A microgrid multi-energy integrated dispatching system

Abstract

The present invention discloses a microgrid multi-energy integrated dispatching system, belonging to the field of microgrid control technology, comprising: a data acquisition module, a data association module, a data cleaning module, a model development module, an energy integrated dispatching module, and a dispatching execution module; the present invention takes into account the impact of the usage loss rate of various equipment in the microgrid on energy output, demand and storage, improves prediction accuracy, that is, improves dispatching accuracy.

Description

Technical Field

[0001] The present invention belongs to the field of microgrid control technology, and in particular relates to a microgrid multi-energy integrated scheduling system. Background Art

[0002] A microgrid refers to a small power system that integrates distributed power generation, load management, monitoring and protection, and automation devices, aiming to achieve internal power self-sufficiency and balance.

[0003] The microgrid multi-energy integrated dispatching system is a key component of microgrid control technology. By integrating various renewable energy power generation devices, energy storage equipment and environmental data, it can achieve precise dispatch and optimized management of microgrid output power.

[0004] Chinese patent application number 202210972927.6 discloses a microgrid multi-energy integrated scheduling system, including a data acquisition module, a cloud platform, a power prediction module, and a power scheduling module. The data acquisition module collects the power generation and environmental data of the renewable energy power generation device, and the cloud platform stores the power generation and environmental data of the renewable energy power generation device, and generates a power generation history change curve, an environmental data history change curve, and an environmental data change curve with future time, respectively. The power prediction module extracts the environmental data change curve with future time in the future time period t1 and marks it as a first change curve, extracts the curve segment with the highest similarity to the first change curve in the environmental data history change curve, marks this curve segment as a second change curve, extracts the power generation history change curve of the same period as the second change curve and marks it as a power generation prediction curve. The power scheduling module obtains the average power generation in the future time period t1 based on the power generation prediction curve, and sets the average power generation as the rated output power of the microgrid in the future time period t1. When the actual power generation is higher than the rated output power, the energy storage device stores excess power generation. When the actual power generation is lower than the rated output power, the energy storage device releases power to maintain the rated output power.

[0005] The above technology has the following problems: the above microgrid multi-energy integrated scheduling system generates a power generation historical change curve, an environmental data historical change curve and an environmental data change curve with future time based on the power generation power and environmental data of the renewable energy power generation device, and then extracts similar segments of the environmental data change curve with future time and the environmental data historical change curve, and extracts the overlapping segments with the similar segments in the power generation historical change curve as the predicted power generation curve, and then uses the average value of the power generation curve as the rated output power of the microgrid, and then generates a scheduling plan based on the rated output power. The generated scheduling plan does not take into account the loss rate of the renewable energy power generation device as the use time increases, resulting in inaccurate prediction results and inaccurate scheduling plan.

[0006] In view of this, a microgrid multi-energy integrated scheduling system is designed to solve the above problems. Summary of the Invention

[0007] To solve the problems raised in the above background technology, the present invention provides a microgrid multi-energy integrated scheduling system, which takes into account the impact of the usage loss rate of various devices in the microgrid on energy output, demand and storage, and improves prediction accuracy, that is, improves scheduling accuracy.

[0008] To achieve the above objectives, the present invention provides the following technical solutions: a microgrid multi-energy integrated dispatching system, comprising:

[0009] Data acquisition module, collects historical operation data of various equipment i =(O1, O2, ..., O n ) and historical environmental data E i =(E1, E2, ..., E n ), where i represents the device type, including distributed energy devices, energy storage systems, and load devices;

[0010] Data association module, which associates the historical operation data and historical environment data of various devices according to the collection date, namely OE i =[(O1,E1),(O2,E2),…,(O n , E n )];

[0011] The data cleaning module estimates the usage loss coefficient L of various equipment based on the historical operation data under the same historical environment data according to the collection date from recent to far. i , retain the historical operation data and historical environmental data of various equipment with the same corresponding loss coefficient, and use the loss coefficient L i The expression is:

[0012]

[0013] Where: O near Indicates the historical operation data of various devices collected recently. far Indicates that the historical operation data collected for various devices is far away;

