characteristic matrix of each photovoltaic system to calculate the predicted power of each photovoltaic system at the current moment; The power prediction model is: P i (t)=W p ·RNN(h N-1 ,[CNN(X i );S i (t)])+b p (2) In formula (2), P i (t) represents the predicted power of the photovoltaic system numbered i at the current time t; W p is the weight matrix; RNN represents recurrent neural network; h N-1 Represents the state of the recurrent neural network at time step N-1; CNN(X i ) means that the feature matrix X numbered i i Using convolutional neural network processing; S i (t) represents the power generation observed by the photovoltaic system numbered i at the current time t; b p is the bias vector.

3. A photovoltaic system control method for a microgrid according to claim 2, characterized in that: In step S2, the expressions of the maximum fluctuation threshold, the minimum fluctuation threshold and the standard fluctuation threshold are respectively: Minimum power fluctuation threshold of photovoltaic system numbered i The formula is: Maximum power fluctuation threshold of photovoltaic system numbered i The formula is: The formula for the standard power fluctuation threshold of the photovoltaic system numbered i is: In formula (3), (4), and (5), α represents the geographic feature vector coefficient; β represents the equipment type feature vector coefficient; γ represents the historical data feature vector coefficient; represents the minimum geographic feature vector of the photovoltaic system numbered i; The minimum equipment type feature vector representing the photovoltaic system numbered i; represents the minimum historical data feature vector of the photovoltaic system numbered i; D i Indicates the dust concentration at the location of the photovoltaic system numbered i; D max Indicates the maximum dust concentration at the location of the photovoltaic system numbered i; P i max Indicates the maximum charging power of the photovoltaic system numbered i; Indicates the historical minimum load of the photovoltaic system numbered i; It represents the historical negative fluctuation standard deviation of the photovoltaic system numbered i; Indicates the charging efficiency of the photovoltaic system numbered i; represents the maximum geographic characteristic vector of the photovoltaic system numbered i; represents the maximum historical data feature vector of the photovoltaic system numbered i; represents the maximum device type feature vector of photovoltaic system number i; W i Indicates the wind speed at the location of the photovoltaic system numbered i; Wmax indicates the maximum wind speed at the location of the photovoltaic system numbered i; Indicates the maximum discharge power of the photovoltaic system numbered i; Indicates the discharge efficiency of the photovoltaic system numbered i; Th max Indicates the maximum ambient temperature of the location of the photovoltaic system numbered i; Th min Indicates the minimum ambient temperature at the location of the photovoltaic system numbered i; Indicates the historical maximum load of the photovoltaic system numbered i; It represents the historical positive fluctuation standard deviation of the photovoltaic system numbered i; represents the standard geographic feature vector of the photovoltaic system numbered i; The standard equipment type feature vector representing the photovoltaic system numbered i; represents the standard historical data feature vector of the photovoltaic system numbered i; Indicates the historical average load of the photovoltaic system numbered i; Represents the historical average fluctuation standard deviation of photovoltaic system numbered i.

4. A photovoltaic system control method for a microgrid according to claim 3, characterized in that: In steps S3 and S4, it is assumed that there are M energy storage devices in the microgrid, numbered 1 to M, including X fast-response energy storage devices and Y long-cycle regulation energy storage devices; The charge state formula of the energy storage device numbered k (k∈[1,M]) at the next time t+1 is: In formula (6), SOC k (t+1) represents the charge state of the energy storage device numbered k at the next time t+1; SOC k (t) represents the charge state of the energy storage device numbered k at the current time t; represents the charging efficiency of the energy storage device numbered k at time t; represents the charging power of the energy storage device numbered k at time t; Δt represents the time interval; represents the discharge efficiency of the energy storage device numbered k at time t; Represents the discharge power of the energy storage device numbered k at time t.

5. A photovoltaic system control method for a microgrid according to claim 4, characterized in that: The formula for the net load demand of the photovoltaic system numbered i is: P i v (t)=P l (t)-P i (t)(7) In formula (7), P l (t) represents the power demand of the user for the photovoltaic system numbered i at time t; P i v (t) represents the net load demand of the user on the photovoltaic system numbered i at time t; P i (t) represents the predicted power of the photovoltaic system numbered i at time t.