[0014] The model development module constructs an energy output prediction model and an energy demand prediction model. The energy output prediction model is trained based on the retained historical operation data and historical environmental data of distributed energy equipment to obtain the energy output prediction model of distributed energy equipment. The energy output is predicted based on the environmental data of distributed energy equipment. The energy demand prediction model is trained based on the retained historical operation data and historical environmental data of load equipment to obtain the energy demand prediction model of load equipment. The energy demand is predicted based on the environmental data of load equipment. The expressions of the energy output prediction model and the energy demand prediction model are:

[0015]

[0016] Where: a n and b are learning parameters, E n Denotes the training sample, G n (E n ,O n ) is represented as a Gaussian kernel function;

[0017] The multi-energy integrated scheduling module collects the future environmental data of distributed energy equipment and the load equipment expected to be used. The energy output prediction model and energy demand prediction model predict the energy output E of distributed energy equipment based on the collected future environmental data. cout and the energy demand E of the load equipment to be used cin Collect the historical operating data of distributed energy equipment and the load equipment expected to be used under the same environmental data in the recent time period, estimate the use loss coefficient, and obtain the final energy output E of the distributed energy equipment based on the predicted energy output and energy demand and the estimated use loss coefficient. dout and the final energy demand E of the load equipment expected to be used din , compare the final energy output and the final energy demand. If the final energy output exceeds the final energy demand, collect the historical energy storage data of the energy storage system under the same environmental data in the recent time period, estimate the use loss coefficient, and calculate the energy storage system rated parameter E based on the energy storage system rated parameter E. cst And the estimated usage loss coefficient, the final storage energy E of the energy storage system is obtained dst , compare the remaining energy output with the final storage capacity of the energy storage system. If the remaining energy output exceeds the final storage capacity of the energy storage system, part of the remaining energy output is stored in the energy storage system and part is transmitted to the large power grid. If the remaining energy output does not exceed the final storage capacity of the energy storage system, the remaining energy output is stored in the energy storage system. If the final energy output does not exceed the final energy demand, part of the electricity from the large power grid is absorbed to generate a multi-energy integrated scheduling plan. The expressions are:

[0018] E dout =E cout ×Li

[0019] E din =E cin ×L i

[0020] E dst =E cst ×L i ;

[0021] The scheduling execution module performs scheduling based on the generated multi-energy integrated scheduling plan.

[0022] Furthermore, in the process of correlating the historical operation data and historical environment data of various devices according to the collection date, any historical operation data O n There is no corresponding historical environmental data E n Or any historical environmental data E n There is no corresponding historical operation data O n , then directly remove it.

[0023] Furthermore, the data cleaning module estimates the usage loss coefficient L of various equipment based on the historical operation data under the same historical environment data according to the collection date from recent to long. i The specific steps include:

[0024] Traverse the historical environmental data E from recent to far according to the collection date i =(E1, E2, ..., E n ) until the historical environmental data E1 that is the same as the most recent historical environmental data E1 is found. n stop;

[0025] Extract the historical environmental data E1 and E n Related historical operation data O1 and O n ;

[0026] Historical operating data O1 and O n According to the use loss coefficient L i The loss coefficient L is calculated using the expression i , where O1 is O near , O n O far .

[0027] Furthermore, the data cleaning module traverses the historical environment data E from recent to far according to the collection date. i =(E1, E2, ..., E n ) process, further comprising the following steps:

[0028] Determine whether the historical environmental data E1 of the most recent date is the same as the historical environmental data E2 of the second most recent date. If they are the same, continue searching until a historical environmental data E1 different from the historical environmental data E1 of the most recent date is found. n Stop and repeat the above estimation of the usage loss coefficient L of various equipment i Steps;

[0029] And so on, search for different historical environment data E multiple times n Make estimates and avoid estimation errors.

[0030] Furthermore, the energy storage system rated parameters of the multi-energy integrated scheduling module are pre-set or obtained through big data retrieval and analysis.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] 1. The present invention first collects historical operation data and historical environmental data of various devices, estimates the usage loss coefficients of various devices based on the differences in historical operation data under the same historical environmental data, retains the historical operation data and historical environmental data of various devices with the same usage loss coefficients as training data to train the constructed energy output prediction model and energy demand prediction model to obtain the energy output prediction model and energy demand prediction model, and then predicts the energy output of distributed energy devices and the energy demand of load devices based on the energy output prediction model and energy demand prediction model, and then obtains the final energy output and energy demand based on the usage loss coefficients of distributed energy devices and load devices, generates a scheduling plan based on the final energy output, energy demand and energy storage, and then performs microgrid multi-energy integrated scheduling based on the scheduling plan. Compared with the existing technology, the impact of the usage loss rate of various devices in the microgrid on energy output, demand and storage is taken into account, thereby improving the prediction accuracy, that is, improving the scheduling accuracy.

[0033] 2. The final energy output, energy demand and energy storage loss rate obtained by the present invention are based on the recent operating data of various equipment, that is, real-time adjustments are made according to the conditions of various equipment to further improve the prediction accuracy, that is, further improve the scheduling accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a framework diagram of the microgrid multi-energy integrated dispatching system of the present invention;

[0035] In the figure: 1. Data acquisition module; 2. Data association module; 3. Data cleaning module; 4. Model development module; 5. Multi-energy integrated scheduling module; 6. Scheduling execution module. DETAILED DESCRIPTION

[0036] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0037] See also Figure 1 The present invention provides the following technical solutions: a microgrid multi-energy integrated dispatching system, comprising:

[0038] Data acquisition module 1, collects historical operation data of various devices i =(O1, O2, ..., O n ) and historical environmental data E i =(E1, E2, ..., E n ), where i represents the device type, including distributed energy devices, energy storage systems, and load devices;

[0039] Data association module 2, which associates the historical operation data of various devices according to the collection date. i =(O1, O2, ..., O n ) and historical environmental data E i =(E1, E2, ..., E n ),Right now

[0040] OE i =[(O1,E1),(O2,E2),…,(O n , E n )];

[0041] Data cleaning module 3, according to the collection date from recent to far, based on the same historical environment data E n Historical operating data under O i =(O1, O2, ..., O n ) Estimate the usage loss coefficient L of various equipment i , retain the loss factor L i The same corresponding historical operation data of various equipment i =(O1, O2, ..., O n ) and historical environmental data E i =(E1, E2, ..., E n ), using the loss factor L i The expression is:

[0042]

[0043] Where: O nearIndicates the historical operation data of various devices collected recently. far Indicates that the historical operation data collected for various devices is far away;

[0044] Model development module 4, building energy output forecasting model and energy demand forecasting model based on the retained historical operation data of distributed energy equipment. i =(O1, O2, ..., O n ) and historical environmental data E i =(E1, E2, ..., E n ) trains the energy output prediction model to obtain the energy output prediction model of distributed energy equipment, based on the environmental data E of distributed energy equipment i =(E1, E2, ..., E n ) Forecast energy output E cout , based on the historical operating data of the retained load equipment O i =(O1, O2, ..., O n ) and historical environmental data E i =(E1, E2, ..., E n ) trains the energy demand prediction model to obtain the energy demand prediction model of the load equipment, based on the environmental data E i =(E1, E2, ..., E n ) Forecast energy demand E cin , where the expressions of energy output forecast model and energy demand forecast model are:

[0045]

[0046] Where: a n and b are learning parameters, E n Denotes the training sample, G n (E n ,O n ) is represented as a Gaussian kernel function;

[0047] Multi-energy integrated scheduling module 5, collects future environmental data E of distributed energy equipment and expected load equipment i =(E1, E2, ..., E n ), the energy output prediction model and energy demand prediction model predict E based on the collected future environmental data i =(E1, E2, ..., E n )Energy output of distributed energy equipment E cout and the energy demand E of the load equipment to be used cin ;

[0048] Collect the same environmental data E of distributed energy equipment and load equipment expected to be used in the recent time periodn Historical operating data under O i =(O1, O2, ..., O n ), estimate the use loss coefficient L i ;

[0049] Based on the predicted energy output E cout and energy demand E cin and the estimated usage loss coefficient L i , and obtain the final energy output E of distributed energy equipment dout and the final energy demand E of the load equipment expected to be used din , the expressions are:

[0050] E dout =E cout ×L i

[0051] E din =E cin ×L i ;

[0052] Compare the final energy output E dout and final energy demand E din , if the final energy output E dout Exceeding final energy demand E din , then collect the same environmental data E of the energy storage system in the recent time period n Historical energy storage data under O i =(O1, O2, ..., O n ), estimate the use loss coefficient L i , based on the energy storage system rated parameters E cst and the estimated usage loss coefficient L i , and the final storage energy E of the energy storage system is obtained dst , the expression is:

[0053] E dst =E cst ×L i ;

[0054] Compare the remaining energy output with the final storage capacity E of the energy storage system dst , if the remaining energy output exceeds the final storage capacity E of the energy storage system dst , then part of the remaining energy output is stored in the energy storage system, and part is transmitted to the large power grid. If the remaining energy output does not exceed the final storage capacity E of the energy storage system dst , the remaining energy output is stored in the energy storage system;

[0055] If the final energy output E dout Does not exceed final energy demand E din, partially absorb the power of the large power grid;

[0056] Generate multi-energy integrated scheduling solutions;

[0057] The scheduling execution module 6 performs scheduling based on the generated multi-energy integrated scheduling plan.

[0058] Specifically, in the process of the data association module 2 associating the historical operation data and historical environment data of various devices according to the collection date, any historical operation data O n There is no corresponding historical environmental data E n Or any historical environmental data E n There is no corresponding historical operation data O n , then directly remove it.

[0059] Specifically, the data cleaning module 3 estimates the usage loss coefficient L of various equipment based on the historical operation data under the same historical environment data according to the collection date from recent to far. i The specific steps include:

[0060] Traverse the historical environmental data E from recent to far according to the collection date i =(E1, E2, ..., E n ) until the historical environmental data E1 that is the same as the most recent historical environmental data E1 is found. n stop;

[0061] Extract the historical environmental data E1 and E n Related historical operation data O1 and O n ;

[0062] Historical operating data O1 and O n According to the use loss coefficient L i The loss coefficient L is calculated using the expression i , where O1 is O near , O n O far .

[0063] Specifically, the data cleaning module 3 traverses the historical environment data E from recent to far according to the collection date. i =(E1, E2, ..., E n ) process, further comprising the following steps:

[0064] Determine whether the historical environmental data E1 of the most recent date is the same as the historical environmental data E2 of the second most recent date. If they are the same, continue searching until a historical environmental data E1 different from the historical environmental data E1 of the most recent date is found. n Stop and repeat the above estimation of the usage loss coefficient L of various equipment i Steps;

[0065] And so on, search for different historical environment data E multiple times n Make estimates and avoid estimation errors;

[0066] Specifically:

[0067] Complete the most recent date of historical environmental data E1 and the same historical environmental data E n Corresponding historical operation data O1 and O n The loss coefficient L i After calculation;

[0068] Determine whether the historical environment data E1 of the most recent date is the same as the historical environment data E2 of the second most recent date. If they are the same, skip. Determine whether the historical environment data E1 of the most recent date is the same as the historical environment data E2 of the second most recent date. If they are the same, skip. Repeat this process until a historical environment data E1 different from the historical environment data E1 of the most recent date is found. n ;

[0069] Based on the historical environmental data E n As a standard, until the historical environmental data E is found n Same historical environmental data E n stop;

[0070] Extract historical environmental data E n and E n Related historical operation data O n and O n ;

[0071] Historical operating data n and O n According to the use loss coefficient L i The loss coefficient L is calculated using the expression i , where O n O near , O n O far ;

[0072] Repeat the above steps to search for multiple historical environmental data E n Use loss coefficient L i calculation to avoid errors caused by single calculation.

[0073] Specifically, the energy storage system rated parameter E of the multi-energy integrated scheduling module 5 is cst Pre-set or obtained through big data retrieval and analysis.