6. A photovoltaic system control method for a microgrid according to claim 5, characterized in that: Step S6 includes the following steps: S6.1: Let the monitoring number be i and the real-time power fluctuation of the photovoltaic system be Comparison number is the real-time power fluctuation of photovoltaic system i and its standard power fluctuation threshold According to the comparison result, the fast-response energy storage device is allowed to discharge to each photovoltaic system, or the photovoltaic system is allowed to charge each fast-response energy storage device, or the operation is maintained; S6.2: If This is the first case, where the fast-response energy storage device needs to discharge to the photovoltaic system; Step S6.2 further comprises the following steps: S6.2.1: Calculate the SOC values ​​of all fast-response energy storage devices, sort and number them in descending order of SOC values, and discharge them into the photovoltaic system in descending order of SOC values; S6.2.2: Then the discharge power of the fast response energy storage device numbered j (j∈[1,X]) to the photovoltaic system numbered i is: In the formula, Indicates the discharge power of the fast-response energy storage device numbered j to the photovoltaic system numbered i; is the maximum power fluctuation threshold of the photovoltaic system numbered i; P md,j is the maximum discharge power of the fast response energy storage device numbered j to the photovoltaic system numbered i; min{*,*} represents the minimum value of the two; During the process of the fast response type energy storage device numbered j discharging to the photovoltaic system numbered i, the SOC value of the fast response type energy storage device numbered j is calculated in real time. If the SOC value of the fast response type energy storage device is lower than 20% of its total state of charge, it stops discharging to the photovoltaic system numbered i, and the fast response type energy storage device numbered (j+1) continues to discharge to the photovoltaic system numbered i; or at the same time, during the process of the fast response type energy storage device numbered j discharging to the photovoltaic system numbered i, the net load demand P of the photovoltaic system numbered i is calculated in real time. i v (t), when the net load demand of photovoltaic system number i meets When , it also stops discharging to the photovoltaic system numbered i, and the fast-response energy storage device numbered (j+1) continues to discharge to the photovoltaic system numbered i; S6.3: If This is the second case, where the photovoltaic system needs to charge the fast-response energy storage device; Step S6.3 further comprises the following steps: S6.3.1: Calculate the SOC values ​​of all fast-response energy storage devices, sort and number them in ascending order of SOC value, and the photovoltaic system charges each fast-response energy storage device in ascending order of SOC value; S6.3.2: The charging power of the photovoltaic system numbered i to the fast-response energy storage device numbered g is: In the formula, represents the charging power of the photovoltaic system numbered i to the fast-response energy storage device numbered g; P mc,g The maximum charging power of the photovoltaic system numbered i to the fast-response energy storage device numbered g; During the process of charging the fast-response energy storage device numbered g by the photovoltaic system numbered i, the SOC value of the fast-response energy storage device numbered g is calculated; when the SOC value of the fast-response energy storage device is higher than 90% of its total state of charge, the photovoltaic system numbered i stops charging the fast-response energy storage device numbered g, and the photovoltaic system numbered i continues to charge the fast-response energy storage device numbered (g+1); or at the same time, during the process of charging the fast-response energy storage device numbered g by the photovoltaic system numbered i, the net load demand P of the photovoltaic system numbered i is calculated. i v (t), when the net load demand of photovoltaic system number i meets When , the photovoltaic system numbered i stops charging the fast-response energy storage device numbered g, and the photovoltaic system numbered i continues to charge the fast-response energy storage device numbered (g+1); S6.4: When When the current operation is maintained.

7. A photovoltaic system control method for a microgrid according to claim 6, characterized in that: Step S7 includes: S7.1: Set the load threshold to 0; compare the net load demand P of the photovoltaic system numbered i i v (t) and load threshold, and according to the result, the long-cycle regulating energy storage device discharges to each photovoltaic system, or the photovoltaic system charges each long-cycle regulating energy storage device, or maintains operation; S7.2: If P i v (t)<0, which is the first case, the photovoltaic system needs to charge the long-cycle regulating energy storage device; S7.2 further comprises the following steps: S7.2.1: Calculate the SOC values ​​of all long-cycle regulating energy storage devices, sort and number them in ascending order of SOC value, and the photovoltaic system charges each long-cycle regulating energy storage device in ascending order of SOC value; S7.2.2: The charging power of the photovoltaic system numbered i to the long-cycle regulating energy storage device numbered r is: In the formula, represents the charging power of the photovoltaic system numbered i to the long-cycle regulating energy storage device numbered r; P mch,r Indicates the maximum charging power of the photovoltaic system numbered i to the long-cycle regulating energy storage device numbered r; When the photovoltaic system numbered i is charging the long-cycle regulating type energy storage device numbered r, the SOC value of the long-cycle regulating type energy storage device is calculated. When its SOC value is greater than 90% of its total state of charge, the photovoltaic system numbered i stops charging the long-cycle regulating type energy storage device numbered r and continues to charge the long-cycle regulating type energy storage device numbered (r+1); or at the same time, the net load demand P of the photovoltaic system numbered i is calculated in real time. i v (t), when the net load demand of the photovoltaic system numbered i meets When , the photovoltaic system numbered i stops charging the long-cycle regulating type energy storage device, and continues charging the long-cycle regulating type energy storage device numbered (r+1); S7.3: If P i v (t)>0, it is case 2; the long-cycle regulating energy storage device needs to discharge to the photovoltaic system; S7.3 further comprises the following steps: S7.3.1: Calculate the SOC values ​​of all long-cycle regulating energy storage devices, sort them in descending order and number them, and each long-cycle regulating energy storage device discharges to the photovoltaic system in sequence; S7.3.2: The discharge power of the long-cycle regulated energy storage device numbered o to the photovoltaic system numbered i is: In the formula, represents the discharge power of the long-cycle regulating energy storage device numbered o to the photovoltaic system numbered i; P mdh,o Indicates the maximum discharge power of the long-cycle regulating energy storage device numbered o to the photovoltaic system numbered i; During the process of the long-cycle regulating type energy storage device numbered o discharging to the photovoltaic system numbered i, the SOC value of the long-cycle regulating type energy storage device is calculated in real time. If its SOC value is less than 20% of its total state of charge, the long-cycle regulating type energy storage device numbered o stops discharging to the photovoltaic system numbered i, and the long-cycle regulating type energy storage device numbered (o+1) discharges to the photovoltaic system numbered i; or at the same time, the net load demand P of the photovoltaic system numbered i is calculated in real time. i v (t), when the net load demand of photovoltaic system number i When , the long-period regulating type energy storage device numbered o stops discharging to the photovoltaic system numbered i, and the long-period regulating type energy storage device numbered (o+1) discharges to the photovoltaic system numbered i; S7.4: When P i v When (t) = 0, maintain the current operation.

Images