[0074] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A microgrid multi-energy integrated dispatching system, characterized in that: include: Data acquisition module (1), collects historical operation data of various devices i =(O1, O2, ..., O n ) and historical environmental data E i =(E1, E2, ..., E n ), where i represents the device type, including distributed energy devices, energy storage systems, and load devices; Data association module (2) associates the historical operation data and historical environment data of various devices according to the collection date, namely OE i =[(O1,E1),(O2,E2),…,(O n , E n )]; Data cleaning module (3) estimates the usage loss coefficient L of various equipment based on the historical operation data under the same historical environment data according to the collection date from recent to far i , retain the historical operation data and historical environmental data of various equipment with the same corresponding loss coefficient, and use the loss coefficient L i The expression is: Where: O near Indicates the historical operation data of various devices collected recently. far Indicates that the historical operation data collected for various devices is far away; The model development module (4) constructs an energy output prediction model and an energy demand prediction model, trains the energy output prediction model based on the retained historical operation data and historical environmental data of the distributed energy equipment, obtains the energy output prediction model of the distributed energy equipment, predicts the energy output based on the environmental data of the distributed energy equipment, trains the energy demand prediction model based on the retained historical operation data and historical environmental data of the load equipment, obtains the energy demand prediction model of the load equipment, and predicts the energy demand based on the environmental data of the load equipment, wherein the expressions of the energy output prediction model and the energy demand prediction model are: Where: a n and b are learning parameters, E n Denotes the training sample, G n (E n ,O n ) is represented as a Gaussian kernel function; The multi-energy integrated scheduling module (5) collects the future environmental data of distributed energy equipment and the load equipment to be used. The energy output prediction model and the energy demand prediction model predict the energy output E of the distributed energy equipment based on the collected future environmental data. cout and the energy demand E of the load equipment to be used cin Collect the historical operating data of distributed energy equipment and the load equipment expected to be used under the same environmental data in the recent time period, estimate the use loss coefficient, and obtain the final energy output E of the distributed energy equipment based on the predicted energy output and energy demand and the estimated use loss coefficient. dout and the final energy demand E of the load equipment expected to be used din , compare the final energy output and the final energy demand. If the final energy output exceeds the final energy demand, collect the historical energy storage data of the energy storage system under the same environmental data in the recent time period, estimate the use loss coefficient, and calculate the energy storage system rated parameter E based on the energy storage system rated parameter E. cst And the estimated usage loss coefficient, the final storage energy E of the energy storage system is obtained dst , compare the remaining energy output with the final storage capacity of the energy storage system. If the remaining energy output exceeds the final storage capacity of the energy storage system, part of the remaining energy output is stored in the energy storage system and part is transmitted to the large power grid. If the remaining energy output does not exceed the final storage capacity of the energy storage system, the remaining energy output is stored in the energy storage system. If the final energy output does not exceed the final energy demand, part of the electricity from the large power grid is absorbed to generate a multi-energy integrated scheduling plan. The expressions are: AND dout =And cout ×L i AND din =And cin ×L i AND dst =And cst ×L i ; The scheduling execution module (6) performs scheduling based on the generated multi-energy integrated scheduling plan.

2. A microgrid multi-energy integrated dispatching system according to claim 1, characterized in that: In the process of correlating the historical operation data and historical environment data of various devices according to the collection date, any historical operation data O n There is no corresponding historical environmental data E n Or any historical environmental data E n There is no corresponding historical operation data O n , then directly remove it.

3. A microgrid multi-energy integrated dispatching system according to claim 1, characterized in that: The data cleaning module (3) estimates the usage loss coefficient L of various equipment based on the historical operation data under the same historical environment data according to the collection date from recent to long. i The specific steps include: Traverse the historical environmental data E from recent to far according to the collection date i =(E1, E2, ..., E n ) until the historical environmental data E1 that is the same as the most recent historical environmental data E1 is found. n stop; Extract the historical environmental data E1 and E n Related historical operation data O1 and O n ; Historical operating data O1 and O n According to the use loss coefficient L i The loss coefficient L is calculated using the expression i , where O1 is O near , O n O far .

4. A microgrid multi-energy integrated dispatching system according to claim 2, characterized in that: The data cleaning module (3) traverses the historical environmental data E from recent to far according to the collection date i =(E1, E2, ..., E n ) process, further comprising the following steps: Determine whether the historical environmental data E1 of the most recent date is the same as the historical environmental data E2 of the second most recent date. If they are the same, continue searching until a historical environmental data E1 different from the historical environmental data E1 of the most recent date is found. n Stop and repeat the above estimation of the usage loss coefficient L of various equipment i Steps; And so on, search for different historical environment data E multiple times n Make estimates and avoid estimation errors.

5. A microgrid multi-energy integrated dispatching system according to claim 1, characterized in that: The energy storage system rated parameters of the multi-energy integrated scheduling module (5) are pre-set or obtained through big data retrieval and analysis.

Images